User:Eugene M. Izhikevich/Proposed/Olfactory oscillations
The olfactory system shows many types of oscillations in the local field potential (LFP) signal. These are near sinusoidal signals of various frequency bands. Gamma oscillations (40-100 Hz in many mammals), are evoked by odorants when an animal inhales Figure 1. Beta oscillations are in the 15-30 Hz range, and are associated with learning and odor sensitization. Respiratory, or theta, oscillations (2-12 Hz) follow the respiratory rhythm and are associated with input to the olfactory bulb from the olfactory nerve. The olfactory bulb shows these oscillations most robustly, but they can be seen often at lower amplitudes in other brain regions that form the extended olfactory and limbic systems.
Odor-evoked oscillations in the mammalian olfactory bulb were first described in the literature in the 1940s in the oft-cited papers by Lord Edgar Douglas Adrian. Adrian reported that under anesthesia in cats, rats and hedgehogs various odors produced an oscillation of the local population spiking and the LFP of about 40 Hz Figure 1. Around the same time several reports appeared noting a slower oscillation that followed the respiratory rhythm Figure 1.
In the late 1950s Walter J. Freeman first described the properties of an oscillatory evoked potential in the piriform (or pyriform) cortex (Freeman, 1959), and in the early 1960s the first computational studies of pyriform cortex oscillations appeared. Many papers by Freeman between 1962 and 1968 helped lay the foundation for computational analysis of cortical oscillations (Freeman, 1975). In 1968, Rall and Shepherd published the first computational interpretation of mammalian OB oscillatory evoked potential.
Basic structure of the olfactory bulb
The mammalian olfactory bulb is a 3-layered paleocortex in which outgoing axons arise from a relatively constrained mitral cell layer and tufted cells that can be displaced from this layer. This same population of neurons projecting out of the olfactory bulb receives input directly from the sensory afferents in the nasal epithelium. These inputs contact the mitral cells in the glomeruli occupying the periphery of the main olfactory bulb. There are many types of small neurons that surround the olfactory bulb glomeruli (juxtaglomerular cells) and provide GABAergic, dopamine and glutamatergic influences on other juxtaglomerular cells, sensory axon terminals and mitral cells. A large population of primarily GABAergic interneurons lies deep to the mitral cell layer; these are mostly granule cells, but also deep short axon cells (Eyre et al., 2008), Blanes cells (Pressler and Strowbridge, 2006) and others. Many granule cells form dendrodendritic synapses with mitral cells in the external plexiform layer. Centrifugal inputs from other brain areas are of many types, but most excitatory incoming fibers are from glutamatergic neurons in many areas of the olfactory and limbic forebrain, including the piriform cortex, entorhinal cortex, some parts of the extended amygdala, and the ventral hippocampus. Other modulatory inputs come from the hypothalamus, the basal forebrain (horizontal nucleus of the diagonal band of Broca) and brainstem neuromodulatory centers (Raphe nuclei and Locus Coeruleus). This circuitry of the olfactory bulb is capable of producing many types of oscillatory signals that have been shown to have specific behavioral, sensory processing and computational properties.
The power spectrum from olfactory bulb local field potential signals conforms to a 1/f noise process, with deviations from this line in the high theta (6-12 Hz), beta (15-30 Hz) and gamma (40-100 Hz) bands.
In the mammalian olfactory bulb there is a slow respiratory-associated rhythm (2-12 Hz; Fig 1A) that overlaps in frequency with the theta band described in animal studies of the hippocampus (4-12 Hz). Because of this, the respiratory oscillation is often referred to as a theta rhythm. This oscillation is primarily driven by the afferent volley from the olfactory nerve exciting the juxtaglomerular and mitral cell populations, and juxtaglomerular cells fire preferentially in the respiratory frequency range, which may support the theta oscillation (Hayar et al., 2004). Respiratory-associated mitral cell firing also has a central component (Ravel and Pager, 1990). This rhythm is considered an intrabulbar measure of the respiratory cycle. The respiratory rhythm can be coupled with hippocampal theta rhythms during some types of olfactory associative learning, but is otherwise independent of the hippocampal rhythm. In anesthetized rats, the respiratory rhythm cycles in and out of coherence with the piriform cortex theta rhythm, similar to the sleep cycling seen in thalamocortical systems (Fontanini and Bower, 2006).
The gamma oscillation (40-100 Hz in rodents, with lower frequencies in other mammals) is the best understood of all the olfactory system oscillations [Figure 2]. It was first described as an odor-evoked oscillation, although it can be seen even in the absence of specific odorants. This oscillation arises from the reciprocal dendrodendritic synapse between glumatergic mitral and GABAergic granule cells in the external plexiform layer in the olfactory bulb. The probability of a mitral cell firing is strongly coupled to the LFP gamma oscillation, so that as the cells in a population become more strongly coupled to each other, the gamma oscillation becomes larger.
Nonmammalian vertebrates also show odor-evoked gamma-like oscillations. Frogs show a 10 Hz oscillation. Study of zebrafish has yielded a great deal of insight into the coding properties of neurons in the olfactory bulb, and this system produces an odor-evoked gamma-like oscillation as well (20-30 Hz).
An ‘’analogous’’ odor-evoked oscillation is seen in many insect olfactory systems, arising from the interaction of projection neurons (PNs) and local neurons (LNs) in the antennal lobe, but measured either intracellularly in PNs as fluctuations in membrane potential or from postsynaptic activity of the PNs in the mushroom body [Figure 2xx]. While the frequency of these oscillations is considerably lower in insects (~20 Hz), the neurons a nd synapses involved and the relationship of the PN firing patterns to the LFP are the same as in the vertebrate systems.
