To elucidate the role of the BM microenvironment in AML development (as we proposed in the project-aim1), I examined, in vivo the transcriptional changes induced by AML by isolating the BM niche cell type populations by flow cytometry (endothelial-EC, mesenchymal-MSC, and osteoblasts-OB) in AML intermediate state. Then, I made RNAseq experiments, specifically MARSseq due to the low number of cells. RNAseq results showed differences between controls and leukemic mice, as well as interesting functions in all BM niche cell type populations. Due to the high transcriptional heterogeneity, I continued with unbiased single-cell RNA-seq (SC). To do SC technology, I used the Chromium Single-Cell 3´ solution from 10XGenomix of non-hematopoietic cells and I detected EC, MSC, and OLN among others. The low number of sequenced cells didn´t allow to resolve the complexity of the BM niche. However, at the same time as us, other groups were also performing the same type of analysis with the same problems, so I decided to integrate three datasets separately for two well-defined populations (MSC and EC) to improve the obtained resolution (Figure 1). This analysis provides the most comprehensive atlas of the cellular composition in the mouse BM niche with the identification and characterization of 14 intermediate reversible cell functional states in the EC compartment and 11 cell subsets defining different stages of differentiation in MSC. Furthermore, this deep characterization allows inferring conserved features in humans (Figure 2A), suggesting that the layers of microenvironmental regulation of hematopoiesis may also be shared between species (Figure 2B). To this end, I prospectively isolated EC and MSC-OLN cells from iliac crest BM aspirates from four healthy young adults using scRNA-seq and this data was integrated with the previous mouse BM atlas. Our resource and methodology are a stepping-stone toward a comprehensive cell atlas of the BM microenvironment.
Candidate genes and pathways deregulated at the transcriptomic level (MARSseq) in specific niche cell populations upon leukemia induction were selected for functional validation by means of gain and loss-of-function studies (as we proposed in the project-aim2). I focused on CXCL12, which is upregulated after AML development. I started setting up an ex-vivo model, a 2D culture system. For in vitro experiments, I used two independent shRNAs to induce a pronounced downregulation of Cxcl12 at the transcriptional level as measured by qPCR and a reduction in the soluble Cxcl12 present in the cell supernatant and I observed a cell viability reduction after Cxcl12 silencing (Figure 3A). To verify this data, I also performed MARSseq sequencing of MLL::AF9 cells from the OP9-shRNA-Ctrl and OP9-shRNA-Cxcl12 co-culture systems and these results revealed enrichment in categories associated with cell cycle, replication stress, and apoptosis as well as mitochondrial respiration (Figure 3B). Then, I continued with in vivo validation creating three in vivo experimental models and also the AML induction on these models. In vivo experiments showed that early MSC, but not mature OLN or EC represents a bona fide nurturing niche for AML cells (Figure 3C). As a consequence, I sought to investigate the cellular and molecular mechanisms that explain this dependency through genome-wide transcriptome analysis. Leukemic cells isolated from mice without Cxcl12 expression in MSC showed downregulation of pathways associated with cell adhesion, cell death, or REDOX among others (Figure 3D).
This work demonstrates the critical contribution of the BM microenvironment in the neoplastic process and points toward the different requirements of normal and malignant stem cells for their maintenance. For this reason, it is very important to plan determined measures to exploit and disseminate these SyLeNCe results. All these results are open access and they have been published in iScience and Leukemia journals (internationally recognized peer-reviewed publications). Furthermore, they have also been presented at a national conference (SEHH and FEHH Meeting) or at a smaller level in different seminars at a departmental level in CIMA. In addition, I recorded a video to disseminate generated knowledge through the internet.
