(648b) ­­Mathematical Modeling Shows That the BMP Signaling Spatiotemporally Regulates Drosophila Ovarian Germline Stem Cell Differentiation through a Negative Feedback Loop.

Bone-Morphogenetic Proteins (BMP) are a class of proteins that derive their name from ubiquitous expression and importance as regulators throughout the body. They are known to regulate growth, development, and adult homeostasis. They play key roles in processes such as bone formation, muscle growth, heart development, and stem cell decisions. An understanding of how BMP regulates processes such as stem cell decisions can help develop new treatments for diseases by potentially directing stem cells to become specific cell types. The BMP pathway is conserved across many species, including Drosophila melanogaster, the common fruit fly. Flies make an excellent model system to study these regulatory pathways due to the availability of genetic manipulation tools, rapid generation time, and accessibility of quantitative imaging.

The Drosophila ovarian germline is a well-characterized model system to study the dynamics of cellular decision-making. The asymmetric division of germline stem cells (GSCs) to form two daughter cells­ — a self-renewed GSC and a differentiated cystoblast (CB) — is the first symmetry-breaking event in the germline. The germline stem cells reside in a microenvironment ("niche”) that comprises terminal filament cells (TF), cap cells (CC), and anterior escort cells (AECs). The CCs reside at the anterior tip of the germline and secrete Dpp (a Drosophila homolog of the growth factor BMP2/4). The differentiated daughters of the GSCs, the CBs lie at the posterior end outside the niche. This forms a short-ranged extra-cellular Dpp gradient from the GSCs to the CBs which remain connected until late telophase.

Within the niche, Dpp forms a heteromeric complex with the serine/threonine transmembrane receptors thus activating them. This activated ligand-receptor complex phosphorylates Mothers against dpp (Mad), a homolog of SMAD1, which forms a hetero-trimeric complex with a transcriptional activator Medea (Med), a homolog of SMAD4. The pMad/Med complex translocates to the nucleus and regulates the expression of genes such as Daughters against dpp (dad), a homolog of SMAD6/7, and represses the expression of Fused (ortholog of STK36), and hence bam, the differentiating factor. Thus, Dpp signal transduction promotes GSC maintenance.

In the GSCs, Dad inhibits Dpp signal transduction by deactivating the receptors, forming a negative feedback loop (NFL), while Fused ubiquitinates the signaling complex in the cytoplasm, forming a positive feedback loop (PFL). However, Dpp concentration diminishes one-cell diameter away in the differentiated progeny. In the CBs, the PFL favors high Fused/low pMad, which enables bam expression, and hence, differentiation. These regulatory feedback loops operating over a ~10-micron distance are robust to the variability in the levels of Dpp that the GSCs experience. We hypothesized that Dad regulates the extent of Dpp signal transduction in the GSC to enable bam expression in the CBs.

To that end, we developed a biologically-informed deterministic mathematical model. In our multi-compartment model, we track Dpp pathway members as the GSC undergoes growth and division. To recapture the developmental dynamics, we simulate the model for a duration of 12h and allow the mass transfer interface connecting the two cells to shrink during cytokinesis. We verified that our model conforms to the system behavior reported in the literature. This allowed us to investigate the dynamic roles Dad and Fused play in determining cell fate.

We tested our hypothesis in wild-type (wt) and dad^KO backgrounds. We found that Fused (in PFL) makes the system bistable, such that in the wt system CB-like levels of Dpp lead to bam expression allowing differentiation. Whereas in dad^KO, the same Dpp levels lead to bam suppression owing to high pMad/Med concentration. Hence, in dad null mutants GSCs were more likely to divide symmetrically. We then subjected our model to small perturbations in the levels of Dpp and observed the response in pMad/Med levels for wt and dad^KO backgrounds. Our simulations show that Dad and Fused make the GSCs and CBs robust, respectively. We also observed that Dad controls the response time of Fused and hence bam expression, ensuring GSC differentiation is temporally regulated. Thus Dad (in NFL) optimally controls the levels of pMad-Med to enable GSC homeostasis and differentiation to be spatiotemporally regulated in the ovarian germline.

To summarize, we developed a mathematical model to infer the role of two Dpp pathway proteins: Dad (in NFL) and Fused (in PFL), in regulating GSC maintenance and differentiation. We found that Dad regulates the level of Dpp signal transduction to ensure robust asymmetric differentiation, while Fused allows GSCs to attain a bistable cell fate on spatial cues. Hence, Dad and Fused work in tandem to ensure the GSC division is robust throughout the division cycle.