Allergic asthma is a significant health burden in western countries, and continues to increase in prevalence. Th2 cells contribute to the development of disease through release of the cytokines IL-4, IL-5, and IL-13, resulting in increased airway eosinophils and mucus hypersecretion. The molecular mechanisms behind the disease pathology remain largely unknown. In this study we investigated a potential regulatory role for the Hox5 gene family, Hoxa5, Hoxb5, and Hoxc5, genes known to be important in lung development within mesenchymal cell populations. We found that Hox5-mutant mice show exacerbated pathology compared with wild-type controls in a chronic allergen model, with an increased Th2 response and exacerbated lung tissue pathology. Bone marrow chimera experiments indicated that the observed enhanced pathology was mediated by immune cell function independent of mesenchymal cell Hox5 family function. Examination of T cells grown in Th2 polarizing conditions showed increased proliferation, enhanced Gata3 expression, and elevated production of IL-4, IL-5, and IL-13 in Hox5-deficient T cells compared with wild-type controls. Overexpression of FLAG-tagged HOX5 proteins in Jurkat cells demonstrated HOX5 binding to the Gata3 locus and decreased Gata3 and IL-4 expression, supporting a role for HOX5 proteins in direct transcriptional control of Th2 development. These results reveal a novel role for Hox5 genes as developmental regulators of Th2 immune cell function that demonstrates a redeployment of mesenchyme-associated developmental genes.

This article is featured in In This Issue, p.379

Allergic asthma is characterized by inflammation and bronchiole constriction, which over time leads to adverse remodeling of the airways. In the United States, 1 in 11 children and 1 in 12 adults has asthma, resulting in over 3000 deaths each year with healthcare costs exceeding $50 billion. Annually, there are nearly 500,000 asthma-related hospitalizations and nearly two million visits to emergency departments due to asthma, with asthma-related doctor visits approaching nine million (1). Treatment is generally limited to anti-inflammatories and inhaled corticosteroids to reduce both the severity and frequency of episodes, without providing a short- or long-term cure. Allergic asthma is characterized by a Th2 response, which is associated with increased levels of the cytokines IL-4, IL-5, and IL-13 (25). The Th2 phenotype is driven by IL-4, which binds to the IL-4 receptor composed of two subunits, IL-4Rα and IL-2Rγ (6, 7). Binding of IL-4 leads to phosphorylation of STAT6, which then homodimerizes and translocates into the nucleus to drive the production of the Th2-associated transcription factors, especially GATA3 (8, 9). GATA3 in turn binds to the promoter regions of Il4, Il5, and Il13 genes. Our laboratory and others have found that characteristic clinical features of allergic asthma can be replicated in mice following sensitization and challenge with allergen (1012). Although the disease pathologies have been primarily related to immune-associated proteins, the regulation of the Th2 responses is incompletely understood.

Hox genes encode transcriptional regulatory factors that are primarily necessary for patterning the body axis and many organ systems (1315). All mammals have four Hox clusters (HoxA, B, C, and D), divided into 13 paralogous groups. Each group has between two and four paralogs, with a total of 39 genes (16). Hox genes within each cluster are tightly linked and expressed sequentially from 3′ to 5′. The genes near the 3′ end are generally expressed earlier during development at anterior segments of the embryo, whereas the genes closer to the 5′ end are expressed later and in more posterior segments of the embryo. All Hox genes contain a homeodomain, which is a helix-turn-helix DNA-binding domain composed of 60 aa (1719). This domain is highly conserved among species and preserves function (20). The three members of the Hox5 paralogous group have been shown to be important in the development of the lung, especially Hoxa5 (21). Mice with complete loss in the Hoxa5 gene are minimally viable, with <20% of offspring surviving past birth due to defects in the respiratory tract (22), with increased alveolar space, and increased numbers of goblet cells in the trachea and primary bronchi (23). Within a paralogous group, there is often significant functional redundancy that allows loss-of-function of one gene to be partially or completely compensated by the remaining members of that group. In the Hox5 family the Hoxb5 and Hoxc5 genes cannot completely compensate for a loss of function of Hoxa5, with mice lacking all three members of the Hox5 paralogous family having severely affected lungs, and are not viable (24).

The role of Hox genes in immune cell differentiation and mature immune cell function has not been explored. The majority of research has been focused on Hoxa9 in leukemia, where translocation of the mixed lineage leukemia gene leads to overexpression of Hoxa9 (25). A role for Hox genes in hematopoiesis has also been suggested including development of erythrocytes, monocytes, and lymphocytes (26, 27). There is less information about the expression of Hox genes in mature immune cells. Expression of various Hox genes, including the Hox5 family, has been detected in both naive and stimulated T lymphocytes (28, 29). However, the function of these genes in T cells remains unexplored. In this study, we examine the role of Hox5 genes in the lungs during the development of allergic airway disease. We further explore Hox5 gene function in Th2 cells, a major driver of allergic disease, and we report a novel role for these genes in T cell differentiation.

Hox5AabbCc mice and wild-type (WT) controls were bred at the University of Michigan. GATA3-GFP reporter mice (Gata3g/+) on the C57BL/6 background were provided by Dr. J. D. Engel (30). The University Committee of Use and Care of Animals (University of Michigan) approved all animal experimental protocols, and experiments were conducted according to the guidelines provided by the committee. Mice 8–12-wk old mice were sensitized s.c. and i.p. with a 1:1 mixture of cockroach Ag (CRA) and IFA. Mice were challenged with 15 μl of CRA (Greer, Lenoir, NC) intranasally at days 14, 18, 22, and 26, then with 50 μl of CRA intratracheally on days 30 and 32. Animals were sacrificed either 6 h after each challenge or 24 h after the final challenge.

RNA was extracted using TRIzol reagent (Invitrogen, Carlsbad, CA). Lung tissue was homogenized in TRIzol prior to extraction. cDNA was synthesized using murine leukemia virus reverse transcriptase (Applied Biosystems, Foster City, CA) and incubated at 37°C for 1 h, followed by 95°C for 5 min to stop the reaction. Real-time quantitative PCR (qPCR) was multiplexed using Taqman primers with a FAM-conjugated probe and GAPDH with a VIC-conjugated probe (Applied Biosystems) to measure transcription of murine Hoxa5, Hoxb5, Hoxc5, and Gata3 and human Gata3 and Il4, IL5, and IL13. Fold change was quantified using the 2−ΔΔ cycle threshold (CT) method or expression was measured relative to GAPDH using the 2−ΔCT method. Custom primers were designed to measure Muc5ac and Gob5 mRNA levels. All reactions were run on an ABI Prism 7500 Sequence Detection System or ViiA 7 Real Time PCR System (both from Applied Biosystems).

Lungs were removed after the final allergen challenge. The large left lobe of each lung was inflated by injection with 4% formaldehyde. Lungs were embedded in paraffin, and 5 μm sections were sectioned and stained with H&E to visualize inflammatory cells or with periodic acid-Schiff stain to visualize mucus production.

