Nonsensory target-dependent organization of piriform cortex

The piriform cortex (PCX) is the largest component of the olfactory cortex and is hypothesized to be the locus of odor object formation. The distributed odorant representation found in PCX contrasts sharply with the topographical representation seen in other primary sensory cortices, making it difficult to test this view. Recent work in PCX has focused on functional characteristics of these distributed afferent and association fiber systems. However, information regarding the efferent projections of PCX and how those may be involved in odor representation and object recognition has been largely ignored. To investigate this aspect of PCX, we have used the efferent pathway from mouse PCX to the orbitofrontal cortex (OFC). Using double fluorescent retrograde tracing, we identified the output neurons (OPNs) of the PCX that project to two subdivisions of the OFC, the agranular insula and the lateral orbitofrontal cortex (AI-OPNs and LO-OPNs, respectively). We found that both AI-OPNs and LO-OPNs showed a distinct spatial topography within the PCX and fewer than 10% projected to both the AI and the LO as judged by double-labeling. These data revealed that the efferent component of the PCX may be topographically organized. Further, these data suggest a model for functional organization of the PCX in which the OPNs are grouped into parallel output circuits that provide olfactory information to different higher centers. The distributed afferent input from the olfactory bulb and the local PCX association circuits would then ensure a complete olfactory representation, pattern recognition capability, and neuroplasticity in each efferent circuit.The olfactory system creates perceptual odor objects from often complex mixtures of diverse airborne chemicals (1, 2). This formidable job is mainly accomplished by a surprisingly “shallow” three-level pathway, comprising the olfactory epithelium, olfactory bulb, and olfactory cortex (3). The olfactory epithelium accommodates millions of olfactory sensory neurons (OSNs), each of which can be defined by the particular receptor protein selected for expression from the ∼1,000 odor receptor genes in the typical mammalian genome (4, 5). Axons from all OSNs expressing the same odor receptor coalesce into a few glomeruli on the surface of the olfactory bulb (6–8). Each glomerulus is therefore dedicated to a particular receptor. The position of each glomerulus appears to vary only slightly from animal to animal, giving rise to speculation that the glomeruli form a spatial map of odor sensitivities.Within the glomeruli, the incoming OSN axons form synapses with the apical dendrites of second-order neurons and the mitral and tufted cells, providing what would seem to be an anatomical basis for topographical odorant representation (9–11). Each of about a dozen mitral or tufted cells innervating only a single glomerulus send their axons to targets in a number of ventral forebrain areas, collectively termed the olfactory cortex (12).However, this seemingly orderly topography of odorant representation is not maintained in the olfactory cortex. Especially in the largest olfactory area—the piriform cortex (PCX)—odorants are represented by sparse, distributed, and spatially overlapping neural ensembles across the cortex (13–18). This nontopographical representation stems largely from the architecture of the PCX, including distributed afferent inputs (19–21), and a similarly distributed intracortical association fiber system, which links single cortical neurons (pyramidal and semilunar cells) with neighboring and distant neurons (22–25). Furthermore, each cortical neuron receives an apparently random collection of glomerular inputs (26, 27). Therefore, a spatial location of the cortex is not predictive of odorant tuning as neighboring neurons may exhibit distinct receptive ranges (14, 18, 27).Compared with existing data on the afferent and association connections, data on the efferent aspects of PCX are limited. Although previous studies have identified a number of higher centers that are targeted by the PCX output neurons (OPNs), including the orbitofrontal cortex (OFC), hippocampus, hypothalamus, and thalamic nuclei (28–31), information regarding organization and spatial distribution of these neurons (pyramidal and semilunar cells) is lacking. Questions such as how the OPNs projecting to different targets are distributed within PCX and whether that may imply any intrinsic or functional organization of the PCX remain unanswered. Given the complexity of the distributed afferent and association fiber system, data from the efferent system may help to discern organizing principles in the PCX and lend some understanding as to how it processes incoming sensory information.To reach this goal, we focused on the projection from the PCX to the OFC, an important center for odor-guided behaviors (19, 30, 32). We injected different cholera toxin B (CTB) subunit fluorescent conjugates into two subdivisions of the OFC, the agranular insular (AI) and the lateral OFC (LO), in mice and examined the PCX for retrograde labeling (33). We found that the OPNs projecting to the AI and the LO are differentially distributed. Interestingly, both OPNs to the AI and LO exhibited topographically specific distributions in the PCX. In addition, they showed distinct distribution patterns along the anterior–posterior axis of the PCX. These two OPN populations had limited overlap within the anterior PCX (aPCX), as double-labeled neurons were extremely rare. These data suggest that the functional organization of PCX may be better understood through its output circuits, shedding new light on the role of olfactory cortex in central odor processing.     

