Constitutive activation of SHP2 protein tyrosine phosphatase inhibits ICSBP-induced transcription of the gene encoding gp91^PHOX during myeloid differentiation

Abstract

The IFN consensus sequence-binding protein (ICSBP; also referred to as IFN regulatory factor 8) is a transcription factor which is expressed in myeloid and B cells. In previous studies, we found that ICSBP activated transcription of the gene encoding gp91^PHOX (the CYBB gene), a rate-limiting component of the phagocyte respiratory burst oxidase expressed exclusively after the promyelocyte stage of myelopoiesis. Previously, we found that CYBB transcription was dependent on phosphorylation of specific ICSBP tyrosine residues. Since ICSBP is tyrosine-phosphorylated during myelopoiesis, this provided a mechanism of differentiation stage-specific CYBB transcription. In the current studies, we found that ICSBP was a substrate for Src homology-containing tyrosine phosphatase 2 (SHP2-PTP) in immature myeloid cells but not during myelopoiesis. Therefore, SHP2-PTP inhibited CYBB transcription and respiratory burst activity in myeloid progenitor cells by dephosphorylating ICSBP. In contrast, we found that ICSBP was a substrate for a leukemia-associated, constitutively active mutant form of SHP2, described previously, throughout differentiation. Consistent with this, constitutive SHP2 activation blocked ICSBP-induced CYBB transcription and respiratory burst activity in differentiating myeloid cells. ICSBP-deficiency and constitutive SHP2 activation have been described in human myelodysplastic syndromes. As these two abnormalities may coexist, our results identified a potential molecular mechanism for impaired phagocyte function in this malignant myeloid disease.

INTRODUCTION

The IFN consensus sequence-binding protein (ICSBP), also known as IFN regulatory factor 8 (IRF8), is a transcription factor which is expressed in myeloid and B cells [1]. ICSBP regulates a number of genes which mediate phagocyte and B cell functional activities. Such ICSBP targets include genes encoding the phagocyte NADPH oxidase proteins gp91^PHOX and p67^PHOX, the TLR4, and the IL-12R [2–5]. The role of ICSBP in hematopoiesis and the immune response was investigated using an ICSBP-deficient murine model [6, 7]. Although phagocyte function has not been studied in these animals, ICSBP-deficient mice exhibited abnormalities of B cell function, as anticipated [6]. Somewhat surprisingly, these mice also developed a myeloproliferative disorder. This aspect of the phenotype was at least partly explained by impaired expression of Neurofibromin 1 (a Ras-GTPase-activating protein) in ICSBP-deficient cells [8–10]. These studies suggested that ICSBP regulates multiple aspects of myelopoiesis. ICSBP expression has also been investigated in human myeloid malignancies, and decreased expression was found in therapy-related myelodysplastic syndrome (tMDS) and acute myeloid leukemia (tAML) and in uncontrolled and progressive chronic myeloid leukemia (CML) [11, 12].

ICSBP binds to DNA cis elements conforming to an IFN-stimulated response element, positive regulatory domain I, or composite ets/IRF consensus sequences [13]. For example, ICSBP activates composite ets/IRF cis elements in the genes encoding gp91^PHOX and p67^PHOX (the CYBB and NCF2 genes, respectively) [2, 3]. These genes are actively transcribed after the promyelocyte stage of differentiation, and transcription is increased further in mature phagocytes by inflammatory mediators [14–16]. A multi-protein complex interacts with the composite ets/IRF cis elements in the CYBB and NCF2 genes in a differentiation stage-specific manner [2, 3]. This complex includes the ets protein PU.1, the IRF proteins IRF1 and ICSBP, and the CREB-binding protein (CBP) [3]. PU.1 and IRF1 interact with the CYBB and NCF2 cis elements in immature myeloid cells but activate transcription poorly [3]. During differentiation, cytokine-induced phosphorylation of a conserved tyrosine residue in the “IRF domain” of ICSBP permits interaction with DNA-bound PU.1 and IRF1 [3]. This complex recruits CBP to the promoter, activating transcription. Therefore, one would anticipate impaired gp91^PHOX and p67^PHOX expression in ICSBP-deficient murine models and human diseases. As gp91^PHOX and p67^PHOX are the rate-limiting oxidase proteins [17], one would also anticipate defective respiratory burst activity under these conditions.

Regulation of transcription of a number of ICSBP target genes has been found to be influenced by ICSBP post-translational modification [2, 9, 18]. We previously found that the hematopoietic-specific Src homology-containing tyrosine phosphatase 1 (SHP1)-protein-tyrosine-phosphatase (PTP) was involved in maintaining ICSBP in a nontyrosine-phosphorylated state in immature myeloid cells [3]. Expression of a dominant-negative (DN) form of SHP1 increased ICSBP tyrosine phosphorylation and CYBB and NCF2 transcription in undifferentiated myeloid cell lines [3]. However, neither DN-SHP1 nor overexpression of wild-type (Wt)-SHP1 altered ICSBP tyrosine phosphorylation or transcription of these genes during cytokine-induced differentiation. This suggested SHP1-PTP activity was regulated during myelopoiesis.

SHP2 is a ubiquitously expressed PTP, which is structurally similar to SHP1. Activation of SHP1 or SHP2 occurs upon interaction of phosphotyrosine residues from substrate or another protein with SH2 domains in these PTPs. This interaction results in a conformational change, which unmasks the PTP domain [19]. Comparative studies of the influence of SHP1 and SHP2 on myelopoiesis have been impaired by the paucity of identified substrates for either PTP. Recently, mutations in the gene encoding SHP2 have been described in human myeloid malignancies, including MDS and AML [20, 21]. These mutations alter protein conformation in a manner that unmasks the PTP domain, resulting in constitutive activation [21, 22]. Such mutations might dysregulate the activity of SHP2 for normal substrates, result in promiscuous substrate selection, or both. We found that constitutively active SHP2 mutants dephosphorylated ICSBP in myeloid cell lines undergoing IFN-γ-induced differentiation and in primary murine progenitor cells undergoing differentiation with M-CSF ± IFN-γ [9].

However, our previous studies did not determine the impact of leukemia-associated, activated SHP2 mutants on ICSBP-induced CYBB transcription during myelopoiesis. The current studies will investigate this issue and also the impact of ICSBP deficiency and constitutive SHP2 activation on respiratory burst activity. ICSBP deficiency and constitutively active SHP2 mutants have been described in human MDS. Subjects with MDS exhibit immunodeficiency, which is partly related to impaired phagocyte function [23]. Results of these studies would therefore have implications for understanding defects in the innate immune response in such subjects.