For section immunohistochemistry, adult lungs were dissected, processed, embedded in paraffin, and sectioned as described above for histopathology. Tissue sections were stained with the following Abs and dilutions: rabbit anti-CCSP (1:1000; Seven Hills Bioreagents); mouse anti-acetylated tubulin (1:1000; Sigma-Aldrich); and mouse anti-Muc5ac (1:75; Novocastra). Primary Abs were detected by using either fluorescently conjugated (Alexa Fluor; Invitrogen) or alkaline phosphatase conjugated (Jackson ImmunoResearch) secondary Abs (1:500).

Lungs and mediastinal lymph nodes were removed and single cells were isolated by enzymatic digestion with 1 mg/ml collagenase A (Roche, Indianapolis, IN) and 20 U/ml DNaseI (Sigma, St. Louis, MO) in RPMI 1640 containing 10% FCS. Cells were resuspended in PBS and live cells were identified using LIVE/DEAD Fixable Yellow Dead Cell Stain kit (Thermo Fisher Scientific, Waltham, MA), then washed and resuspended in PBS with 1% FCS and Fc receptors were blocked with purified anti-CD16/32 (clone 93; BioLegend, San Diego, CA). Surface markers were identified using Abs (clones) against the following Ags, all from BioLegend: CD3ε (17A2), CD4 (RM4-5), CD69 (H1.2F3), CD8 (53–6.7), and Gr-1 (RB6-8C5). SiglecF (E50-2440) was purchased from BD Biosciences (San Jose, CA). Intracellular staining was performed using an anti-IL-4 Ab (11B11; eBioscience, San Diego, CA) and an anti-Gata3 Ab (TWAJ; eBioscience). Total T cells were gated on CD3+, followed by CD4+ and CD8+ single positive, CD4+CD69+ to gate activated T cells, or CD4+IL-4+ to gate IL-4–producing T cells and CD4+Gata3+ to gate Gata3-producing T cells.

Mediastinal lymph nodes were removed and single cells were isolated by enzymatic digestion. Next, 5 × 105 cells were plated in 200 μl of complete medium (RPMI 1640 supplemented with 10% FCS, l-glutamine, penicillin/streptomycin, nonessential amino acids, sodium pyruvate, 2-ME), and were restimulated with CRA for 48 h. Supernatants were collected and levels of the cytokines IL-4, IL-5, IL-13, IL-17, and IFN-γ were measured by BioPlex assay (Bio-Rad, Hercules, CA).

Hox5AabbCc mice and WT controls were irradiated twice with 500 rad using a [137Cs] irradiator, with 4 h between each irradiation. Four hours after the second irradiation, mice were injected intravenously with 5 × 106 bone marrow cells from either Hox5AabbCc or WT donor mice. Bone marrow cells were allowed to engraft for 12 wk before mice were used for experimental models.

Cells were removed from spleens of naive mice and a single-cell suspension was made by homogenization. Following lysis of RBCs, CD4+ T cells were isolated using a negative selection kit (Miltenyi Biotec, San Diego, CA). Cells were cultured in complete medium in 96-well plates that had been coated with 2.5 μg/ml anti-CD3 for 2 h at 37°C. For primary skewing, T cells were activated with 3 μg/ml anti-CD3 for 48 h or 5 d to drive Th0 differentiation. In addition to soluble anti-CD28, the following skewing conditions were used: Th1 cells: rIL-12 (100 ng/ml) and anti-IL-4 (10 μg/ml); Th2 cells: rIL-4 (100 ng/ml) and anti-IL-12 (10 μg/ml), and anti-IFN-γ (10 μg/ml); and Th17 cells: rIL-6 (100 ng/ml), rTGF-β (20 ng/ml), anti-IL-4 (10 μg/ml), anti-IL-12 (10 μg/ml), and anti-IFN-γ (10 μg/ml). For secondary skewing, cells were rested in complete medium for 3 d, then restimulated with plate-bound anti-CD3 and soluble anti-CD28 for 48 h. For skewing of Jurkat cells, the cells were expanded in RPMI 1640 with 10% FCS and penicillin/streptomycin. Following transfection (see the following section), cells were cultured on plate-bound anti-CD3 in the presence of soluble anti-CD28, anti-IL-12, and recombinant human IL-4. All Abs were purchased from eBioscience, and all recombinant cytokines were purchased from R&D Systems (Minneapolis, MN).

Jurkat cells were grown in DMEM containing 10% FCS and penicillin/streptomycin. Cells were then transfected with Lipofectamine 2000 (Invitrogen) following the manufacturer’s instructions. Briefly, cells were plated at 1 × 106 cells/ml in 5 ml DMEM with 10% FCS. Lipofectamine 2000 was incubated in Opti-MEM (Thermo Fisher Scientific) for 5 min at room temperature, then 8 μg of DNA was added for an additional 20 min. The complexes were then added to the cells and incubated at 37°C for 24 h.

Chromatin immunoprecipitation (ChIP) was performed using an assay kit (Millipore, Billerica, MA) with minor modifications. Briefly, cells were fixed in 1% formaldehyde, then lysed in SDS buffer. Cells were then sonicated using a Branson Digital Sonifier 450 (VWR, West Chester, PA) to create 200–1000 bp fragments. The lysate was clarified by centrifugation, and 5% of the supernatant was saved to measure the input DNA. The remaining chromatin was incubated with 1 μg of anti-FLAG Ab (Abcam, Cambridge, MA) or control IgG (Millipore) and incubated at 4°C with rotation overnight. Immune complexes were precipitated with salmon sperm DNA/protein A agarose beads. Crosslinking was reversed by incubation at 65°C and samples were treated with proteinase K. DNA was purified by phenol:chloroform:isoamyl alcohol separation and ethanol precipitation. Primers for putative STAT6 binding sites on the Gata3 gene were designed and DNA was amplified by qPCR using SYBR Green buffer (Applied Biosystems). The primers for amplifying DNA are as follows: STAT6 1: Forward 5′-GCGGAGAAGCATTTTTCATT-3′, Reverse 5′-GTTTCTCCTGAGCCCACTTG-3′; STAT6 2: Forward 5′-GCGTCCTCTACCCTGCTGT-3′, Reverse 5′-CCCCTAAGACACAAAATTCCA-3′; and STAT6 3: Forward 5′-TGTGTGGATTTGCACTTGCT-3′, Reverse 5′ ATAACGTAAGCCCCATGCAC-3′.

Results are expressed as mean ± SE. Statistical significance was measured by one-way or two-way ANOVA as appropriate, followed by post hoc Student t test as appropriate. A p value <0.05 was considered significant.

