A point mutation in the murine Hem1 gene reveals an essential role for Hematopoietic protein 1 in lymphopoiesis and innate immunity

Abstract

Hem1 (Hematopoietic protein 1) is a hematopoietic cell-specific member of the Hem family of cytoplasmic adaptor proteins. Orthologues of Hem1 in Dictyostelium discoideum, Drosophila melanogaster, and Caenorhabditis elegans are essential for cytoskeletal reorganization, embryonic cell migration, and morphogenesis. However, the in vivo functions of mammalian Hem1 are not known. Using a chemical mutagenesis strategy in mice to identify novel genes involved in immune cell functions, we positionally cloned a nonsense mutation in the Hem1 gene. Hem1 deficiency results in defective F-actin polymerization and actin capping in lymphocytes and neutrophils caused by loss of the Rac-controlled actin-regulatory WAVE protein complex. T cell development is disrupted in Hem1-deficient mice at the CD4(-)CD8(-) (double negative) to CD4(+)CD8(+) (double positive) cell stages, whereas T cell activation and adhesion are impaired. Hem1-deficient neutrophils fail to migrate in response to chemotactic agents and are deficient in their ability to phagocytose bacteria. Remarkably, some Rac-dependent functions, such as Th1 differentiation and nuclear factor kappaB (NF-kappaB)-dependent transcription of proinflammatory cytokines proceed normally in Hem1-deficient mice, whereas the production of Th17 cells are enhanced. These results demonstrate that Hem1 is essential for hematopoietic cell development, function, and homeostasis by controlling a distinct pathway leading to cytoskeletal reorganization, whereas NF-kappaB-dependent transcription proceeds independently of Hem1 and F-actin polymerization.

Figure 1.

Generation and identification of Hem1 mutant mice by ENU mutagenesis. (A) Recessive mutant (NTB.1) obtained from a G3 pedigree was found to display a high percentage of CD11b⁺ and a low percentage of B220⁺ cells in peripheral blood. Blood cells were stained with fluorescent α-CD11b and α-B220 followed by flow cytometry. The percentage of CD11b- and B220-positive cells in blood is shown, where each symbol represents an individual mouse and each color (column) represents a unique G3 family pedigree, which is denoted by numbers on the x axis. (B) Genomic structure of the mouse Hem1 gene. The location of a C–T conversion induced by ENU mutagenesis in the 13th exon of the Hem1 gene is shown. (C) A representative immunoblot of two experiments showing the absence of Hem1 protein in hematopoietic tissues of Hem1^−/− mice compared with normal littermates. Whole cell lysates were generated from thymocytes (Th), splenocytes (Sp), and BM cells. TFhem1, positive control from pCDNA3.1-Hem1–transfected 293T cells; −C, negative control from empty pcDNA3.1–transfected 293T cells.

Figure 2.

Loss of Hem1 results in lymphopenia, neutrophilia, and induces tissue-specific pathology. (A–C) CBCs and differential cell counts were performed on peripheral blood from Hem1^−/− and normal littermate control mice (WT). Shown are the mean ± SEM from four mice/group. WBC, total WBC; RBC, total RBC; HGB, hemoglobin; HCT, hematocrit; MCV, mean corpuscular volume; MCH, mean corpuscular hemoglobin, MCHC, mean corpuscular hemoglobin concentration; Baso, basophils; Eos, eosinophils; Poly, neutrophils; Lymph, lymphocytes; Mono, monocytes. An asterisk denotes significant differences between mutant and WT littermates. WBC, P = 0.06; hemoglobin, P = 0.05; hematocrit, P = 0.02; mean corpuscular volume, P = 0.00005; mean corpuscular hemoglobin, P = 0.004; mean corpuscular hemoglobin concentration, P = 0.05; neutrophils, P = 0.03; lymphocytes, P = 0.00006. (D) Hem1^−/− mice exhibit increased immature erythrocytes (reticulocytes) in blood. The number of reticulocytes per microliter of blood are represented (mean ± SEM for four animals per group). (E) Representative blood smears of four mice/group of erythrocytes from WT and Hem1^−/− mice were imaged under 400× magnification. Hem1^−/− erythrocytes show morphological abnormalities such as anisocytosis (variation in size), echinocytes or schistocytes (black arrow), keratocytes (red arrows), and dacryocytes (blue arrow). Bars, 5 μm. (F) Hem1^−/− erythrocytes exhibit increased fragility. RBCs from WT or Hem1^−/− mice were added to decreasing (hypotonic) solutions of NaCl. After centrifugation, the OD of the supernatant at 540 nm was assessed to determine the amount of hemoglobin released into the supernatant. Shown is a bar graph depicting the mean ± SEM of the percent lysis of WT (n = 5) and Hem1^−/− (n = 5) RBCs. (G and H) Hematoxylin and eosin staining was performed in tissue sections from liver and kidney. Images are representative of six mice/group. (G) At necropsy, Hem1^−/− mice commonly exhibit whitish liver margins that are histologically represented as dense accumulations of amyloid interspersed with mineral and fibrous tissue. 10× power; bar, 100 μm. (H) Renal glomeruli from Hem1^−/− mice frequently have an increase in mesangial matrix and cellularity consistent with membranoproliferative glomerulopathy. 40× power; bars, 50 μm.

