ALBERT

Description: The production of placental growth factor (PlGF) increases during normal third trimester pregnancy, a point at which women are at an increased risk of morbid viral infections (e.g. an 8-fold increased risk of mortality during the 2009 H1N1 influenza pandemic). Pathologic increases in PlGF have also been detected in sickle cell disease patients and are associated with similarly increased morbidity (hospitalizations and acute chest syndrome). While treatment of normal donor mononuclear cells with exogenous PlGF reportedly increases cytokine transcripts (including tumor necrosis factor-alpha [TNF]), whether TNF mRNA is translated is unclear, as are: (a) the comparative activity of PlGF isoforms, (b) the scope of inflammatory cytokines involved, (c) the TNF-dependency of other induced cytokines, (d) and the potential effects of PlGF when combined with toll-like receptor (TLR) pathway activators. Consequently, we sought to test the hypothesis that PlGF could enhance pro-inflammatory responses of peripheral blood monocytes to TLR-stimulation and focused on endosomal TLRs which would be expected to be activated by the types of viral infections (ssRNA) for which pregnant women and sickle cell anemia patients are at risk. Normal primary peripheral blood CD14+ cells were exposed to the PlGF isoforms (PlGF-1, -2, -3). None induced TNF secretion. However, when CD14+ cells were exposed to both PlGF-1 (250 ng/ml) and the TLR-7/8 ligand R848 (3 μM), production of TNF mRNA (qRT-PCR) at 4-6 hours (mean 2.52-fold, p=0.05, 10 separate experiments) and TNF protein (ELISA) at 24-hours (mean 2.15-fold, p=0.0007, 24 separate experiments) was significantly enhanced. PlGF-2 and -3 had no effect on TLR-7/8 activated cells. PlGF-1 did not prolong TNF mRNA half-life (Actinomycin D chase results) but did enhance TNF gene transcription. Specifically, in kinetic studies we quantified equivalent increases in both spliced and unspliced TNF mRNA and TNF protein (Figure 1). The same experiments indicated that PlGF-1 prolongs the duration of TNF gene transcription compared to cells treated with R848 alone. The synergistic effect of PlGF with TLR-7/8 was not seen with the TLR-4 activator LPS. VEGF-A exhibited no influence on TLR-7/8 induced TNF gene expression. We sought to identify signaling pathways involved in the PlGF/TLR effect and determined by western blotting (total IKK-α and β, and phospho-IKK-α/β) that PlGF markedly enhanced R848-induced phosphorylation of IKK. Moreover, treatment of primary cells with an IKK inhibitor (BMS 345541; 0.5, 1.5, 5 μM) inhibited the enhancing effect of PlGF on R848-induced TNF production. TNF gene expression was not inhibited by small molecule inhibitors of p38 MAPK, PI3K, MEK-1/2, or ERK-1/2. Gene microarray analysis (Affymetrix HTA 2.0) revealed a PlGF/R848-specific increase in RIPK1 mRNA encoding a known IKK activator. To identify other inflammatory cytokines influenced by PlGF/TLR, and their dependency on TNF, we performed Luminex multiplex cytokine arrays of conditioned culture media from CD14+ cells cultured for 6 and 24-hours with PlGF-1 (250 ng/ml) and R848 (3 μM), and found (n=2 biological replicates) that PlGF enhanced R848-induced protein levels of: TNF, interleukin-6 (IL-6), IL-8, IL-1β, macrophage inflammatory protein 1 alpha (MIP-1α), MIP-1β, monocyte chemoattractant protein-1 (MCP-1), interferon-alpha, G-CSF, and monocyte interferon gamma inducing factor (MIG). To test the idea that the production of these factors was TNF-dependent, CD14+ cells were cultured as above in the presence and absence of soluble TNF receptor (sTNFR) (0.1, 1, 10 μg/ml). Complete inhibition of signal transduction through the TNF receptor had no effect on the production of other inflammatory cytokines (i.e. IL-8). We conclude that PlGF directly contributes to an exaggerated pro-inflammatory response in human mononuclear phagocytes incited by TLR-7/8 activation and suggest that it does so by promoting the expression or activation of an IKK kinase (e.g. RIPK1) and/or suppressing an IKK phosphatase. We also conclude that the production of inflammatory cytokines by PlGF and R848 is independent of autocrine TNF production. Clarification of the effect of PlGF on TLR-pathway hyper-reactivity may lead to the identification of key pathways that can be safely suppressed in individuals with high PlGF levels to achieve the goal of reducing virus related morbidity and mortality. Figure 1. Figure 1. Disclosures Tyner: Constellation Pharmaceuticals: Research Funding; Aptose Biosciences: Research Funding; Array Biopharma: Research Funding; Janssen Pharmaceuticals: Research Funding; Incyte: Research Funding.

