| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
1Whitney Laboratory for Marine Bioscience, 1Center for Smell and Taste, 2Departments of Zoology and Neuroscience, and 1McKnight Brain Institute, University of Florida, Gainesville, Florida
Submitted 18 October 2006; accepted in final form 27 November 2006
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
; Marder et al. 2005; Ramirez et al. 2004), including neural networks associated with sensory processing (Hayar et al. 2004a,b; Krahe and Gabbiani 2004; Zheng et al. 2006). Although there is evidence for repetitive bursting in several types of primary sensory neurons (papers cited in Wiederhold and Carpenter 1982), including thermoreceptors (Iggo 1969), electroreceptors (Braun et al. 1994), and crustacean proprioceptors (Birmingham et al. 1999; Combes et al. 1997), the idea that bursting primary sensory neurons may play a fundamental role in sensory coding has been considerably less well studied, particularly with respect to olfaction. Reports of bursting discharge in primary olfactory receptor neurons (ORNs) have appeared in the literature for many years but have largely been ignored, probably stemming in part from the early association of prolonged bursting discharge with dying or strongly overstimulated cells (Ache, personal knowledge). Yet, more recent evidence for conditional oscillations in amphibian and mammalian ORNs (Frings and Lindemann 1988; Reisert and Matthews 2001a,b) and the rhythmic bursting discharge seen in mammalian vomeronasal receptor cells (e.g., Holy et al. 2000) suggest that the question of rhythmic discharge in primary ORNs be revisited. Is there any evidence, for example, that bursting discharge can be inherent in ORNs and, if so, does bursting discharge at least have the potential for encoding information about odorants? Here, we address this question in lobster ORNs, an established animal model for olfactory research (e.g., Ache and Derby 1985). We show that intrinsically bursting ORNs coexist along with the more traditional tonically discharging ORNs in the lobster olfactory organ and that the spontaneous bursting rhythm of these cells can be modulated by odorants in a concentration-dependent manner. We argue that odor input from bursting ORNs together with input from tonically discharging ORNs provide a novel means for encoding odor information. |
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
Primary lobster (Panulirus argus) ORNs were studied in situ using single annuli sectioned from the filamentous olfactory organ (antennule) mounted in the bottom of a 35-mm plastic culture dish filled with Panulirus saline (PS; see Solutions and odors). The preparation was arranged such that one of two superfusion streams bathed the somata of the ORNs with PS (about 1 ml/min), whereas the other stream independently bathed the olfactory sensilla (aestethascs) that contain the outer dendrites of the ORNs with either PS or PS + odorant (about 1 ml/min; see Solutions and odors). The stream bathing the outer dendrites could be switched briefly from PS to PS + odorant using a stepper motor–driven rapid solution changer (RSC-160, Bio-Logic Instruments, Claix, France). The duration of the odorant pulse delivered to the outer dendrites was increased in steps from 40 to 800 ms to change the intensity of the odorant. Stimuli were always applied randomly, independent of any ongoing action potentials or bursts of action potentials. ORNs recorded in this configuration were viable and remained responsive to odorants for many hours with little or no evidence of decline in responsiveness over the recording period.