A previous study has shown that Hox5 genes (Hoxa5, Hoxb5, Hoxc5) play redundant roles in lung development and that the most severe phenotypes result from the loss of function of all three Hox5 genes (31). Gene expression analysis by qPCR revealed that all three Hox5 genes continue to be expressed during adult stages at levels comparable to those observed during embryonic lung development (Fig. 1A). Because Hox5 triple mutants die shortly after birth, we analyzed surviving compound loss-of-function Hox5 mutants of the genotype Hoxa5+/−;Hoxb5−/−;Hoxc5+/− (Hox5AabbCc) for possible lung defects at adult stages. Immunohistochemistry analyses demonstrated that the proximal airway of adult Hox5AabbCc mutants have fewer acetylated tubulin+ ciliated cells compared with controls (Fig. 1B). Although muc5ac+ goblet cells are normally not observed in the proximal airway of adult WT animals, these cells are present in Hox5AabbCc mutants (Fig. 1B). Thus, Hox5 genes continue to be expressed in the adult lung and play important roles in the proper distribution of the epithelial cells of the proximal airway, a mechanism that may impact lung function and airway disease.

FIGURE 1.

Hox5 gene expression and proximal airway cell distribution in the adult lung. (A) Lung expression of Hoxa5, Hoxb5, and Hoxc5 was determined at different stages of embryonic development and at 8 wk of age by qPCR. Kidney samples were included as a negative control. (B) Sections of adult lung tissue were stained with Muc5ac to measure goblet cells and acetylated tubulin to measure ciliated cells. n = 5 mice per group.

FIGURE 1.

Hox5 gene expression and proximal airway cell distribution in the adult lung. (A) Lung expression of Hoxa5, Hoxb5, and Hoxc5 was determined at different stages of embryonic development and at 8 wk of age by qPCR. Kidney samples were included as a negative control. (B) Sections of adult lung tissue were stained with Muc5ac to measure goblet cells and acetylated tubulin to measure ciliated cells. n = 5 mice per group.

Close modal

Because Hox5 genes are critical for lung development and continue to be expressed in the adult lung, we hypothesized that they would also be redeployed during chronic disease and are important in the development of allergic airway disease. Our studies initially determined whether Hox5 gene expression was altered in the lungs of mice exposed to allergens. Mice were sensitized by i.p. and s.c. injection of CRA, and challenged into the airway at days 14, 18, 22, 26, 30, and 32 (Fig. 2A). Mice were sacrificed 6 h after each challenge, and RNA was extracted from total lung tissue. We found that Hoxa5, Hoxb5, and Hoxc5 genes were differentially upregulated following allergen challenge, with higher levels of gene expression after the final intratracheal challenges, with Hoxc5 especially prominent (Fig. 2B). The upregulation of Hox5 gene expression suggests that these genes play a role during the development of disease.

FIGURE 2.

Hox5 gene expression is upregulated following allergic challenge. (A) Mice were sensitized with CRA mixed with IFA by i.p. and s.c. injections, then challenged with four intranasal and two intratracheal instillments of CRA. (B) Mice were sacrificed 6 h after each challenge, and total RNA was extracted from the lungs. Expression levels of Hox5 genes were assessed using qPCR. Results are representative of one of two independent experiments with five mice per group.

FIGURE 2.

Hox5 gene expression is upregulated following allergic challenge. (A) Mice were sensitized with CRA mixed with IFA by i.p. and s.c. injections, then challenged with four intranasal and two intratracheal instillments of CRA. (B) Mice were sacrificed 6 h after each challenge, and total RNA was extracted from the lungs. Expression levels of Hox5 genes were assessed using qPCR. Results are representative of one of two independent experiments with five mice per group.

Close modal

To determine a potential role for Hox5 genes in the development of chronic allergic disease, we used Hox5AabbCc mice with the CRA model shown in Fig. 2A, and sacrificed mice 24 h following the final challenge. H&E staining of sections was used to visualize inflammation in WT and Hox5AabbCc mice, whereas mucus was visualized by periodic acid-Schiff staining (Fig. 3A). The observed increase in mucus production was verified by PCR analysis of lung tissue, which showed an increase in the mucus-associated genes Muc5ac and Gob5 (Fig. 3B). When treated with allergen, the lungs of these mice demonstrate significantly increased infiltration of inflammatory cells and increased mucus production compared with WT controls. Flow cytometric analyses of the inflammatory cells show a significant increase in the number of CD4+ cells in the lungs, with no change in CD8+ cells (Fig. 3C). As Th2 cells drive allergic inflammation in the lungs, we measured the transcription factor Gata3 in T cells from the lung tissue, and found increased numbers of Gata3+CD4+ T cells in Hox5AabbCc mice (Fig. 3D). In addition, other inflammatory cells associated with allergic airway disease including neutrophils and eosinophils were significantly increased (Fig. 3E). Cells were removed from the mesenteric lymph nodes (MLN) and restimulated ex-vivo with CRA, resulting in increased Th2 cytokines IL-4, IL-5, and IL-13 in the Hox5AabbCc mice, with no change in IFN-γ production (Fig. 3F). Furthermore, the number of IL-4–producing CD4+ T cells was greater in the MLN from Hox5AabbCc mice compared with WT controls, as was the level of IL-4 production from individual T cells as measured by mean fluorescence intensity (Fig. 3G). Together, these data suggest that the Hox5 genes are important in controlling the development of pathology in allergic airway disease.

FIGURE 3.

Hox5AabbCc mice have increased Th2 inflammation following allergic challenge. (A) Hox5AabbCc mice and WT controls were treated with CRA as in Fig. 1A. Lung sections were stained with H&E to visualize inflammation and periodic acid-Schiff to visualize mucus. (B) RNA was extracted from lung tissue and expression of the mucus-associated genes Muc5ac and Gob5 was measured by qPCR. (C) Lung tissue was dissociated using Collagenase A and DNaseI. CD4+ and CD8+ T cells in the lungs were characterized by flow cytometry. (D) Expression of Gata3 in lung CD4+ T cells was measured by intracellular cytokine staining. (E) Neutrophil and eosinophil numbers were measured in lung tissue by flow cytometry following tissue dissociation. (F) Mediastinal lymph nodes were removed and single cells were restimulated with CRA. Cytokine production was measured by BioPlex after 48 h. (G) The number of IL-4–producing CD4+ T cells and the mean fluorescence intensity were determined by intracellular flow cytometry. Results are from one of three independent experiments. n = 4–5 mice per group. *p < 0.05, **p < 0.01, ***p < 0.001 compared with WT.

FIGURE 3.

Hox5AabbCc mice have increased Th2 inflammation following allergic challenge. (A) Hox5AabbCc mice and WT controls were treated with CRA as in Fig. 1A. Lung sections were stained with H&E to visualize inflammation and periodic acid-Schiff to visualize mucus. (B) RNA was extracted from lung tissue and expression of the mucus-associated genes Muc5ac and Gob5 was measured by qPCR. (C) Lung tissue was dissociated using Collagenase A and DNaseI. CD4+ and CD8+ T cells in the lungs were characterized by flow cytometry. (D) Expression of Gata3 in lung CD4+ T cells was measured by intracellular cytokine staining. (E) Neutrophil and eosinophil numbers were measured in lung tissue by flow cytometry following tissue dissociation. (F) Mediastinal lymph nodes were removed and single cells were restimulated with CRA. Cytokine production was measured by BioPlex after 48 h. (G) The number of IL-4–producing CD4+ T cells and the mean fluorescence intensity were determined by intracellular flow cytometry. Results are from one of three independent experiments. n = 4–5 mice per group. *p < 0.05, **p < 0.01, ***p < 0.001 compared with WT.