MATERIALS AND METHODS

Plasmids and PCR mutagenesis

Protein expression vectors

The ICSBP cDNA was obtained from Dr. Ben Zion-Levi (Technion, Haifa, Israel), and the full-length cDNA was generated by PCR and subcloned into the mammalian expression vector pcDNA and the murine stem cell retrovirus vector pMSCVpuro for retroviral production (Stratagene, La Jolla, CA, USA). The latter was used for retroviral production as described [8]. The cDNAs for Wt- and DN (C463S)-SHP2-PTP were obtained from Dr. Stuart Frank (University of Alabama, Birmingham, AL, USA). A leukemia-associated, activating SHP2 mutant (E76K) [20] was generated by site-directed mutagenesis using the Clontech “Quickchange” protocol, as described [9]. Wt- and E76K-SHP2 were subcloned into the mammalian expression vector pSX and the pMSCVneo retroviral vector (Stratagene).

Reporter constructs

The proximal 470 bp of the human CYBB 5′ flank was obtained from Dr. Mary Dinauer (Indiana University, Indianapolis, IN, USA) and subcloned into the pCATE reporter vector (Promega, Madison, WI, USA), as described previously [2]. An artificial promoter construct was generated with four copies of the ICSBP-binding cis element from the CYBB promoter (bp −32 to −69), linked to a minimal promoter and a chloramphenicol acetyltransferase (CAT) reporter (the TATACAT vector). This construct is referred to as cybbTATACAT and has been described previously [2].

Oligonucleotides

Oligonucleotides were custom-synthesized by MWG Biotech (Piedmont, NC, USA). Double-stranded oligonucleotides used in EMSAs represented the −32- to −69-bp CYBB promoter sequence (5′-ctgctgttttcatttcctcattggaagaagaagcatag-3′), which has homology to the ets/IRF consensus, in bold type. Subcloning this oligonucleotide into a plasmid vector to generate radiolabeled EMSA probes has been described [2].

Myeloid cell line culture

The human myelomonocytic cell line U937 [24] was obtained from Andrew Kraft (Hollings Cancer Center, Medical University of South Carolina, Charleston, SC, USA). Cells were maintained and differentiated as described [2]. For differentiation experiments, U937 cells were treated for 48 h with 400 U per ml human recombinant (hr)IFN-γ (Roche, Indianapolis, IN, USA) [2].

Transfections and reporter gene assays

Transfections for promoter analysis

U937 cells were cultured and transfected as described previously [1, 2, 8]. Cells (32×10⁶ per sample) were transfected with a vector to express ICSBP (ICSBP/pcDNAamp) or empty vector control (30 µg); Wt-SHP2 (SHP2/pSX), DN-SHP2 (C463S-SHP2), SHP2 with a leukemia-associated activating mutation (E76K-SHP2/pSX), or empty vector control (30 µg); a reporter vector with the proximal 470 bp of the human CYBB promoter (CYBB-CATE) or empty pCATE vector control (50 µg); and promoter cytomegalovirus β-galactosidase (p-CMVβgal; as a control for transfection efficiency). In other assays, similar transfectants were preformed but with a dose titration of vector to express Wt-, C463S-, or E76K-SHP2, or control vector (15, 30, 60 µg). Results of CYBB-CATE reporter activity were expressed relative to background reporter activity from empty p-CATE. In other experiments, U937 cells were transfected with vectors to overexpress ICSBP or vector control (30 µg); Wt-SHP2, C463S-SHP2, E76K-SHP2/pSX, or empty vector control (30 µg); the minimal promoter/ reporter vector pTATACAT with four copies of the −32- to −69-bp CYBB sequence (cybbTATACAT; 50 µg); and p-CMVβgal. Transfectants were harvested 48 h after transfection, with or without incubation with hrIFN-γ (400 U/ml). Lysates were analyzed for CAT and βgal activity, as described [8].

Transfections for stable pools overexpressing SHP2

U937 cells were transfected with an empty expression vector or a vector to overexpress Wt-SHP2, C463S-SHP2, E76K-SHP2, or control vector, as above. After 24 h, the media were supplemented with G418 to 1 mg/ml, and stable transfectant pools were selected over 14 days, as described [2, 3], and instead of clones to compensate for possible integration site effects. All experiments were repeated with at least two independent transfectant pools for each construct.

Murine bone marrow culture and transduction

Animal studies were performed according to a protocol approved by the Animal Care and Use Committees of Northwestern University and Jesse Brown VHA Medical Center (Chicago, IL, USA). Bone marrow mononuclear cells were obtained from the femurs of Wt or ICSBP−/− C57/BL6 mice. Sca1+ cells were separated using the Miltenyi magnetic bead system, according to the manufacturer’s instructions (Miltenyi Biotechnology, Auburn, CA, USA). Bipotential myeloid progenitor cells were cultured (at a concentration of 2×10⁵ cells per ml) for 48 h in DMEM supplemented with 10% FCS, 1% pen-strep, 10 ng/ml murine (m)GM-CSF (R&D Systems Inc., Minneapolis, MN, USA), and 5 ng/ml mrIL-3 (R&D Systems Inc.). After retroviral transduction, cells were maintained in GM-CSF + IL-3 (myeloid progenitor cells) or switched to DMEM supplemented with 10% FCS, 1% pen-strep, and 10 ng/ml mrM-CSF (R&D Systems Inc.) for 96 h (monocyte differentiation). Cells were harvested, and cell lysates were used in Western blot experiments, as described below.

Retroviral transduction of murine bone marrow myeloid cells

Work with murine stem cell retrovirus was conducted according to a protocol approved by the Safety Committees of Northwestern University and Jesse Brown VHA Medical Center. High titer murine stem cell retroviral supernatants were produced using the pMSCVneo vector (for E76K-SHP2 or empty vector control) or pMSCVpuro vector (for ICSBP or empty vector control) in the PT67 packaging cell line, as per the manufacturer’s instructions (Stratagene). Filtered retroviral supernatants were used immediately or stored at −80°C. Transductions of murine bone marrow myeloid progenitor cells were performed as described previously [8]. Briefly, cells were harvested, and 4.0 × 10⁶ cells were plated in 3 ml DMEM, supplemented with 10% FCS, 10 ng/ml GM-CSF, and 5 ng/ml IL-3. An equal volume of retroviral supernatant was added to each dish, and polybrene was added to a final concentration of 6 µg/ml. Cells were incubated for 8 h at 37°C, 5% CO₂, and then diluted threefold with media supplemented as above. Cells were incubated overnight, and the procedure was repeated the next day. The day after transduction, G418 was added to 250 ng/ml and puromycin to 1.2 ng/ml. Cells were selected in antibiotics for 48 h and then treated with cytokines, as indicated. Each experiment was repeated at least three times. Expression of transduced proteins was verified independently for each experiment by Western blots.

