August 14, 2020—(BRONX, NY)—Chance plays a big part in life—including in how the human body functions. Randomness appears to be inherent in many biological systems and plays an important role even in processes considered tightly controlled or highly coordinated, such as turning genes on and off.

Ulrich G. Steidl, M.D., Ph.D.Faculty ProfileResearch ProfileExpert Profile

In the July 16 issue of Nature, Ulrich Steidl, M.D., Ph.D., and his Einstein colleagues investigated the role of gene activity in differentiating stem cells into more mature cells. They focused on hematopoiesis, the process in which blood-forming stem cells differentiate into the various cells that make up the blood and that requires turning certain genes on and off.

“A fundamental question in stem-cell biology is how stem cells can robustly produce mature, differentiated cells while using intrinsically “noisy” processes such as gene activation—ones that occur somewhat randomly or haphazardly at a molecular level,” said Dr. Steidl, professor of cell biology and of medicine and the Diane and Arthur B. Belfer Faculty Scholar in Cancer Research at Einstein and the associate chair for translational research in oncology at Montefiore Health System. “Specifically, we wanted to study the activation behavior of the genes that code for transcription factors, which are critical determinants of the cell fate choices that stem cells make.”

Genes are “switched on” when specialized proteins called transcription factors attach to DNA close to genes and trigger transcription, in which a gene’s DNA code is copied (“transcribed”) onto messenger RNA (mRNA) molecules. The mRNAs act as blueprints for making the prescribed proteins needed for the cell’s structure or function. This process, known as gene expression, is not continuous but is in part intermittent, in that individual genes alternate rapidly between actively “firing” (i.e. transcribing mRNA) and being “silent.”

Fishing for Answers

The researchers’ efforts were aided by single-molecule RNA fluorescent in situ hybridization (smRNA-FISH). This powerful tool was developed many years ago by study co-author Robert Singer, Ph.D., professor of anatomy and structural biology, of cell biology, and in the  Dominick P. Purpura Department of Neuroscience and the Harold and Muriel Block Chair in Anatomy & Structural Biology at Einstein. smRNA-FISH can detect and accurately measure the abundance of single mRNA molecules in individual living cells. Dr. Steidl and his colleagues, led by Justin C. Wheat, an M.D./Ph.D. student in Einstein’s Medical Scientist Training Program, used smRNA-FISH to detect and measure the number of mRNA molecules derived from seven transcription-factor genes within individual blood-forming stem cells and their direct descendants called progenitor cells. These progenitor cells further mature into all of the blood’s specialized cells.

The findings using smRNA-FISH differed significantly from the same analysis using single-cell RNA sequencing (scRNA-seq), currently the most widely used method for analyzing gene expression in single cells. “SmRNA-FISH was far more sensitive than scRNA-seq at providing quantitative measurements of mRNA during hematopoiesis,” said Dr. Steidl. “We found that transcription driven by “bursts” of stochastic, or random, activity was much more important during stem cell differentiation than had been thought.”

We found that stem cells are not locked into specific cell fates early, but rather follow a highly pliable process in which the intrinsic randomness of bursts of transcription—rather than impeding things—instead lead to the successful production of mature blood cells while at the same time maintains a viable pool of uncommitted stem cells.

Ulrich Steidl, M.D., Ph.D.

Confirming Randomness

The researchers next investigated a key network involving three transcription-factor genes—PU.1, Gata1, and Gata2—which are critical for differentiating blood-forming stem cells and progenitor cells into red cells andwhite cells. The gene PU.1 appears to impede the function of genes Gata1 and Gata2, and the differentiation of stem cells was thought to result from the direct interaction of their battling transcription factors. Yet recent scRNA-seq studies have either failed to detect progenitor cells that co-express PU.1 and Gata1 or have detected co-expression in only a small minority of them, thus challenging the previous model.

“Using smRNA-FISH, we showed that—contrary to all recent scRNA-seq papers—the vast majority of hematopoietic progenitor cells co-express mRNAs for those three antagonistic transcription factors,” Dr. Steidl said. “Importantly, this co-expression involves relatively few mRNA molecules—between 10 and 50 per cell for each gene—and derives from infrequent, stochastic bursts of transcription. For this transcriptional network, we found that stochastic bursts alone were sufficient for enabling stem and progenitor cells to evolve into more differentiated blood cells.”  

In addition, the researchers used time-lapse microscopy of single stem cells as they divided and produced progeny, and determined their kinship with each other. This information was combined with analysis of the transcriptional states of these different cell types using smRNA-FISH, which led to some important insights.

“Prior studies,” said Dr. Steidl, “have reached two conclusions: that hematopoiesis is a highly precisely coordinated process in which a sequence of transcriptional events causes stem cells to progressively differentiate into blood cells; and recent models based on scRNA-seq data have suggested that the majority of stem cells are already transcriptionally primed to become certain cell types very early during this process.  Furthermore, those models assumed that inexactness of molecular processes in general, and transcriptional noise in particular, impeded such precise ‘deterministic’ regulation.

“But our use of smRNA-FISH combined with kinship analysis of stem cells reveals the opposite,” Dr. Steidl said. “We found that stem cells are not locked into specific cell fates early, but rather follow a highly pliable process in which the intrinsic randomness of bursts of transcription—rather than impeding things—instead lead to the successful production of mature blood cells while at the same time maintains a viable pool of uncommitted stem cells. We may well find that stochastic gene expression is a critical component involved in governing other key biological processes as well.”

The paper is titled “Single-molecule imaging of transcription dynamics in somatic stem cells.” Other authors of the paper were Yehonatan Sella, Ph.D., M.D./Ph.D. student Michael Willcockson, Arthur I. Skoultchi, Ph.D., and Aviv Bergman, Ph.D.