Recording
Except where noted otherwise, action potentials (spikes) were recorded from ORNs extracellularly using loose-patch recording. Patch electrodes were pulled from borosilicate capillary glass (Sutter Instrument, BF150-86-10) using a Flaming-Brown micropipette puller (P-87, Sutter Instrument) and filled with PS. Resistance of the electrodes was 2.07 ± 0.05 (n = 67). Currents were measured with an Axopatch 200B patch-clamp amplifier (Axon Instruments) using an AD–DA converter (Digidata 1320A, Axon Instruments), low-pass filtered at 5 kHz, sampled at 5–20 kHz. Data were collected and analyzed with pCLAMP 8.1/9.2 software (Axon Instruments) in combination with SigmaPlot 8/9.0 (SPSS). When necessary, multiunit recordings were sorted into individual unit recordings using the template search procedure provided in pCLAMP 9.0 software. The time of occurrence of the spike was taken as the time of peak current deflection, i.e., the peak of the spike. To detect and analyze bursts we used the interval burst analysis protocol provided in pCLAMP 9.0, in which a burst delimiting spike interval and a minimum number of spikes in a burst were individually specified for each ORN. Interburst intervals (IBIs) were usually calculated as the time between the first spikes of two subsequent bursts. In some cases for incoherent bursts (e.g., Figs. 2B and 3, for comparison see Fig. 1B, top), the last spike in the pair of spikes producing the highest instantaneous frequency within the burst was taken as the reference spike (e.g., "0" in Fig. 3B) and IBIs were estimated as the time between the reference spikes of two subsequent bursts. Standard procedures (e.g., Prinz et al. 2003) were used to construct immediate phase-response curves (PRCs). Briefly, stimulus phase was determined as 
T/P, where T is the time between the odorant pulse and the preceding spontaneous burst and P is the burst period estimated as the mean of a Gaussian approximation to the IBI distribution of spontaneously active ORNs. The change in period (P), caused by an odorant pulse, was defined as the time difference between the first burst after the odorant pulse and P. So, if the normalized period change P/P is randomly scattered above and below 0, the odorant pulse does not affect the spontaneous rhythm. When the odorant pulse causes the first following burst to advance, the spontaneous burst period P and, correspondingly, P/P are negative.
|
|
|
Solutions and odor
PS contained (in mM): 486 NaCl, 5–14 KCl, 13.6 CaCl2, 9.8 MgCl2 and 10 HEPES, pH 7.9. Low Ca2+-PS contained (in mM): 486 NaCl, 10 KCl, 23.4 MgCl2, 10 HEPES, 0.1 mM Ca2+, pH 7.9. Intracellular solution was composed of (in mM): 180 KCl, 30 NaCl, 696 glucose, 10 Hepes, 5 EGTA, pH 7.8 adjusted with Tris-Base. The odorant was an aqueous extract of TetraMarine (TET, Tetra Werke, Melle, Germany), a commercial marine fish food product. The maximum concentration used in all experiments represented 0.1–0.5 mg of the dried powder/ml of PS.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
) they are not considered further here other than to note that in the present recording configuration these cells showed a striking constancy in their response pattern to any given stimulus concentration (Fig. 1A). Our ability to create a recording configuration that for the first time evoked such highly consistent responses from the ORNs over many hours presumably enhanced our ability to clearly identify cells that deviated from this pattern of responsiveness. In contrast to the tonically discharging cells, most of the remaining cells showed spontaneous, rhythmic bursts of action potentials, and responded to odorant stimulation with an evoked burst, although the spontaneous bursting frequency and the structure of the bursts varied across different cells. The discharge pattern of two such cells illustrates the diversity inherent among cells of this type (Fig. 1B). The spontaneous bursting frequency of the different cells ranged from 0.02 to 0.9 bursts/s (mean = 0.22 ± 0.02, n = 92). It is important to note that although different cells had different spontaneous bursting frequencies and burst structure, the frequency of spontaneous bursting and the burst structure for any given cell was consistent, including showing little or no decrement over the period of several hours during which many cells were studied (e.g., Fig. 2A). Nor were cells observed to transition to or from the tonic pattern of discharge described above. Cells showing bursting discharge co-localized with those showing tonic discharge because cells showing both patterns of response could be recorded from the same cluster of ORN somata that innervated a single olfactory (aesthetasc) sensillum.