Hierarchical excitatory synaptic connectivity in mouse olfactory cortex

Topological motifs in synaptic connectivity—such as the cortical column—are fundamental to processing of information in cortical structures. However, the mesoscale topology of cortical networks beyond columns remains largely unknown. In the olfactory cortex, which lacks an obvious columnar structure, sensory-evoked patterns of activity have failed to reveal organizational principles of the network and its structure has been considered to be random. We probed the excitatory network in the mouse olfactory cortex using variance analysis of paired whole-cell recording in olfactory cortex slices. On a given trial, triggered network-wide bursts in disinhibited slices had remarkably similar time courses in widely separated and randomly selected cell pairs of pyramidal neurons despite significant trial-to-trial variability within each neuron. Simulated excitatory network models with random topologies only partially reproduced the experimental burst-variance patterns. Network models with local (columnar) or distributed subnetworks, which have been predicted as the basis of encoding odor objects, were also inconsistent with the experimental data, showing greater variability between cells than across trials. Rather, network models with power-law and especially hierarchical connectivity showed the best fit. Our results suggest that distributed subnetworks are weak or absent in the olfactory cortex, whereas a hierarchical excitatory topology may predominate. A hierarchical excitatory network organization likely underlies burst generation in this epileptogenic region, and may also shape processing of sensory information in the olfactory cortex.Structural and functional plasticity at excitatory synapses in cortical networks represents a fundamental mechanism for encoding sensory representations and memory. As a result, neuronal ensembles that are connected with high probability emerge as functional units to produce a population code of the environment. The topology of such excitatory circuits should contain signatures—as global topological motifs—that reflect the encoding strategy. The cortical column is a well-studied example of such a motif (1). Columnar cortices contain substantial distributed connectivity and some brain areas, such as association cortex, high-order cortices, and the piriform cortex, lack a pronounced columnar structure. In the piriform (olfactory) cortex, there exists only a rudimentary understanding of the relationship between network structure and cortical function. The axons of individual piriform pyramidal neurons ramify widely throughout the olfactory cortex, and only show patchiness on a very broad scale (2–4). Consistent with this architecture, neural activity in response to individual odorants is distributed broadly across the olfactory cortex as detected by 2-deoxyglucose, c-fos expression, multiunit recording, and population calcium imaging (5–8). Likewise, the receptive fields of individual neurons in piriform cortex and anterior olfactory cortex are broad (9, 10). Broad receptive fields in piriform cortex reflect convergence of input from many olfactory bulb glomeruli (11) and are strongly influenced by recurrent connectivity (12).These observations support a highly distributed population representation but reveal little about what processing function the piriform cortex performs. Physiological and anatomical studies have provided some clues. For example, neuronal responses in piriform cortex are specific for category of odorant (13), and odor identity and similarity are separately encoded in anterior and posterior piriform cortex, respectively (14), suggesting hierarchical coding. The endopiriform (EN) and preendopiriform nucleus (pEN), immediately subjacent to the piriform cortex, have dense recurrent connectivity and dense connectivity with overlying areas of piriform cortex (15, 16). The pEN, also called area tempestas, is a highly epileptogenic locus (16, 17). However, the physiological role of its dense connectivity is unknown (15, 18).To probe excitatory connectivity in the olfactory cortex, we isolated excitatory synaptic activity in a tailored brain slice containing the ventral anterior piriform cortex (APC_V), the pEN, the anterior olfactory cortex (AOC; also called anterior olfactory nucleus). Using weak stimulation of the lateral olfactory tract (LOT) input while blocking GABAergic inhibition and NMDA receptors, we evoked transient, all-or-none, network-wide bursts of excitation. Network-wide transient bursts are a dynamic circuit property shared by the hippocampus, neocortex, and piriform cortex in disinhibited recording conditions (19–21). We used the pairwise variance patterns detectable in the fine structure of these bursts as a probe of excitatory network topology. We compared whole-cell recordings from randomly selected pairs of principal neurons in olfactory cortex with patterns generated in simulated networks with a range of network topologies. Our findings suggest that excitatory connectivity in olfactory cortex is neither random nor organized into local or distributed subnetworks. Rather, it shows hierarchical connectivity.     

Identity-specific coding of future rewards in the human orbitofrontal cortex

Nervous systems must encode information about the identity of expected outcomes to make adaptive decisions. However, the neural mechanisms underlying identity-specific value signaling remain poorly understood. By manipulating the value and identity of appetizing food odors in a pattern-based imaging paradigm of human classical conditioning, we were able to identify dissociable predictive representations of identity-specific reward in orbitofrontal cortex (OFC) and identity-general reward in ventromedial prefrontal cortex (vmPFC). Reward-related functional coupling between OFC and olfactory (piriform) cortex and between vmPFC and amygdala revealed parallel pathways that support identity-specific and -general predictive signaling. The demonstration of identity-specific value representations in OFC highlights a role for this region in model-based behavior and reveals mechanisms by which appetitive behavior can go awry.Predictive representations of future outcomes are critical for guiding adaptive behavior. To choose different types of rewards, such as food, shelter, and mates, it is essential that predictive signals contain specific information about the identity of those outcomes. Food rewards differ dramatically in their nutritional composition, and identity-specific cues allow differential foraging depending on current needs of the organism. The absence of precise mappings between predictive reward signals and their intended outcomes would have devastating effects on food-based decisions.Despite the ecological relevance of outcome-specific predictive coding, which can be observed even in Drosophila (1), most research in human and nonhuman primates has focused on “common currency” signals of economic values in the orbitofrontal cortex (OFC) (2, 3) and ventromedial prefrontal cortex (vmPFC) (4–8). These signals, which by definition are independent of the specific nature of the reward, can be used to compare and choose between alternative outcomes, but are unable to inform expectations about the specific identity of the outcome. For this, identity-specific representations that conjointly represent information about both affective value (how good is it?) and outcome identity (what is it?) are necessary. Recent data suggest that the OFC is involved in signaling information about specific outcomes (9–14). For instance, many OFC neurons signal both the value and the identity of the predicted outcome (12), and OFC lesions diminish the effects of outcome identity (but not general affective value) on conditioned behavior (13).Recent imaging work has also begun to address how the human brain encodes predictive information about rewarding outcomes. One study (9) used a functional magnetic resonance imaging (fMRI) adaptation paradigm to provide evidence for identity-based codes for reward in the OFC. Another investigation (4) used fMRI data from a willingness to pay auction combined with decoding techniques to reveal category-dependent and -independent value codes in vmPFC and lateral OFC, respectively. However, neither of these studies varied value independently of identity, and they were therefore unable to test for the presence of identity-specific and -general value codes in the OFC.Here, we combined an olfactory paradigm of classical conditioning with fMRI pattern-based approaches to test the hypothesis that the human OFC simultaneously encodes both the value and the identity of an expected rewarding outcome. Critically, we took advantage of two unique properties of appetizing food odors to reveal identity-specific value representations. First, food odors act as potent rewards (15–17), with pleasantness that can scale with odor intensity (18, 19). Second, different food odors vary widely in identity (e.g., chocolate cake vs. pizza) but may still hold similar value. These distinct features enabled us to systematically manipulate outcome value and identity independently within the same stimulus space.     