Description: Replication and survival of FA-deficient hematopoietic stem and progenitor cells (HSPC) are highly suppressed by exposure to inflammatory cytokines including TNFα, IL-1β, IFNγ, and MIP-1α. Additionally, FA macrophages exposed to specific toll-like receptor (TLR) ligands overproduce the very inflammatory cytokines that exhaust the stem cell pool. Recent evidence using double knockout mice suggests that aldehyde dehydrogenases (ALDH) may play a role in protecting FA HSPC but the mechanisms involved are unclear. We tested the hypothesis that the TLR-dependent overproduction of inflammatory cytokines in FA macrophages results from a loss of FA protein-dependent ALDH function. Control (T-shNT) or FANCC-deficient (T-shFC) THP-1 human monocytic leukemia cells were treated with Alda-1 (a small molecule ALDH agonist known to enhance the activity of both ALDH1A1 and ALDH2), before exposing them to the TLR-7/8 agonist R848 and quantifying the production of TNFα (ELISA of 24- culture supernatants). While R848 alone induced a 6-fold increase in TNFα production by the FANCC-deficient cells, Alda-1 suppressed TLR-induced TNFα production in a dose-dependent manner, by both T-shNT and T-shFC cells (Fig. 1). To identify the ALDH isoform that played a role in the suppression of TNFα production, we ectopically expressed ALDH1A1 or ALDH2 in T-shNT and T-shFC cells. Overexpression of ALDH2 had little effect on R848-induced TNFα production, but ALDH1A1 overexpression completely suppressed TNFα production by FANCC-deficient cells and suppressed (by approximately 50%) the production of TNFα by control cells (Fig. 2). Likewise, in loss-of-function studies, siRNA knockdown of ALDH1A1 (but not ALDH2) enhanced R848-induced production of TNFα (1.8-fold) in T-shNT cells but did not enhance TNFα overproduction in either FANCC-deficient (T-shFC) or FANCA-deficient (T-shFA [FANCA knockdown]) cells. Additionally, ALDH1A1 (but not ALDH2) mRNA was highly inducible in both THP-1 cells (by R848) and in Lin- Sca-1+ Kit+ murine marrow cells (by TNFα). While ALDH1A1 mRNA is expressed similarly in both T-shNT and T-shFC cells, our results indicate that in normal macrophages ALDH1A1 plays a role in the constraining TLR-induced TNF production but is significantly less functional in FANCC-deficient macrophages. In summary, (1) either pharmacological enhancement of ALDH function with Alda-1 or overexpression of ALDH1A1 is sufficient to restore the TLR-suppressive function of ALDH1A1 in FANCC-deficient macrophages, and (2) specific suppression of ALDH1A1, (but not ALDH2) induces an FA-like phenotype in control macrophages. We conclude that (1) optimal function of ALDH1A1 is FANCC- and FANCA-dependent in normal macrophages, (2) TLR-dependent overproduction of inflammatory cytokines by FA-deficient macrophages may result either from an increase in aldehyde load or the loss of a non-canonical signal-suppressive function of ALDH1A1, and (3) enhancement of ALDH activity using small molecule agonists such as Alda-1 may alleviate the FA macrophage/cytokine phenotype and thereby protect HSC from inflammation-induced exhaustion. Disclosures No relevant conflicts of interest to declare.

