through interferon regulatory factor-1 Hematopoietic ...

From bloodjournal.hematologylibrary.org by guest on July 13, 2011. For personal use only.

1995 86: 3373-3380

Hematopoietic inhibition by interferon-gamma is partially mediated through interferon regulatory factor-1 T Sato, C Selleri, NS Young and JP Maciejewski

Information about reproducing this article in parts or in its entirety may be found online at: http://bloodjournal.hematologylibrary.org/site/misc/rights.xhtml#repub_requests Information about ordering reprints may be found online at: http://bloodjournal.hematologylibrary.org/site/misc/rights.xhtml#reprints Information about subscriptions and ASH membership may be found online at: http://bloodjournal.hematologylibrary.org/site/subscriptions/index.xhtml

Blood (print ISSN 0006-4971, online ISSN 1528-0020), is published weekly by the American Society of Hematology, 2021 L St, NW, Suite 900, Washington DC 20036. Copyright 2011 by The American Society of Hematology; all rights reserved.

From bloodjournal.hematologylibrary.org by guest on July 13, 2011. For personal use only.

Hematopoietic Inhibition by Interferon-y Is Partially Mediated Through Interferon Regulatory Factor-l By Tadatsugu Sato, Carmine Selleri, Neal S. Young, and Jaroslaw P. Maciejewski Biologic responses t o cytokines are mediated by intracellular pathways involving induction of signaling and metabolic cascades. Interferon (IFN) regulatory factor-l (IRF-1)is a major transcription factor induced not only by IFN-y but also by other cytokines including tumor necrosis factor-a (TNFa). Possible IRF-1 binding sequence elements have beenlocated in the promoter regions of several genes, including p53, inducible nitric oxide synthase, and cyclin Dl. IFN-y and TNF-acan inhibit hematopoiesis in vitro and have been implicated in the pathophysiology of bone marrow (BM)failure. We investigated whether the inhibitory effects of these cytokines were intracellularly mediated through the expression of IRF-1 or -2 in target cells. In total BM cells, IRF-l mRNA expression increased after stimulation with IFN-yand TNF-a; the stronger effect was observed with IFN-y. In contrast, IRF-2 mRNA expression was constitutive and not altered by cytokine stimulation. By gene amplification, low

levels of IRF-1 mRNA were present in unstimulated, highly purified CD34+ cells;on exposuret o IFN-y and TNF-a, amplified IRF-1 mRNA showed a much stronger signal than control. When CD34’ cells were treated with IFN-y and TNF-a. IRF-1 antisense oligodeoxynucleotide (ODN) partially reversed the suppressive effectson CD34’ cell-derivedcolony formation by IFN-y but not those by TNF-(U.In parallel experiments,IRF-1 antisense ODNdecreased both IRF-1 protein and mRNA expression. The effects of ODN were sequencespecific and concentration-dependent. These results suggest that the inhibitoryhematopoietic effects of IFN-y andTNFa are mediated by different pathways. For IFN-y, IRF-1 is involved in the activation of cellular genes responsible for IFN-y suppressive effects. This is a US government work. There are no restrictions on its use.

I

on hematopoiesis in vitr~~*-’~ andinanimal models~’~25 and they may play a pathophysiologic role in BM failure syndromes including aplastic anemia (AA):629 Fanconi anemia,30 myelody~plasia,3’~~’ andpost-Epstein-Barrvirushemophagocytic syndr0me.3~In laboratory studies, IFN-y and TNF-a exerted inhibitoryactivityacrossawidespectrumofhematopoietic cells,includingphenotypicallyimmature CD34TD38- and thefunctionallymostprimitivelong-termcultureinitiating cell^.^' The effect of E X - y and TNF-a isdirectanddoes not require the presence of accessory cell^.^.^' Both cytokines induceFas-antigenexpressiononearlyhematopoieticcells, rendering them susceptible to Fas-mediated killing.% Recently, we showed that IFN-y- and TNF-a-mediated hematopoietic suppression were associated with induction of apoptotic BM progenitor cell death3’and a block in cell-cycle progression (Selleri C, Maciejewski P, Young NS: submitted). Although much effort has been committed to the analysis of the hematopoietic effects of these cytokines in tissue culture experiments, the molecular mechanism by which IFNy and TNF-a exert their antiproliferative effects in BM is still unclear, and different pathways have been proposed for IFN-y and TNF-a signal transduction. For most cells displaying TNF receptor type I (p55), TNF-a signals are transduced through phosphatidylcholine-specific phospholipase C,37protein kinase C,38transcription factor AP-l,39NFKB,‘“‘and IRF-1 with subsequent activation of various TNF-

NTERFERON (IFN) regulatory factor-l (IRF-1) isa member of a transcription factor family that also includes IRF-2, IFN-a-stimulated gene factor 3y (ISGF-3y), and IFN consensus sequence binding protein (ICSBP).’.’ IRF-1 was originally implicated in IFN-a signal transduction but subsequently has been also associated with IFN-y, tumor necrosis factor (TNF), interleukin-l (IL-l), IL-6, leukemia inhibitory factor (LIF), and ~ r o l a c t i n . ~IRF . ~ binding sequence elements (IRF-Es), which overlap IFN-stimulated response elements (ISREs), have been localized to the promoter regions of several genes including p53, PRADl (parathyroid adenomatosis 1; cyclin Dl), inducible nitric oxide synthase (iNOS), ornithine decarboxylase, cytokine receptors, E-cadherin, intercellular adhesion molecule-1 (ICAM-l), and IRF-2.8”3Many of these genes are involved in the regulation of cell proliferation and metabolism. Thus, IRF-l and -2 mayplayan important role in the pathophysiology of hematologic diseases such as leukemias and lymphomas. IRF-2 may function as an oncogene, whereas anti-oncogenic activity has been proposed for IRF-l.I4 Overexpression of IRF-2 results in transformation of NIH3T3 cells with enhanced tumorigenicity, which can be reversed by concomitant expression of the IRF-1 gene.” Somatic mutations leading to the loss of one or both IRF-1 alleles were implicated in the development of human hematopoietic neoplasms involving 5q3 1.1. l 6 There is a profound reduction of CD8+ T cells in the IRF- 1 knockout mice” and macrophages from IRF-1 knockout mice show amarkeddeficiency in nitric oxide (NO) production.’’ NO may be involved in the mechanism of hematopoietic suppression by IFN-y and TNF-a.” IRF-2-deficient mice show signs of hematopoietic suppression characterized by decreased colony formation by total bone marrow (BM) cells stimulated with IL-3, IL-7, and colony stimulating factor-l (CSF-1).I5 Involvement of IRF-1 and-2 in theIFN-y and TNF-a signal transduction implies that in conditions associated with abemant or elevated expression of these cytokines, IRF-l and -2 could mediate some pathophysiologically important effects of IFNy and TNF-a.Both IFN-y and TNF-a exert inhibitory activity Blood, Vol 86,No 9 (November I), 1995:pp 3373-3380

