Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2013 Oct;14(10):1073-83.
doi: 10.1038/ni.2707. Epub 2013 Sep 8.

Selective Regulation of Lymphopoiesis and Leukemogenesis by Individual Zinc Fingers of Ikaros

Affiliations
Free PMC article

Selective Regulation of Lymphopoiesis and Leukemogenesis by Individual Zinc Fingers of Ikaros

Hilde Schjerven et al. Nat Immunol. .
Free PMC article

Abstract

C2H2 zinc fingers are found in several key transcriptional regulators in the immune system. However, these proteins usually contain more fingers than are needed for sequence-specific DNA binding, which suggests that different fingers regulate different genes and functions. Here we found that mice lacking finger 1 or finger 4 of Ikaros exhibited distinct subsets of the hematological defects of Ikaros-null mice. Most notably, the two fingers controlled different stages of lymphopoiesis, and finger 4 was selectively required for tumor suppression. The distinct defects support the hypothesis that only a small number of genes that are targets of Ikaros are critical for each of its biological functions. The subcategorization of functions and target genes by mutagenesis of individual zinc fingers will facilitate efforts to understand how zinc-finger transcription factors regulate development, immunity and disease.

Figures

Figure 1
Figure 1
Adult B cells are present in mice with germline deletion of Ikzf1 exons encoding zinc fingers 1 or 4. (a) Immunoblot analyses with thymocyte extracts from the Ikzf1ΔF1/ΔF1 strain and the Ikzf1ΔF4/ΔF4 strain. The bands denoted with an asterisk correspond to an unidentified form of Ikaros, most likely a post-transcriptionally modified form of Ik-1 (and Ik-2 after exon 4 deletion). SNP70 is shown as a loading control (bottom). (b) RNA-Seq analyses of thymocyte mRNAs (n = 4) confirmed the absence of reads from exons 4 and 6 in the Ikzf1ΔF1/ΔF1 and Ikzf1ΔF4/ΔF4 strains, respectively. (c) CD11b+ myeloid-cell percentages analyzed in bone marrow from 6-week-old wild-type and mutant mice. One experiment is shown with average ± SEM for wild-type (WT) (n = 21), Ikzf1ΔF1/ΔF1 (n = 10) and Ikzf1ΔF4/ΔF4 (n = 14) strains. (d) CD19+ (left) and CD19+B220+ (right) B lineage cells analyzed in the bone marrow (BM) and spleen of 6-week-old mice from the two mutant strains. Average cell numbers ± SEM (left and middle) and % of live cells (right, horizontal bar represents the mean) from wild-type (n = 12–20), Ikzf1ΔF1/ΔF1 (n = 9–18) and Ikzf1ΔF4/ΔF4 (n = 6–17) mice are shown.
Figure 2
Figure 2
B cell development is disrupted at different stages in Ikzf1ΔF1/ΔF1 and Ikzf1ΔF4/ΔF4 mice. (a). B-cell development was analyzed in whole bone marrow from 6- to 8-week-old mice. One representative experiment out of three or more is shown. Staining is shown for total bone marrow in row 1, pre-pro B to large pre-BII (B220+CD43+) cells in row 2, lineage (Lin) B220+CD43+CD19cells (for analysis of pre-pro B) in row 3, and small pre-BII to mature B (B220+CD43) cells in row 4. The average percentage of the indicated population in total bone marrow is shown. Lineage markers included CD3, TER119, Gr-1, and CD11b. (b) Absolute numbers of the indicated populations in total bone marrow of wild-type (n=13), Ikzf1ΔF1/ΔF1 (n=6) and Ikzf1ΔF4/ΔF4 (n=6-8) mice are shown. Each symbol represents an individual mouse and bar shows the mean. *P<0.05, **P<0.01, ***P<0.001. (c) Cytoplasmic Ig μ expression was analyzed in bone marrow cells, gated on B220+IgM. One representative out of four separate experiments is shown (d) Absolute numbers of B220+IgM+ cells per femur are shown. Each symbol represents an individual mouse and bar shows the mean ± SEM of wild-type (n=5), Ikzf1ΔF1/ΔF1 (n=6) and Ikzf1ΔF4/ΔF4 (n=3) mice. (e-f) Flt3 and IL-7Rα expression was analyzed on c-Kit+ bone marrow cells gated on Linneg–lo cells. One representative experiment is shown with the average percentage of total live bone marrow cells from wild-type (n=12-14), Ikzf1ΔF1/ΔF1 (n=3-4) and Ikzf1ΔF4/ΔF4 (n=8-9) mice.
Figure 3
Figure 3
Selective thymocyte and fetal phenotypes in Ikzf1ΔF1/ΔF1 and Ikzf1ΔF4/ΔF4 mice. Thymocyte development was analyzed with 4-week-old mice, while fetal hematopoietic development was analyzed with E18.5 liver, mesentery, and thymus. (a) Thymic cellularity is shown relative to wild-type littermate controls for Ikzf1ΔF1/ΔF1 (n=8) and Ikzf1ΔF4/ΔF4 (n=9) mice. (b) Early stages of thymocyte development were analyzed gated on CD4CD8 (DN) cells. Numbers in quadrants indicate average percentages of CD44+CD25 (DN1), CD44+CD25+ (DN2), CD44CD25+ (DN3) and CD44CD25 (DN4) cells. (c) The ETP population (LinKit+CD44+CD25) were analyzed in wild-type, Ikzf1ΔF1/ΔF1 (n=11), and Ikzf1ΔF4/ΔF4 mice (n=11). *P<0.05 (d) Analysis of CD4 and CD8 expression in the thymus. In b and d, one representative experiment is shown with average values for the indicated populations for wild-type (n=16), Ikzf1ΔF1/ΔF1 (n=8) and Ikzf1ΔF4/ΔF4 (n=8) mice. (e) Histological analysis with H&E staining of thymi from 4-week-old mice. (f-g) Thymic γδTCR+ cells and thymic B cells (CD19+CD4CD8) were analyzed in Ikzf1ΔF1/ΔF1 mice (n=8-9) and Ikzf1ΔF4/ΔF4 mice (n=8-11). (h-i) Fetal (E18.5) B cells (CD45+CD19+B220+) were analyzed in the Ikzf1ΔF1/ΔF1 strain (n=8) and the Ikzf1ΔF4/ΔF4 strain (n=9). (j) The number of fetal thymocytes was selectively reduced in E18.5 Ikzf1ΔF4/ΔF4 embryos (n=10). (k) Lymphoid tissue inducer cells (LTi, CD45+CD3eCD4+IL-7Ra+) were selectively absent in the fetal mesentery of the Ikzf1ΔF4/ΔF4 strain. (l) Inguinal lymph nodes and other peripheral lymph nodes (not shown) were selectively absent in adult Ikzf1ΔF4/ΔF4 mice, as visualized with Evan's Blue. One representative of two or more experiments are shown (k,l)
