Multimodal histologic atlas of human bone marrow

Project Summary – Overall
Bone marrow produces blood cells whose functions range from oxygen delivery to anti-microbial defense to
hemostasis, all originating from hematopoietic stem cells (HSC). To sustain and regulate this process, bone
marrow stromal cells form multiple niche microenvironments, each tailored to the needs of a particular developing
blood cell population. Using highly-multiplexed imaging technologies, our proposed Bone Marrow Tissue
Mapping Center (TMC) aims to systematically and quantitatively dissect the cellular composition and spatial
organization of human bone marrow microenvironments. The resulting detailed maps will serve as an open and
global platform for understanding which cells and interactions are critical for each branch of hematopoietic
maturation, and how these vary by anatomical site and across diverse patient demographics.
The TMC will define cellular identities and cell states at the transcriptional, translational, and post-
translational levels using Nanostring DSP, Multiplexed Ion Beam Imaging (MIBI), and MALDI-MSI, which
generate quantitative spatial maps of RNA, protein, and N-glycans, respectively. Our cross-disciplinary team not
only includes the inventors of MIBI and a pioneer of MALDI-MSI, but also experts in human HSCs and human
hematopoiesis and a practicing hematopathologist with expertise in histopathologic bone marrow diagnosis. To
overcome the unique challenges of working with hard, mineralized bone, we will leverage parallel, robust,
clinically-validated bone marrow processing pipelines which maximize and standardize sample quality and
compatibility with current and future technologies. Integrating seamlessly into standard clinical workflows, our
pipelines enable convenient sharing of prospectively-collected materials with the Tissue Core. Samples will be
collected from three different sources: (1) prospective, patient-matched multi-site collection from deceased
donors to examine differences between anatomical sites, (2) prospective collection of femoral head from hip
arthroplasty specimens for differences between age ranges, (3) iliac crest bone in the Stanford Pathology archive
for differences between races and genders. These multiple collection strategies, multiple sites, and different
investigational focuses complement prior HuBMAP projects. The Data Analysis Core team has pioneered
multiple novel data processing pipelines, including pixel-based analyses, cell-based analyses including state-of-
the-art cell segmentation and cell clustering and enumeration, and neighborhood analyses. These tools are
broadly-applicable to all highly-multiplexed quantitative imaging technologies. Overall, our team and strategy are
exceedingly well-suited for executing the vision of the proposed Bone Marrow TMC.
The spatial structure of bone marrow reflects the evolutionary mechanisms that terraformed bone to create
unique microenvironments meeting the nutritional needs of developing blood cells with divergent functions. The
interdependence between bone marrow tissue structure and hematopoiesis is informative not just in blood cell
maturation, but for understanding metabolism, aging, and development of cellular therapies.

Grant Number: 5U54HL165445-04
NIH Institute/Center: NIH
Principal Investigator: Sean Bendall