I am a
Home I AM A Search Login

Turning Pluripotent Stem Cells Into a New Population of Sensory Neurons

Researchers use forced transcription method to create a homogenous population of cells, present in humans but not in mice, which detect cold temperature and mechanical force

by Jamie K. Moy


7 May 2020


PRF News

NickollsFeatured

Researchers use forced transcription method to create a homogenous population of cells, present in humans but not in mice, which detect cold temperature and mechanical force

A number of pain labs are now using human tissue, primarily dorsal root ganglia (DRG), to learn more about species differences in pain between animals and people, and to improve translation. Researchers are also taking a complementary approach to overcoming the species divide by putting human induced pluripotent stem cells (iPSCs) to the test, taking these immature cells derived from human tissue or blood and turning them into sensory neurons in vitro. However, doing so is a long, tedious, and arduous task, and it is particularly difficult to obtain a high efficiency of differentiated cells.

 

Now, by adapting a previously reported transcriptional programming method, researchers led by Carsten Bönnemann and Alexander Chesler, National Institutes of Health (NIH), Bethesda, US, differentiate human iPSCs into a homogenous culture of sensory neurons that respond to both cold temperature and mechanical stimuli. They further show that a similar cell population, while not present in mice, exists in humans. They were also able to apply their method to model a rare human sensory disorder in which patients with inactivating mutations in the mechanosensitive ion channel PIEZO2 have deficits in touch and proprioception.

 

“It’s a very interesting paper that uses forced transcription to generate a homogenous population of sensory neurons from iPSCs,” said Jaehoon Shim, a postdoctoral associate in Clifford Woolf’s lab at Boston Children’s Hospital, US. “It’s a good starting point, and I hope that by using the same kind of approach we can generate additional homogenous subpopulations of neurons.” Shim works with iPSCs to learn more about cortical and sensory neurons but was not involved in the new study.

 

This research appeared January 21, 2020, in Cell Reports.

 

Mice and people are different – go figure

The group, including first author Alec Nickolls, originally set out to see if they could improve upon a forced transcription method reported in 2015 (Blanchard et al., 2015Wainger et al., 2015, and PRF related news story). That previous study used two different transcription factors, NGN2 and BRN3A, both important in sensory neuronal maturation, to generate a mixed culture of sensory neurons from mouse embryonic fibroblasts.

 

So in the new study, the group used doxycycline, a compound that can be used to turn on transcription, for 14 days to forcibly induce equal amounts of NGN2 and BRN3A in human iPSCs. By day 21, the iPSCs matured into induced sensory neurons (iSNs) expressing specific neuronal markers such as neurofilament 200 (associated with myelinated fibers), peripherin (which is highly expressed in A-delta and C fibers), and NeuN (a neuronal nuclear marker). Interestingly, with regard to tyrosine receptor kinases (Trk), whose expression in sensory neurons differs according to cell function, only TrkB, which is expressed by gentle touch neurons, was highly abundant, and not TrkA (expressed by temperature-sensing neurons) or TrkC (expressed by neurons that detect body position). Overall, the cells lacked most genes linked to pain, itch, and temperature sensation.

 

RNA sequencing and in-situ hybridization experiments performed on the iSNs further revealed co-expression of PIEZO2 and the cold-sensing ion channel TRPM8, a cell population not found in mice, which express either TRPM8 or PIEZO2 in sensory ganglia but not both. This unforeseen result led the authors to suspect that perhaps these cells were immature, or that the result was an experimental artifact. The species difference begged a further experiment.

 

“We were working on the cells for a couple of years before we decided to actually get some human donor tissue to see if these cells exist in vivo,” said Nickolls. Lo and behold, in-situ mRNA hybridization experiments revealed that, among human adult DRG neurons that expressed PIEZO2 and/or TRPM8, 27% were PIEZO2+/TRPM8+ double-positive neurons.

 

“Go figure that a mouse would turn out to be different than a human,” joked Chesler.