Bursting appeared to be intrinsic in the cells. Spontaneous bursting discharge was observed in acutely dissociated cells. Bursting discharge disappeared within 2 min of breakthrough into the whole cell mode (making whole cell recording impractical, given the slow frequency of bursting), suggesting it was dependent on one or more diffusible intracellular factors. Depolarizing the cells increased the bursting frequency, one of the most general characteristics of rhythmically active neurons despite mechanisms of bursting (e.g., Harris-Warrick 2002). Depolarizing one cell by increasing extracellular [K+] from 5 to 14 mM increased the burst frequency from 0.09 ± 0.002 to 0.27 ± 0.01 Hz and decreased the number of action potentials per burst from 7.6 ± 0.07 to 4.5 ± 0.07 (Fig. 2A). A similar result was obtained for both of the two other bursting ORNs tested (data not shown). Bursting discharge also was Ca2+ dependent as in many other intrinsically bursting neurons, suggesting that burst termination could involve activation of a Ca2+-sensitive, voltage-dependent K+ conductance (e.g., Harris-Warrick 2002). Lowering [Ca2+]o from 16 mM to 100 µM increased the frequency of bursting in one such cell from 0.07 ± 0.01 to 0.1 ± 0.003 Hz, increased the burst duration from 0.54 ± 0.03 to 1 ± 0.03 s, and increased the number of action potentials per burst from 7.9 ± 0.1 to 31.4 ± 0.6 (Fig. 2B). A similar result was obtained for both of the two other ORNs tested (data not shown).
In addition to changing the parameters of bursting, low [Ca2+]o also reversibly decreased the amplitude and modified the shape of the action potentials (Fig. 2B). This effect could be reasonably, although not exclusively, explained by depolarization after a decrease in extra-/intracellular calcium concentration and partial inactivation of voltage-gated sodium channels. However, given the known complex interaction between calcium and many ion channels, understanding of the role of calcium in burst generation in these cells demands more detailed experimental and computational analyses.
As noted earlier (Fig. 1B), bursting cells responded to transient odorant stimulation with an evoked burst. The evoked burst had a similar structure to that of the spontaneous bursts in the same cell (Fig. 3, A and B). The ability to evoke a burst with odorant was phase dependent, i.e., it depended on when the odorant was applied relative to the time of the last spontaneous burst. A phase-response curve (PRC) generated for the same data shown in Fig. 3A quantifies this effect (Fig. 3C). The PRC shows that applying an odorant of a given concentration between phase 0 and 0.3 failed to evoke bursts or produce a measurable phase shift, whereas applying the same odorant later than phase 0.3–0.4 advanced the spontaneous burst period (Fig. 3C). The phase dependency of evoking a burst with an odorant can also be demonstrated in terms of the probability of the cell responding (Fig. 3D, closed circles, left-hand ordinate). Here, the probability is estimated as the number of evoked bursts divided by the number of stimulus applications. Fitting the data with a sigmoid function (solid line through closed circles) gave a time to one-half-maximum of 10.4 s, i.e., that an odorant of that concentration applied 10.4 s after a spontaneous burst (at about 0.3 into the burst cycle) would evoke a burst with a probability of 0.5. The superimposed IBI histogram compares the evoked burst probability with the probability of a spontaneous burst (Fig. 3D, bars, right-hand ordinate). The probability of occurrence of a spontaneous burst even at the mean spontaneous bursting interval of about 30 s, calculated from fitting a Gaussian function to the IBI distribution (solid line over bar histogram), is <0.1. Although shown here in detail for one cell, all 54 rhythmically active ORNs that responded to the odorant (not all cells would necessarily respond to the single odor mixture used in this study) did so by triggering a burst similar to the spontaneous bursts in a phase-dependent manner, even though the cells had different inherent bursting frequencies and burst structure. Thus as in many other rhythmically bursting neurons (e.g., Ramirez et al. 2004), odor inputs can phase-shift the bursting cycle and, in doing so, are effectively amplified in a nonlinear manner by triggering bursts that override (entrain?) the spontaneous bursting rhythm.