Isolation of nuclear proteins and EMSAs

Nuclear extract proteins were isolated from U937 cells by the method of Dignam with protease inhibitors (as described) [25]. In some experiments, U937 cells were differentiated with 400 U/ml IFN-γ before nuclear protein isolation. Oligonucleotide probes were prepared, and EMSA and antibody supershift assays were performed, as described [2, 3, 8]. Antibodies to ICSBP (goat polyclonal) and control GST antibody were obtained from Santa Cruz Biotechnology (Santa Cruz, CA, USA). All experiments were performed with at least three independent batches of nuclear proteins. Integrity of proteins and equality of loading were determined in control EMSA with a probe containing a CCAAT box.

Chromatin immunoprecipitation

U937 cells were cultured with or without IFN-γ for 48 h, as described [2, 3]. Cells for chromatin immunoprecipitation were incubated with formaldehyde and lysed, and lysates were sonicated to generate chromatin fragments with an average size of 2.0 kb [26]. Lysates underwent one round of immunoprecipitation with antiserum to ICSBP or preimmune serum, as described [8]. Antibody to ICSBP (and control preimmune serum) was a kind gift of Dr. Stephanie Vogel (University of Maryland, Baltimore, MD, USA). Coprecipitated chromatin was analyzed by PCR for ICSBP antibody-specific coprecipitation of the CYBB gene.

PCR was performed in two steps. In the first step, template was amplified with primers flanking the proximal CYBB promoter (−240 to −210: 5′-attggtttcattttccactatgtttaattg-3′; +13 to −26: 5′-catggtggcagaggttgaatgtgttgtgtttg-3′). In the second step, 4% of the first-step product was amplified with primers flanking the ICSBP-binding site in the CYBB promoter (−139 to −97: 5′-agcttttcagttgaccaatgattagccaatttctgataaaag-3′;17 to −42: 5′-tgtgttgtgtttgcctttcttctatactatg-3′). Input chromatin was a positive control, and chromatin precipitated by preimmune serum was a negative control. PCR products were analyzed by acrylamide gel electrophoresis.

Immunoprecipitation and Western blots

Western blots of lysate proteins from murine bone marrow cells

Murine bone marrow cells were lysed by boiling in 2× SDS sample buffer. Lysate proteins (50 µg) were separated by SDS-PAGE and transferred to nitrocellulose, according to standard techniques. Western blots were serially probed with antibodies to gp91^PHOX, SHP2, ICSBP, and GAPDH (to control for loading). Antibodies were obtained from Santa Cruz Biotechnology.

Immunoprecipitation and Western blots

Nuclear proteins from U937 cells were diluted in “denaturing lysis buffer.” Radioimmunoprecipitation assay (RIPA) buffer (10 mM Tris, pH 7.4, 150 mM NaCl, 1% Nonidet P-40, 5 mM EDTA) was added to the lysates, and proteins were immunoprecipitated with antibody to phosphotyrosine (Clone 4G10, Upstate Biotechnology, Lake Placid, NY, USA) or irrelevant antibody (anti-GST), as described previously [2, 3]. Precipitated proteins were separated by SDS-PAGE and transferred to nitrocellulose, as above. Western blots were serially probed with antibodies to ICSBP (goat polyclonal from Santa Cruz Biotechnology).

Nitroblue tetrazolium (NBT) slide test

NBT slide tests were performed as described previously [27]. Briefly, 50 µl ex vivo-cultured cells (1×10⁶ cells/ml) are placed on each of two glass slides and incubated at 37°C. Nonadherent cells are washed away with PBS. Krebs-Ringer phosphate with glucose buffer with BSA and NBT is added to each slide, with or without PMA (20 µg/ml). Cells are incubated at 37°C, washed with PBS, fixed with methanol, and counterstained with 1% safranin. Cells are scored microscopically as percent NBT-positive or -negative out of 200 counted.

In vitro-translated proteins and in vitro phosphatase assays

In vitro-transcribed ICSBP or SHP2 mRNA was generated from linearized template DNA using the Riboprobe system, according to the manufacturer’s instructions (Promega). In vitro-translated proteins were generated in rabbit reticulocyte lysate, also according to the manufacturer’s instructions (Promega). Control (unprogrammed) lysates were generated in similar reactions in the absence of input RNA. In vitro-translated ICSBP was treated with in vitro-translated SHP2 or control lysate proteins (20 µl in vitro-translated ICSBP and 5 µl in vitro-translated SHP2 or control lysate), which were incubated for 30 min at 30°C in a 40-µl reaction volume with 1× PTP reaction buffer. The ICSBP tyrosine phosphorylation state was determined by immunoprecipitating the reaction with antiphosphotyrosine or irrelevant control antibody under denaturing conditions, as described previously [3]. Briefly, precipitated proteins were separated by SDS-PAGE, followed by autoradiography to detect in vitro-translated ICSBP that coprecipitated with the phosphotyrosine antibody.

SHP2 immunoprecipitation and PTP assays

SHP2-specific PTP assays were performed as described previously [3]. Briefly, U937 stable transfectants with SHP2, CS453-SHP2, E76K-SHP2, or control vector were lysed in RIPA buffer (in the presence of SDS). Cell lysates (200 µg) were immunoprecipitated with anti-SHP2 antibody (Santa Cruz Biotechnology) or irrelevant control antibody. Immunoprecipitates were collected with Staphylococcus protein A-Sepharose, washed extensively in RIPA buffer (with no SDS), and resuspended in PTP assay buffer (25 mM HEPES, pH 7.2, 50 mM NaCl, 5 mM DTT, 2.5 mM EDTA). Phosphopeptide (RRLIEDAEpYAARG) was added, and assays were performed according to Harder et al. [28] using a commercially available kit (PTP Assay Kit 1, Upstate Biotechnology).

Statistical analysis

Reporter gene assays

For reporter gene assays, groups of transfectants were analyzed by the technique of ANOVA between groups. Conditions with a significant difference in reporter gene expression were identified by calculating the F value and determining the P value for the null hypothesis. For individual pairs of transfectants, Student’s t-tests were used to determine the significance (P value) between individual datasets. Calculations were performed using SigmaPlot and SigmaStat software (Systat Software Inc., Richmond, CA, USA).

RESULTS

SHP2 activity regulates ICSBP-induced activation of the CYBB promoter

We previously found that ICSBP was a substrate for SHP1-PTP in undifferentiated myeloid cells. However, the contribution of the closely related SHP2-PTP to ICSBP-induced CYBB transcription was not established previously, nor was the role of regulated PTP activity during myelopoiesis. To investigate these issues, U937 cells were transfected with a reporter construct with 470 bp of 5′ flank from the CYBB gene or empty vector control. We previously found that overexpressed ICSBP activated this CYBB reporter construct in U937 transfection experiments [2, 3]. These cells were cotransfected with a vector to overexpress ICSBP or empty expression vector (control) and a vector to overexpress Wt-SHP2, a DN form of SHP2 (C463S-SHP2), a constitutively active form of SHP2 (E76K-SHP2), or empty expression vector (control). Transfectants were analyzed for reporter activity with or without IFN-γ differentiation, as we previously found that activation of this construct was increased by differentiation. Expression from the empty reporter vector is consistently low (<50 CPM) and was subtracted from the CYBB promoter vector results as background.

