Single-Cell Multi-omic Integration Compares and Contrasts Features of Brain Cell Identity.

Defining cell types requires integrating diverse single-cell measurements from multiple experiments and biological contexts. To flexibly model single-cell datasets, we developed LIGER, an algorithm that delineates shared and dataset-specific features of cell identity. We applied it to four diverse and challenging analyses of human and mouse brain cells. First, we defined region-specific and sexually dimorphic gene expression in the mouse bed nucleus of the stria terminalis. Second, we analyzed expression in the human substantia nigra, comparing cell states in specific donors and relating cell types to those in the mouse. Third, we integrated in situ and single-cell expression data to spatially locate fine subtypes of cells present in the mouse frontal cortex. Finally, we jointly defined mouse cortical cell types using single-cell RNA-seq and DNA methylation profiles, revealing putative mechanisms of cell-type-specific epigenomic regulation. Integrative analyses using LIGER promise to accelerate investigations of cell-type definition, gene regulation, and disease states.

Nociceptor spontaneous activity is responsible for fragmenting non-rapid eye movement sleep in mouse models of neuropathic pain.

Spontaneous pain, a major complaint of patients with neuropathic pain, has eluded study because there is no reliable marker in either preclinical models or clinical studies. Here, we performed a comprehensive electroencephalogram/electromyogram analysis of sleep in several mouse models of chronic pain: neuropathic (spared nerve injury and chronic constriction injury), inflammatory (Freund's complete adjuvant and carrageenan, plantar incision) and chemical pain (capsaicin). We find that peripheral axonal injury drives fragmentation of sleep by increasing brief arousals from non-rapid eye movement sleep (NREMS) without changing total sleep amount. In contrast to neuropathic pain, inflammatory or chemical pain did not increase brief arousals. NREMS fragmentation was reduced by the analgesics gabapentin and carbamazepine, and it resolved when pain sensitivity returned to normal in a transient neuropathic pain model (sciatic nerve crush). Genetic silencing of peripheral sensory neurons or ablation of CGRP+ neurons in the parabrachial nucleus prevented sleep fragmentation, whereas pharmacological blockade of skin sensory fibers was ineffective, indicating that the neural activity driving the arousals originates ectopically in primary nociceptor neurons and is relayed through the lateral parabrachial nucleus. These findings identify NREMS fragmentation by brief arousals as an effective proxy to measure spontaneous neuropathic pain in mice.

Decreased alertness due to sleep loss increases pain sensitivity in mice.

Extended daytime and nighttime activities are major contributors to the growing sleep deficiency epidemic, as is the high prevalence of sleep disorders like insomnia. The consequences of chronic insufficient sleep for health remain uncertain. Sleep quality and duration predict presence of pain the next day in healthy subjects, suggesting that sleep disturbances alone may worsen pain, and experimental sleep deprivation in humans supports this claim. We demonstrate that sleep loss, but not sleep fragmentation, in healthy mice increases sensitivity to noxious stimuli (referred to as 'pain') without general sensory hyper-responsiveness. Moderate daily repeated sleep loss leads to a progressive accumulation of sleep debt and also to exaggerated pain responses, both of which are rescued after restoration of normal sleep. Caffeine and modafinil, two wake-promoting agents that have no analgesic activity in rested mice, immediately normalize pain sensitivity in sleep-deprived animals, without affecting sleep debt. The reversibility of mild sleep-loss-induced pain by wake-promoting agents reveals an unsuspected role for alertness in setting pain sensitivity. Clinically, insufficient or poor-quality sleep may worsen pain and this enhanced pain may be reduced not by analgesics, whose effectiveness is reduced, but by increasing alertness or providing better sleep.

Specific Targeting of the Basolateral Amygdala to Projectionally Defined Pyramidal Neurons in Prelimbic and Infralimbic Cortex.