Close modal

The alteration of the immune response in the above studies suggests that Hox5 paralogous genes may be contributing to the increased inflammation observed in the Hox5AabbCc mice. To address this, we performed reciprocal bone marrow transplantation experiments, in which bone marrow cells from Hox5AabbCc mice were transplanted into irradiated WT mice, and WT bone marrow cells were transplanted into Hox5AabbCc mice. Following a 12-wk engraftment period, these mice were then sensitized and challenged with CRA as in Fig. 1A. H&E staining of lung sections demonstrated that those mice that received Hox5AabbCc bone marrow cells had increased inflammation over mice that received WT bone marrow, regardless of whether the cells were transplanted into WT or Hox5AabbCc mice (Fig. 4A). Furthermore, the increased inflammation again corresponded with an increase in CD4+ T cells, including activated CD4+CD69+ cells (Fig. 4B), whereas no changes were seen in the numbers of CD8+ T cells or Ag-presenting dendritic cells (data not shown). Restimulation of MLN cells with CRA resulted in increased production of IL-4, IL-5, and IL-13, but no changes in IFN-γ (Fig. 4C). Similar to the results from the CRA-treated Hox5AabbCc mice, those mice that received Hox5AabbCc bone marrow had an increase in expression of the mucus-associated genes Muc5ac and Gob5 in the lungs over mice that were injected with WT cells (Fig. 4D), which corresponded to increased mucus in the lungs of these mice (Fig. 4E).

FIGURE 4.

Increased inflammation in Hox5AabbCc mice is driven by differences in immune cell function. WT and Hox5AabbCc mice were irradiated and reconstituted with bone marrow cells. (A) Inflammation was visualized on lung sections stained with H&E. (B) Lung cells were analyzed by flow cytometry. (C) Cells from the mediastinal lymph node were restimulated in vitro with CRA for 48 h. Levels of cytokines in the supernatants were measured by BioPlex. (D) RNA was extracted from lung tissue and the genes Muc5ac and Gob5 were measured by qPCR. (E) Lung sections were stained with periodic acid-Schiff to visualize mucus. Results represent one of two independent experiments. n = 3–5 mice per group. *p < 0.05, **p < 0.01 compared with WT→ WT mice.

FIGURE 4.

Increased inflammation in Hox5AabbCc mice is driven by differences in immune cell function. WT and Hox5AabbCc mice were irradiated and reconstituted with bone marrow cells. (A) Inflammation was visualized on lung sections stained with H&E. (B) Lung cells were analyzed by flow cytometry. (C) Cells from the mediastinal lymph node were restimulated in vitro with CRA for 48 h. Levels of cytokines in the supernatants were measured by BioPlex. (D) RNA was extracted from lung tissue and the genes Muc5ac and Gob5 were measured by qPCR. (E) Lung sections were stained with periodic acid-Schiff to visualize mucus. Results represent one of two independent experiments. n = 3–5 mice per group. *p < 0.05, **p < 0.01 compared with WT→ WT mice.

Close modal

We observed an increase in the number of CD4+ T cells in both the Hox5AabbCc mice and WT mice that had been injected with Hox5AabbCc bone marrow cells. In addition, many of these cells had an activated phenotype. We therefore asked whether the CD4+ T cells from these mice produced increased levels of Th2 cytokines. We began by investigating whether the expression of the three Hox5 genes was increased in Th2 cells. We sorted naive CD4+ T cells from the spleens of WT mice and skewed them in vitro toward Th0, Th1, Th2, and Th17 lineages using standard skewing assays (32, 33). Cells were harvested at 2, 6, 24, and 48 h after skewing, and the results showed that Hox5 gene expression was increased in cells skewed toward the Th2 lineage, but not under other skewing conditions, although this increase was not significant in Hoxa5-transfected cells (Fig. 5A). We found that Hoxc5 expression is increased to the greatest degree in these cells, and that gene expression occurs early following IL-4 stimulation, but remains elevated through 48 h (Fig. 5A). To determine whether the expression of the Hox5 genes affected Th2 cytokine production, naive CD4+ T cells were isolated from the spleens of WT and Hox5AabbCc mice, and were again skewed to Th2 cells in vitro. We found that 5 d after skewing cells in the presence of IL-4, the levels of IL-4, IL-5, and IL-13 were increased in the cultures from the Hox5AabbCc T cells compared with WT T cells (Fig. 5B). The cells were rested for 3 d, and then restimulated with anti-CD3 and anti-CD28 to determine whether they maintained increased Th2 cytokine production upon secondary activation. After an additional 48 h of activation, T cells isolated from Hox5AabbCc mice continued to produce increased levels of Th2 cytokines (Fig. 5C). The increase in cytokine production from Hox5AabbCc T cells was limited to Th2 cells, as no differences were seen upon primary activation of T cells skewed toward Th1 or Th17 with regard to production of IFN-γ or IL-17, respectively (Supplemental Fig. 1). Furthermore, the transcription of Gata3, a master transcription factor for Th2 cytokine production, was increased in T cells from Hox5AabbCc mice compared with WT controls at both 48 h and 5 d (Fig. 5D). Gata3 transcription is activated by the IL-4–induced transcription factor STAT6, which becomes phosphorylated and binds to Gata3 and other genes. However, other STAT6-activated genes such as c-maf and Irf4 were not increased, suggesting that the activation and phosphorylation of STAT6 was not affected (Supplemental Fig. 2). Finally, to confirm the increased expression of Gata3, we crossed both WT and Hox5AabbCc mice with Gata3-GFP reporter mice and skewed splenic naive CD4+ T cells toward the Th0 and Th2 lineages. We found that Gata3 gene expression was higher in Hox5AabbCc mice compared with WT mice 5 d after skewing (Fig. 5E). Representative plots from Th2-skewed cells are shown in Fig. 5F.

FIGURE 5.

CD4+ T cells from Hox5AabbCc mice produce more Th2 cytokines under IL-4 skewing conditions. (A) Naive T cells were isolated from the spleens of WT mice and were skewed toward Th0, Th1, Th2, Th17, and regulatory T cell lineages. Hox gene expression was measured in each lineage by qPCR, and expression was compared with GAPDH using the 2−ΔCT method. Analysis was done by two-way ANOVA, followed by post hoc t test. *p < 0.05, **p < 0.01 for Th2 compared with Th0 at each time point. (B) Purified T cells from WT or Hox5AabbCc mice were cultured under Th2 skewing conditions for 5 d, and supernatant cytokines were measured by BioPlex. *p < 0.05 compared with WT cells. (C) T cells were skewed for 5 d, then washed and rested for 3 d. The cells were then further stimulated with anti-CD3 and anti-CD28 for 48 h to reactivate the cells. Cytokines were measured by BioPlex. *p < 0.05 compared with WT. (D) T cells were skewed for 48 h or 5 d in Th2 conditions. RNA was extracted and Gata3 gene expression was measured by qPCR. Results are representative of three independent experiments. **p < 0.01 compared with WT cells. (E) WT and Hox5AabbCc mice were crossed with Gata3-GFP reporter mice, then were cultured under Th2 skewing conditions for 5 d. GFP expression was measured by flow cytometry. Representative plots are shown in (F). Results are from one of two independent experiments. Three samples were run per group, and samples were run in triplicate. ***p < 0.001 compared with WT.