Memory formation and retrieval of neuronal silencing in the auditory cortex

Sensory stimuli not only activate specific populations of cortical neurons but can also silence other populations. However, it remains unclear whether neuronal silencing per se leads to memory formation and behavioral expression. Here we show that mice can report optogenetic inactivation of auditory neuron ensembles by exhibiting fear responses or seeking a reward. Mice receiving pairings of footshock and silencing of a neuronal ensemble exhibited a fear response selectively to the subsequent silencing of the same ensemble. The valence of the neuronal silencing was preserved for at least 30 d and was susceptible to extinction training. When we silenced an ensemble in one side of auditory cortex for conditioning, silencing of an ensemble in another side induced no fear response. We also found that mice can find a reward based on the presence or absence of the silencing. Neuronal silencing was stored as working memory. Taken together, we propose that neuronal silencing without explicit activation in the cerebral cortex is enough to elicit a cognitive behavior.Cortical neurons exhibit spontaneous activity without explicit external stimuli (1–3), which may not only increase, but also be suppressed, by sensory stimuli (4, 5). For example, auditory stimuli suppress a subset of auditory cortical neurons in a frequency-dependent manner (5). Synaptic inhibition in the cerebral cortex is fundamental for neuronal modulation (6), including gain control (7), response selectivity (8, 9), and synchronized activities (10, 11). Inhibition-based modulations may contribute to stimulus-driven behaviors and associative memories of sensory stimuli (12); however, it remains unclear whether neuronal silencing (i.e., a transient reduction in firing rates from their spontaneous level) by itself can serve as a memory trace and bring about behavioral expressions. In this study, we tested this possibility by optogenetically silencing auditory cortical neurons.     

Extinction reverses olfactory fear-conditioned increases in neuron number and glomerular size

Although much work has investigated the contribution of brain regions such as the amygdala, hippocampus, and prefrontal cortex to the processing of fear learning and memory, fewer studies have examined the role of sensory systems, in particular the olfactory system, in the detection and perception of cues involved in learning and memory. The primary sensory receptive field maps of the olfactory system are exquisitely organized and respond dynamically to cues in the environment, remaining plastic from development through adulthood. We have previously demonstrated that olfactory fear conditioning leads to increased odorant-specific receptor representation in the main olfactory epithelium and in glomeruli within the olfactory bulb. We now demonstrate that olfactory extinction training specific to the conditioned odor stimulus reverses the conditioning-associated freezing behavior and odor learning-induced structural changes in the olfactory epithelium and olfactory bulb in an odorant ligand-specific manner. These data suggest that learning-induced freezing behavior, structural alterations, and enhanced neural sensory representation can be reversed in adult mice following extinction training.Increasing evidence suggests that the cellular, neuroanatomical, and receptive field organizations of vertebrate sensory systems are continually reshaped throughout adulthood by cues from the external environment. Activity-dependent changes are known to occur both during critical periods of development and also in the adult brain, allowing the animal to optimally perform behaviors based on the demands of the surrounding environment. Postmitotic organizational changes, along with activity-dependent plasticity, have been largely implicated in shaping sensory circuits from development through adulthood (1–4). In particular, the olfactory sensory system of adult mice exhibits functional and neuroanatomical learning-dependent changes following olfactory fear conditioning in adulthood (5–7). The M71-LacZ transgenic mouse line expresses LacZ under the M71 odorant receptor (OR) promoter (encoded by the olfactory receptor 151 gene, Olfr151) (8) in the M71 OR-expressing, acetophenone-responsive population of olfactory sensory neurons (OSNs). Using this line, we previously demonstrated an increased number of M71-expressing OSNs in the main olfactory epithelium (MOE) of adult mice following olfactory fear conditioning to acetophenone (5, 7), an odorant that activates the M71/M72 ORs (9, 10). This increase in receptor-specific OSNs within the MOE was directly correlated with an increase in the area of M71+ axons innervating the M71 glomeruli within the olfactory bulbs (OBs). Behaviorally, these olfactory fear-conditioned mice also exhibited enhanced fear-potentiated startle (FPS) and freezing specific to the conditioned odor stimulus. Notably such changes were never seen with equivalent odorant exposure alone but only when the odorant was paired with an aversive or appetitive cue (5, 7), suggesting the critical importance of behavioral learning facilitating these structural and functional alterations.Reversing the behavioral and neuroanatomical effects of such emotional learning is important for our understanding of disorders such as posttraumatic stress disorder (PTSD), in which exposure-based psychotherapy is widely used for treatment. Notably, extinction training in rodent fear-conditioning models closely parallels many aspects of exposure-based psychotherapy in humans where exposure to nonreinforced presentations of the previously acquired conditioned stimulus (CS) reduces acquired fear responses such as freezing to the CS (11, 12). In the current study, we demonstrate that previously acquired structural changes within the primary olfactory system are reversed with olfactory fear extinction specific to the conditioned odorant cue.     