From the Hematology Branch, National Heart, Lung and Blood Institute, Bethesda, MD. Submitted February 6, 1995; accepted June 28, 1995. Address reprint requests to Jaroslaw P. Maciejewski, MD, Hematology Branch, National Heart, Lung and Blood Institute, Bldg 10, Room 7C106, Bethesda, MD 20892-1652 The publication costs ofthis article were defrayed in part by page charge payment. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. section 1734 solely to indicate this fact. This is a US govemment work. There are no resrrictions on its use. 0006-4971/95/8609-00$0.00/0 3373

From bloodjournal.hematologylibrary.org by guest on July 13, 2011. For personal use only. 3374

SAT0 ET AL

inducible genes.4' IFN-y signaling pathways differ from those described for TNF-a: they include JAK tyrosine kinases:' STAT proteins, and the IRF family of protein^.^' However, IFN-y and TNF-cu signal transduction can show convergent effects resulting in activation of common signaling and metabolic pathways as, for example, the generation of reactive oxygen intermediate^^^.^' and N0.46.47 All these data suggest that both IRF-l and -2 could play roles in the regulation of normal hematopoiesis, the pathophysiology of BM failure syndromes, and possibly in the pathogenesis of clonal evolution of hematopoietic malignancies. Our study was designed to investigate whether IFN-yand TNF-a-mediated hematopoietic inhibition was associated with induction of I R F - 1 or -2 and the role of these transcription factors in cytokine-mediated inhibition of hematopoiesis in vitro. MATERIALS ANDMETHODS Bone marrow cell preparation. BM was obtained from healthy volunteers by aspiration from the posterior iliac crest into syringes containing media supplemented 1: 10 with heparin (O'Neill and Feldman, St Louis, MO). Informed consent was obtained according to a protocol approved by the Institutional Review Board of the National Heart, Lung and Blood Institute. Mononuclear BM cells were isolated by density gradient centrifugation using lymphocyte separation medium (Organon, Durham, NC). After washing in Hanks' balanced salt solution (HBSS; Live Technologies, Gaithersburg, MD), cells were resuspended in Iscove's modified Dulbecco's medium (IMDM; Live Technologies) supplemented with 20% fetal calf serum (FCS; Live Technologies). Isolation of CD34' cells. CD34+ cells were separated using an affinity column (Cellpro, Bathell, WA) and microfluorometry. Briefly, nonadherent BM cells were incubated at room temperature with murine antihuman CD34 IgM, washed in phosphate-buffered saline (PBS; Live Technologies), followed by incubation with streptavidin-conjugated goat F(ab')z antimouse IgM. After washing with PBS supplemented with 1% human albumin, cells were applied to an affinity column containing biotin-coated beads. and the CD34+ cell fraction was eluted with PBS. An aliquot of eluted cells was stained with phycoerythrin (PE)-conjugated anti-CD34 HPCA-2 monoclonal antibody (MoAb; Becton Dickinson, Mountain View, CA) to assess the purity of this fraction: usually, 70% to 90% were CD34'. For higher purity preparations, cells were further fractionated: column-purified cells were stained with fluorescein isothiocyanate-labeled (FITC) anti-CD34 MoAb (Becton Dickinson), washed with PBS, and sorted by microfluorometry (Epics V; Coulter, Hiahleah, FL). The purity of cells obtained by combining affinity chromatography and flow cytometry was 97% to 99%. Hematopoietic cell culture. For short-term suspension cultures, BM cells were cultured in 24 flat-bottom well plates at a density of 1 X lo6 cells/mL in media consisting of IMDM, 10% FCS, 20 ng/ mL TNF-m, 2,000 U/mL IFN-y (both from Boehringer, Indianapolis, IN), and 1 pg/mL human anti-Fas IgM (clone CH-11; Amac, Westbrook, ME). Several IRF-1 antisense oligodeoxynucleotides (ODNs) were designed4* and used in tissue-culture experiments. IRF-1 antisense ODN 1 (ASl) had sequence 5"CGAGTGATGGGCATGTTGGC3' for targeting of the translation initiation site in IRF-1 mRNA , IRFI ODN 2 (AS2) 5'-GTGGCGAGCTCTGCCAGGGC-3' targeted the transcription initiation site in the IRF-l gene, and the missense (MS) All control ODN was 5'-GCATGCCGTACCGCAGGC'IT-3'. ODNs were sythetized by the solid-phase technique and purified using high-pressure liquid chromatography (Operon Technologies