Figure 3.

Hem1–deficient neutrophils exhibit defective migration, phagocytosis, and F-actin polymerization. (A and B) Total BM and splenocytes were stained with fluorescent-conjugated α-MAC1 and α-GR1 or α-Ter119 and α-CD61 antibodies followed by flow cytometry. The percentage of monocyte (Mac1⁺), granulocyte (GR1⁺), erythroid precursor (Ter119⁺), and megakaryocyte (CD61⁺) populations are shown in the representative histogram of eight mice per group. (C) BM-derived neutrophils were stimulated for 45 min with different concentrations of MLP to determine chemotactic ability across a membrane. The number of migrated cells (at 400×) was counted for three random fields and the mean ± SEM values are displayed on y axis. P-values are shown. Data are representative of three independent experiments. (D) 10⁵ neutrophils from WT and Hem1^−/− BM were cultured with fluorescent 1-μm diameter microspheres for 5 min. Phagocytosis was measured by flow cytometry. The single-parameter histograms are representative of four experiments. (E) Peritoneal macrophages were incubated with fluorescent-conjugated E. coli for 0, 2, and 4 h. Phagocytosis was measured by flow cytometry. Bars represent the mean ± SEM from three independent experiments. P-values are shown. (F) BM-derived neutrophils were stimulated with 10 nM MLP. F-actin polymerization was determined at the time points indicated by flow cytometry after intracellular Alexa Fluor 488 phalloidin staining. The mean ± SEM of three animals per group (120-s time point) are shown in the bar graph. (G) BM-derived neutrophils from WT and Hem1^−/− mice were stimulated with FMLP for 15 s. F-actin polymerization and capping formation were determined by confocal microscopy using Alexa Fluor 488 phalloidin (green) and DAPI (blue) staining. Bar, ∼12–15 μm.

Figure 4.

T cell development is severely impaired in Hem1-deficient mice. Total thymocytes (A) and splenocytes (B) from Hem1^−/− and WT littermate mice were stained with fluorescent-conjugated α-CD4, α-CD8, α-CD3ε, α-CD25, α-CD44, α-B220, α-GR1, α–TCR-γδ, α–TCR-β, and intracellular α-Foxp3 and analyzed by flow cytometry. Representative histograms of eight mice per group of cells that fall within a forward scatter (FSC) and side scatter (SSC) lymphocyte gate are shown. Total thymocyte and splenocyte number were multiplied by the percentage of cells that fell within each quadrant to obtain the total number of cells within each developmental subset (right). Cells negative for B220, GR1, CD4, and CD8 were analyzed with CD44 and CD25 staining to determine the DN1-DN4 fractions. Bars represent the mean ± SEM from eight mice. P-values are shown. The high side scatter cells in Hem1^−/− spleens are myeloid and progenitors cells.

Figure 5.

Impaired T and B cell development in Hem1 mutant mice is cell autonomous. Lin⁻ cells from Hem1^−/− (CD45.2), WT (CD45.1), or mixed WT/Hem1^−/− (1:1) were injected i.v. into Rag2^−/−γc^−/− recipient mice. Thymocytes, BM, and splenocytes were harvested from the recipient mice 8–15 wk after transplantation and analyzed by flow cytometry using fluorescent-conjugated α-CD4, α-CD8, α-CD45.1, α-B220, α-IgM, α-CD45.2, and α–TCR-β. (A) Representative dot-plot histograms of three independent experiments showing CD4, CD8, and TCR-β expression in conjunction with CD45.2 markers on thymocytes and splenocytes from the indicated chimeric mice. (B) Representative dot-plot histograms showing B220 and IgM, in conjunction with CD45.2 markers on BM and spleen cells from the indicated chimeric mice.

Figure 6.