Description: Fanconi anemia, complementation group C (FANCC)–deficient hematopoietic stem and progenitor cells are hypersensitive to a variety of inhibitory cytokines, one of which, TNFα, can induce BM failure and clonal evolution in Fancc-deficient mice. FANCC-deficient macrophages are also hypersensitive to TLR activation and produce TNFα in an unrestrained fashion. Reasoning that suppression of inhibitory cytokine production might enhance hematopoiesis, we screened small molecules using TLR agonist–stimulated FANCC- and Fanconi anemia, complementation group A (FANCA)–deficient macrophages containing an NF-κB/AP-1–responsive reporter gene (SEAP). Of the 75 small molecules screened, the p38 MAPK inhibitor BIRB 796 and dasatinib potently suppressed TLR8-dependent expression of the reporter gene. Fanconi anemia (FA) macrophages were hypersensitive to the TLR7/8 activator R848, overproducing SEAP and TNFα in response to all doses of the agonist. Low doses (50nM) of both agents inhibited p38 MAPK–dependent activation of MAPKAPK2 (MK2) and suppressed MK2-dependent TNFα production without substantially influencing TNFα gene transcription. Overproduction of TNFα by primary FA cells was likewise suppressed by these agents and involved inhibition of MK2 activation. Because MK2 is also known to influence production and/or sensitivity to 2 other suppressive factors (MIP-1α and IFNγ) to which FA hematopoietic progenitor cells are uniquely vulnerable, targeting of p38 MAPK in FA hematopoietic cells is a rational objective for preclinical evaluation.

Description: Key PointsTLR-activated FANCA- and FANCC-deficient macrophages overproduce IL-1β. IL-1β suppresses in vitro expansion of Fancc-deficient multipotent hematopoietic progenitor cells.

Description: Hematopoietic progenitor cells (HPC) from mice nullizygous at the Fanconi anemia (FA) group C locus and children with Fanconi anemia group C (FA-C) are hypersensitive to interferon-gamma (IFN-γ) and tumor necrosis factor-α. This hypersensitivity results, in part, from the capacity of these cytokines to prime the fas pathway. Because fas-mediated programmed cell death in many cells involves sequential activation of specific caspases, we tested the hypothesis that programmed cell death in FA HPC involves the ordered activation of specific caspase molecules. Lysates from lymphoblasts treated with both agonistic anti-fas antibody and IFN-γ contained activated caspase 3 family members (caspases 3, 6, and 7), as well as caspase 8, whereas activation of caspases 1, 2, 4, 9, and 10 was not detected. The apoptotic effects of fas agonists in IFN-γ-treated human and murine FA-C cells were blocked when pretreated with inhibitors (ac-DEVD-cho, CP-DEVD-cho, Z-DEVD-FMK) of the caspase 3 protease. Inhibitors (ac-YVAD-cho, CP-YVAD-cho, Z-YVAD-FMK) of caspase 1 did not block apoptosis or caspase 3 activation. Treatment of FA cells with the fluoromethyl ketone tetrapeptide caspase 8 inhibitor (ac-IETD-FMK) did suppress caspase 3 activation. A 4-fold greater fraction of IFN-induced FA-C cells expressed caspase 3 than FA-C cells complemented by retroviral-mediated transfer of FANCC. Therefore fas-induced apoptosis in Fanconi anemia cells of the C type involves the activation of caspase 8, which controls activation of caspase 3 family members and one direct or indirect function of the FANCC protein is to suppress apoptotic responses to IFN-γ upstream of caspase 3 activation.

Description: The Fanconi anemia (FA) group C gene product (FANCC) functions to protect cells from cytotoxic and genotoxic effects of cross-linking agents. FANCC is also required for optimal activation of STAT1 in response to cytokine and growth factors and for suppressing cytokine-induced apoptosis by modulating the activity of double-stranded RNA-dependent protein kinase. Because not all FANCC mutations affect STAT1 activation, the hypothesis was considered that cross-linker resistance function of FANCC depends on structural elements that differ from those required for the cytokine signaling functions of FANCC. Structure-function studies were designed to test this notion. Six separate alanine-substituted mutations were generated in 3 highly conserved motifs of FANCC. All mutants complemented mitomycin C (MMC) hypersensitive phenotype of FA-C cells and corrected aberrant posttranslational activation of FANCD2 in FA-C mutant cells. However, 2 of the mutants, S249A and E251A, failed to correct defective STAT1 activation. FA-C lymphoblasts carrying these 2 mutants demonstrated a defect in recruitment of STAT1 to the interferon γ (IFN-γ) receptor and GST-fusion proteins bearing S249A and E251A mutations were less efficient binding partners for STAT1 in stimulated lymphoblasts. These same mutations failed to complement the characteristic hypersensitive apoptotic responses of FA-C cells to tumor necrosis factor-α (TNF-α) and IFN-γ. Cells bearing a naturally occurring FANCC mutation (322delG) that preserves this conserved region showed normal STAT1 activation but remained hypersensitive to MMC. The conclusion is that a central highly conserved domain of FANCC is required for functional interaction with STAT1 and that structural elements required for STAT1-related functions differ from those required for genotoxic responses to cross-linking agents. Preservation of signaling capacity of cells bearing the del322G mutation may account for the reduced severity and later onset of bone marrow failure associated with this mutation.