Theta burst stimulation dissociates attention and action updating in human inferior frontal cortex

Everyday circumstances require efficient updating of behavior. Brain systems in the right inferior frontal cortex have been identified as critical for some aspects of behavioral updating, such as stopping actions. However, the precise role of these neural systems is controversial. Here we examined how the inferior frontal cortex updates behavior by combining reversible cortical interference (transcranial magnetic stimulation) with an experimental task that measures different types of updating. We found that the right inferior frontal cortex can be functionally segregated into two subregions: a dorsal region, which is critical for visual detection of changes in the environment, and a ventral region, which updates the corresponding action plan. This dissociation reconciles competing accounts of prefrontal organization and casts light on the neural architecture of human cognitive control.     

Unitary response of mouse olfactory receptor neurons

The sense of smell begins with odorant molecules binding to membrane receptors on the cilia of olfactory receptor neurons (ORNs), thereby activating a G protein, G(olf), and the downstream effector enzyme, an adenylyl cyclase (ACIII). Recently, we have found in amphibian ORNs that an odorant-binding event has a low probability of activating sensory transduction at all; even when successful, the resulting unitary response apparently involves a single active Gα(olf)-ACIII molecular complex. This low amplification is in contrast to rod phototransduction in vision, the best-quantified G-protein signaling pathway, where each photoisomerized rhodopsin molecule is well known to produce substantial amplification by activating many G-protein, and hence effector-enzyme, molecules. We have now carried out similar experiments on mouse ORNs, which offer, additionally, the advantage of genetics. Indeed, we found the same low probability of transduction, based on the unitary olfactory response having a fairly constant amplitude and similar kinetics across different odorants and randomly encountered ORNs. Also, consistent with our picture, the unitary response of Gα(olf)(+/-) ORNs was similar to WT in amplitude, although their Gα(olf)-protein expression was only half of normal. Finally, from the action potential firing, we estimated that ≤19 odorant-binding events successfully triggering transduction in a WT mouse ORN will lead to signaling to the brain.     

Dedicated olfactory neurons mediating attraction behavior to ammonia and amines in Drosophila

Animals across various phyla exhibit odor-evoked innate attraction behavior that is developmentally programmed. The mechanism underlying such behavior remains unclear because the odorants that elicit robust attraction responses and the neuronal circuits that mediate this behavior have not been identified. Here, we describe a functionally segregated population of olfactory sensory neurons (OSNs) and projection neurons (PNs) in Drosophila melanogaster that are highly specific to ammonia and amines, which act as potent attractants. The OSNs express IR92a, a member of the chemosensory ionotropic receptor (IR) family and project to a pair of glomeruli in the antennal lobe, termed VM1. In vivo calcium-imaging experiments showed that the OSNs and PNs innervating VM1 were activated by ammonia and amines but not by nonamine odorants. Flies in which the IR92a⁺ neurons or IR92a gene was inactivated had impaired amine-evoked physiological and behavioral responses. Tracing neuronal pathways to higher brain centers showed that VM1-PN axonal projections within the lateral horn are topographically segregated from those of V-PN and DC4-PN, which mediate innate avoidance behavior to carbon dioxide and acidity, respectively, suggesting that these sensory stimuli of opposing valence are represented in spatially distinct neuroanatomic loci within the lateral horn. These experiments identified the neurons and their cognate receptor for amine detection, and mapped amine attractive olfactory inputs to higher brain centers. This labeled-line mode of amine coding appears to be hardwired to attraction behavior.     

