NEW YORK — A single-cell transcriptomic analysis has traced the origin of neuroblastomas back to developing adrenal neuroblasts.
Neuroblastomas are common pediatric cancer that affect the adrenal glands or sympathetic ganglia. Some neuroblastomas are high-risk tumors — often harboring MYCN amplifications, TERT rearrangements, or alterative telomere lengthening — but others are low-risk tumors. Neuroblastomas have been suspected to arise from neural crest-derived precursor cells of sympathetic neurons and adrenal chromaffin cells during development, though that hasn't been fully clear.
By comparing the single-cell transcriptomes of normal, developing adrenal glands and neuroblastomas, researchers led by Frank Westermann from Hopp Children's Cancer Center Heidelberg in Germany found that neuroblastomas transcriptionally resemble normal fetal adrenal neuroblasts. As they reported in Nature Genetics on Thursday, they further noted that certain clinical neuroblastoma phenotypes corresponded with different developmental time points, suggesting that targeting differentiation blocks could be a potential treatment approach.
"Our study resolves the lineage relationships and differentiation dynamics of human adrenal medullary cell types during development," Westermann and his colleagues wrote in their paper
They analyzed samples from 17 fresh-frozen developing human adrenal glands using droplet-based single-nucleus RNA-seq. These samples represented seven developmental timepoints ranging from seven weeks post-conception to 17 weeks post-conception. They clustered the cells and assigned them to major cell types based on the markers they expressed, but focused much of their analysis on adrenal medullary cells such as Schwann cell precursors, chromaffin cells, and neuroblasts.
As development progresses, these cells began to express additional differentiation markers, the researchers noted. They further found that some cells like Schwann cell precursors and chromaffin cells had a higher proliferative capacity than bridge cells — cells falling in the transcriptional space between the three cell types — and neuroblasts.
By comparing these normal developing human adrenal gland cells to cells from 14 neuroblastomas also analyzed by snRNA-seq, the researchers found that the tumors resembled differentiating adrenal neuroblasts.
But they also noticed some differences by tumor type. For instance, MYCN-amplified neuroblastoma cells were most similar to normal neuroblasts from seven or eight weeks post-conception, while lower-risk neuroblastomas included more cells resembling late neuroblasts.
This suggested to the researchers that low-risk tumors might develop from cells further along in the development and differentiation process. They confirmed this finding by projecting single neuroblastoma cells onto diffusion maps of normal adrenal medullary cells to again find neuroblastoma cells mapped to normal neuroblasts and that low-risk tumors were more similar to differentiated neuroblasts and high-risk ones to earlier-state neuroblasts. They additionally found that differentiation markers varied between high-risk and low-risk tumors.
This prompted the researchers to examine whether MYCN — often amplified among high-risk tumors — can suppress differentiation. In an inducible MYCN knock-down model of MYCN-amplified neuroblastomacells, they found that elevated MYCN can induce de-differentiation and activate proliferation.
At the same time, though, activating TFAP2B — a transcription factor that is highly expressed in normal neuroblasts but not in high-risk neuroblastomas — restores differentiation signatures.
This finding could, according to the researchers, be the basis for designing future neuroblastoma treatments.
"Our identification of tumor-related transcriptional changes and molecular mechanisms underlying impaired differentiation may guide future studies on the functional evaluation of candidate genes, refined risk classification and generation of clinically relevant neuroblastoma models," they wrote in their paper. "Moreover, we have provided the framework to evaluate therapeutic concepts that are based on induction of differentiation."