FIGURE 5.

CD4+ T cells from Hox5AabbCc mice produce more Th2 cytokines under IL-4 skewing conditions. (A) Naive T cells were isolated from the spleens of WT mice and were skewed toward Th0, Th1, Th2, Th17, and regulatory T cell lineages. Hox gene expression was measured in each lineage by qPCR, and expression was compared with GAPDH using the 2−ΔCT method. Analysis was done by two-way ANOVA, followed by post hoc t test. *p < 0.05, **p < 0.01 for Th2 compared with Th0 at each time point. (B) Purified T cells from WT or Hox5AabbCc mice were cultured under Th2 skewing conditions for 5 d, and supernatant cytokines were measured by BioPlex. *p < 0.05 compared with WT cells. (C) T cells were skewed for 5 d, then washed and rested for 3 d. The cells were then further stimulated with anti-CD3 and anti-CD28 for 48 h to reactivate the cells. Cytokines were measured by BioPlex. *p < 0.05 compared with WT. (D) T cells were skewed for 48 h or 5 d in Th2 conditions. RNA was extracted and Gata3 gene expression was measured by qPCR. Results are representative of three independent experiments. **p < 0.01 compared with WT cells. (E) WT and Hox5AabbCc mice were crossed with Gata3-GFP reporter mice, then were cultured under Th2 skewing conditions for 5 d. GFP expression was measured by flow cytometry. Representative plots are shown in (F). Results are from one of two independent experiments. Three samples were run per group, and samples were run in triplicate. ***p < 0.001 compared with WT.

Close modal

Our previous results suggest that Hox5 deficiency results in increased Gata3 expression that modifies Th2 cytokine production. To understand how HOX5 proteins regulate Th2 cells, we used the human Jurkat T cell line to overexpress WT HOX5 proteins. Plasmids expressing FLAG-tagged Hox5 constructs were transfected into Jurkat cells either individually or together in equal concentrations. We then pulled down the HOX5 proteins by ChIP using an anti-FLAG Ab. We designed primers to amplify putative consensus STAT6 binding sites conserved in humans and mice on the Gata3 gene (34) at sites that also contain the consensus binding sequence for HOX5 proteins. The location of these sites relative to the transcription start site and gene exons is shown in Fig. 6A. We found that HOX5 proteins bound significantly to two of the three STAT6 binding sites that we assayed (Fig. 6B). When we transfected the Hox5-expressing plasmids individually, we found that HOXC5 protein bound more efficiently to the Gata3 promoter at both STAT6 binding sites relative to HOXA5 and HOXB5 (Fig. 6C). We then confirmed that the binding of the HOX5 proteins resulted in decreased expression of Gata3 and IL-4. The Hox5-expression plasmids were transfected into Jurkat cells and the cells were cultured under Th2 skewing conditions for 24 h. Decreased Gata3 expression was observed in cells transfected with Hoxc5 alone, as well as with all three Hox5-expression plasmids (Fig. 6D). Downstream of Gata3, we found a similar expression pattern of IL-4 and IL-13 in the transfected cells, although IL-5 was only decreased in cells that were transfected with all three Hox5-expressing plasmids (Fig. 6E). Together, these data suggest that the HOX5 proteins bind to the Gata3 promoter and that HOXC5 binds the most efficiently, and thereby affects Th2 cytokine production.

FIGURE 6.

HOX5 proteins interact with the Gata3 gene at STAT6 binding sites. (A) The Gata3 gene exons are shown in blue boxes and STAT6 binding sites assayed are shown as red circles. The proximal and distal promoters are marked, as well as the transcriptional start site (TSS). (B) Equal concentrations of FLAG-tagged Hox5-expressing plasmids were transfected into Jurkat cells for 24 h, then ChIP was performed using an anti-FLAG Ab. Primers were designed for putative STAT6 binding sites on the Gata3 gene, and DNA bound to the HOX proteins was amplified by qPCR. (C) Individual Hox5 plasmids were transfected into Jurkat cells and ChIP was done with an anti-FLAG Ab. qPCR was performed on the first two STAT6 binding sites identified in Fig. 5A. *p < 0.05, **p < 0.01 compared with IgG control. (D) Jurkat cells were transfected with Lipofectamine only (control), or with plasmids expressing Hoxa5, Hoxb5, Hoxc5, or with all three together (Hoxa5+b5+c5). After 24 h, cells were then cultured under Th2 skewing conditions. Gata3 gene expression was measured after a further 24 h. (E) IL-4, IL-5, and IL-13 gene expression were measured after 24 h of Th2 skewing conditions. Results are representative of three independent experiments. Three samples were run per group, and samples were run in triplicate. *p < 0.05 compared with Lipofectamine control.

FIGURE 6.

HOX5 proteins interact with the Gata3 gene at STAT6 binding sites. (A) The Gata3 gene exons are shown in blue boxes and STAT6 binding sites assayed are shown as red circles. The proximal and distal promoters are marked, as well as the transcriptional start site (TSS). (B) Equal concentrations of FLAG-tagged Hox5-expressing plasmids were transfected into Jurkat cells for 24 h, then ChIP was performed using an anti-FLAG Ab. Primers were designed for putative STAT6 binding sites on the Gata3 gene, and DNA bound to the HOX proteins was amplified by qPCR. (C) Individual Hox5 plasmids were transfected into Jurkat cells and ChIP was done with an anti-FLAG Ab. qPCR was performed on the first two STAT6 binding sites identified in Fig. 5A. *p < 0.05, **p < 0.01 compared with IgG control. (D) Jurkat cells were transfected with Lipofectamine only (control), or with plasmids expressing Hoxa5, Hoxb5, Hoxc5, or with all three together (Hoxa5+b5+c5). After 24 h, cells were then cultured under Th2 skewing conditions. Gata3 gene expression was measured after a further 24 h. (E) IL-4, IL-5, and IL-13 gene expression were measured after 24 h of Th2 skewing conditions. Results are representative of three independent experiments. Three samples were run per group, and samples were run in triplicate. *p < 0.05 compared with Lipofectamine control.