Origins of correlated spiking in the mammalian olfactory bulb

Mitral/tufted (M/T) cells of the main olfactory bulb transmit odorant information to higher brain structures. The relative timing of action potentials across M/T cells has been proposed to encode this information and to be critical for the activation of downstream neurons. Using ensemble recordings from the mouse olfactory bulb in vivo, we measured how correlations between cells are shaped by stimulus (odor) identity, common respiratory drive, and other cells’ activity. The shared respiration cycle is the largest source of correlated firing, but even after accounting for all observable factors a residual positive noise correlation was observed. Noise correlation was maximal on a ∼100-ms timescale and was seen only in cells separated by <200 µm. This correlation is explained primarily by common activity in groups of nearby cells. Thus, M/T-cell correlation principally reflects respiratory modulation and sparse, local network connectivity, with odor identity accounting for a minor component.Mitral/tufted cells (M/Ts) of the olfactory bulb (OB) receive odor-evoked activity from sensory neurons and transmit it to central brain structures. Thus, understanding how odor information is represented by these neurons’ activity is essential to understanding olfactory coding. Studying coding properties at this stage in the olfactory system is particularly interesting because the small number of M/Ts (∼50,000) compared with sensory neurons (∼10 million) or olfactory cortical neurons (∼2 million) suggests that this stage represents a bottleneck (1).Odor information is encoded in the spatial pattern of activity across the OB (2). However, the timing of M/T activity may also play a crucial role in odor representation. Individual M/Ts fire odor-specific patterns of spikes (3), and spike timing across populations of M/Ts relative to the respiration cycle has been proposed as an olfactory code (4, 5). However, whether odor identity influences the correlation of M/T activity (i.e., the tendency of neurons to spike together) has not been specifically addressed.Ensemble firing patterns better predict odorant identity than do single neuron firing rates alone (6, 7), suggesting the utility of a population timing code. Additionally, learned olfactory behaviors are associated with increased M/T spike synchrony (8), and disrupting this synchrony in insect M/T analogs reduces odor discriminability (9). Furthermore, analysis of neural correlations has informed our understanding of the relationship between neural circuits and population activity and has constrained hypotheses concerning “decoding” of incoming population activity by downstream areas (10).Here, we evaluated how relative M/T timing depends upon odor identity and timing, respiration phase (inhalation/exhalation), and other neurons’ spiking. Correlated spiking in the OB is familiar (11, 12), but how these correlations depend on such variables is unknown. Correlations may originate in common stimulus or respiration phase preferences (“signal correlation”). Cell pairs’ spiking may also exhibit covariation beyond that predicted from such preferences (“noise correlation,” R_noise) and may reflect correlated input noise or synaptic coupling between cells (13, 14). In Xenopus and Drosophila, M/Ts and their analogs exhibit significant noise correlation (15, 16). However, the origins, magnitude, and scope of such correlations have not been described in the mammalian OB.Critically, correlation driven by respiration or population activity in the local circuit has not been estimated, yet this is required to understand the sources and possible functions of OB correlations, and the theoretical coding capacity and mechanisms of OB neural ensembles (17, 18). We contrast our analysis to the computation of trial-averaged population response correlations (i.e., “pattern correlations”). Our approach is more analogous to that of, for example, Kazama and Wilson (16): We address within-trial spike-timing correlations between cell pairs rather than correlations between trial-averaged responses to different odorants (19).We made ensemble recordings from mouse OB during odor presentation. From these recordings we isolated contributions of several olfactory variables to spiking in individual neurons and to intercell correlation. Respiration phase tuning accounts for much correlation, whereas some nearby cell pairs exhibit small, positive R_noise, independent of the stimulus. Conditional on the activity of the larger population, functional coupling between cells is sparse overall, with significant implications for olfactory coding.     

轻度认知功能障碍患者海马及嗅皮质微结构变化的扩散峰度成像研究

目的应用扩散峰度成像(DKI)技术探讨伴脑白质高信号(WMH)的轻度认知功能障碍(MCI)患者海马区和嗅皮质区微结构的变化。方法选择2018年1月~2019年1月于我院西院神经内科病房住院的82例WMH患者,应用蒙特利尔认知评估量表(MoCA)评价患者认知功能,并将患者分为MCI组48例和无MCI组34例。患者行常规MRI形态学及DKI扫描。选择患者海马区和嗅皮质区为感兴趣区,测量DKI相关参数,包括平均扩散系数(MD)、轴向扩散张量(AD)、垂直扩散张量(RD)、各向异性分数(FA)、平均峰度(MK)、轴向峰度(AK)、径向峰度(RK)和峰度各向异性分数(KFA),比较2组各参数差异,且对多个兴趣区多次比较用Bonferroni校正P值,并行相关性分析。结果与无MCI组比较,MCI组重度WMH比例明显增高(37.5%vs 14.7%,P=0.038),logistic回归分析显示,重度WMH是MCI独立危险因素(OR=0.134,95%CI:0.283~0.614,P=0.01)。经Bonferroni校正P值后显示,左侧海马区,MCI组MD和RD显著高于无MCI组,AK和MK显著低于无MCI组(P=0.005、0.006、0.011和0.002)。左侧海马区,FA、MK、RK和KFA与MoCA评分呈正相关(r=0.374,0.370,0.392,0.242,P<0.05),而MD、RD与MoCA评分呈负相关(r=-0.227,-0.255,P<0.05)。结论DKI技术可评估伴WMH的MCI患者海马区和嗅皮质区微结构的变化,DKI参数对MCI的临床影像学评估有一定的应用价值。     

Budesonide irrigation with olfactory training improves outcomes compared with olfactory training alone in patients with olfactory loss

Background

Olfactory training (OT) helps many patients with olfactory loss, but unfortunately it is ineffective for a significant number of patients. Budesonide irrigations are widely used to help patients with paranasal sinus inflammation, but have never been tested as a treatment for olfactory loss. We sought to examine the effect of adding budesonide irrigation to olfactory training on patients with olfactory loss without any visible sign of sinonasal inflammation.

Methods

In this randomized, controlled trial, 138 patients with olfactory loss and without any visible sign of sinonasal inflammation were randomized to either OT with saline irrigations or OT with budesonide irrigations. The University of Pennsylvania Smell Identification Test (UPSIT) was administered at the beginning of the study and at 6 months.

Results

A total of 133 patients completed the study. Forty‐seven patients (35.3%) had a clinically significant change in UPSIT score. Among those in the budesonide irrigation + olfactory therapy group, 43.9% improved, compared with 26.9% in the saline irrigation + olfactory therapy group (p = 0.039); this corresponds to an odds ratio of 3.93 (95% confidence interval, 1.20‐12.88) in a fully adjusted model (p = 0.024). Younger age and shorter duration of olfactory loss were also significant predictors of improvement.

Conclusion

Adding budesonide irrigation to olfactory training significantly improved olfactory ability compared with olfactory training plus saline irrigation.