Inc, Alameda, CA). Computer searches did not show any significant sequence similarity between ODNs and anyof the sequences in GenBank. We also used sense ODNs to target the DNA binding site in IRF1 protein.' Their sequences were as follows: IRE-l, 5"AAACTGAAAGC-3'; IRE-2, 5'-AAACCGAAAGC-3'; IRE-3, S'-AAAGCGAAAGC-3'; IRE-4, 5'-AAAGTGAAAGT-3'; and IRE-missense, S'WAGATATAGAA-3'.CD34+ cells were treated for 16 hours with I O pmoliL of each ODN, and then fresh ODN was added to adjust the final concentration to I O pmoVL. Colony formation by human hematopoietic progenitors was measured in standard methylcellulose cultures: CD34+ cells resuspended in IMDM supplemented with FCS and mixed with methylcellulose (Stem Cell Technologies, Vancouver, Canada) containing 50 ng/mL IL-3 (Genzyme, Boston, MA), 20 ng/mL granulocyte-macrophage colony-stimulating factor (GM-CSF Boehringer), SO ng/mL stem cell factor (SCF; Amgen, Thousand Oaks, CA), and 2 U/mL erythropoietin (EPO; Amgen). CD34' cells were plated at a density of 1 X IO3 cells/0.5 mLI methylcellulose. TNF-a and IFN-y were added to the culture at appropriate concentrations. All cultures were performed in duplicate. All experimental procedures were performed in endotoxin-free plasticware and, according to the manufacturer, the levels of endotoxin contamination in the cytokine preparations were 3 endotoxin units/ mgby the limulus assay. Northern blot analysis and reversetranscription-polymerase chain reaction (RT-PCR). Total RNA was extracted from constant numbers of mononuclear marrow cells and CD34' highly purified cells using RNAsol (CinnaBiotecx, Friendswood, TX). After contaminating DNAwas digested using RNAse-free DNAseI (Boehinger), RNA was reextracted with phenol and chloroform, precipitated with ethanol, and diluted in RNAse-free water. After RT using oligo d(T),, primer, a specific 805-bp fragment of I R l - I cDNA was amplified using the upstream primer, 5"AGACCAGAGCAGGAACAAGG-3', and the downstream primer, S'GGTCACACTTGGCTGTTGAG-3'. For human @-actin, specific as an primer pairs, S'-CA'ITGTGATGGACTCCGGAGACGG-3' upstream primer and 5'-CATCTGCTGCTCGAAGTCTAGAGC-3' as downstream primer, were used. PCR products were electrophoresed on 1.2% agarose gels, transferred onto Nytran filters (Schleicher & Schuell, Keene, NH), cross-linked using UV-light, and the blots were hybridized with full-length IRF-I cDNA, BamHU Hind111 fragment from pHIRF31 plasmid (kindly gifted from Prof T. Taniguchi, Osaka University, Osaka, Japan),"' or internal ODN for human @-actin. IRF-l cDNA was labeled with [ ~ Y - ~ * P ] ~ C T P (3,000 pCi; Amersham, Arlington Heights, IL) using a random priming reaction (Stratagene, La Jolla, CA). Actin internal ODN was '*Plabeled using [y-'*P]dATP (3,000 pCi; Amersham) and T4-kinase reaction (Stratagene). After washing under stringent conditions (0.1 x SSC. 0.1 % sodium dodecyl sulfate [SDS]), membranes were exposed to x-ray film and developed. For Northern blotting, 20 pg of total RNA was resuspended in diethyl pyrocarbonate (DEPC)-treated water and a denaturing solution of formamide, formaldehyde, and 3-(N-Morpholino) propanesulfonic acid (MOPS) buffer was added. RNA was denatured by heating at65"C, electrophoresed under denaturing conditions, stained with ethidium bromide, and transferred onto Nytran filters. After UV-cross-linking, the membranes were prehybridized in Hybrisol I (Oncor, Gaithersburg, MA) for 2 hours. To avoid crosshybridization, the entire IRF-l cDNA and the Xba I fragment from pHIRF4S51 plasmid (also donated by Prof T. Tanig~chi),"~ which is the 3'-untranslated region of IRF-2 cDNA, were '2P-labeled by random priming reaction and used for hybridization. The membranes were then hybridized with a total of S X lo7 cpm of radiolabeled probes, washed under stringent conditions (0.1 X SSC, 0. I % SDS), and exposed to x-ray films.

From bloodjournal.hematologylibrary.org by guest on July 13, 2011. For personal use only. IRF-1EXPRESSION

3375

IN HEMATOPOIETIC CELLS

A

B n a

Fig 1. Effech Of IFN-y, TNFa,and anti-Fas MoAb on theexpressionofIRF-1 and -2 mRNA in total BM cells. Northern blot analysis of IRF-1(AI and IRF-2 (B1 mRNA expression in cells stimulatedwith cytokines. The concentration of cytokine used was 2,000 U/mL for IFN-y, 20 ng/mL for TNF-a, and 1 pg/mL for antiFas MoAb, respectively. Upper panels, autoradiographs; lower panels, ethidium-bromide staining of agarose gels showing intensity of ribosomal bands and integrity of RNA. ( + l Control, total RNA from U-937 cells stimulated by IFN-y.

