CD166 Engagement Augments Mouse and Human Hematopoietic Progenitor Function via Activation of Stemness and Cell Cycle Pathways

Abstract

Hematopoietic stem (HSC) and progenitor (HPC) cells are regulated by interacting signals and cellular and noncellular elements of the hematopoietic niche. We previously showed that CD166 is a functional marker of murine and human HSC and of cellular components of the murine niche. Selection of murine CD166⁺ engrafting HSC enriched for marrow repopulating cells. Here, we demonstrate that CD166-CD166 homophilic interactions enhance generation of murine and human HPC in vitro and augment hematopoietic function of these cells. Interactions between cultured CD166⁺ Lineage^- Sca-1⁺ c-Kit⁺ (LSK) cells and CD166⁺ osteoblasts (OBs) significantly enhanced the expansion of colony-forming units (CFUs). Interactions between CD166⁺ LSK cells and immobilized CD166 protein generated more CFU in short-term cultures than between these cells and bovine serum albumin (BSA) or in cultures initiated with CD166^- LSK cells. Similar results were obtained when LSK cells from wildtype (WT) or CD166 knockout (KO) (CD166^-/- ) mice were used with immobilized CD166. Human cord blood CD34⁺ cells expressing CD166 produced significantly higher numbers of CFUs following interaction with immobilized CD166 than their CD166^- counterparts. These data demonstrate the positive effects of CD166 homophilic interactions involving CD166 on the surface of murine and human HPCs. Single-cell RNA-seq analysis of CD150⁺ CD48^- (signaling lymphocyte activation molecule (SLAM)) LSK cells from WT and CD166^-/- mice incubated with immobilized CD166 protein revealed that engagement of CD166 on these cells activates cytokine, growth factor and hormone signaling, epigenetic pathways, and other genes implicated in maintenance of stem cell pluripotency-related and mitochondria-related signaling pathways. These studies provide tangible evidence implicating CD166 engagement in the maintenance of stem/progenitor cell function. Stem Cells 2019;37:1319-1330.

Keywords: CD166; Hematopoietic stem and progenitor cells; Homophilic interaction; Single-cell RNA-seq analysis.

Figure 1.. HPC generating capabilities of CD166⁺ LSK and CD166⁻ LSK cells cultured with OBs.

(A) Illustration of OB, CD166⁺ and CD166⁻ LSK cell isolation, co-cultures and analysis. Preparation of OBs was initiated 7 days before seeding at time 0 (D0) with 1000 freshly sorted C57BL/6–derived CD166⁺ or CD166⁻ LSK cells per well. OBs were cultured in a-MEM medium with 10% FBS, 1% Penicillin-Streptomycin, 1% L-glutamine. Once CD166⁺ and CD166⁻ LSK cells were seeded, the medium was changed to a 1:1 mixture of OB culture medium (a-MEM as described above) and LSK cell culture medium (IMDM medium with 10% FBS, 10 ng/mL recombinant murine SCF, 10 ng/mL IL-3, 20 ng/mL insulin-like growth factor 1 (IGF-1), 20 ng/mL TPO, 25 ng/mL IL-6 and 25 ng/mL Flt3L). CFU-GM (B), BFU-E (C), CFU-GEMM (D), total CFU (E), and CFU fold change (F) between D0 numbers and those obtained on D7 from the progeny of 1000 LSK cells. CFU fold increase was calculated relative to that obtained from 1000 freshly isolated CD166⁺ and CD166⁻ LSK cells plated on D0. Data are pooled from 2 independent experiments, each performed in triplicates. *p < 0.05, **p < 0.01.

Figure 2.. Effects of recombinant mouse CD166 protein (rmCD166) on the generation of HPCs from CD166⁺ and CD166⁻ LSK cells.

