Targeted gene deletion demonstrates that the cell adhesion molecule ICAM-4 is critical for erythroblastic island formation

Abstract

Erythroid progenitors differentiate in erythroblastic islands, bone marrow niches composed of erythroblasts surrounding a central macrophage. Evidence suggests that within islands adhesive interactions regulate erythropoiesis and apoptosis. We are exploring whether erythroid intercellular adhesion molecule 4 (ICAM-4), an immunoglobulin superfamily member, participates in island formation. Earlier, we identified alpha(V) integrins as ICAM-4 counterreceptors. Because macrophages express alpha(V), ICAM-4 potentially mediates island attachments. To test this, we generated ICAM-4 knock-out mice and developed quantitative, live cell techniques for harvesting intact islands and for re-forming islands in vitro. We observed a 47% decrease in islands reconstituted from ICAM-4 null marrow compared to wild-type marrow. We also found a striking decrease in islands formed in vivo in knock-out mice. Further, peptides that block ICAM-4/alpha(V) adhesion produced a 53% to 57% decrease in reconstituted islands, strongly suggesting that ICAM-4 binding to macrophage alpha(V) functions in island integrity. Importantly, we documented that alpha(V) integrin is expressed in macrophages isolated from erythroblastic islands. Collectively, these data provide convincing evidence that ICAM-4 is critical in erythroblastic island formation via ICAM-4/alpha(V) adhesion and also demonstrate that the novel experimental strategies we developed will be valuable in exploring molecular mechanisms of erythroblastic island formation and their functional role in regulating erythropoiesis.

Figure 1.

Deletion of Icam4 gene. Targeting strategy for homologous recombination in ES cells. Restriction map of wild-type Icam4 allele (top), targeting vector (middle), and targeted allele (bottom). In the wild-type Icam4 allele the black boxes represent the 3 exons. In the targeting vector, hGh-N indicates human growth hormone-N gene; PGK, phosphoglycerate kinase; Neo, bacterial neomycin resistance gene.

Figure 2.

Targeted disruption of Icam4. (A) Southern blot analysis of NsiI-digested DNA derived from tail vein samples of offspring from heterozygous mating pair. Blot probed with 3′ probe (shown in Figure 1) depicts homozygous animals containing only a 5.2-kb band derived from the targeted allele and the neomycin cassette (lanes 1 and 2). Heterozygote possesses both the 5.2-kb band and the endogenous DNA migrating at 12.8 kb (lane 3). Wild-type animal contains only the endogenous 12.8-kb band (lane 4). (B) PCR analysis of tail gDNA. Primers binding to Icam4 exon 1 and exon 2 amplified a 528-bp fragment; primers binding to the Neo gene amplified a 381-bp fragment. Molecular weight markers (lane 1); gDNA from wild-type mouse generated a 528-bp fragment (lane 2); gDNA from heterozygote generated 528-bp and 381-bp fragments (lane 3); gDNA from homozygous mouse generated a 381-bp fragment (lane 4). (C) Western blot analysis of erythrocyte membranes. Equivalent amounts of erythrocyte membranes from wild-type and knock-out mice probed with antibody recognizing mouse ICAM-4 produced a band of appropriate size for ICAM-4 in wild-type membranes, which was absent in knock-out membranes. As a positive control, human erythrocyte membranes were probed with BS56, a well-characterized antibody to ICAM-4 and produced an immunoreactive band migrating at a similar molecular weight as the band observed in wild-type mouse erythrocyte membranes.

Figure 3.

Reconstituted erythroblastic islands. Bright field (A) and immunofluorescent standard (B) and confocal (C) micrographs of typical erythroblastic islands formed from single cell suspensions of MacGreen mouse bone marrow. Immunofluorescent micrographs of islands show cells stained for erythroid-specific marker GPA (Ter119; red), macrophage marker M-CSF receptor GFP transgene expression (green), and DNA (Hoechst 33342; blue). In the confocal image some of the cells appear blurred because they are not in the plane of focus. However, macrophage staining is apparent in various regions of the island. Reticulocytes, arrowheads; macrophage, arrows; bars represent 10 μm. (D) Histogram shows number of erythroblastic islands formed from 1 × 10⁵ single cells; n = 10. Results are shown as mean ± SD.

Figure 4.

Erythroblastic islands from ICAM-4 null and wild-type mouse bone marrow. (A) Islands reconstituted from B6,129 ICAM-4 null and wild-type mouse bone marrow cells. Histogram of number of erythroblastic islands formed from 1 × 10⁵ single cells obtained from wild-type (n = 10) and ICAM-4 null (n = 10). *P < .001 when compared to islands formed from wild-type marrow. (B) Erythroblastic islands formed in vivo in ICAM-4 null and wild-type mice. Histogram of number of erythroblastic islands obtained from wild type (n = 6) and ICAM-4 null (n = 6). *P < .001 when compared to islands formed from wild-type marrow. Results are shown as mean ± SD.

Figure 5.

Erythroblastic island formation in the presence and absence of peptides blocking adhesion of ICAM-4 to α_V integrins. Model of extracellular domain of ICAM-4 is shown revealing its solvent exposed surface in 3 orientations rotated 120° to each other. The region of ICAM-4 involved in adhesion to α_V integrins is shown in yellow and green. Yellow designates area of FVW peptide sequence and green depicts location of ATSR peptide sequence (A). Histogram of percentage of islands formed in the presence of 0.5 to 1.0 mM FWV peptide (n = 11), 1.0 to 2.0 mM FWV peptide (n = 16), and 2.0 to 3.0 mM FWV peptide (n = 8) compared to islands formed in media alone. *P = .02, **P < .001, ***P < .001 when compared to islands formed in media alone (B). Histogram of percentage of islands formed in the presence of 0.5 to 1.0 mM ATSR peptide (n = 3), 1.0 to 2.0 mM ATSR peptide (n = 8), and 2.0 to 3.0 mM ATSR peptide (n = 4) compared to islands formed in media alone. *P = .1, **P = .004, ***P < .001 when compared to islands formed in media alone (C). Results are shown as mean ± SD.

Figure 6.

Expression of F4/80 and α_V integrin on central macrophages. Central macrophages isolated from erythroblastic islands were analyzed by immunofluorescence microscopy in double antibody label experiments using macrophage-specific probe antibody F4/80 (green), anti-CD51 (red), and DNA (Hoechst 33342; blue). Two representative cells are shown, one in the top 3 panels and the other in the bottom 3 panels. Merged image signals showed colocalization (yellow) of the 2 antibody markers indicating that central macrophages isolated from erythroblastic islands express the α_V integrin subunit. Bar represents 10 μm.