For these studies, we chose the U937 myeloid leukemia line. These cells undergo terminal differentiation with various cytokines, including IFN-γ [24]. Functional differentiation occurs over 48 h and is characterized by acquisition of mature phagocyte activities, including respiratory burst competence and phagocytosis. CYBB and NCF2 transcription, which requires new protein synthesis, begins to increase at 24 h and is maximal by 48 h. The cells also undergo proliferation arrest at 24 h and apoptosis at 72 h. U937 differentiation is characterized by increasing ICSBP tyrosine phosphorylation, which is detectable at 12 h and maximal at 48 h [3].

We found that ICSBP overexpression, with or without IFN-γ differentiation, significantly increased CYBB promoter activity (P≤0.001, n=3), as shown previously (Fig. 1A). This result was not altered by overexpression of Wt-SHP2 under these assay conditions. We found that coexpression of C463S-SHP2 significantly increased CYBB promoter activity in undifferentiated U937 cells overexpressing ICSBP (P=0.01, n=3). CYBB promoter activity in these transfectants was not significantly different than promoter activity in ICSBP-overexpressing, IFN-γ-differentiated transfectants (P=0.6, n=3). This result suggested that IFN-γ differentiation or SHP2 inhibition increases ICSBP-tyrosine phosphorylation and therefore, CYBB transcription. Consistent with this hypothesis, IFN-γ treatment did not further increase CYBB promoter activity in cells cotransfected with C463S-SHP2 and ICSBP expression vectors.

Fig. 1.

SHP2-PTP activity influences ICSBP-induced CYBB transcription. (A) Activation of the CYBB promoter by ICSBP was increased by a DN form of SHP2 and impaired by a constitutively active form of SHP2. U937 cells were cotransfected with reporter vector with the proximal 470 bp of the CYBB 5′ flank or empty reporter vector, a vector to express ICSBP or empty expression vector, and a vector to express SHP2, a DN form of SHP2 (C463S-SHP2), a constitutively active form of SHP2 (E76K-SHP2), or empty expression vector. Reporter assays were performed with or without IFN-γ differentiation. CYBB promoter results are presented with background from the empty reporter vector subtracted. The bars designated “Control” indicate the activity from the CYBB reporter construct with empty expression vectors only. Statistically significant differences in reporter activity (P≤0.01, n=3) are indicated by *, **, ^#, ^##, ^&, and ^&&. (B) Overexpressed Wt-SHP2 and E76K-SHP2 influenced CYBB promoter activity in a dose-dependent manner. Studies were performed to determine the relative impact of SHP2 overexpression and constitutive activation on ICSBP-induced CYBB promoter activity. U937 cells were transfected with the CYBB/reporter vector or empty reporter vector and a vector to overexpress ICSBP or empty expression vector, as above. Cells were also transfected with a dose titration (15, 30, or 60 µg) of vectors to express C463S-SHP2, E76K-SHP2, Wt-SHP2, or empty expression vector. Reporter assays were performed on differentiated transfectants, and CYBB promoter results are presented with background from the empty reporter vector subtracted. The middle dose in these studies was equivalent to the dose used in the studies above. The bars designated Control indicate the activity from the CYBB reporter construct with empty expression vectors only. Statistically significant differences in reporter activity (P≤0.02, n=3) are indicated by *, **, ^#, and ^##. (C) Activation of the composite ets/IRF cis element in the CYBB promoter by ICSBP was increased by a DN form of SHP2 and impaired by a constitutively active form of SHP2. U937 cells were cotransfected with reporter vector with the four copies of the composite ets/IRF CYBB cis element linked to a minimal promoter and reporter (cybbTATACAT) or control vector (TATACAT), a vector to overexpress ICSBP or empty expression vector, and vectors to express C463S-SHP2, E76K-SHP2, Wt-SHP2, or empty expression vector (at 30 µg). Reporter gene assays were performed with or without IFN-γ differentiation. The bars designated Control indicate the activity from cybbTATACAT or pTATACAT with empty expression vectors only. Statistically significant differences in reporter activity (P≤0.03, n=3) are indicated by *, **, ^#, and ^##.

Conversely, we found that expression of E76K-SHP2 decreased CYBB promoter activity in undifferentiated transfectants and blocked induction of CYBB promoter activity during IFN-γ differentiation. We also found that E76K-SHP2 expression significantly decreased ICSBP-induced activation of the CYBB promoter during IFN-γ differentiation and in undifferentiated transfectants (P≤0.007, n=3). This latter result was consistent with our previous observation that overexpressed ICSBP is somewhat tyrosine-phosphorylated in U937 cells, and phosphorylation further increases with differentiation [3]. None of these overexpressed proteins influenced the reporter expression from the empty vector.

These studies suggested the hypothesis that SHP2-PTP activity influenced CYBB transcription by inhibiting ICSBP-tyrosine phosphorylation. At the level of overexpression in these studies, we found that Wt-SHP2 did not have the same impact on this process as a constitutively active form. However, in planning further experiments to investigate such activating SHP2 mutations, it was important to determine the impact of SHP2 overexpression on these events. Therefore, we investigated the role of SHP2 “dose” in additional U937 transfections. These experiments were similar to those above, except with a dose titration of the vectors to express Wt-SHP2, C463S-SHP2, E76K-SHP2, or empty expression vector control (15, 30, or 60 µg; Fig. 1B). Because of the potential for activated SHP2 mutants to influence ICSBP-induced CYBB transcription during myelopoiesis, reporter activity was assayed in differentiated transfectants.

We found that increasing amounts of Wt-SHP2 overexpression decreased ICSBP-induced CYBB promoter activity at the highest dose tested (P=0.02, n=3). Expression of E76K-SHP2 also decreased ICSBP-induced CYBB promoter activity in a dose-dependent manner. However, the activated mutant impaired ICSBP-induced CYBB promoter activity at the lowest dose tested (P=0.01, n=3). In contrast, the impact of C463S-SHP2 on CYBB promoter activity was not significantly different at the various doses (P=0.8, n=3). This suggested that even the lowest dose was adequate for the DN effect. As above, none of these overexpressed proteins influenced activity from the empty reporter vector, which was subtracted as background.

These studies indicated the importance of titrating E76K-SHP2 expression to study the impact of the activating mutation, not of overexpression. For further investigations, we chose a dose of expression vector, for which E76K-SHP2 expression impaired ICSBP activation of the CYBB promoter, but Wt-SHP2 expression did not (30 µg). This issue is considered further below.