Adjacent prelimbic (PL) and infralimbic (IL) regions in the medial prefrontal cortex have distinct roles in emotional learning. A complete mechanistic understanding underlying this dichotomy remains unclear. Here we explored targeting of specific PL and IL neurons by the basolateral amygdala (BLA), a limbic structure pivotal in pain and fear processing. In mice, we used retrograde labeling, brain-slice recordings, and adenoviral optogenetics to dissect connectivity of ascending BLA input onto PL and IL neurons projecting to the periaqueductal gray (PAG) or the amygdala. We found differential targeting of BLA projections to PL and IL cortex. Activating BLA projections evoked excitatory and inhibitory responses in cortico-PAG (CP) neurons in layer 5 (L5) of both PL and IL cortex. However, all inhibitory responses were polysynaptic and monosynaptic BLA input was stronger to CP neurons in IL cortex. Conversely, the BLA preferentially targeted corticoamygdalar (CA) neurons in layer 2 (L2) of PL over IL cortex. We also reveal that BLA input is projection specific by showing preferential targeting of L5 CP neurons over neighboring L3/5 CA neurons in IL cortex. We conclude by showing that BLA input is laminar-specific by producing stronger excitatory responses CA neurons in L3/5 compared with L2 in IL cortex. Collectively, this study reveals differential targeting of the BLA to PL and IL cortex, which depends both on laminar location and projection target of cortical neurons. Overall, our findings should have important implications for understanding the processing of pain and fear input by the PL and IL cortex.

Highly differentiated cellular and circuit properties of infralimbic pyramidal neurons projecting to the periaqueductal gray and amygdala.

The infralimbic (IL) cortex is a key node in an inter-connected network involved in fear and emotion processing. The cellular and circuit-level mechanisms whereby IL neurons receive, filter, and modulate incoming signals they project onward to diverse downstream nodes in this complex network remain poorly understood. Using the mouse as our model, we applied anatomical labeling strategies, brain slice electrophysiology, and focal activation of caged glutamate via laser scanning photostimulation (glu-LSPS) for quantitative neurophysiological analysis of projectionally defined neurons in IL. Injection of retrograde tracers into the periaqueductal gray (PAG) and basolateral amygdala (BLA) was used to identify cortico-PAG (CP) and cortico-BLA (CA) neurons in IL. CP neurons were found exclusively in layer 5 (L5) of IL whereas CA neurons were detected throughout layer 2, 3, and 5 of IL. We also identified a small percentage of IL neurons that project to both the PAG and the BLA. We found that L5 CP neurons have a more extensive dendritic structure compared to L5 CA neurons. Neurophysiological recordings performed on retrogradely labeled neurons in acute brain slice showed that CP and CA neurons in IL could be broadly classified in two groups: neuronal resonators and non-resonators. Layer 2 CA neurons were the only class that was exclusively non-resonating. CP, CA, and CP/CA neurons in layers 3 and 5 of IL consisted of heterogeneous populations of resonators and non-resonators showing that projection target is not an exclusive predictor of intrinsic physiology. Circuit mapping using glu-LSPS revealed that the strength and organization of local excitatory and inhibitory inputs were stronger to CP compared to CA neurons in IL. Together, our results establish an organizational scheme linking cellular neurophysiology with microcircuit parameters of defined neuronal subclasses in IL that send descending commands to subcortical structures involved in fear behavior.

Probing NMDA receptor GluN2A and GluN2B subunit expression and distribution in cortical neurons.

The spatial distribution of N-methyl-d-aspartate receptor (NMDAR) subunits in layer 5 (L5) neurons of the medial prefrontal cortex (mPFC) is important for integrating input-output signals involved in cognitive functions and motor behavior. In this study, focal laser scanning photostimulation of caged glutamate, slice electrophysiology, and small peptide pharmacology, were used to map the distribution of functional GluN2A and GluN2B subunits of the NMDAR from L5 neurons of wild-type (WT) and GluN2A(-/-) mice. Focal uncaging of glutamate evoked spatially-restricted glutamatergic responses on various dendritic locations of pyramidal neurons in the mPFC. Analyses of the spatial arrangements of the GluN2A and GluN2B subunits were performed by comparing inhibition of glutamatergic responses in the presence of the GluN2A-selective pharmacological antagonist, NVP-AAM077 (NVP), and the GluN2B-selective peptidic antagonist, conantokin-G (con-G). We found that apical and basal expression and distribution of GluN2A and GluN2B were similar in L5 mPFC neurons of WT mice. However, the inhibition of glutamatergic responses by NVP in brain slices of GluN2A(-/-) mice were dramatically decreased, while con-G inhibition remained similar to that observed in WT brain slices. The data obtained show that expression and spatial arrangement of GluN2B subunits is independent of GluN2A in L5 neurons of the mPFC. These findings have important ramifications for NMDAR organization and function in L5 pyramidal neurons of the mPFC, and show that specific populations of NMDARs can be antagonized, while sparing other subgroups of NMDARs, thus preserving selective NMDAR functions, an important therapeutic advantage.