Close modal

Although Hox genes have been identified in some hematopoietic populations, little has been done to describe their function or role in immune cell populations. In this study, we demonstrate a novel role for the HOX5 paralogous group of proteins in regulating Th2 inflammation in a model of allergic lung disease. These studies highlight the ability of this gene family, primarily described as a mesenchyme-associated developmental transcription factor, to be redeployed in T cells for the regulation of Th2 cell development. The association of Hox5 proteins with Gata3 expression is especially key, as it designates the Hox5 proteins to the Th2 lineage, without directly altering any of the other T cell responses or total number of T cells. The functional significance of this fidelity is not presently clear, however, it appears to be associated with STAT6 binding sites on the Gata3 gene, an aspect that we continue to explore with the generation of additional tools. Together, these studies define three important and novel aspects: 1) Hox5 paralogous proteins provide a discrete regulatory function for the generation of Th2 cells; 2) the absence of Hox5 regulatory control in T cells leads to enhanced Th2 disease pathology in vivo; and 3) the function of Hox5 genes in regulating Th2 cell development is, at least, via the repression of Gata3 expression. These striking data establish a novel paradigm to be further investigated for the role of HOX proteins in immune cell regulation that controls disease progression.

The HOX5 proteins are critical for lung development, particularly HOXA5. Hoxa5−/− mice do not thrive, primarily due to faults in alveolar development due to mesenchyme-associated defects (2123). The mice used in this study have one mutated Hoxa5 allele, and these mice thrive without evidence of major defects. Although they do have a noticeable defect in alveolarization, it does not lead to an altered immune response to CRA as evidenced by our data using bone marrow chimeras. These mice do not have functional Hoxb5 and previous studies have shown that there is some functional redundancy between Hoxb5 and Hoxa5 during lung development (24). Although Hoxc5 has conflicting roles in lung development (24, 31), we found Hoxc5 to be most highly upregulated in the lung tissue after allergen challenge. We also found that Hoxc5 was the most highly upregulated in the Th2 cell cultures. And, although both Hoxa5 and Hoxb5 were able to bind to Gata3, our ChIP results show that Hoxc5 was able to bind at a higher rate. These data may suggest some compartmentalization of Hox5 paralog function that could separate their roles based upon cell-specific effects and the maturation state of the host.

Our results suggest a role for HOX5 proteins in regulating the activation and/or differentiation of Th2 cells specifically, as other subsets of T cells were not altered upon differentiation. In the absence of HOX5 proteins, an increase in Gata3 gene expression and Th2 cytokine production was observed. Furthermore, transfecting plasmids expressing Hox5 genes into Jurkat cells results in decreased Gata3 expression, indicating that HOX5 proteins are able to repress Gata3. Several other studies have identified a repressive role for HOX proteins, particularly during development. For example, activation of Hoxa13 has been shown to downregulate signaling through the Wnt/β-catenin pathway, which is important in controlling elongation of the anteroposterior axis during development, a result that is replicated by over-expression of Hoxa13 (35). Other work in both Drosophila and mice has shown that the HOX proteins activated during the development of the posterior embryo are able to repress genes that are needed for anterior development (3638). In addition, different HOX proteins can activate or repress the same gene in different tissues, and this function has been attributed to the nonhomeodomain of the protein (39). The mechanism of HOX5 protein function in this study is presently unclear but presents a potentially important and new paradigm.

The mechanism by which HOX proteins affect target gene activation or repression is poorly understood. All HOX proteins recognize a similar AT-rich sequence in vitro. Studies in which the homeodomain of one protein was swapped for the domain of another have demonstrated that these domains are at least partially able to confer some degree of specificity in vivo (40, 41). Additionally, the sequences flanking the homeodomain have also been shown to be important in DNA binding (42). Others have shown that the N-terminus of various HOX proteins is important in determining gene specificity in both flies and mammals (39, 43). Furthermore, cofactors that bind to HOX proteins also affect the specificity of the HOX protein for certain target genes and not others (4446). Finally, the ability of HOX proteins to interact with other transcription factors to target the binding to specific promoter and/or enhancer elements has been suggested to be important for tissue- and cell-specific functions (47, 48). Our results demonstrate that HOX5 proteins are able to bind to Gata3-associated gene elements. Although it has been shown that STAT6 binds to the Gata3 gene in mouse Th2 cells, the activation of Gata3 in human T cells is not as well defined (34, 49, 50). Our results suggest that HOX5 proteins binding to the Gata3 gene at STAT6 binding sites downregulate its activation, although whether this is directly due to disrupted STAT6 binding or occurs through indirect mechanisms is unknown. It is likely that the HOX5 proteins are acting in concert with other elements to target them to the Gata3 locus. The identity of these elements remains a subject of investigation and may include STAT6 itself.

This study demonstrates a role for HOX5 proteins in controlling lung inflammation during a model of allergic asthma. The increased inflammation in the Hox5-deficient mice was mediated by increased Th2 cell development, which was associated with regulation of Gata3 expression. These results imply a novel role for the HOX5 paralogs in T cell function leading to regulation of allergic disease, and opens up a new area of study in T cell gene regulation. Further studies may lead to not only a better understanding of Th2-associated disease progression but also identify unappreciated roles for HOX proteins beyond development.

We thank Judith Connett for critical reading of the manuscript.

This work was supported by National Institutes of Health Grants R01 HL036302 and R01 HL119215, and by an American Lung Association Biomedical Research Grant.

The online version of this article contains supplemental material.

Abbreviations used in this article:

ChIP

chromatin immunoprecipitation

CRA

cockroach Ag

CT

cycle threshold

MLN

mesenteric lymph node

qPCR

quantitative PCR

WT

wild type.