The probability of evoking a burst with an odorant was also concentration dependent. More intense stimuli caused the cell to discharge proportionally earlier in the cycle. As shown for one cell in Fig. 4A, applying an odorant at three different concentrations (left three panels, only 40 of 80 responses at each concentration shown for brevity) caused the cell to discharge proportionally earlier in its discharge cycle. This effect can be seen more clearly in the corresponding immediate PRCs (right three panels), which show the relative period change (P/P) at different stimulus phases (T/P). Note the shift of the PRC to an earlier time in the bursting cycle as the concentration of the odorant was increased, showing the efficacy of entrainment increased with increasing odorant concentration. As done for a single odorant concentration in Fig. 3D, the data from the cell shown in Fig. 4A can also be expressed in terms of the probability of generating an evoked burst in response to odorant stimulation (Fig. 4B). Note the progressive left shift of the average evoked burst probability with increasing odorant concentration (Fig. 4D, left-hand ordinate, open circles vs. light solid circles vs. dark solid circles). Fitting a sigmoid function to the evoked burst probability at each concentration indicates that a twofold increase in odorant concentration extended the time within which the cell could trigger evoked bursts by about 6 s, i.e., about 0.5 of a burst cycle. The maximum stimulus concentration used applied 3.2 s after a spontaneous burst (i.e., about 0.27 into the burst cycle) would evoke a burst with a probability of 0.5. In comparison, the probability of occurrence of a spontaneous burst even at the mean spontaneous bursting interval for this particular cell of about 11 s, calculated from fitting a Gaussian function to the IBI distribution (solid line over bar histogram), is about 0.2. Although shown here in detail for one cell, all of the 18 rhythmically active ORNs tested with different odorant concentrations responded to the odorant by triggering a burst similar to the spontaneous bursts in a concentration-dependent manner, even though the relative odor sensitivity and therefore the corresponding magnitude of the phase shift varied across different ORNs. Again, as in many other rhythmically bursting neurons (e.g., Prinz et al. 2003), increasing the strength of the input—in this case odorant concentration—produces a greater phase shift in the bursting cycle.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
). That, together with the fact that we never saw any of the over 400 cells surveyed transition between tonic and bursting activity, sometimes during recording periods lasting several hours, argue that cells showing tonic and bursting patterns of responsiveness may represent distinct subpopulation of ORNs in the lobster olfactory organ. However, the utility of having the two types of input to the olfactory CNS would not necessarily require that the inputs come from distinct subpopulations of ORNs.
We assume that the periodicity is not imposed and that the bursting is intrinsic to the cells. Bursting behavior was recorded with nondisruptive extracellular recording. It persisted in acutely dissociated cells. The structure of the bursts and the frequency of bursting were consistent for any given cell over hours. As is typical of intrinsically rhythmic neurons (e.g., Harris-Warrick 2002), bursting was sensitive to the imposed membrane potential of the cell and was Ca2+ dependent. Intrinsic bursting would be consistent with the understanding that lobster ORNs (Grünert and Ache 1988), as with ORNs in most animals (Ache and Young 2005), do not make peripheral synapses and project independently to the CNS. It would also be consistent with recent evidence that the discharge pattern of insect ORNs is an inherent property of the cell in question (Hallem et al. 2004). Intrinsic bursting of course does not exclude the possibility that the periodicity could be initiated or modulated by one or more activity-dependent factors, such as blood-borne factors that could be delivered to the olfactory organ from neurohemal organs potentially associated with the circulation of the olfactory organ or elsewhere in the animal.