     

Developmental and evolutionary constraints on olfactory circuit selection

Across species, neural circuits show remarkable regularity, suggesting that their structure has been driven by underlying optimality principles. Here we ask whether we can predict the neural circuitry of diverse species by optimizing the neural architecture to make learning as efficient as possible. We focus on the olfactory system, primarily because it has a relatively simple evolutionarily conserved structure and because its input- and intermediate-layer sizes exhibit a tight allometric scaling. In mammals, it has been shown that the number of neurons in layer 2 of piriform cortex scales as the number of glomeruli (the input units) to the 3/2 power; in invertebrates, we show that the number of mushroom body Kenyon cells scales as the number of glomeruli to the 7/2 power. To understand these scaling laws, we model the olfactory system as a three-layer nonlinear neural network and analytically optimize the intermediate-layer size for efficient learning from limited samples. We find, as observed, a power-law scaling, with the exponent depending strongly on the number of samples and thus on longevity. The 3/2 scaling seen in mammals is consistent with observed longevity, but the 7/2 scaling in invertebrates is not. However, when a fraction of the olfactory circuit is genetically specified, not learned, scaling becomes steeper for species with a small number of glomeruli and recovers consistency with the invertebrate scaling. This study provides analytic insight into the principles underlying both allometric scaling across species and optimal architectures in artificial networks.

