Periodic Reporting for period 3 - ImmuNiche (Identifying spatial determinants of immune cell fate commitment)
Reporting period: 2022-10-01 to 2024-03-31
The bone marrow is the adult tissue where all cells of the blood are generated from hematopoietic stem cells in a process termed hematopoiesis. The differentiation and maturation of blood cells is controlled by a variety of bone marrow-resident cell populations within their microenvironment, the so called niche. While a healthy bone marrow niche supports normal hematopoiesis, perturbations of the niche could give rise to hematological malignancies and vice versa. In particular, a prevailing model of blood cancer postulates niche competition between healthy hematopoietic stem cells and tumor cells. According to this model, tumor cells remodel the microenvironment to gain a competitive advantage over healthy stem cells, which eventually leads to a dominance of tumor cells in the bone marrow niche and the loss of healthy stem cells. Therefore, the bone marrow microenvironment is an important driver of healthy and malignant hematopoiesis and a better understanding of niche interactions underlying blood cell differentiation could reveal novel therapeutic targets for the treatment of blood diseases, such as myelodysplastic syndromes (MDS), representing one of the most frequent blood cell malignancies, commonly giving rise to leukaemia with poor prognosis.
To understand the role of the microenvironment during hematopoiesis, spatial single-cell resolution analysis of bone marrow tissue would be required. To address this challenge within the ImmuNiche project, we are combining state-of-the-art single-cell RNA-sequencing technology with large-scale single-molecule resolution imaging of hundreds of genes in bone marrow tissue sections to create a single-cell resolution spatial map of the bone marrow . We are establishing tailored machine learning methods to integrate these data types and to infer cell-cell interactions and the underlying molecular pathways controlling blood cell differentiation. We will first investigate the bone marrow microenvironment in healthy mice as a model organism, and then apply the same approach to mouse models of myelodysplastic syndromes (MDS) and leukemia.
In parallel, we will collect bone marrow samples from MDS patients across different stages of the disease and perform a similar analysis on these samples to identify conserved disease-related changes of cell-cell interactions, which we prioritise as candidates for functional validation in the mouse model. With this strategy we hope to acquire a better understanding of the emergence of MDS and leukemia by elucidating the involvement of the microenvironment, and, in particular, to identify novel candidates for treating these malignancies.
To understand the role of the microenvironment during hematopoiesis, spatial single-cell resolution analysis of bone marrow tissue would be required. To address this challenge within the ImmuNiche project, we are combining state-of-the-art single-cell RNA-sequencing technology with large-scale single-molecule resolution imaging of hundreds of genes in bone marrow tissue sections to create a single-cell resolution spatial map of the bone marrow . We are establishing tailored machine learning methods to integrate these data types and to infer cell-cell interactions and the underlying molecular pathways controlling blood cell differentiation. We will first investigate the bone marrow microenvironment in healthy mice as a model organism, and then apply the same approach to mouse models of myelodysplastic syndromes (MDS) and leukemia.
In parallel, we will collect bone marrow samples from MDS patients across different stages of the disease and perform a similar analysis on these samples to identify conserved disease-related changes of cell-cell interactions, which we prioritise as candidates for functional validation in the mouse model. With this strategy we hope to acquire a better understanding of the emergence of MDS and leukemia by elucidating the involvement of the microenvironment, and, in particular, to identify novel candidates for treating these malignancies.
During this period we could establish single-cell RNA-sequencing of all common cell types in the mouse bone marrow to create a reference atlas for the project. We also achieved progress with microscopic imaging of bone marrow tissue, but this method has not been fully established yet. Development of tailored computational analysis methods is ongoing. We already published VarID, a method for the quantification of gene expression variability (Grün, 2020, Nature Methods), and in this period we developed the second generation of this method for the quantification of biological noise in cell state space (manuscript in preparation). This method helps to identify previously unknown cell states and to discover noisy regulators of hematopoietic cell fate choice. We furthermore developed a computational method for the integration of single-cell RNA-sequencing data and large-scale single-molecule resolution microscopy data to reconstruct tissue cell type architecture and to reveal molecular pathways covarying in neighbouring cell types (manuscript in preparation). This method will allow us to infer functional crosstalk of neighbouring cells.
As a major step towards our goal to study perturbation of the microenvironment during emergence of leukemia, we established a single-cell RNA-sequencing time course dataset of a leukemia mouse model. Finally, we were able to perform first single-cell RNA-sequencing pilot experiments on human patient bone marrow biopsies, and could recover a rich spectrum of cell types.
As a major step towards our goal to study perturbation of the microenvironment during emergence of leukemia, we established a single-cell RNA-sequencing time course dataset of a leukemia mouse model. Finally, we were able to perform first single-cell RNA-sequencing pilot experiments on human patient bone marrow biopsies, and could recover a rich spectrum of cell types.
Our computational methods are completely novel and go beyond state-of-the-art. The algorithms we have developed during the last period will allow us to identify determinants in the microenvironment driving gene expression variability and fate of a blood cell.
We expected to finalize establishment of our high-resolution spatial bone marrow imaging, which will enable us to go beyond previous spatial profiling attempts of bone marrow tissue. We will use this method in conjunction with the single-cell RNA-sequencing data, that we keep producing for the mouse model and human patient data, to gain unprecedented insights into the cellular architecture of the bone marrow niche in health and disease.
We expected to finalize establishment of our high-resolution spatial bone marrow imaging, which will enable us to go beyond previous spatial profiling attempts of bone marrow tissue. We will use this method in conjunction with the single-cell RNA-sequencing data, that we keep producing for the mouse model and human patient data, to gain unprecedented insights into the cellular architecture of the bone marrow niche in health and disease.