SHP2 activity regulates ICSBP-induced activation of the composite ets/IRF cis element in the CYBB promoter

The 470-bp CYBB promoter construct includes not only an ICSBP-activated cis element but also a negative cis element, which is repressed by HoxA10 in a tyrosine phosphorylation-dependent manner. Therefore, we investigated whether SHP2 inhibition or constitutive activation specifically influenced the composite ets/IRF cis element in the CYBB promoter. For these experiments, we used an artificial promoter construct with four copies of this positive cis element linked to a minimal promoter and a CAT reporter (referred to as cybbTATACAT) [2, 3]. U937 cells were cotransfected with this reporter vector or empty TATACAT control, a vector to overexpress ICSBP or empty expression vector control, and a vector to express Wt-SHP2, C463S-SHP2, E76K-SHP2, or empty expression vector control (30 µg). Transfectants were assayed for reporter activity, with or without IFN-γ differentiation.

We found that the CYBB cis element was significantly activated by IFN-γ differentiation or ICSBP overexpression (P≤0.03, n=3), as previously shown (Fig. 1C) [2, 3]. Similar to our studies with the intact promoter, we found that overexpression of Wt-SHP2 at this level did not influence activity of the ets/IRF cis element, with or without ICSBP overexpression or differentiation. We also found that expression of C463S-SHP2 in undifferentiated transfectants significantly increased ICSBP-induced CYBB cis element activity (P=0.03, n=3). As with the complete promoter, DN-SHP2 expression or IFN-γ differentiation induced an equivalent increase in CYBB cis element activity in transfectants overexpressing ICSBP (P=0.9, n=3). Also similar to our results with the intact promoter, ICSBP-induced CYBB cis element activity was not further increased by IFN-γ differentiation of transfectants co-expressing C463S-SHP2 (P=0.8, n=3). Conversely, we found that E76K-SHP2 expression significantly decreased ICSBP-induced activation of the CYBB cis element in IFN-γ-differentiated transfectants (P=0.007, n=3). In control experiments, none of the overexpressed proteins significantly altered expression from the empty TATACAT vector.

The results of these experiments were consistent with the hypothesis that SHP2-PTP activity inhibits CYBB transcription in undifferentiated myeloid cells by dephosphorylating ICSBP and that activated SHP2 mutants impair CYBB transcription in differentiating myeloid cells by inhibiting ICSBP-tyrosine phosphorylation. To investigate these hypotheses, we evaluated the influence of endogenous SHP2 using a model expressing a DN-SHP2. We also investigated the impact leukemia-associated mutant on differentiation using a model expressing constitutively active SHP2. However, our studies above indicated that it was important to distinguish between the effect of overexpression and of the activating mutation.

Therefore, we generated pools of U937 cells that were stably transfected with a vector to express Wt-SHP2, C463S-SHP2, E76K-SHP2, or empty expression vector control. At least six independent pools were generated for each construct. These pools were analyzed to identify those that were functionally equivalent to the transient transfections above. To do this, the stable transfectant pools were cotransfected with the CYBB promoter/reporter vector (or empty reporter vector control) and an ICSBP expression vector (or empty expression vector control). Pools were chosen for further analysis in which ICSBP-induced CYBB promoter activity was increased by C463S-SHP2 expression in undifferentiated cells, blocked by E76K-SHP2 expression in undifferentiated and differentiating cells, or not altered by Wt-SHP2 expression.

Nuclear proteins isolated from these stable transfectant pools were analyzed by Western blot (three to four for each construct). Blots were serially probed with antibody to SHP2, ICSBP, and Erk2, and a representative blot is shown (Fig. 2A). We identified functionally relevant pools in which Wt-SHP2, C463S-SHP2, and E76K-SHP2 were equivalently overexpressed, and overexpressing these proteins did not influence expression levels of endogenous ICSBP. In these experiments, we studied nuclear proteins, as there is little ICSBP in the cytoplasmic fractions of U937 cells, with or without differentiation [3]. In previous studies, we found SHP2 in the fraction we define as “nuclear” [9, 29]. An anti-Erk2 antibody was used to control for nuclear protein loading, as abundance of this protein is not altered by IFN-γ differentiation of U937 cells.

Fig. 2.

SHP2-PTP activity is altered by expression of DN or constitutively active SHP2 in U937 cells. (A) Overexpression of Wt-SHP2, C463S-SHP2, or E76K-SHP2 did not alter ICSBP expression in U937 transfectants. Stable transfectants of U937 cells were generated with a vector to express Wt-SHP2, C463S-SHP2, E76K-SHP2, or with empty expression vector control. Nuclear proteins from these transfectants were analyzed by Western blot (WB), which were serially probed with antibodies to SHP2, ICSBP, and Erk2 (to control for nuclear protein loading). Relative abundance of SHP2 protein under each of these conditions was determined by densitometry of anti-SHP2 Western blots. (B) SHP2-PTP activity in U937 cells was altered by expression of C463-SHP2 or E76K-SHP2. The U937 stable transfectants, described above, were also analyzed for SHP2-specific PTP activity. Nuclear proteins were isolated from these stable transfectants and immunoprecipitated under denaturing conditions with an anti-SHP2 antibody or irrelevant control antibody. Immunoprecipitates were renatured and analyzed for PTP activity. Statistically significant differences (P≤0.04, n=3) are indicated by *, **, ***, ^#, or ^##.

We next determined the impact of IFN-γ differentiation on SHP2-specific PTP activity in these stable transfectants. In these experiments, SHP2 was immunoprecipitated from cell lysates under denaturing conditions, immunoprecipitated proteins were renatured, and PTP activity was determined using a colormetric assay (as in our previous studies [3]). We found that IFN-γ differentiation significantly decreased SHP2-PTP activity in control cells (P=0.04, n=3; Fig. 2B). These results were not significantly different in stable transfectants overexpressing Wt-SHP2 (P≥0.2, n=3). SHP2-PTP activity was significantly greater in E76K-SHP2-overexpressing transfectants (P<0.0001, n=3). It was of interest that SHP2-PTP activity was not significantly altered by IFN-γ differentiation of E76K-SHP2-overexpressing cells (P=0.3, n=3). Consistent with the functional data, expression of DN-SHP2 significantly decreased SHP2-PTP activity in undifferentiated transfectants (P=0.04, n=3). These stable transfectant pools were used in further investigations of the impact of SHP2-PTP activity on ICSBP expression and CYBB transcription in differentiating myeloid cells.