Brains exhibit a large range of cell types, connectivity patterns, and organizational structures, at both micro- and macroscales. There is a rich history in neuroscience of explaining these structures from a normative point of view (1–3). Most of that work focused on computation, in the sense that it asked what circuit, and connection strengths, leads to optimal performance on a particular task. However, the connection strengths have to be learned, and model selection theory tells us that the efficiency of learning depends crucially on architecture, especially when a limited number of trials are available (4–8). This is also true for deep networks, where the choice of neural architecture plays a critical role in both learning speed and performance (9). Here we attempt to understand the organizational structure of the brain from a model selection perspective, hypothesizing that evolution optimized the brain for efficient learning.We build a model inspired by the olfactory circuitry and study its allometric scaling analytically. We focus on the olfactory system primarily because it has a relatively simple, evolutionarily conserved, predominantly feedforward structure (10–12). In particular, odorants are first detected by olfactory sensory neurons; from there, olfactory information is transmitted to glomeruli. The number of glomeruli, however, varies widely across species, from between 10 and 100 in insects to ∼1,000 in mammals. The question we address is, How does the number of glomeruli affect downstream circuitry? And in particular, what downstream circuitry would best help the animal survive? The tradeoffs that go into answering this question are in principle straightforward: More complicated circuitry (i.e., more parameters) can do a better job accurately predicting reward and punishment, but, because there are more parameters, there is a danger of overfitting (4, 7, 8). And even if learning is performed with sample-by-sample updates to avoid overfitting, learning tends to be slower in complicated circuitry, as typically more samples are required (13, 14). Navigating these tradeoffs requires that we choose an architecture, which must come from biology. For that we take inspiration from the olfactory system of both mammals and invertebrates.In the mammalian olfactory system, information from the glomeruli is transmitted to mitral/tufted cells, then to layer 2 of piriform cortex among others, and then mainly to layer 3; after that, information is passed on to higher-order cortical areas (10, 12). Thus, although many studies suggest that reciprocal interactions between mitral/tufted cells and granule cells (15, 16), as well as feedback from the cortex (17, 18), are also important for olfactory processing, as a first-order approximation the olfactory system can be modeled as a feedforward neural network. Moreover, because sister mitral cells receiving input from the same glomeruli are highly correlated, both with each other and with the glomeruli from which they receive input (19), the olfactory network essentially has three layers: an input layer corresponding to glomeruli, a hidden layer corresponding to layer 2 of piriform cortex, and an output layer corresponding to layer 3.Based on this picture, in our analysis we use an architecture corresponding to a three-layer feedforward network. The size of the input layer is the number of glomeruli, and we assume that each unit of the output layer is extracting a different feature of the olfactory input, such as expected reward or punishment, or a behaviorally relevant concept. Consequently, we focus on the hidden layer. For that we ask, How many units should the hidden layer have? That question was chosen partly because its answer provides insight into learning principles in general and partly because it was recently addressed experimentally: Srinivasan and Stevens (20) found, based on six mammalian species, a very tight relationship between the number of glomeruli and the number of neurons in layer 2 of piriform cortex (Fig. 1A; data taken from ref. 20). More precisely, using L_x to denote the input-layer size (the number of glomeruli) and L_h to denote the hidden-layer size (the number of neurons in layer 2 of piriform cortex), they found the approximate scaling law Lh∼Lx3/2.Open in a separate windowFig. 1.(A) Scaling law in mammalian olfactory circuits. Data points were taken from supplementary tables S2 and S3 of Srinivasan and Stevens (20). (B) Scaling law in invertebrate olfactory circuits. See SI Appendix, section 1.1 for details.Motivated by this result, we asked whether a similar scaling law holds for the invertebrate olfactory system. Like their mammalian counterparts, odors detected by olfactory sensory neurons converge to glomeruli. After that, though, the circuitry differs. Glomeruli send information to the projection neurons (12), which mainly extend synapses onto mushroom body Kenyon cells and lateral horn neurons (21). The latter is mostly related to innate olfactory processing (22), so we focus on the mushroom body, which transmits information to higher-order regions through mushroom body output neurons and is considered to be the learning center of the insect brain (23, 24). Insect olfactory circuits also contain various nonfeedfoward connections, such as lateral inhibition between the projection neurons (12). But, as with the mammalian olfactory system, as a first-order approximation we omit them from the model. Thus, the invertebrate olfactory system can also be modeled as a three-layer neural network: an input layer corresponding to glomeruli, a hidden layer corresponding to Kenyon cells, and an output layer corresponding to mushroom body output neurons (3, 25).A literature survey of the number of glomeruli and Kenyon cells of various insects (23, 24, 26–35) (see SI Appendix, section 1.1 for details) yielded a scaling law, as in the mammalian olfactory system, but with an exponent of about 7/2 rather than 3/2 (Lh∼Lx7/2, as shown in Fig. 1B). Drone (male) bees are the clear outlier. That might be because the caste system of honey bees puts the drones under unique ecological pressure; for instance, the drones are the only ones among the seven insects listed that do not engage in foraging. It should be noted that the data were not properly controlled, as they were collected from different sources and in some cases in different eras. Moreover, for the locust, we used the number of olfactory receptor genes instead of the number of glomeruli; that is because their microglomeruli structure makes direct comparison with other species difficult (36). In addition, the mushroom body also takes part in visual processing in bees and cockroaches (37).Several normative hypotheses have been offered to explain the population size of sensory circuits. One line of theoretical work showed that expansion in the hidden layer is beneficial for sensory coding (3, 38, 39), but it remains elusive how much expansion is optimal, because in these studies, more expansion was in principle always better. Other studies estimated the optimal population size in multiple layers from a width–depth tradeoff, assuming that the total number of neurons is fixed (40, 41) by external factors such as a constraint on energy (42). However, this energy constraint should be violated if increasing the number of neurons improves foraging ability, resulting in a better energy budget (43). Evaluation of the optimal population size was also attempted from other biological constraints, such as synaptic (44) and neuronal (45) noise. While these models provided insight into circuit structure, none were able to provide a quantitative explanation for the population sizes of circuits across different species. Srinivasan and Stevens (20), on the other hand, offered a quantitative derivation of the 3/2 power law observed in mammals. Their derivation was based on the hypothesis that not much information is lost between areas. While this is a reasonable hypothesis, their derivation relied on several implicit assumptions; in particular, they assumed that the noise between different neurons is uncorrelated, and the olfactory signals are not mixed as they propagate across layers. However, both correlations and mixing are likely to exist, and that will affect the number of neurons required in downstream areas (46). In addition, their theory does not explain the 7/2 scaling seen in invertebrates.Here we develop a mechanistic explanation of the scaling laws, focusing on the fact that the transformation from glomeruli to piriform cortex (for mammals), or from glomeruli to mushroom body output neurons (for invertebrates), has to be learned from a limited number of samples. That explanation draws on model selection theory, in which the primary constraint is the poverty of teaching signals and resultant overfitting (4, 7, 8). Because the olfactory circuit has to tune its numerous synaptic weights from very sporadic, low-dimensional reward signals in the natural environment (47, 48), this constraint should be highly relevant. Therefore, we formulate the problem of olfactory circuit design as a model selection problem and then derive the optimal hidden-layer size under various learning rules and nonlinearities.Our derivation proceeds in two steps: First, we expand the covariance matrix in the hidden layer in powers of the average hidden-layer correlation coefficient; then we use random matrix theory to compute the generalization error. That enabled us to determine, analytically, the factors that control the optimal hidden-layer size, thus bypassing the intensive numerical optimization typically used in deep-learning settings (9, 40, 49). Our analysis shows that the optimal hidden-layer size follows an allometric scaling with the input-layer size and reveals the factors that control the scaling exponent. Not surprisingly (because learning takes time) we find that the optimal hidden-layer size, and thus the scaling exponent, depends on the lifetime of the organism. The 3/2 scaling found in mammals is, though, largely consistent with observed lifetimes. This scaling relationship is robust against the choice of nonlinearity, activity sparseness, and the noise level and also against the optimization method: It holds under both maximum-likelihood estimation and stochastic gradient descent with cumulative error minimization.Our theory was not, however, able to capture the 7/2 power law found in invertebrates. That is because traditional model selection theory fails to take into account the fact that neural circuits are at least partially genetically specified. In particular, rich innate connectivity structure is known to exist in the invertebrate olfactory systems (22, 50). Thus, we extend the framework to the case where a fixed genetic budget can be used to specify connections and consider how that affects scaling. The budget we used—about 2,000 bits—had little effect on the scaling of the mammalian circuit, primarily because mammals have a large number of glomeruli, for which a complicated downstream circuit is needed to achieve good performance—far more complicated than could be constructed by 2,000 bits. However, it had a large effect on invertebrates, which contain far fewer glomeruli. In particular, the scaling became steeper, making it possible to replicate the observed 7/2 power law without disrupting the 3/2 power law in mammals. These results shed light on potential constraints on the development and evolution of neural circuitry.     

Collapse of complexity of brain and body activity due to excessive inhibition and MeCP2 disruption

Complex body movements require complex dynamics and coordination among neurons in motor cortex. Conversely, a long-standing theoretical notion supposes that if many neurons in motor cortex become excessively synchronized, they may lack the necessary complexity for healthy motor coding. However, direct experimental support for this idea is rare and underlying mechanisms are unclear. Here we recorded three-dimensional body movements and spiking activity of many single neurons in motor cortex of rats with enhanced synaptic inhibition and a transgenic rat model of Rett syndrome (RTT). For both cases, we found a collapse of complexity in the motor system. Reduced complexity was apparent in lower-dimensional, stereotyped brain–body interactions, neural synchrony, and simpler behavior. Our results demonstrate how imbalanced inhibition can cause excessive synchrony among movement-related neurons and, consequently, a stereotyped motor code. Excessive inhibition and synchrony may underlie abnormal motor function in RTT.