1
Winer
,
R. A.
,
X.
Qin
,
T.
Harrington
,
J.
Moorman
,
H.
Zahran
.
2012
.
Asthma incidence among children and adults: findings from the behavioral risk factor surveillance system asthma call-back survey–United States, 2006-2008.
J. Asthma
49
:
16
22
.
2
Swain
,
S. L.
,
D. T.
McKenzie
,
R. W.
Dutton
,
S. L.
Tonkonogy
,
M.
English
.
1988
.
The role of IL4 and IL5: characterization of a distinct helper T cell subset that makes IL4 and IL5 (Th2) and requires priming before induction of lymphokine secretion.
Immunol. Rev.
102
:
77
105
.
3
Cherwinski
,
H. M.
,
J. H.
Schumacher
,
K. D.
Brown
,
T. R.
Mosmann
.
1987
.
Two types of mouse helper T cell clone. III. Further differences in lymphokine synthesis between Th1 and Th2 clones revealed by RNA hybridization, functionally monospecific bioassays, and monoclonal antibodies.
J. Exp. Med.
166
:
1229
1244
.
4
Mosmann
,
T. R.
,
H.
Cherwinski
,
M. W.
Bond
,
M. A.
Giedlin
,
R. L.
Coffman
.
1986
.
Two types of murine helper T cell clone. I. Definition according to profiles of lymphokine activities and secreted proteins.
J. Immunol.
136
:
2348
2357
.
5
de Waal Malefyt
,
R.
,
C. G.
Figdor
,
R.
Huijbens
,
S.
Mohan-Peterson
,
B.
Bennett
,
J.
Culpepper
,
W.
Dang
,
G.
Zurawski
,
J. E.
de Vries
.
1993
.
Effects of IL-13 on phenotype, cytokine production, and cytotoxic function of human monocytes. Comparison with IL-4 and modulation by IFN-gamma or IL-10.
J. Immunol.
151
:
6370
6381
.
6
Kondo
,
M.
,
T.
Takeshita
,
N.
Ishii
,
M.
Nakamura
,
S.
Watanabe
,
K.
Arai
,
K.
Sugamura
.
1993
.
Sharing of the interleukin-2 (IL-2) receptor gamma chain between receptors for IL-2 and IL-4.
Science
262
:
1874
1877
.
7
Idzerda
,
R. L.
,
C. J.
March
,
B.
Mosley
,
S. D.
Lyman
,
T.
Vanden Bos
,
S. D.
Gimpel
,
W. S.
Din
,
K. H.
Grabstein
,
M. B.
Widmer
,
L. S.
Park
, et al
.
1990
.
Human interleukin 4 receptor confers biological responsiveness and defines a novel receptor superfamily.
J. Exp. Med.
171
:
861
873
.
8
Zheng
,
W.
,
R. A.
Flavell
.
1997
.
The transcription factor GATA-3 is necessary and sufficient for Th2 cytokine gene expression in CD4 T cells.
Cell
89
:
587
596
.
9
Takeda
,
K.
,
T.
Tanaka
,
W.
Shi
,
M.
Matsumoto
,
M.
Minami
,
S.
Kashiwamura
,
K.
Nakanishi
,
N.
Yoshida
,
T.
Kishimoto
,
S.
Akira
.
1996
.
Essential role of Stat6 in IL-4 signalling.
Nature
380
:
627
630
.
10
Berlin
,
A. A.
,
C. M.
Hogaboam
,
N. W.
Lukacs
.
2006
.
Inhibition of SCF attenuates peribronchial remodeling in chronic cockroach allergen-induced asthma.
Lab. Invest.
86
:
557
565
.
11
Campbell
,
E. M.
,
I. F.
Charo
,
S. L.
Kunkel
,
R. M.
Strieter
,
L.
Boring
,
J.
Gosling
,
N. W.
Lukacs
.
1999
.
Monocyte chemoattractant protein-1 mediates cockroach allergen-induced bronchial hyperreactivity in normal but not CCR2-/- mice: the role of mast cells.
J. Immunol.
163
:
2160
2167
.
12
Jang
,
S.
,
S.
Morris
,
N. W.
Lukacs
.
2013
.
TSLP promotes induction of Th2 differentiation but is not necessary during established allergen-induced pulmonary disease.
PLoS One
8
:
e56433
.
13
Tümpel
,
S.
,
L. M.
Wiedemann
,
R.
Krumlauf
.
2009
.
Hox genes and segmentation of the vertebrate hindbrain.
Curr. Top. Dev. Biol.
88
:
103
137
.
14
Wellik
,
D. M.
2009
.
Hox genes and vertebrate axial pattern.
Curr. Top. Dev. Biol.
88
:
257
278
.
15
Zakany
,
J.
,
D.
Duboule
.
2007
.
The role of Hox genes during vertebrate limb development.
Curr. Opin. Genet. Dev.
17
:
359
366
.
16
Graham
,
A.
,
N.
Papalopulu
,
R.
Krumlauf
.
1989
.
The murine and Drosophila homeobox gene complexes have common features of organization and expression.
Cell
57
:
367
378
.
17
Qian
,
Y. Q.
,
M.
Billeter
,
G.
Otting
,
M.
Müller
,
W. J.
Gehring
,
K.
Wüthrich
.
1989
.
The structure of the Antennapedia homeodomain determined by NMR spectroscopy in solution: comparison with prokaryotic repressors.
Cell
59
:
573
580
.
18
McGinnis
,
W.
,
M. S.
Levine
,
E.
Hafen
,
A.
Kuroiwa
,
W. J.
Gehring
.
1984
.
A conserved DNA sequence in homoeotic genes of the Drosophila Antennapedia and bithorax complexes.
Nature
308
:
428
433
.
19
Laughon
,
A.
,
M. P.
Scott
.
1984
.
Sequence of a Drosophila segmentation gene: protein structure homology with DNA-binding proteins.
Nature
310
:
25
31
.
20
Santini
,
S.
,
J. L.
Boore
,
A.
Meyer
.
2003
.
Evolutionary conservation of regulatory elements in vertebrate Hox gene clusters.
Genome Res.
13
(
Suppl.
6A
):
1111
1122
.
21
Grier
,
D. G.
,
A.
Thompson
,
T. R.
Lappin
,
H. L.
Halliday
.
2009
.
Quantification of Hox and surfactant protein-B transcription during murine lung development.
Neonatology
96
:
50
60
.
22
Aubin
,
J.
,
M.
Lemieux
,
M.
Tremblay
,
J.
Bérard
,
L.
Jeannotte
.
1997
.
Early postnatal lethality in Hoxa-5 mutant mice is attributable to respiratory tract defects.
Dev. Biol.
192
:
432
445
.
23
Mandeville
,
I.
,
J.
Aubin
,
M.
LeBlanc
,
M.
Lalancette-Hébert
,
M. F.
Janelle
,
G. M.
Tremblay
,
L.
Jeannotte
.
2006
.
Impact of the loss of Hoxa5 function on lung alveogenesis.
Am. J. Pathol.
169
:
1312
1327
.
24
Boucherat
,
O.
,
S.
Montaron
,
F. A.
Bérubé-Simard
,
J.
Aubin
,
P.
Philippidou
,
D. M.
Wellik
,
J. S.
Dasen
,
L.
Jeannotte
.
2013
.
Partial functional redundancy between Hoxa5 and Hoxb5 paralog genes during lung morphogenesis.
Am. J. Physiol. Lung Cell. Mol. Physiol.
304
:
L817
L830
.
25
Faber
,
J.
,
A. V.
Krivtsov
,
M. C.
Stubbs
,
R.
Wright
,
T. N.
Davis
,
M.
van den Heuvel-Eibrink
,
C. M.
Zwaan
,
A. L.
Kung
,
S. A.
Armstrong
.
2009
.