2.4 kb

28 S

18 S

Immunoblorring of IRF-l protein. For immunoblotting, after culture for 2 days with or without ODNs as described above, cells were washed in PBS and lysed by radioimmunoprecipitation buffer (20 mmol/L TRIS-HCI [pH 8.0].3% Nonident P-40, 150 mmol/L NaCI, 50 mmol/L NaF, 1 mmol/L 4-[2-aminoethyl]-benzenesulfonylfluoride hydrochlorine [AEBSF; ICN Biomedicals, Aurora, OH], leupeptin hemisulfate [ICN] at 10 pg/mL, aprotinin [ICN] at IO pgImL, pepstatin A [ICN] at 1 pglmL) for 30 minutes on ice. The wholecell lysates were obtained after centrifugation and fractionated by 10% SDS-polyacrylamide gel electrophoresis (SDS-PAGE). Gels were equilibrated in transfer buffer (125 mmol/L TRIS-base, 960 mmol/L glycine, 20%methanol) and separated proteins were electrophoretically transferred to PVDF membrane filter (Immobilon-P Millipore, Bedford, MA). Membranes were blocked in TBST-milk ( I O mmol/L TRIS-HCI [pH 8.01, 150 mmol/L NaCI, 0.5% Tween20, 1% nonfat dry milk,1 % bovine serum albumin [Cohn fraction V; Miles, Kankakee, IL]) and treated with rabbit anti-IRF-l polyclonal antibody (clone C-20, 1 pg/mL in TBST-milk; Santa Cruz Biotechnology, Santa Cruz, CA) at room temperature. After washing three times by TBST-milk, membranes were incubated with alkaline phosphatase-labeled goat-antirabbit IgG (H+L) (1 :1 ,OOO dilution in TBST-milk; Pierce, Rockford, IL) at room temperature and subsequently with mouse antialkaline phosphatase MoAb (Dako, Carpinteria, CA). Specific bands were detected using nitro blue tetrazolium (NBT)/5-bromo-4-chloro-3-indolylphosphate (BCIP) substrate (Pierce). According to manufacturer's data, this anti-IRF-l antibody does not cross-react with either IRF-2 or ISGF-37.

RESULTS

Induction of expression of IRF-I and IRF-2 mRNA in human BM. We tested if stimulation of BM cells with IFNy and TNF-a induced IRF-l and -2 gene expression. The kinetics of IRF-I and -2 mRNA expression were determined in the histiocytic cell line U-937. Although IRF-2 mRNA expression was constitutive in this cell line, IRF-I mRNA expression could be induced with both cytokines: after addi-

tion of either cytokine, the first effects on IRF-I mRNA expression were seen after 2 hours, achieved a plateau after 6 hours, and decreased after 18 hours (data not shown). In further experiments performed with total BM, we measured IRF-1mRNA expression after 18hours of culture in the presence of IFN-y and TNF-a. Both IFN-y and TNF-CYenhanced constitutive levels of IRF-l mRNA expression in total BM cells (Fig 1). Expression levels of IRF-2 mRNA were not affected by IFN-y and TNF-a. On the other hand, anti-Fas MoAb,which mimics Fas-ligand andcaninduce suppression of colony formation when combined with IFNy and TNF-CY,~~ could not affect on mRNA expression levels of either IRF-l or IRF-2. IFN-y induces expression of IRF-I mRNA in highly purified human CD34' cells. Since Northernblot analysis showed constitutive expression of IRF-2 mRNA and inducible IRF-l transcription in BM cells by IFN-y and TNF-a, we tested whether IRF-I mRNA was present in highly purified CD34'cells. The CD34' population includes committed and immature progenitor cells, and stem ~ells."~~' Using RTPCR, we demonstrated verylow levels of IRF-I mRNA expression in fresh or unstimulated CD34' cells (Fig 2B). However, strong signals were obtained after stimulation of CD34' cells for 18 hours with IFN-y or TNF-a, suggesting that IRF-I mRNA expression was induced in CD34' cells. Similar results were obtained when IRF-1 protein expression was studied in CD34' cells using immunoblotting (data not shown; see also Fig 3). To exclude the possibility that IRF1 mRNA expression was related to the differentiation of CD34' population to more mature cells during the culture period and notinduced in immature CD34' cells, BM CD34' cells were enriched, stimulated with IFN-y or TNF-a, and subsequently resorted by the presence of CD34 antigen (Fig 2A). These results showed IRF-l mRNA expression to be

From bloodjournal.hematologylibrary.org by guest on July 13, 2011. For personal use only. SAT0 ET AL

3376

induced in CD34' cells, and this effect is neither related to differentiation nor to the presence of accessory cells in the cultures. Reversal of IFN-y mediated inhibition of colonyformation by human CD34' cells using IRF-I antisense oligodeoxynucleotide. Expression of the IRF-l gene in total BM cells was induced by IFN-y and TNF-a, and IRF-I mRNA was increased in CD34' cells upon stimulation with these cytokines. Therefore, we tested whether the inhibitory effects of IFN-y and TNF-a on hematopoietic colony formation were modulated by intracellular IRF-l. A series of IRF- 1 antisense oligodeoxynucleotides was synthesized and used to inhibit the expression of IRF-l and the effects of IFN-y or TNF-a on colony formation by highly purified CD34' cells. After treatment of highly purified CD34' cells with ODNs, we measured IRF-l protein levels by immunoblotting (Fig 3B). AS1 ODN markedly decreased IRF-l protein levels whereas MS ODN did not affect steady-state levels. To test whether the decrease in the IRF-l protein levels was caused by antisense ODN-mediated inhibition of translation or mRNA transcription, we measured IRF-l mRNA expression. Using RT-PCR,we showed AS 1 ODN decreased IRF-l mRNA expression levels in highlypurifiedCD34' cells cultured

A

% 1 Log PE (CD34) Fluorescence

116.3 97.466.3 55.4 -&K 36.5 -

31-

!

-3K

Fig 3. Effect ofIRF-1antisenseoligodeoxynucleotide (ODN) on IRF-1 protein levels in whole-celllysates. (A) Histogramsshowing CD34' cell purity after resorting following tissue culture (97%; log red fluorescence W cell count). (B) Protein analysis for IRF-1 byimmunoblot (left) and Coomasie-blue staining (right). M.W.M., molecular weight markers; MS,missense ODN; AS1, IRF-1 antisense ODN; untreated control, cells cultured in the absence of ODNs.

