Archives

  • 2018-07
  • 2018-10
  • 2018-11
  • 2019-04
  • 2019-05
  • 2019-06
  • 2019-07
  • 2019-08
  • 2019-09
  • 2019-10
  • 2019-11
  • 2019-12
  • 2020-01
  • 2020-02
  • 2020-03
  • 2020-04
  • 2020-05
  • 2020-06
  • 2020-07
  • 2020-08
  • 2020-09
  • 2020-10
  • 2020-11
  • 2020-12
  • 2021-01
  • 2021-02
  • 2021-03
  • 2021-04
  • 2021-05
  • 2021-06
  • 2021-07
  • 2021-08
  • 2021-09
  • 2021-10
  • 2021-11
  • 2021-12
  • 2022-01
  • 2022-02
  • 2022-03
  • 2022-04
  • 2022-05
  • 2022-06
  • 2022-07
  • 2022-08
  • 2022-09
  • 2022-10
  • 2022-11
  • 2022-12
  • 2023-01
  • 2023-02
  • 2023-03
  • 2023-04
  • 2023-05
  • 2023-06
  • 2023-08
  • 2023-09
  • 2023-10
  • 2023-11
  • 2023-12
  • 2024-01
  • 2024-02
  • 2024-03
  • 2024-04
  • 2024-05

    • Hematopoietic stem cells HSCs maintain the turnover of red b

    2018-10-24

    Hematopoietic stem abacavir (HSCs) maintain the turnover of red blood cells, platelets, and immune cells. They reside in the bone marrow, where several cell types contribute to their regulation (Lo Celso et al., 2011; Morrison and Scadden, 2014). How the concerted action of multiple niche components regulates HSC fate is not clear, and understanding the localization of HSCs relative to multiple surrounding cellular and structural constituents of the bone marrow microenvironment is the first step toward solving this puzzle. Single-cell resolution intravital microscopy of fluorescently labeled HSCs and niche components allows direct observation of HSCs in mouse bone marrow. We, and others, have successfully used fluorescent dyes to label HSCs prior to transplantation to achieve their detection in vivo through the bone of anesthetized recipient mice (Köhler et al., 2009; Lo Celso et al., 2009). The endosteal niche is a bone marrow microenvironment proximal to trabecular/cortical bone, lined by cells of the osteoblast lineage, and associated with effective HSC engraftment as well as maintenance of their long-term survival and function (Calvi et al., 2003; Kunisaki et al., 2013; Lane et al., 2011). Two-photon microscopy is essential to detect second harmonic generation (SHG) signal emitted by bone collagen and to recognize the location of bone marrow cavities and their endosteal surface (Lo Celso et al., 2011). Transgenic reporter mice expressing GFP under the control of an osteoblast-specific promoter (herein referred to as Col2.3GFP mice) (Kalajzic et al., 2002) are a powerful tool for visualizing the HSC endosteal niche. Using these and other reporter strategies, manual analysis of HSC in vivo images has provided the indication that normal HSCs localize near vasculature, endosteum, osteoblastic cells, and nestin-positive mesenchymal progenitor cells (Lane et al., 2011; Lo Celso et al., 2009; Méndez-Ferrer et al., 2010; Sanchez-Aguilera et al., 2011). However, such analysis has the following limitations: (1) it is time-consuming and, therefore, limited to measuring a few parameters; and (2) it is subject to human error, leading to intra- and inter-researcher inconsistencies. We reasoned that specialized image analysis tools would simplify the biological interpretation of 3D HSC microenvironment images, and they not only would provide unbiased data analysis but also expand the number of measurable parameters with the potential to uncover new aspects of HSC biology. Although several segmentation and classification methods have been developed for 2D (Chung and Vese, 2009; Saikumar et al., 2012; Yeo et al., 2011) and 3D datasets (Lou et al., 2014; Nandy et al., 2014; Pop et al., 2013), their application to in vivo bone marrow images is not straightforward due to the complexity of the structures of interest. In conjunction, light scattering caused by the surrounding tissue, especially by overlaying bone, limits the resolution of in vivo microscopy of bone marrow compared to that of other tissues or ex vivo techniques. To overcome these issues, we propose a local heterogeneity-based image segmentation (LH-SEG) approach that utilizes multi-resolution segmentation (Mallinis et al., 2008) and mean intensity difference to neighbor thresholding. This approach measures local morphological and intensity homogeneity and combines these values with neighborhood distance features to segment and threshold the objects of interest. To ensure reliable edge detection across objects with high-intensity heterogeneity, LH-SEG is applied on each 2D slice prior to 3D rendering. HSCs labeled ex vivo using lipophilic membrane dyes such as 1,1’-Dioctadeciyl-3,3,3′,3′-Tetramethylindodicarbocyanine (DiD) generate a bright fluorescent signal (Lo Celso et al., 2009, 2011). However, these dyes lead to background signal from cell debris and aggregates (Progatzky et al., 2013), posing the extra challenge of filtering the dye signal to identify bona fide HSCs, a task that is not trivial even for the experienced user. In order to standardize HSC recognition, we used a machine learning protocol based on morphological and textural features to recognize and classify all segmented DiD signal. Finally, once HSCs, osteoblasts, and bone are identified, the minimum distance between each HSC and osteoblast/endosteum objects can be calculated in 3D. We used the proposed approach to examine the localization of HSCs in multiple in vivo datasets and tested its performance compared to other widely used segmentation methods as well as manual benchmarking data. We demonstrate that the method is robust and applicable to a variety of datasets that are challenging to analyze manually.