One part of the proposal relates to the identification of the cellular basis of somatosensation and pain (Objective 1 and 2). For this purpose, we identified and classified sensory cell types using single-cell RNA sequencing (scRNA-seq) in rodents and non-human primates. The dorsal horn of the spinal cord is critical to processing distinct modalities of noxious and innocuous sensation, but little was known of the neuronal subtypes involved. (doi: 10.1038/s41593-018-0141-1). We used scRNA-seq to classify sensory neurons in the mouse dorsal horn and identified 15 inhibitory and 15 excitatory subtypes of neurons. We validated the classification and existence of all the different neuronal types in vivo. We also developed a computational strategy to map back into a reference spinal cord the identified neuronal types into their correct anatomical location. Neuron types, when combined, define a multilayered organization with like neurons layered together. This led to the discovery that the spinal cord is built by the layering of the different types of excitatory neurons. By this, we evidenced that the spinal cord has a much more refined organization than previously described.
In Bartesaghi et al., (doi: 10.1016/j.celrep.2019.02.098) we describe how pain transducing sensory neurons form during development. In Kupari et al., 2019 (doi: 10.1016/j.celrep.2019.04.096.) we showed that somatosensation is conserved regardless of axial position: After scRNA-seq of the nodose ganglion complex which includes both somatosensory jugular neurons and viscerosensory nodose neurons, we used machine learning to show that the strategy of somatosensation is shared between jugular ganglion and dorsal root ganglion (DRG). We also showed that viscerosensation is governed by an entirely different strategy with no resemblance to somatosensation.
In addition to the above, we scRNA sequenced the non-primate (Macaca mulatta) DRG neurons , published in Kupari et al., 2021 (doi:10.1038/s41467-021-21725-z) and successfully classified non-human primate sensory neurons based on their transcriptome. We identified nine neuronal types and used machine learning to expose an overall inter-species conserved strategy and shared taxonomy for nociception, highlighting the importance of incorporating primate knowledge for meaningful translation of discoveries in mouse models. We also explored how each cell type in the macaque DRGs contributes to heritability of painful phenotypes in humans. To do so, we employed a large cohort (UK Biobank), in which information about eight chronic pain sites is available. Among the nine neuron types identified in the macaque DRGs, we found that common-variants associated with chronic pain mapped to two distinct neuron types due to sets of genes that were specifically expressed in each of these cell types. Interestingly, seven of the nine neuron types did not contribute significantly in any pain sites. This shows that each of the two pain neuron types carries different heritability to chronic pain and thus, that the vulnerability to chronic pain is cell type specific, with different genes and underlying mechanism in the different cell types.
For Objective 2, we established the role of mouse cell types for acute and chronic pain. Peng et al., (doi: 10.1126/science.aam7671) showed that both basal mechanical and neuropathic pain are controlled by the microRNA-183-cluster in mice. This single cluster controls more than 80% of neuropathic pain-regulated genes and scales basal mechanical sensitivity and mechanical allodynia .Interestingly, while basal sensitivity was found to be controlled by nociceptors, allodynia was found to involve TrkB+AD light-touch mechanoreceptors, which are normally involved in sensing hair deflection and pressure. However, we report in this paper that these neurons change function from touch sensation to pain during neuropathy. This is due to changes in connectivity in the spinal cord.
Objective 2 mainly centred around developing a new strategy that allows a systematic mapping of all cell types involved in acute and chronic pain. This strategy is based on genetic
tagging of functionally active neurons by means of immediate early genes. The technology to do this has been technically very challenging and because it was highly wet-lab intensive, thus this objective was the one most affected by Covid. Nevertheless, we have successfully developed and validated this technology. We are preparing a manuscript now, where we use genetic capture, activity manipulation and scRNA seq, to identify distinct neural ensembles in the spinal cord encoding mechanical and heat pain. Re-activation or silencing these assembles potentiated or stopped, respectively, affective but not reflex behavior without altering pain behavior to cross stimuli modality. Within ensembles, excitatory neurons encoded quality and a single molecular type of polymodal inhibitory neuron type gated affective pain regardless of modality. Following peripheral nerve injury, a marked circuit-wide molecular perturbation was caused by inflammation and the ensembles failed to respect noxious information quality and to resolve allodynia and hypersensitivity in mice. These results reveal the existence of a spinal representation of cutaneous noxious heat and mechanical stimuli which forms the neural basis of the affective qualities of acute pain perception.
In objective 3 we proposed to determine if specialized cutaneous glia exists and form a sensory organ in the skin that contributes to pain sensation. In Abdo et al., (doi: 10.1126/science.aax6452) we describe a specialized cutaneous glial cell type with extensive processes forming a mesh-like network in the subepidermal border of the skin that conveys noxious thermal and mechanical sensitivity. We demonstrated a direct excitatory functional connection to sensory neurons and thus, provided evidence of a previously unknown organ that has an essential physiological role in sensing noxious stimuli. We showed that these glial cells are inherently mechanosensitive and transmit nociceptive information to the nerve.
All results and generated data have been or will be disseminated by scientific publications, lectures at conferences, seminars and invited presentations at universities in different countries. Non-scientific audience has been accessed through interviews with reporters. The raw data and annotated data have been deposited in open access repositories.