LOG R

FIT

-R

"-

+ + + - + - + -

Fig 2. IRF-1 mRNA analysis of CD34' cells treated with IFN-y and

TNF-a. (A) CD34' cell purity. Shown are scattergrams of cells after 18 hoursof tissue culture (top;forward scatter wlogred fluorescence) and histograms after resorting (bottom; log red fluorescence W cell count). (B) Autoradiographs of RT-PCR performed with primers specific for IRF-1 (top) and P-actin (bottom). (+)Control, total RNA from U-937 and KG-la cells stimulated by IFN-y.

with or without IFN-y (Fig 4). MS ODN decreased mRNA expression when cultured without IFN-y but not with IFNy . suggesting a nonspecific effect. IRF-l AS1ODN was capable of partially abrogating IFN-y-suppressed colony formation (Table 1, Fig 5). This effect was specific anddosedependent by comparison with MS ODN of an equal length (Fig 5). When we tested another sequence of IRF-l mRNA (AS2 ODN), designed to target the transcription-initiation site of the IRF-l gene, no effects on colony formation were seen (data not shown). We also attempted to block the effects of IRF-l by competitive inhibition of the DNA binding site of the protein using ODNs homologous to the IRE site, but this strategy was not effective, possibly due to competitive binding of these ODNs to other IRF family proteins (data not shown). Using AS I ODN, which wascapable of abrogating the effects of IFN-y on colony formation, we were not able to restore TNF-a-suppressed colony formation by CD34' cells (Table l), suggesting thatincreased IRF-l mRNA after stimulation with TNF-a maybe a secondary effect or that another pathway was involvedin the inhibitory action of this cytokine. DISCUSSION

Studies on IRF- 1 and -2 expression in celllines transfected with IRF-I or -2 genes or induced to express these proteins

From bloodjournal.hematologylibrary.org by guest on July 13, 2011. For personal use only. IRF-1EXPRESSION IN HEMATOPOIETICCELLS

3377

A

B

Erythroid

Myeloid

70 -

m

++-+-+-+-+-+-

--In

-

50

-

80-

40

-

60-

30

-

20

-

Q V

“c+ 100 -

1 0 0 bpmbp-

l_

60

120 -

i

~9 1 mM AS1 W 10 mM AS1

I T r

m O 0

n

0Control

ta i

5

5c -g 40 0

600 bp-

0

20

-

10 -

100 bp-

0Fig 4. Effects of IRF-1 antisense oligodeoxynucleotide onthe IRF1 mRNA expression by highly purifiedBM CD34’ cells. Ethidium bromide staining of RT-PCR products performed with primers specific for IRF-l (top) andp a c t i n (bottom). The concentration of IFN-yused was 2,000 U/mL. ( + l Control, total RNA from KG-lacells stimulated by IFN-7.

by cytokines have suggested that both transcription factors may play an important role in the regulation of hematopoietic proliferation and differentiation. However, little experimental data existed on the JRFs in primary BM cells. We have shown that both cytokines, IFN--y and TNF-a, induced the expression of IRF-I in total BM cells. In

Table 1. Effects of Antisense Oligodeoxynucleotide on Colony Formation by Highly PurifiedCD34’ Cells No. of Colonies 1%) ~~

CFU-GM

Control MS 100 ODN (10 pmol/L) +IFN-y +TNF-a

AS1 ODN (10 pmol/L) +IFN-y +TNF-a

113 2 3

48

BFU-E

118 2 8’

100 47 2 11’ 22 t- 15’

-t 19* 30 t- 15’

113 t- 8 105 5 20t 23 t- 14*

107 t- 7 77 5 13t 23 t- 9*

Summary of results from all fiveexperiments performed. The concentration of cytokines used was 2,000 U/mL for IFN-y and 20 ng/ mL for TNF-a.Values represent mean percentage t- SD of colony formation. To exclude unspecific effects of MS oligodeoxynucleotide (ODN), colony formationin the presence of thisODN was set as 100%. Statistical analysis (paired t-test): P < .01 compared with MS ODN: t P < .05 compared to IFN-y with MS ODN. Abbreviations: CFU-GM, colony-forming unit-granulocyte-macrophage: BFU-E, burst-forming unit-erythroid.

0

500

1000

0-

0

500

1000

IFN-y U/mL Fig 5. Effects of IRF-1 antisense oligodeoxynucleotide IODNI on colony formation by highly purified BM CD34’ cells. Dose-dependent effects of IRF-1 antisense ODN on erythroid (AIand myeloid (B) colony formation were observed. Bars represent results fromone representative experiment (two dishes per sample).

agreement with our findings, IRF-I mRNA was found to be expressed in mouse BM c e k J Constitutively high expression of IRF-2 appeared to be largely independentof cytokine stimulation. Induction of IRF-I was not restricted to mature hematopoietic cells, as IFN--y and TNF-a increased the expression of IRF-I in highly purified CD34’ cell populations containing hematopoietic progenitor andstem cells. This finding isconsistent with other data indicating that theeffects of IFN--y and TNF-a on BM progenitor and stem cells are direct and do not require accessory Because specific antisense ODN designed to target IRF-I transcription abrogated hematopoietic inhibition by JFN-7, negative effects of this cytokine may be directly mediated through the induction of TRF-I. The positive effect of IRF-I ODN on colony formationreduced byIFN--y correlated with the decrease in IRF-I protein and mRNAexpression. Lack of effects of IRF1 antisense ODN on TNF-a-mediated inhibition of colony formation suggest that increased IRF-I transcription seen on TNF-a stimulation is secondary and that the inhibitory activity of TNF-a is mediated along a different signaling pathway; for TNF-a, signal transduction may involve NF-KB.”.‘” The IRF family of transcription factors includes also ICSBP, which can be induced byJFN-y.’? ICSBP was reported to be expressed in macrophages andlymphocytes (non-B non-T and B cells) but not in myeloid cells (K-S62 and HL-60).” Similar to IRF-2, ICSBP could repress IRFI -mediated induction of transcription.”” In our study, we showed that IRF-I, butnot IRF-2, played a major role in


