Specific Stromal Populations Research Articles

Abstract LB-219: Higher levels of mast cells associate with favorable outcomes in non-small cell lung cancer and correlate with lower malignant cell proliferation

Abstract The tumor microenvironment (TME) involves complex interactions between malignant and stromal cell types. Much of our knowledge of cancer biology has been derived from studying molecular mechanisms underlying bulk tumors, with a focus on specific malignant pathways that have become dysregulated during tumorigenesis and tumor progression. Although studies have identified interactions between malignant and stromal cells within the TME, few have sought to comprehensively identify such relationships. Here, we performed RNA-seq on bulk and flow-sorted non-small cell primary human lung tumors enriching for malignant cells, endothelial cells, immune cells, and fibroblasts. We derived a map of cell-specific differential gene expression of prognostically associated secreted factors and cell surface markers, and computationally reconstructed pairwise cross-talk between cell types. We found significant novel associations between transcriptional profiles of malignant populations and specific stromal populations, focusing here on mast cells. We identified presence of infiltrating mast cells to be negatively correlated with malignant cell proliferation. Expression of TPSAB1 (Tryptase Alpha/Beta 1) is largely confined to mast cells, and its high expression in both adenocarcinoma and squamous cell carcinoma is favorably prognostic in both histologies across multiple datasets based on gene expression data. We validated the prognostic relevance of mast cells in NSCLC by immunohistochemical (IHC) staining of a lung tumor tissue microarray (TMA) for MCT (mast cell tryptase). Higher mast cell count was associated with better overall survival across all NSCLC, or when considering either adenocarcinoma or squamous cell carcinoma alone. When mast cell counts were quantified as “High”, “Intermediate”, “Low”, and “Negative” levels without reference to clinical outcomes, these were statistically significantly associated with survival. Negative-, low-, and intermediate-levels of mast cells all conferred worse prognosis than high mast-cell levels. Mast cell levels remained prognostic in multivariate analysis with independent clinical factors including age, stage, and gender. Finally, we directly evaluated the relationship of mast cell levels to tumor proliferation by staining the same samples for the proliferation marker KI67. The percentage of tumor cells staining positive for KI67 was lower in tumors with high vs low numbers of mast cells (p=0.048, Wilcox rank-sum test). In summary, we identified and validated a specific inverse relationship between levels of infiltrating mast cells and malignant cell proliferation. These results illustrate the utility of transcriptomic profiling of flow-sorted subpopulations from solid tumors in order to identify tumor-microenvironment interactions that may have prognostic and therapeutic relevance. Citation Format: Andrew J. Gentles, Angela Hui, Weiguo Feng, Ramesh V. Nair, Alice Yu, Majid Shafiq, Erna Forgo, Amanda Khuong, Yue Xu, Chuong D. Hoang, Robert B. West, Matt van de Rijn, Maximilian Diehn, Sylvia K. Plevritis. Higher levels of mast cells associate with favorable outcomes in non-small cell lung cancer and correlate with lower malignant cell proliferation [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr LB-219. doi:10.1158/1538-7445.AM2017-LB-219

Targeting Bone Marrow Mesenchymal Stromal Cells Using Cre-Recombinase Transgenes

The bone marrow microenvironment contains hematopoietic niches that regulate the proliferation, differentiation, and trafficking of hematopoietic stem/progenitors cells (HSPCs). These hematopoietic niches are comprised of a heterogeneous population of stromal cells that include, endothelial cells, osteoblasts, CXCL12-abundant reticular (CAR) cells, mesenchymal stem cells (MSCs), arteriolar pericytes, and sympathetic nerves. Emerging data suggest that specific stromal populations may regulate distinct types of HPSCs. Thus, it is important to have validated approaches to interrogate and target specific stromal cell populations. Prior studies have shown that Prx1-Cre, Osx-Cre, Lepr-Cre, and Nes-Cre broadly target mesenchymal stromal cells in the bone marrow. Here, we rigorously define the stromal cell populations targeted by two Cre-transgenes that are commonly used to target osteolineage cells (Ocn-Cre, and Dmp1-Cre) and introduce a new Cre-transgene (Tagln-Cre) that efficiently targets bone marrow pericytes. For each Cre-transgene, we performed lineage mapping using ROSA26Ai9/Ai9 mice, in which cells that have undergone Cre-mediated recombination express tdTomato. In some cases, we further crossed these mice to introduce the Cxcl12gfp transgene, which can be used to define GFP-bright CAR cells. Immunostaining of bone sections and flow cytometry were used to define the target stromal cell population(s) in these mice. Osteocalcin (Bglap, Ocn) is primarily expressed in mature osteoblasts. Accordingly, Ocn-Cre is widely used to specifically target osteoblasts. However, our lineage mapping studies show that Ocn-Cre targets not only all osteoblasts, but also 72 ± 4.0% of CAR cells. Ocn-Cre also targets a subset of NG2+ arteriolar pericytes. Dentin matrix acidic phosphoprotein 1 (Dmp1) is expressed primarily in osteocytes, and Dmp1-Cre has been widely used to specifically target osteocytes. However, we show that Dmp1-Cre also efficiently targets endosteal osteoblasts and approximately 40% of CAR cells. To target bone marrow pericytes, we tested several Cre-transgenes, ultimately focusing on Tagln-Cre. Transgelin (Tagln, SM22a) is broadly expressed in pericytes, smooth muscle cells, and cardiomyocytes. Lineage-mapping studies show that Tagln-Cre targets all arteriolar and venous sinusoidal pericytes in the bone marrow. It also targets osteoblasts and 75 ± 5.2% of CAR cells. There are several recent studies that have ascribed specific functions to osteoblasts or osteocytes based on targeting using Ocn-Cre or Dmp1-Cre, respectively. In light of our data, these conclusions need to be re-evaluated. Ocn-Cre, Dmp1-Cre, and Tagln-Cre each target a subset of CAR cells. Studies are underway to determine whether these CAR subsets have unique expression profiles and functions. Finally, Talgn-Cre represents a new tool for investigators in the field to efficiently target bone marrow pericytes. Disclosures No relevant conflicts of interest to declare.