(A) Schematic diagram of rmCD166 coating, CD166⁺ and CD166⁻ LSK cell isolation, culture and analysis. Coating of rmCD166 (or BSA as control) was initiated 1 day before seeding at time 0 (D0) with 500 freshly sorted C57BL/6–derived CD166⁺ or CD166⁻ LSK cells per well. CD166⁺ and CD166⁻ LSK cells were cultured in IMDM medium with 10% FBS. Medium was supplemented with exogenous cytokines as described in the legend of Figure 1. CFU-GM (B), BFU-E (C), CFU-GEMM (D), total CFU (E), and CFU fold change (F) between D0 numbers and those obtained on D7 from the progeny of 500 LSK cells. CFU fold increase was calculated relative to that obtained from 500 freshly isolated CD166⁺ LSK and CD166⁻ LSK cells assayed on D0. (G) Flow cytometric analyses of the Lin⁻ Sca-1⁺ population and comparison of absolute numbers of Lin⁻ Sca-1⁺ cells on D7 of culture. Data pooled from 4 independent experiments, each performed in triplicates. *p < 0.05, **p < 0.01. Please note that all cultures were supplemented with exogenous cytokines, and therefore it is expected to observe a CFU-increase regardless of what substrate is used.

Figure 3.. Impact of homophilic CD166 interactions on WT LSK and CD166^−/− LSK cells.

(A) Schematic illustration of rmCD166 protein treatment, wide type (WT) and CD166 knockout (CD166^−/−) LSK cell isolation, culture and analysis. Coating of 48-well culture plates with 10 μg/mL rmCD166 (BSA as control) in Eagle’s balanced salt solution overnight at 4 °C was initiated 1 day before seeding at time 0 (D0) with 500 freshly sorted WT LSK and CD166^−/− LSK cells per well. Cultures were incubated at 5% O₂, 5% CO₂ and supplemented with exogenous cytokines as detailed in Figure 1. CFU-GM (B), BFU-E (C) CFU-GEMM (D), total CFU (E), and CFU fold change (F) between D0 numbers and those obtained on D7 from the progeny of 500 LSK cells. CFU fold increase was calculated relative to that obtained from 500 freshly isolated WT LSK and CD166^−/− LSK cells assayed on D0. Data are pooled from 3 independent experiments, each performed in triplicates. *p < 0.05, **p < 0.01.

Figure 4.. Impact of recombinant human CD166 protein (rhCD166) on generation of HPCs from human CB CD34⁺ cells.

(A) Schematic representation of rhCD166 protein treatment, CD34⁺, CD34⁺CD166⁺, CD34⁺CD166⁻ cell isolation, culture and analysis. Coating of 48-well culture plates with 10 μg/mL rhCD166 (BSA as control) in Eagle’s balanced salt solution overnight at 4 °C was initiated 1 day before seeding at time 0 (D0) with 4000 freshly sorted CD34⁺, CD34⁺CD166⁺, CD34⁺CD166⁻ cells per well. Cultures were cultured in StemSpan™ SFEM II supplemented with human cytokines: 50 ng/ml TPO, 100 ng/ml SCF, and Flt3L and incubated at 5% O₂, 5% CO₂. Analysis of CFU-GM (B), CFU-GEMM (C), total CFU (D), and CFU fold change (E) between D0 numbers and those obtained on D7 from the progeny of 4000 cells from each phenotype. CFU fold increase was calculated relative to that obtained from 4000 freshly isolated CD34⁺, CD34⁺CD166⁺, and CD34⁺CD166⁻ cells plated on D0. Data is pooled from at least 4 independent experiments, each performed in triplicates. *p < 0.05, **p < 0.01.

Figure 5.. Single cell data analysis.

(A) Analysis of the 3 data sets from 3 separate cell types as indicated in the Figure legend on a tSNE plot. (B) PCA plot of significantly expressed genes after cell cycle correction. (C and D) PCA and tSNE plots of the differentially expressed genes suggest distinct differences between CD166^−/− and WT cells. (E and F) Predicted cell trajectories using genes of mitochondria-mediated signaling, Epigenetic, Cytokine,growth factor and hormone signaling, Cell communication, Cell cycle signaling, Metabolic, Translational regulation, and Stem Cell Pluripotency. (G) Gene expression profile of selected genes. (H, I) Gene expression profile of differentially expressed stem cell pluripotency and metabolic genes. CD166 KO and WT samples are colored by green and pink, respectively, on the column side color bar.

Figure 6.. Chimerism data from in vivo transplantation studies.

Data from two independent experiments showing chimerism levels at 8 weeks and 16 weeks post-transplantation. Each symbol represents a single mouse. CD166+ LSK cells were cultured for 5 days on plates covered with immobilized CD166 protein or BSA then co-transplanted with 250,000 BoyJ (CD45.1) BM mononuclear competitor cells. n=17 for rmCD166 and n=18 for BSA groups.