 

The authors next tested whether the iSNs could detect cold and mechanical stimuli, as would be suggested by the presence of TRPM8 and PIEZO2 messenger RNA, using calcium imaging as an indirect measure of neuronal activity. The iSNs showed increased calcium activity in response to cold temperature (4°C), which was reduced with a TRPM8 blocker. The iSNs also responded to menthol, a natural ligand of TRPM8. The cells were also able to detect mechanical force, based on findings from cell membrane indentation studies. Little response was seen to TRP channel ligands such as capsaicin (a TRPV1 ligand) or allyl isothiocyanate (a TRPA1 ligand). In sum, the findings showed that both PIEZO2 and TRPM8 were functional in the iSNs.

 

Does skipping steps matter?

One problem in using a forced transcription method to create mature sensory neurons, in comparison to the more common method of using small-molecule inhibitors to do so, is that it skips steps. That is, when using small-molecule inhibitors, the iPSCs go through the natural stages of embryonic development. That includes the neural crest (NC) progenitor stage, after which the cells terminally differentiate. The forced transcription method used by the study authors misses this stage, which could have explained the unusual gene expression patterns of the iSNs.

 

To tackle this issue, the group first differentiated the human iPSCs into NC cells. Then they used their BRN3A and NGN2 forced transcription protocol for either 14 days (referred to as NC-iSN1 cells), or only 24 hours (NC-iSN2).

 

The investigators found that both the NC-iSN1 and NC-iSN2 neurons resembled mature peripheral sensory neurons. But the NC-iSN2s, in which NGN2-BRN3A expression was only briefly induced, showed larger cell soma sizes and expressed only PIEZO2, in comparison to the NC-iSN1 cells, which mimicked the profile of the original iSNs derived from the human iPSCs, co-expressing both TRPM8 and PIEZO2. Additional RNA sequencing experiments revealed significant transcriptomic differences between the NC-iSN1 and NC-iSN2 populations, with the NC-iSN1 cells once again appearing just like the original iSNs.

 

“Disentangling the developmental events that allow one cell type to turn into two different cell types would be fun to do in this setting,” Nickolls said.

 

The dream and the promise

One of the best advantages of using iPSC-derived sensory neurons, according to the researchers, is the opportunity it offers to understand what goes awry in people with diseases.

 

“Model systems are great, animal systems are interesting, but we’re always interested in figuring out what happens in the human situation. Can we use human-derived materials to model the disease and also look for what goes wrong, and what can be improved in the human situation?” asked Bönnemann.

 

The current study provided an encouraging answer to that question, with the researchers looking to patients they had previously reported to have impairments in proprioception and gentle touch due to loss-of-function mutations in PIEZO2 (Chesler et al., 2016PRF related news story). The team reprogrammed iPSCs from the two patients, as well as from an unaffected sibling and other control lines, differentiating the cells into iSNs. The iSNs from the patients were mechanically insensitive, showing little current in response to cell soma indentation with a pipette. In addition, using CRISPR-Cas9 gene editing to genetically repair the PIEZO2 loss-of-function mutations in the iSNs from one of the patients, the group was able to rescue the mechanically activated currents. The patient iSNs otherwise resembled the authors’ newly created iSN cell line, including the ability to respond to menthol.

 

The researchers believe the iPSC-derived sensory neuron approach has vast potential to help patients.

 

“We have one subject who came through NIH and has a lot of phenotypic similarities to the PIEZO2 deficiency subjects, but she doesn’t carry any exome mutations in PIEZO2,” said Chesler, referring to the coding portions of the genome. “But Alec [first study author] can now make neurons, and we can actually study those neurons and see if we can find any deficits that exome sequencing didn’t reveal. We could make new discoveries that are really patient focused.”

 

Chesler said the ideal is to “have a bunch of cells in the freezer, and you hit them with your doxycycline and you get a full panel of human sensory neurons in a dish – but each dish has just thermo-nociceptors, or A-delta nociceptors; you get them all, like an array of human DRG in dishes, scale them up, and then you can do biochemistry, physiology, and drug screening directly on these human cells. You could have the full repertoire of sensory fibers, and that’s the dream and promise of this technology.”

 

Jamie Moy is a postdoctoral pain researcher at the University of Pittsburgh, US.

 

Featured image credit: Ethan Tyler and Alan Hoofring, NIH Medical Arts Branch.

Share This