Given that odorants activate bursting ORNs by triggering a concentration-independent, all-or-nothing burst of action potentials rather than a concentration-dependent train of action potentials as in the tonically discharging ORNs, detection of weak stimuli would be selectively enhanced in bursting ORNs. That is to say, even a near-threshold stimulus would elicit an intrinsic, "stereotyped" burst if it arrived at the correct phase of the bursting cycle of the ORN. Although the evoked bursts themselves were independent of stimulus concentration, it is important in the context of having a population of parallel bursting cells that the probability of eliciting a burst was concentration dependent. This would allow odor concentration to be encoded across the population of bursting cells because the more intense the stimulus, the earlier in its bursting cycle the cell would discharge in response to the odorant. As a result, the number of ORNs bursting synchronously would increase as the stimulus intensity increases, which presumably could be read by downstream circuitry in the olfactory lobe and beyond. Stimulus acquisition in the lobster olfactory system is periodic as a result of antennular flicking, a reflex shown to be functionally equivalent to sniffing in mammals (Schmitt and Ache 1979). Flicking intermittently and rapidly exposes the ORNs to odorants and could be expected to synchronize the exposure of all ORNs in the population to odorants to within several tens of milliseconds (Koehl et al. 2001; Schmitt and Ache 1997), thereby enhancing burst synchrony across the population. Thus the all-or-none nature of burst-dependent coding, together with synchronization of the bursting neurons, could be expected to selectively enhance the detection and amplification of weak signals, generally assumed to be one of the hallmarks of olfaction.
Bursts are well known to be transmitted across synapses more reliably than isolated action potentials (Lisman 1997). Although speculative in the context of the present study, the synchronized input from bursting cells could also potentially activate olfactory projection neurons and/or local interneurons, the central targets of ORNs, and prepare them to receive more linear input from the tonically active cells, as proposed for single neurons in the lateral geniculate (Sherman 2001). Also, bursts of action potentials potentially could provide an effective mechanism for selective communication with specific subsets of projection neurons or local interneurons (Izhikevich et al. 2003), although it remains to be determined whether ORNs showing the two patterns of responsiveness work ensemble or whether each codes selective subsets of odor information.
Overall, our findings strongly imply that intrinsically oscillatory ORNs occur in the lobster olfactory system. They also show for the first time that rhythmically bursting ORNs at least have the potential to play a fundamental role in coding odor information. The presence of receptor cells with rhythmic bursting discharge in the mammalian vomeronasal organ (Holy et al. 2000) as well as the conditional oscillations seen in amphibian and mammalian ORNs induced by prolonged odorant stimulation (Frings and Lindemann 1988; Reisert and Matthews 2001a,b) suggest our finding may be of general utility and help further our understanding of coding in other olfactory systems.
|
GRANTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
FOOTNOTES |
|---|
Address for reprint requests and other correspondence: Y. V. Bobkov, Whitney Laboratory for Marine Bioscience, University of Florida, 9505 Ocean Shore Blvd., Saint Augustine, FL 32080 (E-mail: bobkov{at}whitney.ufl.edu)
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
Ache BW, Young JM. Olfaction: diverse species, conserved principles. Neuron 48: 417–430, 2005.[CrossRef][ISI][Medline]
Birmingham JT, Szuts ZB, Abbott LF, Marder E. Encoding of muscle movement on two timescales by a sensory neuron that switches between spiking and bursting modes. J Neurophysiol 82: 2786–2797, 1999.
Bobkov YV, Ache BW. Pharmacological properties and functional role of a TRP-related ion channel in lobster olfactory receptor neurons. J Neurophysiol 93: 1372–1380, 2005.
Braun HA, Wissing H, Schafer K, Hirsch MC. Oscillation and noise determine signal-transduction in shark multimodal sensory cells. Nature 367: 270–273, 1994.[CrossRef][Medline]
Buzsáki G, Draguhn A. Neuronal oscillations in cortical networks. Science 304: 1926–1929, 2004.
Combes D, Simmers J, Moulins M. Conditioned dendritic oscillators in a lobster mechanoreceptor neurone. J Physiol 499: 161–177, 1997.[CrossRef][ISI]
Frings S, Lindemann B. Odorant response of isolated olfactory receptor-cells is blocked by amiloride. J Membr Biol 105: 233–243, 1988.[CrossRef][ISI][Medline]
Grünert U, Ache BW. Ultrastructure of the aesthetasc (olfactory) sensilla of the spiny lobster, Panulirus argus. Cell Tissue Res 251: 95–103, 1988.