Figure 4
Figure 4
Differential DNA binding of Ikaros in Ikzf1ΔF1/ΔF1 and Ikzf1ΔF4/ΔF4 thymocytes. (a-d) Ikaros ChIP-Seq was performed with thymocytes from 4-week-old wild-type and mutant mice. UCSC genome browser tracks are shown for four genomic regions that contain repetitive Ikaros recognition motifs, thereby yielding unusually strong ChIP-Seq peaks. Examples of interactions that do not depend on neither finger 1 nor finger 4 (Snx29), that depend on finger 1 (Nr3c2), or that depend on finger 4 (Chr19 and Snx25) are shown. Input genomic DNA was sequenced as a negative control. The repetitive genomic sequences are shown below the tracks, with putative Ikaros recognition sites in yellow (GGGAA or GGAA) and green (TTCC). Position weight matrices of the putative Ikaros recognition sequences that were repeated are shown at the bottom, with the zinc fingers predicted to contact or lie in close proximity to the nucleotides indicated (F1-F4). (e) Electrophoretic mobility shift assays were performed with oligonucleotide probes from three repetitive sequences and recombinant proteins containing Ikaros fingers 1-4, 1-3, or 2-4. Zinc fingers predicted to contact or lie in close proximity to the nucleotides flanking the GGGAA core recognition sequence are indicated at the bottom (F1 or F4). (f) Bioinformatics analysis of a wild-type Ikaros ChIP-Seq dataset revealed the localization of the core GGGAA sequence relative to the center of each ChIP-Seq peak.
Figure 5
Figure 5
Deregulation of distinct sets of genes in Ikzf1ΔF1/ΔF1 and Ikzf1ΔF4/ΔF4 DP thymocytes. (a) RNA-Seq was performed with FACS-sorted DP thymocyte mRNA from 4-week-old wild-type and mutant mice in duplicate. A Venn diagram shows the number of genes that exhibited increased or decreased mRNA levels of at least 3-fold (P≤0.001) in one of the mutant strains. Genes with RPKM≥4 in at least one of the six samples were included in this analysis. (b) A Venn diagram shows the number of genes from the RNA-Seq experiment described above that exhibited increased or decreased mRNA levels of at least 10-fold (P≤0.001) in one of the mutant strains. (c) Gene Ontology analysis of genes upregulated more than 3-fold (P≤0.001) in Ikzf1ΔF4/ΔF4 DP thymocytes. (d) The distribution of RPKMs (mRNA levels) is shown for all annotated genes (left) and all genes that were upregulated by at least 3-fold (P≤0.001) in Ikzf1ΔF1/ΔF1 (middle) or Ikzf1ΔF1/ΔF1 (right) mutant mice.
Figure 6
Figure 6
Ikzf1ΔF4/ΔF4 but not Ikzf1ΔF1/ΔF1 mice develop spontaneous thymic lymphoma. (a) Kaplan-Meier survival curves reveal premature death of Ikzf1ΔF4/ΔF4 mice due to thymic lymphoma. (b) Histochemical analysis with H&E stain of an Ikzf1ΔF4/ΔF4 mouse where the thymic lymphoma engulfed the heart (right panel) (n=3). A wild-type control is also shown (left panel). (c) Histochemical analysis with H&E stain of an Ikzf1ΔF4/ΔF4 mouse (right panel) revealed thymic lymphoma infiltrating the lung. (d) Variable CD4 and CD8 expression was observed in lymphomas that had metastasized to the spleen in three representative Ikzf1ΔF4/ΔF4 mice. Wild-type spleen cells are shown as a control. Figures c-d display one representative for each of six or more sick mice analyzed.
Figure 7
Figure 7
Selective synergy between BCR-ABL and the Ikzf1ΔF4/Δ F4 mutation in vitro and in vivo. (a) In vitro growth curves from one representative experiment (of three) are shown for bone marrow cells from wild-type, Ikzf1ΔF1/ΔF1, and Ikzf1ΔF4/ΔF4 mice transduced with a BCR-ABL-expressing retrovirus and grown under B-ALL culture conditions. (b) Kaplan-Meier survival curves are shown for irradiated recipient mice transplanted with 106 BCR-ABL-transduced bone marrow cells from wild-type (n=15), Ikzf1ΔF1/ΔF1 (n-8), and Ikzf1ΔF4/ΔF4 (n=14) mice. (c) RNA-Seq was performed with mRNA from sorted pro B cells (pro-B) and pre-BI+large pre-BII cells (pre-BI-IIL), as well as from BCR-ABL-transformed wild-type, Ikzf1ΔF1/ΔF1, and Ikzf1ΔF4/ΔF4 cells harvested at day 21 (d21) or day 28 (d28) in culture. Genes whose mRNA levels differed by 3-fold or more between any two samples among the 12 samples analyzed were clustered by k-means clustering. (d) Genes that were selectively upregulated or downregulated in BCR-ABL-transformed Ikzf1ΔF4/ΔF4 cells were identified by k-means clustering of the data sets from the six BCR-ABL-transformed cultures. Only the two clusters containing genes that were selectively upregulated or downregulated in the Ikzf1ΔF4/ΔF4 samples are shown. Expression data from pro B cells (pro-B) and pre-BI+large pre-BII cells (pre-BI-IIL) were aligned after the cluster analysis was completed. (e) One representative flow cytometry experiment out of three or more is shown for the in vitro cultures described in panel a. Cells are gated on YFP+ cells. (f) Reduced Il2ra (CD25) and increased Kit mRNA levels are shown from the RNA-seq data (n=2).

