Machine learning assisted real-time deformability cytometry of CD34+ cells allows to identify patients with myelodysplastic syndromes.

Diagnosis of myelodysplastic syndrome (MDS) mainly relies on a manual assessment of the peripheral blood and bone marrow cell morphology. The WHO guidelines suggest a visual screening of 200 to 500 cells which inevitably turns the assessor blind to rare cell populations and leads to low reproducibility. Moreover, the human eye is not suited to detect shifts of cellular properties of entire populations. Hence, quantitative image analysis could improve the accuracy and reproducibility of MDS diagnosis. We used real-time deformability cytometry (RT-DC) to measure bone marrow biopsy samples of MDS patients and age-matched healthy individuals. RT-DC is a high-throughput (1000 cells/s) imaging flow cytometer capable of recording morphological and mechanical properties of single cells. Properties of single cells were quantified using automated image analysis, and machine learning was employed to discover morpho-mechanical patterns in thousands of individual cells that allow to distinguish healthy vs. MDS samples. We found that distribution properties of cell sizes differ between healthy and MDS, with MDS showing a narrower distribution of cell sizes. Furthermore, we found a strong correlation between the mechanical properties of cells and the number of disease-determining mutations, inaccessible with current diagnostic approaches. Hence, machine-learning assisted RT-DC could be a promising tool to automate sample analysis to assist experts during diagnosis or provide a scalable solution for MDS diagnosis to regions lacking sufficient medical experts.

Bone marrow stromal cells from MDS and AML patients show increased adipogenic potential with reduced Delta-like-1 expression.

Myelodysplastic syndromes (MDS) and acute myeloid leukemia (AML) are clonal hematopoietic stem cell disorders with a poor prognosis, especially for elderly patients. Increasing evidence suggests that alterations in the non-hematopoietic microenvironment (bone marrow niche) can contribute to or initiate malignant transformation and promote disease progression. One of the key components of the bone marrow (BM) niche are BM stromal cells (BMSC) that give rise to osteoblasts and adipocytes. It has been shown that the balance between these two cell types plays an important role in the regulation of hematopoiesis. However, data on the number of BMSC and the regulation of their differentiation balance in the context of hematopoietic malignancies is scarce. We established a stringent flow cytometric protocol for the prospective isolation of a CD73+ CD105+ CD271+ BMSC subpopulation from uncultivated cryopreserved BM of MDS and AML patients as well as age-matched healthy donors. BMSC from MDS and AML patients showed a strongly reduced frequency of CFU-F (colony forming unit-fibroblast). Moreover, we found an altered phenotype and reduced replating efficiency upon passaging of BMSC from MDS and AML samples. Expression analysis of genes involved in adipo- and osteogenic differentiation as well as Wnt- and Notch-signalling pathways showed significantly reduced levels of DLK1, an early adipogenic cell fate inhibitor in MDS and AML BMSC. Matching this observation, functional analysis showed significantly increased in vitro adipogenic differentiation potential in BMSC from MDS and AML patients. Overall, our data show BMSC with a reduced CFU-F capacity, and an altered molecular and functional profile from MDS and AML patients in culture, indicating an increased adipogenic lineage potential that is likely to provide a disease-promoting microenvironment.

Direct modulation of the bone marrow mesenchymal stromal cell compartment by azacitidine enhances healthy hematopoiesis.

Mesenchymal stromal cells (MSCs) are crucial components of the bone marrow (BM) microenvironment essential for regulating self-renewal, survival, and differentiation of hematopoietic stem/progenitor cells (HSPCs) in the stem cell niche. MSCs are functionally altered in myelodysplastic syndromes (MDS) and exhibit an altered methylome compared with MSCs from healthy controls, thus contributing to disease progression. To determine whether MSCs are amenable to epigenetic therapy and if this affects their function, we examined growth, differentiation, and HSPC-supporting capacity of ex vivo-expanded MSCs from MDS patients in comparison with age-matched healthy controls after direct treatment in vitro with the hypomethylating agent azacitidine (AZA). Strikingly, we find that AZA exerts a direct effect on healthy as well as MDS-derived MSCs such that they favor support of healthy over malignant clonal HSPC expansion in coculture experiments. RNA-sequencing analyses of MSCs identified stromal networks regulated by AZA. Notably, these comprise distinct molecular pathways crucial for HSPC support, foremost extracellular matrix molecules (including collagens) and interferon pathway components. Our study demonstrates that the hypomethylating agent AZA exerts its antileukemic activity in part through a direct effect on the HSPC-supporting BM niche and provides proof of concept for the therapeutic potential of epigenetic treatment of diseased MSCs. In addition, our comprehensive data set of AZA-sensitive gene networks represents a valuable framework to guide future development of targeted epigenetic niche therapy in myeloid malignancies such as MDS and acute myeloid leukemia.

Azacitidine combined with the selective FLT3 kinase inhibitor crenolanib disrupts stromal protection and inhibits expansion of residual leukemia-initiating cells in FLT3-ITD AML with concurrent epigenetic mutations.

Effectively targeting leukemia-initiating cells (LIC) in FLT3-ITD-mutated acute myeloid leukemia (AML) is crucial for cure. Tyrosine kinase inhibitors (TKI) have limited impact as single agents, failing to eradicate LIC in the bone marrow. Using primary AML samples and a patient-derived xenograft model, we investigated whether combining the FLT3-selective TKI crenolanib with the hypomethylating agent azacitidine (AZA) eliminates FLT3-ITD LIC and whether efficacy of this combination depends on co-existing mutations. Using multiparameter flow cytometry, we show FLT3-ITD occurs within the most primitive Lin-/CD33(+)/CD45dim/CD34+CD38- LIC compartment. Crenolanib alone could not target FLT3-ITD LIC in contact with niche cells while addition of AZA overcame stromal protection resulting in dramatically reduced clonogenic capacity of LIC in vitro and severely impaired engraftment in NSG mice. Strikingly, FLT3-mutated samples harboring TET2 mutations were completely resistant to crenolanib whereas neither NPM1 nor DNMT3A mutations influenced response. Conversely, primary AML LIC harboring either TET2, DNMT3A or NPM1 mutations did not show increased sensitivity to AZA. In summary, resistance of FLT3-ITD LIC to TKI depends on co-existing epigenetic mutations. However, AZA + crenolanib effectively abrogates stromal protection and inhibits survival of FLT3-ITD LIC irrespective of mutations, providing evidence for this combination as a means to suppress residual LIC.

siPools: highly complex but accurately defined siRNA pools eliminate off-target effects.

Short interfering RNAs (siRNAs) are widely used as tool for gene inactivation in basic research and therapeutic applications. One of the major shortcomings of siRNA experiments are sequence-specific off-target effects. Such effects are largely unpredictable because siRNAs can affect partially complementary sequences and function like microRNAs (miRNAs), which inhibit gene expression on mRNA stability or translational levels. Here we demonstrate that novel, enzymatically generated siRNA pools-referred to as siPools-containing up to 60 accurately defined siRNAs eliminate off-target effects. This is achieved by the low concentration of each individual siRNA diluting sequence-specific off-target effects below detection limits. In fact, whole transcriptome analyses reveal that single siRNA transfections can severely affect global gene expression. However, when complex siRNA pools are transfected, almost no transcriptome alterations are observed. Taken together, we present enzymatically produced complex but accurately defined siRNA pools with potent on-target silencing but without detectable off-target effects.