The Tumor Microenvironment-Dependent Transcription Factors AHR and HIF-1α Are Dispensable for Leukemogenesis in the Eµ-TCL1 Mouse Model of Chronic Lymphocytic Leukemia

Abstract

Chronic lymphocytic leukemia (CLL) is the most frequent leukemia in the elderly and is characterized by the accumulation of mature B lymphocytes in peripheral blood and primary lymphoid organs. In order to proliferate, leukemic cells are highly dependent on complex interactions with their microenvironment in proliferative niches. Not only soluble factors and BCR stimulation are important for their survival and proliferation, but also the activation of transcription factors through different signaling pathways. The aryl hydrocarbon receptor (AHR) and hypoxia-inducible factor (HIF)-1α are two transcription factors crucial for cancer development, whose activities are dependent on tumor microenvironment conditions, such as the presence of metabolites from the tryptophan pathway and hypoxia, respectively. In this study, we addressed the potential role of AHR and HIF-1α in chronic lymphocytic leukemia (CLL) development in vivo. To this end, we crossed the CLL mouse model Eµ-TCL1 with the corresponding transcription factor-conditional knock-out mice to delete one or both transcription factors in CD19+ B cells only. Despite AHR and HIF-1α being activated in CLL cells, deletion of either or both of them had no impact on CLL progression or survival in vivo, suggesting that these transcription factors are not crucial for leukemogenesis in CLL.

Keywords: AHR; HIF1α; chronic lymphocytic leukemia; tumor microenvironment.

Figure 1

RNA sequencing of B cells from C57BL/6 and Eµ-TCL1 mice. Splenic B cells were FACS-sorted from three mice of each genotype, and mRNA was sequenced. (A) Principal component analysis of individual animals. (B) K-means clustering and Gene Ontology enrichment analysis. (C) Scatterplot depicting the expression of genes in the groups. (D) Unsupervised hierarchical clustering showing 1416 genes upregulated and 1041 genes downregulated in TCL1. (E,G) Gene Set Enrichment Analysis showing the enrichment of hypoxia (E) and AHR (G) signatures in Eµ-TCL1 versus C57BL/6 mice. (F) Protein–protein interactions network (STRING) for transcription factors involved in enriched hallmark pathways (GSEA). (H) Transcription factor activity (Z-values, ISMARA) for HIF-1α and AHR motifs in WT and TCL1 B cells.

Figure 2

Knocking out HIF-1α does not show an effect on the leukemogenesis of CLL cells in the murine Eµ-TCL1 model. (A) Scheme of the breeding strategy to generate Eµ-TCL1 CD19^Cre/WT Hif1a^fl/fl (cKO). (B–D) Validation of the knock-out of exon 2 of Hif1a in isolated B cells from cKO mice incubated with CoCl₂ at the DNA (B), RNA (C), and protein (D) levels compared to control mice (n = 3). ** p < 0.01, **** p < 0.0001. (E) Survival curve of cKO compared to control mice shows no significant difference. (F–G) Flow cytometry analysis of CLL cells (CD19+ CD5+) in the peripheral blood (PB) of cKO and control mice over time (Ctrl: n = 45, cKO: n = 39) (F) and at month 8 (Ctrl: n = 42, cKO: n = 35) (G). (H–I) CLL cells from cKO and control mice (n = 3) were subjected to RNA sequencing. (H) Unsupervised hierarchical clustering showing the top 10,000 most expressed genes. (I) Volcano plot showing DEG between cKO and control mice. (J) CLL development of the adoptive transfer of Eµ-TCL1 CLL cells in CD19^Cre/WT Hif1a^fl/fl mice (Ctrl: n = 5, cKO: n = 6).

Figure 3

AHR knock-out does not influence CLL development in the murine Eµ-TCL1 model. (A) Scheme of the breeding strategy to generate Eµ-TCL1 CD19^Cre/WT Ahr^fl/fl (cKO) mice. (B–D) Validation of the knock-out of exon 2 of Ahr in isolated B cells from cKO mice incubated with FICZ at the DNA (B), RNA (C), and protein (D) levels compared to control mice (n = 3). ** p < 0.01. (E) Survival curve of cKO compared to control mice shows no significant difference. (Ctrl: n = 45, cKO: n = 38). (F,G) Flow cytometry analysis of CLL cells (CD19+ CD5+) in the peripheral blood (PB) of cKO and control mice over time (Ctrl: n = 45, cKO: n = 38) (F) and at month 8 (Ctrl: n = 42, cKO: n = 34) (G). (H,I) CLL cells from cKO and control mice (n = 3) were subjected to RNA sequencing. (H) Unsupervised hierarchical clustering showing the top 10,000 most expressed genes. (I) Volcano plot showing DEG between cKO and control mice. (J) CLL development of the adoptive transfer of Eµ-TCL1 CLL cells in CD19^Cre/WT Ahr^fl/fl mice (Ctrl: n = 5, cKO: n = 5).

Figure 4

Double knock-out of AHR and HIF-1α does not appear to have an impact on the development of neoplastic B cells. (A) Scheme of the breeding strategy to generate Eµ-TCL1 CD19^Cre/WT Hif1a^fl/flAhr^fl/fl (cDKO) mice. (B–F) Validation of the knock-out of exon 2 of Hif1a and Ahr, respectively, in isolated B cells from cDKO mice incubated with CoCl₂ and FICZ at the DNA (B,C), RNA (D,E), and protein (F) levels compared to control mice (n = 3). ** p < 0.01, *** p < 0.001, **** p < 0.0001. (G) Survival curve of cDKO compared to control mice shows no significant difference. (Ctrl: n = 45, cDKO: n = 21). (H,I) Flow cytometry analysis of CLL cells (CD19+ CD5+) in the peripheral blood (PB) of cDKO and control mice over time (Ctrl: n = 45, cDKO: n = 21) (H) and at month 8 (Ctrl: n = 35, cDKO: n = 16) (I). (J,K) CLL cells from cDKO and control mice (n = 3) were subjected to RNA sequencing. (L) Unsupervised hierarchical clustering showing the top 10,000 most expressed genes. (I) Volcano plot showing DEG between cDKO and control mice. (J) CLL development of the adoptive transfer of Eµ-TCL1 CLL cells in CD19^Cre/WT Hif1a^fl/fl Ahr^fl/fl mice (Ctrl: n = 4, cDKO: n = 4).