Comment in

  • Fingerprinting ikaros.
    Gounari F, Kee BL. Gounari F, et al. Nat Immunol. 2013 Oct;14(10):1034-5. doi: 10.1038/ni.2709. Nat Immunol. 2013. PMID: 24048134 Free PMC article.
  • Two versions of the Ikaros tale.
    Papatriantafyllou M. Papatriantafyllou M. Nat Rev Immunol. 2013 Nov;13(11):772-3. doi: 10.1038/nri3550. Epub 2013 Oct 7. Nat Rev Immunol. 2013. PMID: 24096335 No abstract available.

Similar articles

See all similar articles

Cited by 51 articles

See all "Cited by" articles

References

    1. Wang JH, et al. Selective defects in the development of the fetal and adult lymphoid system in mice with an Ikaros null mutation. Immunity. 1996;5:537–549. - PubMed
    1. Georgopoulos K. Haematopoietic cell-fate decisions, chromatin regulation and ikaros. Nat Rev Immunol. 2002;2:162–174. - PubMed
    1. Yoshida T, Ng SY, Georgopoulos K. Awakening lineage potential by Ikaros-mediated transcriptional priming. Curr Opin Immunol. 2010;22:154–160. - PMC - PubMed
    1. Yoshida T, Ng SY, Zuniga-Pflucker JC, Georgopoulos K. Early hematopoietic lineage restrictions directed by Ikaros. Nat Immunol. 2006;7:382–391. - PMC - PubMed
    1. Kirstetter P, Thomas M, Dierich A, Kastner P, Chan S. Ikaros is critical for B cell differentiation and function. Eur J Immunol. 2002;32:720–730. - PubMed

Publication types

MeSH terms

Substances

Associated data

Feedback