Because of differing frequencies among species, while gamma is used to refer to a frequency band, ‘’it is the relationship of the mitral cells to the odor-evoked oscillation that defines the gamma oscillation’’.
Several important studies have pointed at a functional role for gamma oscillations. Schneider and Freeman showed that activity in the gamma band over a spatial array on the surface of the rabbit olfactory bulb represents meaning-related activity associated with an odorant. MacLeod and Laurent showed that ablating this rhythm in the insect system by blocking antennal lobe GABA receptors leaves the slow temporal firing patterns of the PNs intact but abolishes the fast temporal structure represented by the odor-evoked 20 Hz oscillation. This same treatment leaves honeybees unable to discriminate between closely related odorants (fine discrimination), but discriminating unrelated odorants (coarse discrimination) is unaffected (Stopfer, Smith, Laurent and colleagues). Nusser and colleagues showed an opposite effect in the beta3 knockout mouse. These mice have enhanced olfactory bulb gamma oscillations and are better at fine odor discrimination than control littermates. Finally, Beshel and colleagues showed that rats increase the amount of gamma oscillatory activity by as much as an order of magnitude when performing fine odor discriminations as opposed to coarse ones.
A lower frequency rhythm in the rodent gamma range has also been described [Fig xx; ref], and this oscillation predominates when mammals are attentive and motionless. This rhythm has been called ‘’gamma 2’’, with the odor-evoked gamma labeled ‘’gamma 1’’. There is some evidence that inhibition of granule cells plays a role in the gamma 2 rhythm, but little else is known about the functionality of this oscillatory process.
Beta oscillations (15-30 Hz in rats) have been described in many systems, and in the olfactory system they have recently been associated with odor learning and odor sensitization [refs; fig xx]. The mechanisms of this oscillation are less well understood than for the gamma 1 oscillation, but they appear to require an intact bidirectional loop between the olfactory bulb and the rest of the brain [martin et al, 2006; Neville & Haberly, 2003]. Beta oscillations can be quite strong in both the olfactory bulb and the piriform cortex, showing a very robust coherence profile. In odor discrimination learning, beta oscillation power rises with the onset of criterion performance in a Go / No-Go task and is seen in the olfactory bulb, piriform cortex, entorhinal cortex, and both the septal and temporal poles of the hippocampus. In odor sensitization, highly volatile pure odorants evoke an odor-specific beta oscillation after several presentations [zibrowski] accompanied by enhanced coupling in the theta and beta frequency bands between the olfactory bulb and piriform cortex [Lowry & Kay]. It is not known whether the learning-associated and odor sensitization beta oscillatory events use the same mechanism.
Implications for neural coding
Some important unsolved questions
- Fontanini, A. and Bower, J.M. (2006) Slow-waves in the olfactory system: an olfactory perspective on cortical rhythms. Trends in Neurosciences, 29(8): 429-437.
- Ravel, N. and Pager, J. (1990) Respiratory patterning of the rat olfactory bulb unit activity: nasal versus tracheal breathing. Neuroscience Letters, 115(2-3): 213-218.
- Hayar, A., Karnup, S., Shipley, M.T., and Ennis, M. (2004) Olfactory bulb glomeruli: external tufted cells intrinsically burst at theta frequency and are entrained by patterned olfactory input. Journal of Neuroscience, 24(5): 1190-1199.
- Eyre, M.D., Antal, M., and Nusser, Z. (2008) Distinct deep short-axon cell subtypes of the main olfactory bulb provide novel intrabulbar and extrabulbar GABAergic connections. Journal of Neuroscience, 28(33): 8217-8229.
- Pressler, R.T. and Strowbridge, B.W. (2006) Blanes cells mediate persistent feedforward inhibition onto granule cells in the olfactory bulb. Neuron, 49(6): 889-904.
- Freeman, W.J. (1959) Distribution in time and space of prepyriform electrical activity. Journal of Neurophysiology, 22(6): 644-665.
- Neville, K.R. and Haberly, L.B. (2003) Beta and gamma oscillations in the olfactory system of the urethane-anesthetized rat. Journal of Neurophysiology, 90(6): 3921-30.
- Bathellier, B., Lagier, S., Faure, P., and Lledo, P.M. (2006) Circuit properties generating gamma oscillations in a network model of the olfactory bulb. Journal of Neurophysiology, 95(4): 2678-2691.
- Laurent, G. and Davidowitz, H. (1994) Encoding of olfactory information with oscillating neural assemblies. Science, 265(5180): 1872-1875.
- MacLeod, K. and Laurent, G. (1996) Distinct mechanisms for synchronization and temporal patterning of odor-encoding neural assemblies. Science, 274(5289): 976-979.
- Stopfer, M., Bhagavan, S., Smith, B.H., and Laurent, G. (1997) Impaired odour discrimination on desynchronization of odour- encoding neural assemblies. Nature, 390(6655): 70-74.
- Nusser, Z., Kay, L.M., Laurent, G., Homanics, G.E., and Mody, I. (2001) Disruption of GABA(A) receptors on GABAergic interneurons leads to increased oscillatory power in the olfactory bulb network. Journal of Neurophysiology, 86(6): 2823-2833.
- Kay LM, Beshel J, Brea J, Martin C, Rojas-Libano D and Kopell N(2009) Olfactory oscillations: the what, how and what for. Trends in Neurosciences, 32(4): 207-214.
- Kay LM and Stopfer M (2006) Information processing in the olfactory systems of insects and vertebrates. Seminars in Cell & Developmental Biology, 17(4): 433-442.
- Rojas-Libano D and Kay LM (2008) Olfactory system gamma oscillations: the physiological dissection of a cognitive neural system. Cognitive Neurodynamics, 2(3): 179-194.
- Freeman WJ, Mass Action in the Nervous System. 1975, New York: Academic Press. 489 pp.