Peripheral sensory neurons are classified based on characteristics such as anatomical structure, physiological responses to stimuli, and expression of gene markers. In a further attempt at classification, a new study uses RNA sequencing to identify transcriptome profiles of dorsal root ganglion (DRG) neuronal subtypes; these profiles predicted ion channel contributions to the intrinsic electrophysiological properties of the neurons.
Led by David Ginty, Harvard Medical School, Boston, US, researchers performed RNA sequencing of eight well-known, physiologically distinct subtypes of DRG neurons in mice. The transcriptome data revealed differences in ion channel expression among the eight subtypes. Guided by these findings, the researchers performed electrophysiological recordings in dissociated mouse DRG neurons and found that the expression of different potassium (Kv) channels shaped specific firing patterns of the different DRG subtypes.
“This is an important study in the pain field,” said Patrik Ernfors, Karolinska Institute, Stockholm, Sweden, who studies somatosensation but was not involved in the new research. “Peripheral sensory neurons have previously been characterized and classified based on their molecular characteristics, but this study is the first effort to link molecular characteristics of DRG neuronal subtypes to physiological properties.”
The study appeared August 21, 2019, in Neuron.
RNA sequencing takes center stage
A recent explosion of RNA sequencing approaches in the pain field has led to a new understanding of the molecular diversity of cell populations in the somatosensory system, from the skin to the brain. In the pain field, Ernfors led the first effort toward single cell RNA sequencing of DRG neurons in 2015 (Usoskin et al., 2015).
Since then, several labs have performed similar sequencing of cell types in the skin, trigeminal ganglia, and the spinal cord (Joost et al., 2016Lopes et al., 2017Häring et al., 2018). The transcriptome databases that result from such efforts are proving to be valuable toolboxes for other research groups to correlate gene expression profiles with neuronal function.
Ginty has studied how the anatomical structure of receptors in the skin determines the function of DRG neurons and the sense of touch (Abraira and Ginty 2013). He and his colleagues’ new paper focuses on the transcriptome profiles of diverse DRG neuronal subtypes.
Previous classification efforts using similar methods have taken unbiased approaches to determine gene expression differences in sensory neuron populations. The novelty in the new study is that the authors focus on how gene expression variability could determine functional differences in well-known DRG neuronal populations, in particular with regard to the cells’ electrophysiological properties.
“This paper provides a resource of transcriptome profiles as well as distinct in vitro physiological properties of eight functionally and genetically defined somatosensory neuron subtypes,” said co-first author Yang Zheng, who has since left the Ginty lab for a postdoc position in Elizabeth Hong’s group at Caltech, Pasadena, US.
Common and unique characteristics of DRG neurons
The researchers began by using available genetic mouse lines that separately label eight subtypes of DRG neurons, including the major classes of nociceptors and touch receptors that innervate the skin. This enabled classification of around 85 percent of all DRG neurons, including MrgD+ non-peptidergic C-fibers and TrkB+ Aβ-fiber low-threshold mechanoreceptors (LTMRs).
In vitro electrophysiological recordings revealed distinct electrical stimulus-evoked signatures for each of the eight populations. C-LTMRs, for example, displayed delayed firing patterns that were distinct from the strong adaptation of Aδ-LTMRs.
The authors then performed RNA sequencing, which resulted in unique transcriptome profiles for the eight DRG neuronal subtypes. Predictably, these subtypes shared common molecular features, especially when considering anatomically similar fiber populations, as well as unique combinations of upregulated and downregulated genes. The researchers also observed uniquely enriched genes in the different subtypes. For example, one gene, Calbindin, was upregulated in Aβ rapidly adapting (RA)-LTMRs and could be used as a novel molecular marker of these touch receptors.
“The overall neuronal subtypes seem to be the same, with similar markers, as described in previous work,” Ernfors said, referring to research from his group and colleagues (Usoskin et al., 2015Zeisel et al., 2018). “One difference is that, due to the high quality of the RNA sequencing, the authors found a [new] subcluster among the Aβ touch receptors,” said Ernfors, referring to the Calbindin-expressing neurons.
Differential potassium channel subtype expression
The next aim of the researchers, including co-first author Pin Liu, was to investigate if the molecular composition of the DRG neurons could account for their different electrophysiological signatures. The RNA sequencing data would reveal unique expression patterns of ion channels in the eight DRG neuronal subtypes.
The authors focused on voltage-gated potassium (Kv) channel expression because of the genetic diversity of these channels and their functional importance in setting the intrinsic excitability of neurons. The eight DRG neuronal populations showed remarkably different Kv channel subtype expression patterns. Aβ-fiber touch receptors, for example, showed high expression of Kv1.1 and Kv1.2, while C-LTMRs had high expression of Kv4.3.
The authors then used this information to test whether blocking these different Kv channels would affect action potential firing properties. This approach revealed that Kv1.1 and Kv1.2 channels enriched in Aβ-fiber touch receptors were important for determining non-repetitive versus repetitive firing patterns. Specifically, blocking Kv1 channels increased repetitive spiking in slowly-adapting (SA) Aβ-LTMRs, but not in Aβ RA-LTMRs. Kv4.3 expression in C-LTMRs was found to be crucial for the delayed firing signatures of this neuronal subtype.
These data were complemented by computational modeling of the contributions of Kv channels to DRG neuron current properties. The researchers found that Kv4 delayed the onset of action potentials in a model neuron because of rapid Kv4 current activation at subthreshold voltages. This feature of Kv4 channels is particularly effective at delaying firing patterns in C-LTMRs because of a lack of Nav1.6, which is a channel, including in Aβ-fibers, that would normally override the effect of opening Kv4 and cause rapid depolarization.
“In general, computational models are great for understanding convoluted systems that are hard to conceptualize,” said Zheng. “For example, each cell expresses many different ion channels that work in harmony to achieve certain firing properties. With different channel kinetics and expression levels, it may be difficult to envision how this changes the firing properties.”
Kv channels as analgesic targets?
One important next step is to investigate how gene regulation in DRG subtypes may be linked to pain conditions.
“It would be interesting to generate DRG neuron subtype-specific transcriptome profiles in different pain models. With the transcriptome profile in this study serving as a reference, genes that are differentially regulated in peripheral neurons after pain induction may be revealed,” said Zheng.
Another exciting prospect is that targeting particular Kv channels to modulate the firing of specific DRG neurons could potentially be used to treat chronic pain. DRG neuronal hyperexcitability is a hallmark of neuropathic pain, and Kv channels have been considered as attractive analgesic targets for decades (Tsantoulas and McMahon, 2014).
“With a bit of luck, there is a [Kv] channel that is selective for sensory neurons and used by the particular neuron types that become sensitized during chronic pain,” Ernfors said.
In the meantime, many questions remain. For example, it is not fully known how the firing activity of each of the eight (or more) DRG neuronal subtypes changes during chronic pain. It is also unclear how modulating the activity of specific Kv channels in these DRG populations would alter nociceptive signaling, especially if the affected neuron participates in an inhibitory circuit.
Fred Schwaller, PhD, is a postdoctoral researcher at the Max Delbrück Center for Molecular Medicine, Berlin, Germany.
Image credit: Patrick Guenette/123RF Stock Photo.