Loss of Hem1 results in enhanced IL-17 production but normal IL-2 and IFN-γ production (A) Splenocytes from WT and Hem1^−/− mice were stimulated with different doses of α-CD3/α-CD28 for 18 h. Cells were stained with fluorescent-conjugated α-CD4, α-CD8, and α-CD69 and analyzed by flow cytometry. Representative dot-plot histograms of three independent experiments are shown. (B) Purified WT and Hem1^−/− T cells were labeled with CFSE dye and were stimulated with α-CD3/CD28 in presence of IL-2 for 4 d. Cells were stained with α-CD4, α-CD8, and the vital dye TO-PRO-3. Cell divisions were determined by flow cytometry on live-gated (TO-PRO-3 negative) CD4⁺ or CD8⁺ cells. Representative single-parameter histograms of three independent experiments are shown (each division is separated by vertical lines). Because of the paucity of peripheral T cells in Hem1^−/− mice, all purified T cells from Hem1^−/− mice are used for the stimulated condition. (C) Splenic T cells were stimulated with α-CD3/α-CD28 for 72 h. Culture media were harvested and IL-2, IFN-γ, IL-17, IL-6, and TNF-α production were measured by ELISA. Error bars represent the mean ± SEM of three mice per group. P-values are shown. (D) FACs-sorted CD4⁺CD25⁻CD44^lowCD62L⁺ naive T cells from WT and Hem1^−/− mice were cultured for 5 d under the indicated conditions (Th17), followed by intracellular staining for IL-17 and IFN-γ. A representative dot-plot histogram of three independent experiments is shown. (E) After 5-d culture under Th1 or Th17 conditions, cells were restimulated with α-CD3 and secreted cytokines were measured by ELISA. Data were obtained from two independent experiments. Error bars show the mean ± SEM.

Figure 7.

Loss of Hem1 results in defective F-actin polymerization and actin cap formation, impaired cell adhesion, and loss of WAVE complex components. (A) Thymocytes from WT and Hem1^−/− mice were stimulated with α-CD3/α-CD28 for 24 h, followed by PMA for 15 min. Cells were stained with Alexa Fluor 488 phalloidin, and F-actin polymerization was measured by flow cytometry. A single-parameter histogram, representative of three experiments, is shown. (B) Hem1^−/− T cells display defective TCR-induced capping formation and adhesion. Thymocytes or T cells were incubated with hamster IgG- or α–CD3-ε-coated beads for 30 min and then stained with Alexa Fluor 488 phalloidin and analyzed by confocal microscopy. (B, left) A representative image of capping formation between the T cells or thymocytes and beads. Bars, 5 μm. (B, right) The percentages of positive cell–bead contacts are graphed. Results are representative of two independent experiments. The mean ± SEM is shown. (C) Defective adhesion capacity of Hem1^−/− T cells. Thymocytes and T cells were incubated on fibronectin-coated plates and stimulated with α-CD3 for 30 min. Results are representative of two independent experiments. The mean + SEM is shown. (D–F) Loss of WAVE components in Hem1^−/− hematopoietic cells. Thymocytes and T cells from WT and Hem1^−/− mice were stimulated on α-CD3–coated plates for the indicated time points, and neutrophils were stimulated with MLP for 0 and 15 s. Cells were lysed and subjected to Western blot analysis using antibodies against Hem1, Lck, Zap70, Wave2, Sra1, Abi1, Abi2, β-Actin, phosphotyrosine, total Erk, and phospho-Erk. To assess the Pak1 binding activity of activated Rac proteins, cell extracts from either WT or Hem^−/− mice were incubated as described in Materials and methods. Active Rac proteins were analyzed by Western blotting with α-Rac antibodies. The relative ratio of Rac1 in the immunoprecipitation relative to input is shown below each blot. There is no significant difference in Rac1 activity in Hem1^−/− versus WT samples (time 0, P = 0.2; 15 min, P = 0.09). Blots are representative of three independent experiments.

Figure 8.

Model of Hem1 functions in F-actin polymerization and hematopoietic cell biology. In mammalian cells, the WAVE complex consists of multiple subunits including WAVE (1, 2, or 3), Abi (1 or 2), Hem (Hem2 [also known as Nap 1] or Hem1), and Sra1 (19). The WAVE complex alone does not induce actin polymerization but is stimulated in response to Rac GTP, which is activated by many receptors, and is brought into the WAVE complex via associations with Sra1 and Hem. Hem1 is the essential Hem family member in hematopoietic cells. Association of Rac GTP with WAVE may lead to WAVE-induced activation or relocalization of the actin-regulatory protein complex (not depicted), resulting in F-actin polymerization. Hem1 deficiency results in loss of WAVE complex proteins, thus specifically blocking processes dependent on F-actin polymerization, whereas Rac GTP activation of NF-κB–dependent transcription proceeds normally.