Expression of DN-SHP2 increased ICSBP-tyrosine phosphorylation and binding to the CYBB cis element in undifferentiated myeloid cells

We next determined the role of endogenous SHP2-PTP activity on ICSBP-tyrosine phosphorylation. In these studies, we isolated nuclear proteins from the U937 stable transfectants overexpressing C463S-SHP2 or with empty expression vector control, described above. These proteins were immunoprecipitated under denaturing conditions with an antiphosphotyrosine antibody, immunoprecipitates were separated by SDS-PAGE, and Western blots were probed with an ICSBP-specific antibody (Fig. 3A). This experiment was repeated at least twice with two different stable transfectant pools, and a representative blot is shown. We found increased ICSBP-tyrosine phosphorylation in undifferentiated transfectants expressing C463S-SHP2. As anticipated by the functional assays, DN-SHP2 expression did not further increase ICSBP-tyrosine phosphorylation in differentiating U937 cells.

Fig. 3.

Expression of a DN form of SHP2 increased ICSBP-tyrosine phosphorylation and binding to the CYBB cis element in undifferentiated myeloid cells. (A) C463S-SHP2 expression increased ICSBP-tyrosine phosphorylation in undifferentiated U937 cells. Stable transfectants of U937 cells, described above, were generated with a vector to express C463S-SHP2 or with empty expression vector control. Nuclear proteins from these transfectants were immunoprecipitated (IP) under denaturing conditions with an antibody to phosphotyrosine (P-Y) residues or irrelevant control antibody. Immunoprecipitates were separated by SDS-PAGE, and Western blots were probed with an antibody to ICSBP. (B) Expression of C463S-SHP2 increased in vitro ICSBP binding to the composite ets/IRF CYBB cis element in undifferentiated U937 cells. In vitro ICSBP binding to the composite ets/IRF CYBB cis element was determined using nuclear proteins from the U937 stable transfectants overexpressing C463S-SHP2 or empty expression vector control, described above. EMSAs were performed with a radiolabeled probe representing the −32- to −69-bp sequence of the CYBB promoter. The top arrow indicates the tertiary complex (PU.1+IRF1+ICSBP); the second arrow from the top, the secondary complex (PU.1 + IRF1); and the third arrow from the top, the primary complex (PU.1) binding to the probe (described in refs. [2, 3]). (C) Expression of C463S-SHP2 in U937 cells increased in vivo ICSBP binding to the CYBB cis element. The stable U937 transfectants, described above, were analyzed for in vivo binding of ICSBP to the composite ets/IRF cis element in the CYBB promoter, with or without IFN-γ differentiation. Chromatin was coimmunoprecipitated from lysates of cells overexpressing C463S-SHP2 or with an empty expression vector using an antibody to ICSBP or preimmune serum. Coprecipitating chromatin was amplified by PCR with nested primer sets flanking the composite ets/IRF CYBB cis element. Nonprecipitated chromatin was a control for DNA content in these studies.

We also investigated whether increased ICSBP-tyrosine phosphorylation in undifferentiated cells expressing DN-SHP2 increased binding to the CYBB cis element. For these studies, nuclear proteins from these stable transfectants were used in EMSAs with a probe representing the CYBB cis element (−32 to −69 bp). Previously, we determined that nuclear proteins from undifferentiated U937 cells generate two major protein-DNA complexes in such experiments: a fast mobility complex, which represents PU.1 (primary complex), and a slower mobility complex, which represents a heterodimer of PU.1 + IRF1 (secondary complex) [3]. In assays with nuclear proteins from IFN-γ-differentiated U937 cells, this probe generates three DNA-bound complexes: the two described above and a slower mobility, tertiary complex, which represents PU.1 + IRF1 + ICSBP [3].

Assays with nuclear proteins from control U937 transfectants demonstrated binding of the expected complexes, with or without IFN-γ differentiation (Fig. 3B). In contrast, we found binding of the primary, secondary, and tertiary complexes in assays with nuclear proteins from U937 transfectants expressing C463S-SHP2, with or without differentiation. To determine if ICSBP was a component of the tertiary complex in EMSA with nuclear proteins from undifferentiated transfectants expressing C463S-SHP2, additional experiments were performed. For these studies, binding assays were preincubated with anti-ICSBP antibody or preimmune serum. We found that the ICSBP antibody disrupted binding of the tertiary complex in these studies (Fig. 3B). In control experiments, we demonstrated equivalent protein loading by control EMSA with a CCAAT box-containing sequence (not shown).

We were interested in determining if this in vitro assay reflected in vivo interaction of ICSBP with the CYBB promoter. Therefore, chromatin coimmunoprecipitation assays were performed with U937 stable transfectants expressing C463S-SHP2 or with control vector. Lysates were precipitated with ICSBP antibody or control preimmune serum, and CYBB chromatin was amplified by PCR. In these studies, expression of DN-SHP2 increased interaction of ICSBP with the CYBB promoter in undifferentiated, but not differentiated, transfectants (Fig. 3C). This result was consistent with the in vitro studies and the functional assays.

Constitutive activation of SHP2 decreased ICSBP-tyrosine phosphorylation and binding to the CYBB cis element in differentiating myeloid cells

We similarly investigated the impact of constitutive SHP2 activation on ICSBP-tyrosine phosphorylation and interaction with the CYBB promoter in differentiated myeloid cells. For these experiments, nuclear proteins were isolated from the U937 stable transfectants with the E76K-SHP2 vector or control expression vector, described above. These proteins were immunoprecipitated under denaturing conditions with an antibody to phosphotyrosine residues. Western blots of immunoprecipitates were probed with an anti-ICSBP antibody. As previously described [9], we found that expression of constitutively active SHP2 impaired ICSBP-tyrosine phosphorylation during differentiation of U937 myeloid cells (Fig. 4A).

Fig. 4.

Constitutive activation of SHP2 blocks ICSBP-tyrosine phosphorylation and binding to the CYBB cis element during myeloid differentiation. (A) Expression of E76K-SHP2 decreased ICSBP-tyrosine phosphorylation in U937 cells. Stable transfectants of U937 cells, described above, were generated with a vector to express an activated form of E76K-SHP2 or with empty expression vector control. Nuclear proteins from these transfectants were immunoprecipitated under denaturing conditions with an antibody to phosphotyrosine residues or irrelevant control antibody. Immunoprecipitates were separated by SDS-PAGE, and Western blots were probed with antibody to ICSBP. *, Tyrosine-phosphorylated ICSBP. (B) Expression of E76K-SHP2 decreased in vitro ICSBP binding to the composite ets/IRF CYBB cis element in differentiated U937 cells. In vitro ICSBP binding to the composite ets/IRF CYBB cis element was determined using nuclear proteins from the U937 stable transfectants overexpressing E76K-SHP2 or with empty expression vector control, described above. EMSAs were performed with a radiolabeled probe representing the −32- to −69-bp sequence of the CYBB promoter. The top arrow indicates the tertiary complex (PU.1+IRF1+ICSBP); the second arrow from the top, the secondary complex (PU.1+IRF1); and the third arrow from the top, the primary complex (PU.1) binding to the probe. (C) Expression of E76K-SHP2 in U937 cells decreased in vivo ICSBP binding to the CYBB cis element. The stable U937 transfectants, described above, were analyzed for in vivo binding of ICSBP to the composite ets/IRF cis element in the CYBB promoter, with or without IFN-γ differentiation. Chromatin was coimmunoprecipitated from lysates of cells overexpressing E76K-SHP2 or with control vector using an antibody to ICSBP or preimmune serum. Coprecipitating chromatin was amplified by PCR with nested primer sets flanking the composite ets/IRF CYBB cis element. Nonprecipitated chromatin was a control for DNA content in these studies.