mediating IFN-y -induced hematopoietic suppression. Because inhibition of colony formation wasnot completely abrogated by IRF- 1 antisense ODN, it ispossible that ICSBP or other factors may also be involved in the mechanism of IFN-y-mediated hematopoietic suppression. IRF-1 gene expression can be induced in CD34+ cells by IFN-y and TNF-a. To ensure that the expression of IRF-l in CD34+ cell preparations was not caused by the presence of more differentiated cells, weused highly purified cell populations, and CD34+ cells were also resorted before RNA extraction. Although unlikely, we cannot rigorously exclude the possibility that IFN-y and TNF-a induced IRF-I selectively only ina subset of cells within the CD34’ cell population. Because more immature progenitor cells and stem cells are characterized by function rather than phenotype, selective expression analysis in these cells is not feasible. Our finding of increased IRF- 1 expression in CD34+ cells correlated with the effects of antisense ODNs on colony cultures ofCD34’ cells treatedby IFN-y. IRF-1 antisense ODN was reported to partially reverse IL-6- and LIF-mediated growth inhibition of murine M1 leukemic cells: Antisense ODNs have been used to decrease the expression of viral and cellular genes and may be potentially useful as future therapeutic agents.53However, no general consensus exists about the mechanism of antisense ODN action, and some postulated modes of action include inhibition of translation, RNase H activation, and block of tran~cription.’~ The action of antisense ODN generally is highly concentration- and sequence-dependent, but selection of targeted sequences and length of ODN are still mainly empirical. In our study, IRF-l antisense ODN (AS1) targeted the translation initiation site and decreased both IRF-1 protein levels and IRF-l mRNA expression. Sense-ODNs designed to target the DNA-binding site of the IRF-1 protein were not effective. The effects of antisense ODN (ASI) on IRF-l expression correlated with the increased colony formation by CD34+ cells treated with IFN-y. Increased concentrations of IRF-1 antisense ODN was associated with nonspecific inhibition, and nonspecific stimulation was observed with MS ODN, which mightbe due to decreased IRF-1 mRNA level as shown in Fig 4. IRF-1 antisense ODN (ASI) might bind to IRF-1 mRNA and disturb IRF-1 translation. IRF- 1 mRNA/ DNA hybrids may be susceptible to RNase H action decreasing IRF-1 mRNA stability. The restoration of IFW-y-suppressed colony formation by IRF-1 antisense ODN (AS1) was partial. Incomplete activity could be related to the incomplete abrogation of IRF-1 protein level as shown by immunoblotting, or could be caused by the specific pharmacologic features of the synthetic DNA (affinity, half-life, uptake, etc). Involvement of additional IRF-1 -independent signals or other IFN-y-inducible transcription factors in hematopoietic inhibition by IFN-y might be additionally responsible for partial effects of IRF-1 antisense ODN. However, the effects of IRF-1 antisense ODN (ASl) in our study were significant because they were cytokine- and sequence-specific, dose-dependent, and resulted in a positive biologic effect as detected by increased colony formation. IRF-1 activates several effector genes that may be respon-

SAT0 ET AL

sible for the functional consequences of IFN-y at the cellular level,” such as a blockin cell cycling and apoptosis, as recently shown in early hematopoietic cells.’5 Effects of IFNy could also be attributed to IRF- I -mediated induction of metabolic machinery that negatively affects cellular physiology. Recently, ornithine decarboxylase, a key enzyme in polyamine metabolism, and iNOS, required for NO synthesis, were shown to be activated by IFN-y through IRF1,lO.ll.l3NO, a gaseous radical, functions both as a messenger signal and cytotoxic molecule involved in the inflammatory reaction. NO may cause a spectrum of effects ranging from DNA damage to apoptosis, and some recent studies suggest that NO may also play a role in hematopoietic suppression because competitive inhibition of iNOS in vitro was associated with partially reversed antiproliferative effects of I F N y and TNF-LY on BM progenitor cells.” Strong experimental evidence exists that IFN-y and TNFLY can mediate hematopoietic inhibition in BM failure syndromes, including AA. IRF-1 expression might be a potential marker for the action of these cytokines in aplastic marrow. REFERENCES I . Harada H, Fujita T, Miyamoto M, Kimura Y, Maruyama M, Furia A, Miyata T, Taniguchi T: Structurally similar but functionally distinct factors, IRF-I and IRF-2, bind to the same regulatory elements of IFN and IFN-inducible genes. Cell 58:729, 1989 2. Nelson N, Marks MS, Driggers PH, Ozato K: Interferon consensus sequence-binding protein, a member of the interferon regulatory factor family, suppresses interferon-induced gene transcription. Mol Cell Biol 13:588, 1993 3. Reid LE, Brasnett AH, Gilbert CS, Porter ACG, Gewert DR, Stark GR, Kerr IM: A single DNA response element can confer inducibility by both a- and y-interferons. Proc Natl Acad Sci USA 86:840, 1989 4. Abdollahi A, Lord KA, Hoffman-Liebermann B, Liebermann DA: Interferon regulatory factor 1 is a myeloid differentiation primary response gene induced by interleukin 6 and leukemia inhibitory factor: Role in growth inhibition. Cell Growth Differ 2:401, 1991 5. Fujita T, Reis LF, Watanabe Y, Kimura T, Taniguchi T, Vilcek J: Inductionofthe transcription factor IRF-1and interferon-beta mRNAs by cytokines and activators of second-messenger pathways. Proc Natl Acad Sci USA 86:9936, 1989 6. Watanabe N, Sakakibara J, Hovanessian A, Taniguchi T, Fujita T: Activation of IFN-p promotor element by IRF-I requires a posttranslational event in addition to IRF-I synthesis. Nucleic Acids Res 16:4421, 1991 7. Stevens AM, Yu-Lee LY: Multiple prolactin-responsive elements mediate G1 and S phase expression of the interferon regulatory factor-l gene. Mol Endcrinol 8:345, 1994 8. Tanaka T, Kawakami T, Taniguchi T: Recognition DNA sequences of interferon regulatory factor 1 (IRF-I) and IRF-2, regulators of cell growth and the interferon system. Mol Cell Biol 13:453I , 1993 9. Look DC, Pelletier MR, Holtzman MJ: Selective interaction of a subset of interferon-gamma response element-binding proteins with the intercellular adhesion molecule-l (ICAM-l) gene promotor controls the pattern of expression on epithelial cells. J Biol Chem 2693952, 1994 IO. Martin E, Nathan C, Xie Q: Role of interferon factor 1 in induction of nitric oxide synthase. J Exp Med 180977, 1994 I I. Kamijo R, Harada H, Matsuyama T, Bosland M, Gereciano J, Shapiro D, Le J, Koh SI, Kimura T, Green SJ, Mak T W , Taniguchi

