Rapid eye movement (REM) sleep consists of a dreaming state in which there is activation of the cortical and hippocampal electroencephalogram (EEG), rapid eye movements, and loss of muscle tone. Although REM sleep was discovered more than 50 years ago, the neuronal circuits responsible for switching between REM and non-REM (NREM) sleep remain poorly understood. Here we propose a brainstem flip-flop switch, consisting of mutually inhibitory REM-off and REM-on areas in the mesopontine tegmentum. Each side contains GABA (gamma-aminobutyric acid)-ergic neurons that heavily innervate the other. The REM-on area also contains two populations of glutamatergic neurons. One set projects to the basal forebrain and regulates EEG components of REM sleep, whereas the other projects to the medulla and spinal cord and regulates atonia during REM sleep. The mutually inhibitory interactions of the REM-on and REM-off areas may form a flip-flop switch that sharpens state transitions and makes them vulnerable to sudden, unwanted transitions-for example, in narcolepsy.

SUMMARYAlthough microglia are implicated in nerve injury-induced neuropathic pain, how injured sensory neurons engage microglia is unclear. Here we demonstrate that peripheral nerve injury induces de novo expression of colony-stimulating factor 1 (CSF1) in injured sensory neurons. The CSF1 is transported to the spinal cord where it targets the microglial CSF1 receptor (CSF1R). Cre-mediated sensory neuron deletion of Csf1 completely prevented nerve injury-induced mechanical hypersensitivity and reduced microglia activation and proliferation. In contrast, intrathecal injection of CSF1 induces mechanical hypersensitivity and microglial proliferation. Nerve injury also upregulated CSF1 in motoneurons, where it is required for ventral horn microglial activation and proliferation. Downstream of CSF1R, we found that the microglial membrane adapter protein DAP12 is required for both nerve injury- and intrathecal CSF1-induced upregulation of pain-related microglial genes and the ensuing pain, but not for microglia proliferation. Thus, both CSF1 and DAP12 are potential targets for the pharmacotherapy of neuropathic pain.

It is generally acknowledged that humans display highly variable sensitivity to pain, including variable responses to identical injuries or pathologies. The possible contribution of genetic factors has, however, been largely overlooked. An emerging rodent literature documents the importance of genotype in mediating basal nociceptive sensitivity, in establishing a predisposition to neuropathic pain following neural injury, and in determining sensitivity to pharmacological agents and endogenous antinociception. One clear finding from these studies is that the effect of genotype is at least partially specific to the nociceptive assay being considered. In this report we begin to systematically describe and characterize genetic variability of nociception in a mammalian species, Mus musculus. We tested 11 readily-available inbred mouse strains (129/J, A/J, AKR/J, BALB/cJ, C3H/HeJ, C57BL/6J, C58/J, CBA/J, DBA/2J, RIIIS/J and SM/J) using 12 common measures of nociception. These included assays for thermal nociception (hot plate, Hargreaves' test, tail withdrawal), mechanical nociception (von Frey filaments), chemical nociception (abdominal constriction, carrageenan, formalin), and neuropathic pain (autotomy, Chung model peripheral nerve injury). We demonstrate the existence of clear strain differences in each assay, with 1.2 to 54-fold ranges of sensitivity. All nociceptive assays display moderate-to-high heritability (h2 = 0.30-0.76) and mediation by a limited number of apparent genetic loci. Data comparing inbred strains have considerable utility as a tool for understanding the genetics of nociception, and a particular relevance to transgenic studies.

In humans, trauma to a peripheral nerve may be followed by chronic pain syndromes which are only relieved by blockade of the effects of sympathetic impulse traffic. It is presumed that, after the lesion, noradrenaline released by activity of sympathetic postganglionic axons excites primary afferent neurons by activating alpha-adrenoceptors, generating signals that enter the 'pain pathways' of the central nervous system. The site of coupling is unclear. In some patients local anaesthesia of the relevant peripheral nerve does not alleviate pain, implying that ectopic impulses arise either within the central nervous system, or in proximal parts of the primary afferent neurons. In experimentally lesioned rats, activity can originate within the dorsal root ganglia. Here we report that, after sciatic nerve ligation, noradrenergic perivascular axons in rats sprout into dorsal root ganglia and form basket-like structures around large-diameter axotomized sensory neurons; sympathetic stimulation can activate such neurons repetitively. These unusual connections provide a possible origin for abnormal discharge following peripheral nerve damage. Further, in contrast to the sprouting of intact nerve terminals into nearby denervated effector tissues in skin, muscle, sympathetic ganglia and sweat glands, the axons sprout into a target which has not been partially denervated.