A diverse and complex repertoire of body movements requires diverse and complex neural activity among cortical neurons. Moreover, interactions between movement-related neurons and the body must be sufficiently high dimensional to carry these movement signals with high fidelity. The complexity of movement-related neural activity and neuron–body interactions can be compromised if synchrony among neurons is excessive. Indeed, it is well understood theoretically that excessive correlations can limit the information capacity of any neural code (1–3)—if all neurons are perfectly synchronized, then different neurons cannot encode different motor signals. Synchrony is also known to play a role in pathophysiology of movement-related disorders, like Parkinson’s disease (4–6). However, synchrony and correlations also contribute to healthy function in the motor system (7–14). For instance, particular groups of synchronized neurons seem to send control signals to particular muscle groups (7, 8) and propagation of correlated firing contributes to motor planning (10). Synchrony can also play a role in motor learning (12–14). These findings suggest that correlated activity among specific subsets of neurons encodes specific motor functions. Thus, it stands to reason that if this synchrony became less selective and more stereotyped across neurons, then the motor code would become less complex and lose specificity, resulting in compromised motor function.Here we explored this possibility in rats, in the caudal part of motor cortex where neurons associated with hindlimb, forelimb, and trunk body movement are located (15–17). We focused on two conditions. First, we studied a transgenic rat model of Rett syndrome (RTT), which has disrupted expression of the MeCP2 gene. Second, we studied normal rats with acutely altered inhibitory neural interactions. Both of these cases are associated with abnormal motor behavior and, possibly, abnormal synchrony. Abnormal synchrony is a possibility, because both of these cases are linked to an imbalance between excitatory (E) and inhibitory (I) neural interactions, which in turn is likely to result in abnormal synchrony. For instance, many computational models suggest that synchrony is strongly dependent on E/I interactions (18–21). Likewise, in experiments, pharmacological manipulation of E/I causes changes in synchrony (19, 22, 23) and the excessive synchrony that occurs during epileptic seizures is often attributed to an E/I imbalance (24, 25). Similarly, the majority of people with RTT suffer from seizures (26) and many previous studies establish E/I imbalance as a common problem in RTT (27). MeCP2 dysfunction, which is known to cause RTT, seems to be particularly important in inhibitory neurons (28). For instance, two studies have shown that disrupting MeCP2 only in specific inhibitory neuron types can recapitulate the effects of brain-wide disruption of MeCP2 (29, 30). However, whether the E/I imbalance favors E or I at the population level seems to vary across different brain regions in RTT. Studies of visual cortex (29) and hippocampus (31) suggest that the balance tips toward too much excitation (perhaps explaining the prevalence of seizures), while studies of somatosensory cortex (32, 33) and a brain-wide study of Fos expression (34) suggest that frontal areas, including motor cortex, are tipped toward excessive inhibition. These facts motivated our choice to study pharmacological disruption of inhibition here. While it is clear that E/I imbalance is important in RTT, it is much less clear how it manifests at the level of dynamics and complexity of neural activity that is responsible for coordinating body movements. Thus, in addition to pursuing the general questions about synchrony and complexity in the motor system discussed above, a second goal of our work was to improve understanding of motor dysfunction due to MeCP2 disruption.Taken together, these facts led us to the following questions: How does MeCP2 disruption impact the complexity of body movements, movement-related neural activity, and motor coding? Are abnormalities in the MeCP2-disrupted motor system consistent with excessive inhibition in motor cortex? We hypothesized that both MeCP2 disruption and excessive inhibition lead to reduced complexity of interactions between cortical neurons and body movements, excessive cortical synchrony, and reduced complexity of body movements. Our findings confirmed this hypothesis and suggest that RTT-related motor dysfunction may be due, in part, to excessive synchrony and inhibition in motor cortex.     

Glutamate is an inhibitory neurotransmitter in the Drosophila olfactory system

Glutamatergic neurons are abundant in the Drosophila central nervous system, but their physiological effects are largely unknown. In this study, we investigated the effects of glutamate in the Drosophila antennal lobe, the first relay in the olfactory system and a model circuit for understanding olfactory processing. In the antennal lobe, one-third of local neurons are glutamatergic. Using in vivo whole-cell patch clamp recordings, we found that many glutamatergic local neurons are broadly tuned to odors. Iontophoresed glutamate hyperpolarizes all major cell types in the antennal lobe, and this effect is blocked by picrotoxin or by transgenic RNAi-mediated knockdown of the GluClα gene, which encodes a glutamate-gated chloride channel. Moreover, antennal lobe neurons are inhibited by selective activation of glutamatergic local neurons using a nonnative genetically encoded cation channel. Finally, transgenic knockdown of GluClα in principal neurons disinhibits the odor responses of these neurons. Thus, glutamate acts as an inhibitory neurotransmitter in the antennal lobe, broadly similar to the role of GABA in this circuit. However, because glutamate release is concentrated between glomeruli, whereas GABA release is concentrated within glomeruli, these neurotransmitters may act on different spatial and temporal scales. Thus, the existence of two parallel inhibitory transmitter systems may increase the range and flexibility of synaptic inhibition.