From bloodjournal.hematologylibrary.org by guest on July 13, 2011. For personal use only. IRF-1 EXPRESSION IN HEMATOPOIETIC CELLS

T, Vilcek T: Requirement for transcription factor IRF-1 in NO synthase induction in macrophages. Science 263:1612, 1994 12. Harada H, Takahashi E, Itoh S, Harada K, Hori T, Taniguchi T: Structure and regulation of the human interferon regulatory factor 1 (IRF-1) and IRF-2 genes: Implications for gene network in the interferon system. Mol Cell Biol 14:1500, 1994 13. Benninghoff B, Lehmann V, Eck HP, Droge W: Production of citrulline and ornithine by interferon-gamma treated macrophages. Int Immunol 3:413, 1993 14. Harada H, Kitagawa M, Tanaka N, Yamamoto H, Harada K, Ishihara M, Taniguchi T: Anti-oncogenic and oncogenic potentials of interferon regulatory factors-l and -2. Science 259:971, 1993 15. Matsuyama T, Kimura T, Kitagawa M, Pfeffer K, Kawakami T, Watanabe N, Kundig TM, Amakawa R, Wakeham A, Potter J, Furlonger CL, Narendran A, Suzuki H, Ohashi PS, Paige CJ, Taniguchi T, Mak T W : Targeted disruption of IRF-1 or IRF-2 results in abnormal Type I I F N gene induction and aberrant lymphocyte development. Cell 75:83, 1993 16. Willman CL, Sever CE, Pallavicini MG, Harada H, Tanaka N, Slovak ML, Yamamoto K, Harada K, List AF, Taniguchi T: Deletion of IRF-l, mapping to chromosome 5q3 1.1, inhuman leukemia and preleukemic myelodysplasia. Science 259:968, 1993 17. Maciejewski JP, Selleri C, Sat0 T, Cho HJ, Keefer LK, Nathan CF. Young NS: Nitric oxide suppression of human hematopoiesis in vitro: Involvement in interferon-y and tumor necrosis factora mediated inhibition. J Clin Invest 96:1085, 1995 18. Raefski EL, Platanias LC, Zoumbos NC, Young NS: Studies of interferon as a regulator of hematopoietic cell proliferation. J Immunol 1352507, 1985 19. Roodman GD, Bird A, Hutzler D, Montgomery W: Tumor necrosis factor-alpha and hematopoietic progenitors: Effects of tumor necrosis factor on the growth of erythroid progenitors CFU-E and BFU-E and the hematopoietic cell line K562, HL60 and HEL cells. Exp Hematol 15:928, 1987 20. Zoumbos NC, Djeu JY, Young NS: Interferon is the suppressor of hematopoiesis generated by stimulated lymphocytes in vitro. J Immunol 133:769, 1984 21. Broxmeyer HE, Williams DE, Lu L: The suppressive influences of human tumor necrosis factors on bone marrow progenitor cells from normal donors and from patients with leukemia: Synergism of tumor necrosis factor and interferon-y. J Immunol 136:4487, 1986 22. Johnson RA, Waddelow TA, Car0 J, Oliff A, Roodman CD: Chronic exposure to tumor necrosis factor in vivo preferentially inhibits erythropoiesis in nude mice. Blood 74:130, 1989 23. van der Poll T, van Deventer SJH, Hack CE, Wolbink CE, Aarden LA, Buller HR, ten Cate JW: Effects on leucocytes after injection of tumor necrosis factor into healthy humans.Blood 79:693, 1992 24. Saks S, Rosemblum M: Recombinant human TNF-alpha: Preclinical studies and results from early clinical trials. Immunol Ser 56:567, 1992 25. Quesada JR, Talpaz M, Rios A, Kurzrock R, Gutterman JU: Clinical toxicity of interferons in cancer patients. J Clin Onc 4234, 1986 26. Zoumbos NC, Gascon P, Djeu J, Young NS: Interferon is a mediator of hematopoietic suppression in aplastic anemia in vitro and possibly in vivo. Roc Natl Acad Sci USA 82:188, 1985 27. Nakao S, Yamaguchi M, Shiobara S: Interferon-y gene expression in unstimulated bone marrow mononuclear cells predicts a response to cyclosporine therapy in aplastic anemia. Blood 79:2532, 1992 28. Nisticb A, Young NS: y-Interferon gene expression in the bone marrow of patients with acquired aplastic anemia. Ann Intern Med 120:463, 1994

