Multiplexed fluorescence microscopy reveals heterogeneity among stromal cells in mouse bone marrow sections.
Karolin HolzwarthRalf KöhlerLars PhilipsenKoji TokoyodaValeriia LadyhinaCarolina WählbyRaluca A NiesnerAnja Erika HauserPublished in: Cytometry. Part A : the journal of the International Society for Analytical Cytology (2018)
The bone marrow (BM) consists of multiple, structured micro-environmental entities-the so called niches, which contain hematopoietic cells as well as stromal cells. These niches fulfill a variety of functions, such as control of the hematopoietic stem cell pool, differentiation of hematopoietic cells, and maintenance of immunological memory. However, due to the molecular and cellular complexity and a lack of suitable histological multiplexing methods, the composition of the various BM niches is still elusive. In this study, we apply multiepitope-ligand-cartography (MELC) on bone sections from mice. We combine multiplexed immunofluorescence histology data with various object-based segmentation approaches in order to define irregularly shaped, net-like structures of stromal cells. We confirm MELC as a robust histological method and validate our automated segmentation algorithms using flow cytometry and manual evaluation. By means of MELC multiplexing, we reveal heterogeneous expression of leptin receptor (LpR), BP-1, and VCAM-1 in the stromal network. Moreover, we demonstrate by quantification a preferential contact of B cell subsets as well as of plasma cells to processes of CXCL12-expressing stromal cells, compared with stromal somata. In summary, our approach is suitable for spatial analysis of complex tissue structures.
Keyphrases
- bone marrow
- induced apoptosis
- deep learning
- cell cycle arrest
- single cell
- machine learning
- mesenchymal stem cells
- high resolution
- flow cytometry
- high throughput
- endoplasmic reticulum stress
- oxidative stress
- cell death
- gene expression
- type diabetes
- poor prognosis
- signaling pathway
- binding protein
- dna methylation
- adipose tissue
- metabolic syndrome
- risk assessment
- body composition
- skeletal muscle
- big data
- bone mineral density
- long non coding rna
- label free
- network analysis
- cell adhesion