HOXA9 is required for survival in human MLL-rearranged acute leukemias.
Blood
113
:
2375
2385
.
26
Fuller
,
J. F.
,
J.
McAdara
,
Y.
Yaron
,
M.
Sakaguchi
,
J. K.
Fraser
,
J. C.
Gasson
.
1999
.
Characterization of HOX gene expression during myelopoiesis: role of HOX A5 in lineage commitment and maturation.
Blood
93
:
3391
3400
.
27
Morgan
,
R.
,
K.
Whiting
.
2008
.
Differential expression of HOX genes upon activation of leukocyte sub-populations.
Int. J. Hematol.
87
:
246
249
.
28
Taghon
,
T.
,
K.
Thys
,
M.
De Smedt
,
F.
Weerkamp
,
F. J.
Staal
,
J.
Plum
,
G.
Leclercq
.
2003
.
Homeobox gene expression profile in human hematopoietic multipotent stem cells and T-cell progenitors: implications for human T-cell development.
Leukemia
17
:
1157
1163
.
29
Carè
,
A.
,
U.
Testa
,
A.
Bassani
,
E.
Tritarelli
,
E.
Montesoro
,
P.
Samoggia
,
L.
Cianetti
,
C.
Peschle
.
1994
.
Coordinate expression and proliferative role of HOXB genes in activated adult T lymphocytes.
Mol. Cell. Biol.
14
:
4872
4877
.
30
Hosoya
,
T.
,
T.
Kuroha
,
T.
Moriguchi
,
D.
Cummings
,
I.
Maillard
,
K. C.
Lim
,
J. D.
Engel
.
2009
.
GATA-3 is required for early T lineage progenitor development.
J. Exp. Med.
206
:
2987
3000
.
31
Hrycaj
,
S. M.
,
B. R.
Dye
,
N. C.
Baker
,
B. M.
Larsen
,
A. C.
Burke
,
J. R.
Spence
,
D. M.
Wellik
.
2015
.
Hox5 genes regulate the Wnt2/2b-Bmp4-signaling axis during lung development.
Cell Reports
12
:
903
912
.
32
McKenzie
,
G. J.
,
C. L.
Emson
,
S. E.
Bell
,
S.
Anderson
,
P.
Fallon
,
G.
Zurawski
,
R.
Murray
,
R.
Grencis
,
A. N.
McKenzie
.
1998
.
Impaired development of Th2 cells in IL-13-deficient mice.
Immunity
9
:
423
432
.
33
Gondek
,
D. C.
,
N. R.
Roan
,
M. N.
Starnbach
.
2009
.
T cell responses in the absence of IFN-gamma exacerbate uterine infection with Chlamydia trachomatis.
J. Immunol.
183
:
1313
1319
.
34
Onodera
,
A.
,
M.
Yamashita
,
Y.
Endo
,
M.
Kuwahara
,
S.
Tofukuji
,
H.
Hosokawa
,
A.
Kanai
,
Y.
Suzuki
,
T.
Nakayama
.
2010
.
STAT6-mediated displacement of polycomb by trithorax complex establishes long-term maintenance of GATA3 expression in T helper type 2 cells.
J. Exp. Med.
207
:
2493
2506
.
35
Denans
,
N.
,
T.
Iimura
,
O.
Pourquié
.
2015
.
Hox genes control vertebrate body elongation by collinear Wnt repression.
eLife 4
:
e04379
.
36
Coiffier
,
D.
,
B.
Charroux
,
S.
Kerridge
.
2008
.
Common functions of central and posterior Hox genes for the repression of head in the trunk of Drosophila.
Development
135
:
291
300
.
37
Struhl
,
G.
,
R. A.
White
.
1985
.
Regulation of the ultrabithorax gene of Drosophila by other bithorax complex genes.
Cell
43
:
507
519
.
38
Karch
,
F.
,
W.
Bender
,
B.
Weiffenbach
.
1990
.
abdA expression in Drosophila embryos.
Genes Dev.
4
:
1573
1587
.
39
Yallowitz
,
A. R.
,
K. Q.
Gong
,
I. T.
Swinehart
,
L. T.
Nelson
,
D. M.
Wellik
.
2009
.
Non-homeodomain regions of Hox proteins mediate activation versus repression of Six2 via a single enhancer site in vivo.
Dev. Biol.
335
:
156
165
.
40
Gibson
,
G.
,
A.
Schier
,
P.
LeMotte
,
W. J.
Gehring
.
1990
.
The specificities of sex combs reduced and Antennapedia are defined by a distinct portion of each protein that includes the homeodomain.
Cell
62
:
1087
1103
.
41
Kuziora
,
M. A.
,
W.
McGinnis
.
1989
.
A homeodomain substitution changes the regulatory specificity of the deformed protein in Drosophila embryos.
Cell
59
:
563
571
.
42
Florence
,
B.
,
R.
Handrow
,
A.
Laughon
.
1991
.
DNA-binding specificity of the fushi tarazu homeodomain.
Mol. Cell. Biol.
11
:
3613
3623
.
43
Zeng
,
W.
,
D. J.
Andrew
,
L. D.
Mathies
,
M. A.
Horner
,
M. P.
Scott
.
1993
.
Ectopic expression and function of the Antp and Scr homeotic genes: the N terminus of the homeodomain is critical to functional specificity.
Development
118
:
339
352
.
44
Lelli
,
K. M.
,
B.
Noro
,
R. S.
Mann
.
2011
.
Variable motif utilization in homeotic selector (Hox)-cofactor complex formation controls specificity.
Proc. Natl. Acad. Sci. USA
108
:
21122
21127
.
45
Slattery
,
M.
,
T.
Riley
,
P.
Liu
,
N.
Abe
,
P.
Gomez-Alcala
,
I.
Dror
,
T.
Zhou
,
R.
Rohs
,
B.
Honig
,
H. J.
Bussemaker
,
R. S.
Mann
.
2011
.
Cofactor binding evokes latent differences in DNA binding specificity between Hox proteins.
Cell
147
:
1270
1282
.
46
Chan
,
S. K.
,
L.
Jaffe
,
M.
Capovilla
,
J.
Botas
,
R. S.
Mann
.
1994
.
The DNA binding specificity of Ultrabithorax is modulated by cooperative interactions with extradenticle, another homeoprotein.
Cell
78
:
603
615
.
47
Choe
,
S. K.
,
F.
Ladam
,
C. G.
Sagerström
.
2014
.
TALE factors poise promoters for activation by Hox proteins.
Dev. Cell
28
:
203
211
.
48
Choe
,
S. K.
,
P.
Lu
,
M.
Nakamura
,
J.
Lee
,
C. G.
Sagerström
.
2009
.
Meis cofactors control HDAC and CBP accessibility at Hox-regulated promoters during zebrafish embryogenesis.
Dev. Cell
17
:
561
567
.
49
Wei
,
L.
,
G.
Vahedi
,
H. W.
Sun
,
W. T.
Watford
,
H.
Takatori
,
H. L.
Ramos
,
H.
Takahashi
,
J.
Liang
,
G.
Gutierrez-Cruz
,
C.
Zang
, et al
.
2010
.
Discrete roles of STAT4 and STAT6 transcription factors in tuning epigenetic modifications and transcription during T helper cell differentiation.
Immunity
32
:
840
851
.
50
Elo
,
L. L.
,
H.
Järvenpää
,
S.
Tuomela
,
S.
Raghav
,
H.
Ahlfors
,
K.
Laurila
,
B.
Gupta
,
R. J.
Lund
,
J.
Tahvanainen
,
R. D.
Hawkins
, et al
.
2010
.
Genome-wide profiling of interleukin-4 and STAT6 transcription factor regulation of human Th2 cell programming.
Immunity
32
:
852
862
.

The authors have no financial conflicts of interest.

Supplementary data