3319

29. Katevas P, Maciejewski JP, Sorrentino B,YoungNS: Increased expression of TNF-beta in patients with aplastic anemia. Blood 82:345a, 1993 (abstr, suppl 1) 30. Rosselli F, Sanceau J, Gluckman E, Wietzerbin J, Moustacchi E: Abnormal lymphokine production: A novel feature of the genetic disease FancoN anemia. In vitro and in vivo spontaneous overproduction of tumor necrosis factor alpha. Blood 83:1216, 1994 31. Seipelt G, Ganser A, Duranceyk H, Maurer A, Ottman OG, Holzer D: Induction of TNF-alpha in patients with myelodysplastic syndromes undergoing treatment with interleukin-3. Br J Haematol 84:749, 1993 32.Verhoef GE, Schouwer P, Ceuppens JL, Van Damme JO, Goossens W, Boogaerts MA: Measurement of serum cytokine levels in patients with myelodysplastic syndromes. Leukemia 6:1268, 1992 33. Akashi K, Hayashi S, Gondo H, Mizuno SI, Tamura K, Yamasaki K, Shibuya T, Uike N, Okamura T, Miyamoto T, Niho Y: Involvement of interferon-gamma, and macrophage colony-stimulating factor in pathogenesis of hemophagocytic lymphohistiocytosis in adults. Br J Haematol 87:243, 1994 34. Snoek HV, Van Bockstaele DR. Nys G, Lenjou M, Lardon F, Haenem L, Rodrigus I, Peetermans ME, Berneman ZN: Interferon-? selectively inhibit primitive CD34’+CD38and not mature CD34+CD38+ human hematopoietic progenitor cells. J Exp Med 180:1177, 1994 35. Selleri C, Sat0 T, Anderson A, Young NS, Maciejewski J P Interferon-? and tumor necrosis factor-a suppress both early and late stages of hematopoiesis and induce programmed cell death. J Cell Physiol (in press) 36. Maciejewski JP, Selleri C, Anderson A, Young NS: Fas antigen expression on CD34+ human marrow cells is induced by interferon-y and potentiates cytokine-mediated hematopoietic suppression in vitro. Blood 85:3183, 1995 37. Schutze S, Potthoff K, Machleidt T, Berkovic D, Wiegmann K, Kronke M: TNF activates NF-kappa B by phosphatidylcholinespecific phospholipase C-induced acidic sphingomyelin breakdown. Cell 71:765, 1992 38. Brenner DA, O’Hara M, Angel P, Chojkier M, Karil M: Prolonged activation of jun and collagenase genes by tumor necrosis factor-a. Nature 337:661, 1989 39. Muegge K, Vila M, Gusella GL, Musso T, Herrlich P, Stein B, Durum S K Interleukin-l induction of the c-jun promoter. Proc Natl Acad Sci USA 90:7054, 1993 40. Lenardo MJ, Baltimore D: NF-kappa B: A pleiotropic mediator of inducible and tissue-specific gene control. Cell 58:227, 1989 41. Sluss HK, Barrett T, Derijard B, Davis RJ: Signal transduction by tumor necrosis factor mediated by JNK protein kinases. Mol Cell Biol 14:8376, 1994 42. Igarashi K, Garrota G, Ozmen L, Ziemiecki A, Wilks AF, Harpur AG, Lamer AC, Finbloom DS: Interferon-? induces tyrosine phosphorylation of interferon-y receptor and regulated association of protein tyrosine kinases, Jak 1 and Jak 2, with its receptor. J Biol Chem 269:14333, 1994 43. Damell JE Jr, Kerr I M , Stark GR: Jak-STAT pathways and transcriptional activation in response to IFNs and other extracellular signaling proteins. Science 264:1415, 1994 44. Schulze-Osthoff K, Beyaert R, Vandevoorde V, Haegeman G, Fiers W: Deletion of the mitochondrial electron transport abrogates the cytotoxic and gene-inducible effects of TNF. EMBO J 12:3095, 1993 45. Drapier JC, Wietzbin J, Hibbs JB: Interferon-gamma and tumor necrosis factor induce the L-arginine dependent cytotoxic effector mechanisms in murine macrophages. Eur J Immunol 18:1587, 1988 46. Fast D, Lynch RC, Leu RW: Nitric oxide production by tumor targets in response to TNF Paradoxical correlation with susceptibil-


ity to TNF-mediated cytotoxicity without direct involvement in the cytotoxic mechanism. J Leukoc Biol 52:255, 1992 47. Colasanti M, Mollace V, Cundari E, Massoud R, Nistico G , Lauro GM: The generation of nitric oxide participates in gamma IFN-induced MHC class I1 antigen expression by cultured astrocytoma cells. Int J Immunopharmacol 15:763, 1993 48. Maruyama M, Fujita T, Taniguchi T: Sequence of a cDNA coding for human IRF-l. Nucleic Acids Res 17:3292, 1989 49. Itoh S, Harada H, Fujita T, Mimura T, Taniguchi T Sequence of a cDNA codingfor human IRF-2. Nucleic Acids Res 20:8372, 1989 50. Civin CJ, Strauss LC, Brovall C, Fackler MJ, Schwartz JF, Shaper JH: Antigenic analysis of hematopoiesis 111. A hematopoietic progenitor cell surface antigen defined by monoclonal antibody raised against KG-la cells. J Immunol 133:157, 1984

SAT0 ET AL

5 1. Civin CI, Banquerigo ML, Strauss LC, Loken MR: Antigenic analysis of hematopoiesis. VI. Characterization of my- 10-positive progenitor cells in normal human bone marrow. Exp Hematol 15:10, 1987 52. Weisz A, Mark P, Sharf R, Appella E, Driggers PH, Ozato K, Levi BZ: Human interferon consensus sequence binding protein is a negative regulator of enhancer elements common to interferoninducible genes. J Biol Chem 267:25589, 1992 53. Milligan JF, Jones RJ, Froehler BC, Matteucci MD: Development of antisense therapeutics. Implications for cancer gene therapy. Ann NY Acad Sci 716:228, 1994 54. Herschlag D: Implication of ribozyme kinetics for targeting the cleavage of specific RNA molecules in vitro: More isn’t always better. Proc Natl Acad Sci USA 88:6921, 1991