Hallem EA, Ho MG, Carlson JR. The molecular basis of odor coding in the drosophila antenna. Cell 117: 965–979, 2004.[CrossRef][ISI][Medline]
Harris-Warrick RM. Voltage-sensitive ion channels in rhythmic motor systems. Curr Opin Neurobiol 12: 646–651, 2002.[CrossRef][ISI][Medline]
Hayar A, Karnup S, Ennis M, Shipley MT. External tufted cells: a major excitatory element that coordinates glomerular activity. J Neurosci 24: 6676–6685, 2004.
Hayar A, Karnup S, Shipley MT, Ennis M. Olfactory bulb glomeruli: external tufted cells intrinsically burst at theta frequency and are entrained by patterned olfactory input. J Neurosci 24: 1190–1199, 2004.
Holy TE, Dulac C, Meister M. Responses of vomeronasal neurons to natural stimuli. Science 289: 1569–1572, 2000.
Iggo A. Cutaneous thermoreceptors in primates and sub-primates. J Physiol 200: 403–430, 1969.
Izhikevich EM, Desai NS, Walcott EC, Hoppensteadt FC. Bursts as a unit of neural information: selective communication via resonance. Trends Neurosci 26: 161–167, 2003.[CrossRef][ISI][Medline]
Koehl MAR, Koseff JR, Crimaldi JP, Mccay MG, Cooper T, Wiley MB, Moore PA. Lobster sniffing: antennule design and hydrodynamic filtering of information in an odor plume. Science 294: 1948–1951, 2001.
Krahe R, Gabbiani F. Burst firing in sensory systems. Nat Rev Neurosci 5: 13–23, 2004.[CrossRef][ISI][Medline]
Lisman JE. Bursts as a unit of neural information: making unreliable synapses reliable. Trends Neurosci 20: 38–43, 1997.[CrossRef][ISI][Medline]
Marder E, Bucher D, Schulz DJ, Taylor AL. Invertebrate central pattern generation moves along. Curr Biol 15: R685–R699, 2005.[CrossRef][ISI][Medline]
Prinz AA, Thirumalai V, Marder E. The functional consequences of changes in the strength and duration of synaptic inputs to oscillatory neurons. J Neurosci 23: 943–954, 2003.
Ramirez JM, Tryba AK, Pena F. Pacemaker neurons and neuronal networks: an integrative view. Curr Opin Neurobiol 14: 665–674, 2004.[CrossRef][ISI][Medline]
Reisert J, Matthews HR. Responses to prolonged odour stimulation in frog olfactory receptor cells. J Physiol 534: 179–191, 2001a.
Reisert J, Matthews HR. Response properties of isolated mouse olfactory receptor cells. J Physiol 530: 113–122, 2001b.
Schmitt BC, Ache BW. Olfaction—responses of a decapod crustacean are enhanced by flicking. Science 205: 204–206, 1979.
Sherman SM. Tonic and burst firing: dual modes of thalamocortical relay. Trends Neurosci 24: 122–126, 2001.[CrossRef][ISI][Medline]
Steullet P, Cate HS, Derby CD. A spatiotemporal wave of turnover and functional maturation of olfactory receptor neurons in the spiny lobster Panulirus argus. J Neurosci 20: 3282–3294, 2000.
Wiederhold ML, Carpenter DO. Possible role of pacemaker mechanisms in sensory systems. In: Cellular Pacemakers, edited by Carpenter DO. New York: Wiley, 1982, p. 27–58.
Zheng JJ, Lee S, Zhou ZJ. A transient network of intrinsically bursting starburst cells underlies the generation of retinal waves. Nat Neurosci 9: 363–371, 2006.[CrossRef][ISI][Medline]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