Single units were recorded in dorsal roots or in the sciatic nerve of anaesthetised rats. It was shown by making sections, by stimulation and by collision that some ongoing nerve impulses were originating from the dorsal root ganglia and not from the central or peripheral ends of the axons. In a sample of 2731 intact or acutely sectioned myelinated sensory fibres, 4.75% +/- 3.7% contained impulses generated within the dorsal root ganglia. In 2555 axons sectioned in the periphery 2-109 days before, this percentage rose to 8.6% +/- 4.8%. There was a considerable variation between animals; 0-14% in intact and acutely sectioned nerves and 1-21% in chronically sectioned nerves. The conduction velocity of the active fibres did not differ significantly from the conduction velocity of unselected fibres. The common pattern of ongoing activity from the ganglion was irregular and with a low frequency (about 4 Hz) in contrast to the pattern of impulses originating in a neuroma which usually have a higher frequency with regular intervals. Slight mechanical pressure on the dorsal root ganglion increased the frequency of impulses. Unmyelinated fibres were also found to contain impulses originating in the dorsal root ganglion. In intact or acutely sectioned unmyelinated axons, the percentage of active fibres 4.4% +/- 3.5% was approximately the same as in myelinated fibres but there were no signs of an increase following chronic section. Fine filament dissection of dorsal roots and of peripheral nerves and collision experiments showed that impulses originating in dorsal root ganglia were propagated both orthodromically into the root and antidromically into the peripheral nerve. It was also shown that the same axon could contain two different alternating sites of origin of nerve impulses: one in the neuroma or sensory ending and one in the ganglion. These observations suggest that the dorsal root ganglion with its ongoing activity and mechanical sensitivity could be a source of pain producing impulses and could particularly contribute to pain in those conditions of peripheral nerve damage where pain persists after peripheral anaesthesia or where vertebral manipulation is painful.

We examined the relation between ectopic afferent firing and tactile allodynia in the Chung model of neuropathic pain. Transection of the L5 spinal nerve in rats triggered a sharp, four- to six-fold increase in the spontaneous ectopic discharge recorded in vivo in sensory axons in the ipsilateral L5 dorsal root (DR). The increase, which was not yet apparent 16 h postoperatively, was complete by 24 h. This indicates rapid modification of the electrical properties of the neurons. Only A-neurons, primarily rapidly conducting A-neurons, contributed to the discharge. No spontaneously active C-neurons were encountered. Tactile allodynia in hindlimb skin emerged during precisely the same time window after spinal nerve section as the ectopia, suggesting that ectopic activity in injured myelinated afferents can trigger central sensitization, the mechanism believed to be responsible for tactile allodynia in the Chung model. Most of the spike activity originated in the somata of axotomized DRG neurons; the spinal nerve end neuroma accounted for only a quarter of the overall ectopic barrage. Intracellular recordings from afferent neuron somata in excised DRGs in vitro revealed changes in excitability that closely paralleled those seen in the DR axon recordings in vivo. Corresponding changes in biophysical characteristics of the axotomized neurons were catalogued. Axotomy carried out at a distance from the DRG, in the mid-portion of the sciatic nerve, also triggered increased afferent excitability. However, this increase occurred at a later time following axotomy, and the relative contribution of DRG neuronal somata, as opposed to neuroma endings, was smaller. Axotomy triggers a wide variety of changes in the neurochemistry and physiology of primary afferent neurons. Investigators studying DRG neurons in culture need to be alert to the rapidity with which axotomy, an inevitable consequence of DRG excision and dissociation, alters key properties of these neurons. Our identification of a specific population of neurons whose firing properties change suddenly and synchronously following axotomy, and whose activity is associated with tactile allodynia, provides a powerful vehicle for defining the specific cascade of cellular and molecular events that underlie neuropathic pain.

(1) When hindlimb peripheral nerves are cut across in rats and mice, there is a tendency for the animal to attack the anaesthetic limb. We have called this attack "autotomy". In this paper we describe the time course and degree of autotomy following various types of nerve injury. (2) Four different types of lesion were applied to the sciatic nerve of rats. The most serious autotomy was produced by section of the nerve and encapsulation of its cut end in a polythene tube. Section followed by immediate resuturing also produced serious autotomy. Simple ligation of the nerve end was followed by less autotomy than encapsulation or cut and resuture. A crush lesion caused only minimal attack. (3) Section of the saphenous branch of the femoral nerve produced no autotomy. However, if the saphenous and sciatic nerves were ligated at the same time so that the entire foot became anaesthetic there was a great increase of autotomy over that seen when the sciatic nerve alone was ligated. This increase with the double lesion occurred even if the saphenous nerve was ligated more than 100 days after the sciatic nerve had been cut. (4) Mice showed autotomy very similar to that seen in rats but the onset was somewhat faster. (5) Reasons are given to propose that autotomy is triggered by an abnormal afferent barrage generated in the cut end of the nerve. Autotomy from peripheral nerve lesions is a different phenomenon from that seen after dorsal root section. Autotomy occurs under conditions which produce anaesthesia dolorosa in man. This simple model may be suitable for studies of the prevention of irritations originating from chronic lesions of peripheral nerves.

Ectopic discharge in axotomized dorsal root ganglion neurons is a key driver of neuropathic pain. However, the bulk of this activity is generated and carried centrally in large diameter myelinated Abeta afferents, a cell type that normally signals touch and vibration sense. Evidence is considered suggesting that following axotomy, Abeta afferents undergo a change in their electrical characteristics and also in the neurotransmitter complement that they express. This dual phenotypic switching renders them capable of (1) directly driving postsynaptic pain signaling pathways in the spinal cord, and (2) triggering and maintaining central sensitization.