Abstract
Chronic lymphocytic leukemia (CLL) cells can be subdivided into fractions based on the time since cell birth/division. The fraction of CLL B cells bearing a CXCR4DimCD5Bright phenotype divided recently (proliferative fraction, PF), while those cells that divided earlier and are quiescent bear a CXCR4BrightCD5Dim phenotype (resting fraction, RF). Prior studies suggest that cells of the PF are recent emigrants of solid tissue, whereas cells of the RF did so earlier and are most adept at homing to lymphoid niches. Here we report gene expression and functional analyses of the PF and RF that address the signaling pathway most relevant for the PF and the relative abilities of the PF and RF to migrate using the pathway.
When analyzing genes significantly differentially expressed (DEGs) between PF and RF of 26 CLL patients (13 U-CLL and 13 M-CLL), we found 198 genes upregulated in the PF and 88 in the RF. Using Ingenuity Pathway Analysis (IPA), the DEGs of the PF overlapped best with the integrin signaling pathway (Z-score = 2.887). Moreover, of the other identified pathways with Z scores > 2.0, each impacted integrin expression and function (upregulated: Actin Nucleating by ARP-WASP Complex and Regulation of Actin-based Motility by Rho; downregulated: RhoGDI signaling). To validate the transcriptomic data, mRNA and protein levels of various upregulated integrin subunits were confirmed by Q-PCR and cell surface staining by flow cytometry. When comparing the DEGs between the PF and the RF in U-CLL vs M-CLL patients, a core of 144 genes were shared by the two and 358 were unique to U-CLL and 35 to M-CLL. Notably, IPA indicated that the shared genes overlapped with the integrin signaling pathway and the U-CLL unique genes resembled pathways used by integrins to carry out their functions, such as migration. When the 3 sets of DEGs were subjected to Upstream Regulator analysis using IPA, multiple cytokines were identified as potential regulators of the DEGs shared by U-CLL and M-CLL (Z scores > 2.0) clones, whereas for the DEGs unique to the PF of U-CLL, only IL-4 was identified (Z scores = 2.6).
To corroborate the above, we cultured CLL B cells for 3 days with the of putative upstream regulators of the shared and U-CLL unique DEGs. For the former, we found that [1] IFNα upregulated CD11c in all CLL, [2] IL-1 and IL-2 upregulated CD11a in all CLL, and [3] IL-1, IL-2, IFNα and IFNγ upregulated CD29 in M-CLL only. When testing a regulatory role for IL-4 in the U-CLL unique genes, we found that IL-4 upregulated CD11a, CD11c, and CD18 in U-CLL (P < 0.05), but not in M-CLL.
Finally, we evaluated several functions mediated by integrin molecules. First, we tested the binding of the PF and RF to fibronectin, an important step in cell migration. This indicated that the PF bound better to fibronectin than RF (P < 0.01), and led to extended survival (P < 0.05). Consistent with this, co-culture with anti-CD49d antibody significantly blocked CLL cell adhesion to fibronectin (P < 0.01). Next, we addressed CLL cell migration. In vitro migration studies indicated that PF cells migrated significantly better than the RF with (P < 0.0001) or without (P < 0.001) CXCL12/SDF-1. In line with this, 24-hours after transferring PF and RF into NSG mice, significantly more PF than RF cells were found in the spleen (P < 0.05). Notably, the ability to more rapidly enter solid tissue associated with the rate of disease progression of individual patients (P < 0.001). Since these were unexpected findings, we questioned if sorting PF and RF altered their ability to migrate in vitro. Based on the analysis of 16 patients (U-CLL = 8, M-CLL = 8), simultaneous preincubation of CLL B cells with mAbs to CD5 plus CXCR4 did not change spontaneous nor CXCL12/SDF-1-mediated migration.
Collectively, our findings indicate that integrins and the integrin signaling pathway are key elements in the biology of the recently divided fraction of CLL cells. Moreover, those DEGS that are shared by U-CLL and M-CLL cells and those unique to U-CLL are regulated by distinct upstream influences, with IL-4 standing out as a crucial cytokine for the development of the U-CLL PF. Finally, in vitro and in vivo studies suggest a key role for the integrin pathway in CLL migration, and pinpoint, surprisingly, the PF, not the RF, as the better migrater and document that the rate of migration correlates with disease activity.
Disclosures
Rhodes:SeaGen: Consultancy; Velosbios: Research Funding; TG Therapeutics: Consultancy; Beigene: Consultancy; Genentech: Consultancy; Genmab: Consultancy; Janssen: Consultancy, Research Funding; Pharmacyclics: Consultancy, Research Funding; Loxo Oncology: Research Funding; Morphosys: Consultancy; Oncternal: Research Funding; Verastem: Consultancy; Abbive: Consultancy; Epizyme: Research Funding. Barrientos:Merck, Oncternal, Pharmacyclics/Abbvie: Research Funding; Beigene, AstraZeneca, Pharmacyclics/Abbvie: Consultancy. Allen:BMS: Current equity holder in publicly-traded company; C4 Therapeutics: Current equity holder in private company; Sanofi Genzyme: Membership on an entity's Board of Directors or advisory committees; Alexion: Consultancy, Research Funding.
Author notes
Asterisk with author names denotes non-ASH members.