Unique macrophage phenotypes activated by BMP signaling in breast cancer bone metastases
Claire L. Ihle, Desiree M. Straign, Johana A. Canari, Kathleen C. Torkko, Kathryn L. Zolman, Elizabeth E. Smith, Philip Owens
Claire L. Ihle, Desiree M. Straign, Johana A. Canari, Kathleen C. Torkko, Kathryn L. Zolman, Elizabeth E. Smith, Philip Owens
View: Text | PDF
Research Article Immunology Oncology

Unique macrophage phenotypes activated by BMP signaling in breast cancer bone metastases

Abstract

Metastatic breast cancer (mBC) tissue in bone was systematically profiled to define the composition of the tumor microenvironment. Gene expression identified a high myeloid signature of patients with improved survival outcomes. Bone metastases were profiled by spatial proteomics to examine myeloid populations within the stroma that correlated with macrophage functions. Single-cell spatial analysis uncovered macrophage activation in the stroma of mBC bone lesions. Matched BC patient samples of primary breast tumor and bone metastasis tissues were compared for gene expression in the bone, where bone morphogenetic protein 2 (BMP2) was most significantly upregulated. Immune cell changes from breast to bone demonstrated a loss of lymphoid cells but a consistent population of macrophages. BMP-activated macrophages were increased uniquely in bone. Bone marrow–derived macrophage activation coupled with BMP inhibition increased inflammatory responses. Using experimental mouse models of mBC bone metastasis and trained immunity, we found that BMP inhibition restricts progression of metastases early in the macrophage activation state but not after tumors were established in the bone. This study revealed unique myeloid BMP activation states that are distinctly integrated with bone metastases.

Authors

Claire L. Ihle, Desiree M. Straign, Johana A. Canari, Kathleen C. Torkko, Kathryn L. Zolman, Elizabeth E. Smith, Philip Owens

×

Figure 1

The diverse immune microenvironment of human breast cancer in bone.

Options: View larger image (or click on image) Download as PowerPoint
The diverse immune microenvironment of human breast cancer in bone. (A) ...
(A) Myeloid compartment signature scores from PanCancer IO 360 gene expression analysis of archival FFPE mBC patient bone metastases (n = 14). Cohort separated into samples exhibiting high myeloid compartment gene expression (n = 7, red region) and samples exhibiting low myeloid compartment gene expression (n = 7, blue region). (B) Volcano plot of differential gene expression analysis of high myeloid compartment patients compared with the baseline of low myeloid compartment patients. Genes upregulated in high myeloid compartment patients are shown on the right (red region), and genes upregulated in low myeloid compartment patients are shown on the left (blue region). (C) Undirected gene set enrichment pathway analysis of high versus low myeloid compartment patient gene expression. Global significance scores greater than 1 revealed pathways significantly upregulated in high myeloid compartment patient samples. (D) Heatmap of immune checkpoint gene expression from mBC bone metastases. (E) Cell type gene signature scores for low myeloid compartment and high myeloid compartment patient samples. Data are presented as mean ± SEM. Statistical values determined by Student’s t test. (F) Kaplan-Meier curve of overall survival from mBC bone metastasis diagnosis to death in patient cohort based on high or low myeloid compartment gene expression. Statistical value determined by log-rank test. (G) Representative image of GeoMx Digital Spatial Profiling of FFPE archival mBC patient bone metastases (n = 12). Each patient was stamped with 4 PanCK ROIs, 4 CD68 ROIs, and 4 CD3 ROIs. Scale bar: 100 μM. (H) Correlations of myeloid marker protein expression and functional proteins in PanCK, CD68, and CD3 ROIs from mBC bone samples (n = 12). R2 value for simple linear regression reported. Statistical analysis determined by Pearson Correlation r value and P value. Scale bars: 100 μM.