Therefore, we investigated the impact of constitutive SHP2 activation on ICSBP binding to the positive cis element in the CYBB gene in vitro. As in the section above, EMSAs were performed with a radiolabeled probe representing the −32- to −69-bp sequence from the CYBB promoter and nuclear proteins from U937 stable transfectants expressing E76K-SHP2 or vector control (Fig. 4B). In EMSAs with nuclear proteins from E76K-SHP2-expressing U937 transfectants, only the primary and secondary complexes were generated, with or without IFN-γ treatment. As ICSBP-tyrosine phosphorylation is necessary for participation in the CYBB activation complex, this result supports our hypothesis that constitutive SHP2 activation impairs ICSBP-induced CYBB transcription during myelopoiesis.

To determine the impact of constitutive activation of SHP2 on ICSBP binding to the CYBB promoter in vivo, chromatin coimmunoprecipitation experiments were performed using these stable transfectants. We found that E76K-SHP2 expression blocked differentiation-induced ICSBP binding to the CYBB promoter in U937 cells in vivo (Fig. 4C), consistent with our in vitro results.

Constitutive activation of SHP2 impaired ICSBP-induced gp91^PHOX expression and respiratory burst activity in differentiating murine myeloid progenitor cells

Oher investigators [6] previously generated an ICSBP-deficient murine model. These mice were found to have B cell abnormalities [6]. Based on our previous studies and the results above, we hypothesized that phagocytes from ICSBP-deficient mice would have impaired expression of gp91^PHOX and therefore, impaired respiratory burst activity. We also hypothesized that ICSBP-induced gp91^PHOX expression and NADPH-oxidase activity would also be impaired by constitutive activation of SHP2.

To investigate these hypotheses, bone marrow myeloid progenitors were isolated from ICSBP−/− or Wt mice (control littermates) and cultured in GM-CSF and IL-3 to generate granulocyte-monocyte progenitors (or CFU-GM) [9, 29]. Cells were ex vivo-differentiated with M-CSF (described previously) [9, 29]. Previous studies in our laboratory and by others demonstrated that the majority of Wt and ICSBP−/− cells become Sca1⁻CD14⁺Mac1⁺ under these conditions. RNA was isolated from cultured cells and analyzed by Northern blot for gp91^PHOX and ICSBP mRNA expression (Fig. 5A). We found significantly less gp91^PHOX expression in ICSBP−/− monocytes in comparison with Wt cells. ICSBP mRNA was undetectable in ICSBP−/− cells, as expected. We also tested whether re-expression of ICSBP would rescue gp91^PHOX expression in ex vivo-differentiated ICSBP−/− cells and the impact of constitutive SHP2 activation on this process. For these studies, myeloid progenitors from ICSBP−/− murine bone marrow were transduced with a retroviral vector to express ICSBP or with control vector and a vector to express E76K-SHP2 or control vector. Transduced cells were selected in antibiotics, and progenitors were harvested or ex vivo-differentiated with M-CSF. Western blots of cell lysates were serially probed with antibodies to gp91^PHOX, ICSBP, SHP2, and GAPDH (to control for protein loading).

Fig. 5.

ICSBP deficiency and constitutive SHP2 activation inhibit gp91^PHOX expression and respiratory burst activity in differentiating myeloid cells. (A) Expression of gp91^PHOX is decreased in ex vivo-differentiated monocytes from ICSBP−/− mice. Myeloid progenitor cells from the bone marrow of Wt or ICSBP−/− C57BL/6 mice were ex vivo-differentiated with M-CSF. Total RNA was extracted and analyzed by Northern blot. Expression of ICSBP was a negative control and 18S, a control for RNA loading in this experiment. (B) Rescue of gp91^PHOX expression in ex vivo-differentiating ICSBP−/− monocytes by re-expression of ICSBP is blocked by coexpression of constitutively active SHP2. Myeloid progenitor cells were isolated from the bone marrow of ICSBP−/− mice and transduced with vectors to express ICSBP or empty expression vector control and E76K-SHP2 or empty expression vector control. Cells were cultured in GM-CSF, IL-3, and stem cell factor or differentiated with M-CSF. Total cell lysates were analyzed by Western blots probed with antibodies to gp91^PHOX, ICSBP, SHP2, and GAPDH (to control for loading). Relative abundance of SHP2 protein under each of these conditions was determined by densitometry of anti-SHP2 Western blots. (C) Expression of constitutively active SHP2 prevented re-expression of ICSBP from rescuing NADPH-oxidase activity in ICSBP−/− cells during ex vivo monocyte differentiation. Myeloid progenitor cells were isolated from the bone marrow of Wt or ICSBP−/− mice, as described above. ICSBP−/− cells were transduced with a vector to express ICSBP or empty expression vector control and a vector to express E76K-SHP2 or empty expression vector control. Wt cells were transduced with empty expression vector control. Cells were ex vivo-differentiated with M-CSF, treated with PMA (to activate the NADPH oxidase), and analyzed for NADPH-oxidase activity by a NBT slide test. Statistically significant differences (P≤0.001, n=4) are indicated by *, **, and ^#. (D) ICSBP is dephosphorylated by SHP2 in vitro, and in vitro-translated (IVT) ICSBP and SHP2 were used to determine if ICSBP is a SHP2 substrate. ICSBP and SHP2 were in vitro-translated in rabbit reticulocyte lysate. ³⁵S-Methionine was added to the ICSBP-translation reaction to label the protein. In vitro-translated ICSBP was incubated with in vitro-translated SHP2 or control lysate. Proteins were immunoprecipitated under denaturing conditions with antibody to phosphotyrosine residues or control antibody. Immunoprecipitates were separated by SDS-PAGE, and ICSBP was detected by autoradiography.

We found that transduction with an ICSBP-expression vector increased gp91^PHOX expression in ICSBP−/− myeloid progenitors and ex vivo-differentiated monocytes (Fig. 5B). However, coexpression of E76K-SHP2 blocked gp91^PHOX expression during monocyte differentiation of ICSBP−/− cells transduced with ICSBP expression vector. These results were consistent with our previous studies, in which expression of E76K-SHP2 in Wt murine myeloid progenitors inhibited tyrosine phosphorylation of endogenous ICSBP during ex vivo M-CSF differentiation [9].