Description: Cells from individuals with Fanconi anemia (FA) arrest excessively in the G2/M cell cycle compartment after exposure to low doses of DNA cross-linking agents. The relationship of this abnormality to the fundamental genetic defect in such cells is unknown, but many investigators have speculated that the various FA genes directly regulate cell cycle checkpoints. We tested the hypothesis that the protein encoded by the FA group C complementing gene (FAC) functions to control a cell cycle checkpoint and that cells from group C patients (FA[C]) have abnormalities of cell cycle regulation directly related to the genetic mutation. We found that retroviral transduction of FA(C) lymphoblasts with wild-type FAC cDNA resulted in normalization of the cell cycle response to low-dose mitomycin C (MMC). However, when DNA damage was quantified in terms of cytogenetic damage or cellular cytotoxicity, we found similar degrees of G2/M arrest in response to equitoxic amounts of MMC in FA(C) cells as well as in normal lymphoblasts. Similar results were obtained using isogenic pairs of uncorrected, FAC- or mock-corrected (neo only) FA(C) cell lines. To test the function of other checkpoints we examined the effects of hydroxyurea (HU) and ionizing radiation on cell cycle kinetics of FA(C) and normal lymphoblasts as well as with isogenic pairs of uncorrected, FAC-corrected, or mock-corrected FA(C) cell lines. In all cases the cell cycle response of FA(C) and normal lymphoblasts to these two agents were identical. Based on these studies we conclude that the aberrant G2/M arrest that typifies the response of FA(C) cells to low doses of cross-linking agents does not represent an abnormal cell cycle response but instead represents a normal cellular response to the excessive DNA damage that results in FA(C) cells following exposure to low doses of cross-linking agents.

Description: Hematopoietic progenitor cells (HPC) from mice nullizygous at the Fanconi anemia (FA) group C locus and children with Fanconi anemia group C (FA-C) are hypersensitive to interferon-gamma (IFN-γ) and tumor necrosis factor-α. This hypersensitivity results, in part, from the capacity of these cytokines to prime the fas pathway. Because fas-mediated programmed cell death in many cells involves sequential activation of specific caspases, we tested the hypothesis that programmed cell death in FA HPC involves the ordered activation of specific caspase molecules. Lysates from lymphoblasts treated with both agonistic anti-fas antibody and IFN-γ contained activated caspase 3 family members (caspases 3, 6, and 7), as well as caspase 8, whereas activation of caspases 1, 2, 4, 9, and 10 was not detected. The apoptotic effects of fas agonists in IFN-γ-treated human and murine FA-C cells were blocked when pretreated with inhibitors (ac-DEVD-cho, CP-DEVD-cho, Z-DEVD-FMK) of the caspase 3 protease. Inhibitors (ac-YVAD-cho, CP-YVAD-cho, Z-YVAD-FMK) of caspase 1 did not block apoptosis or caspase 3 activation. Treatment of FA cells with the fluoromethyl ketone tetrapeptide caspase 8 inhibitor (ac-IETD-FMK) did suppress caspase 3 activation. A 4-fold greater fraction of IFN-induced FA-C cells expressed caspase 3 than FA-C cells complemented by retroviral-mediated transfer of FANCC. Therefore fas-induced apoptosis in Fanconi anemia cells of the C type involves the activation of caspase 8, which controls activation of caspase 3 family members and one direct or indirect function of the FANCC protein is to suppress apoptotic responses to IFN-γ upstream of caspase 3 activation.