To determine the physiologic relevance of decreased gp91^PHOX expression for NADPH-oxidase activity, NBT slide tests were performed. In these studies, only cells with NADPH-oxidase activity are NBT-positive. Wt cells were a positive control for NADPH-oxidase competence in these studies. We found a significantly lower percent of NBT-positive ICSBP−/− cells in comparison with Wt after M-CSF differentiation (P=0.001, n=4; Fig. 5C). We also found that re-expression of ICSBP in ICSBP−/− cells significantly increased the percent of NBT-positive ex vivo monocytes (P<0.001, n=4). This increase was significantly inhibited by coexpression of E76K-SHP2 (P=0.02, n=4). These results were consistent with the impact of ICSBP reconstitution and E76K-SHP2 expression on gp91^PHOX protein abundance in ICSBP−/− cells.

In previous studies, we demonstrated that ICSBP was a substrate for constitutively active SHP2 in an assay using recombinant proteins [9]. However, we had not shown previously that ICSBP was a substrate for Wt-SHP2. This is an important issue, as it is possible that leukemia-associated, constitutively active mutants of SHP2 have altered substrate specificity. Therefore, ICSBP might be a substrate for E76K-SHP2 but not the Wt protein. To address this, we performed experiments similar to those prior studies [9]. ICSBP was in vitro-translated in rabbit reticulocyte lysate (in the presence of ³⁵S-methionine) and incubated with in vitro-translated SHP2 (unlabeled) or control lysate. After incubation, reactions were immunoprecipitated under denaturing conditions with an antiphosphotyrosine antibody (or irrelevant control antibody), and immunoprecipitates were separated by SDS-PAGE and were ICSBP-detected by autoradiograph (Fig. 5D). As in our previous studies, we found that in vitro-translated ICSBP was tyrosine-phosphorylated in the control sample [9]. However, incubation with recombinant SHP2 decreased the abundance of tyrosine-phosphorylated, in vitro-translated ICSBP, without altering the abundance of total ICSBP.

DISCUSSION

We previously found that CYBB transcription was activated by ICSBP in a differentiation stage-specific and tyrosine phosphorylation-dependent manner [2, 3]. In additional studies, we determined that SHP1-PTP was involved in maintaining ICSBP in a nontyrosine-phosphorylated state in myeloid progenitors [3]. In the current study, we demonstrated that ICSBP was also dephosphorylated by the closely related SHP2-PTP in myeloid progenitors but not in maturing myeloid cells. Therefore, our results suggested that the combined effects of SHP1 and SHP2 influence the ICSBP tyrosine-phosphorylation state in myeloid progenitors but not mature phagocytes. Our studies also demonstrated that regulation of SHP2-PTP activity was necessary for ICSBP target-gene transcription during myeloid differentiation. We found that constitutive SHP2 activation blocked ICSBP-tyrosine phosphorylation and ICSBP-induced CYBB transcription during myelopoiesis. In additional studies, we determine that respiratory burst activity was impaired in primary monocytes from ICSBP−/− mice. Consistent with the results above, we found that coexpression of constitutively active SHP2 prevented ICSBP expression from rescuing gp91^PHOX expression or phagocyte oxidase activity in these cells.

SHP1 and SHP2 are structurally similar and are expressed in myeloid and B cells. Despite these similarities, little is known about the relative roles of these PTPs in myelopoiesis and phagocyte function. Some studies have suggested that SHP1 and SHP2 exhibit antagonistic functions during myelopoiesis. Our studies indicate that the situation may be more complex. Most of the information regarding SHP1 function was derived from naturally occurring murine mutants with partial or complete SHP1 deficiency (viable moth-eaten and moth-eaten phenotype) [30–32]. These animals are characterized by the accumulation of activated phagocytes in the circulation and tissues and accelerated differentiation to mature phagocyte [30, 31]. Moth-eaten mice die at an early age of bone marrow exhaustion [32] and viable moth-eaten mice at an older age of pulmonary inflammation [30]. In contrast, most of the information about SHP2 and hematopoiesis has been derived from constitutively active mutants. Constitutive SHP2 activation leads to dysregulated myeloid progenitor proliferation in response to various cytokines, including GM-CSF and M-CSF. This results in leukocytosis and accumulation of progenitor cells in the bone marrow [33]. Therefore, although SHP1 deficiency and constitutive activation of SHP2 induce peripheral neutrophilia, the mechanisms involved in these processes appear to be different.

We previously found that the homeodomain protein HoxA10 was also a substrate for SHP1 and SHP2 in immature myeloid cells [29, 34]. Similar to ICSBP, the impact of these PTPs on HoxA10-tyrosine phosphorylation also decreased during myelopoiesis [29]. Also similar to ICSBP, we found that constitutively active SHP2 dephosphorylated HoxA10 throughout the differentiation process. These results are significant to the current studies, as HoxA10 also influences CYBB and NCF2 transcription in a tyrosine phosphorylation-dependent manner [34, 35]. In undifferentiated myeloid cells, HoxA10 repressed homologous cis elements in the CYBB and NCF2 genes [35]. These cis elements contain Hox/Pbx-DNA-binding consensus sequences and are distinct from the composite ets/IRF cis elements in the genes [35]. HoxA10-tyrosine phosphorylation during myeloid differentiation decreased DNA-binding affinity for these cis elements, relieving repression [35]. Therefore, activated SHP2 mutants blocked CYBB transcription by two mechanisms: impaired assembly of the PU.1/IRF1/ICSBP/CBP activation complex and sustained binding of the HoxA10/Pbx repression complex. This results in respiratory burst incompetence and functional differentiation block.

These results have implications for understanding the innate immune response in human myeloid malignancy. ICSBP deficiency is found in tMDS/tAML [11]. Activating SHP2 mutants were described in a different study of subjects with tMDS/tAML [36]. It will be of interest to investigate simultaneous coexistence of these two leukemia-associated mutations in the same subject. Phagocytes from MDS subjects have decreased secondary granules and impaired respiratory burst activity [37]. Additionally, MDS phagocytes do not function normally, even in subjects with normalization of circulating neutrophil numbers as a result of G-CSF treatment [38]. ICSBP deficiency is also found in uncontrolled, refractory, and progressive CML [12]. Interestingly, constitutive activation of Wt-SHP2 is found in cells expressing the bcr/abl kinase [39, 40]. Therefore, ICSBP deficiency and SHP2 dysregulation may also coexist in the same subject with human CML. Respiratory burst activity is impaired in phagocytes from CML subjects, but remission is associated with increased ICSBP expression and normalization of phagocyte function [41–43]. Our studies identify potential molecular mechanisms for these previously observed phagocyte function defects and may have implications for therapeutic approaches to the infection complications in these myeloid malignancies.