Inhibitory effect on natural killer activity of microphthalmia transcription factor encoded by the mutant mi allele of mice

doi:10.1182/blood.V97.7.20752001 97: 2075-2083   

 Akihiko Ito, Tatsuki R. Kataoka, Dae-Ki Kim, Yu-ichiro Koma, Young-Mi Lee and Yukihiko Kitamura encoded by the mutant mi allele of miceInhibitory effect on natural killer activity of microphthalmia transcription factor

http://bloodjournal.hematologylibrary.org/content/97/7/2075.full.htmlUpdated information and services can be found at:

(5020 articles)Immunobiology   � (3131 articles)Hematopoiesis and Stem Cells   �

Articles on similar topics can be found in the following Blood collections

http://bloodjournal.hematologylibrary.org/site/misc/rights.xhtml#repub_requestsInformation about reproducing this article in parts or in its entirety may be found online at:

http://bloodjournal.hematologylibrary.org/site/misc/rights.xhtml#reprintsInformation about ordering reprints may be found online at:

http://bloodjournal.hematologylibrary.org/site/subscriptions/index.xhtmlInformation about subscriptions and ASH membership may be found online at:

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

For personal use only. by guest on June 4, 2013. bloodjournal.hematologylibrary.orgFrom

IMMUNOBIOLOGY

Inhibitory effect on natural killer activity of microphthalmia transcription factorencoded by the mutantmi allele of miceAkihiko Ito, Tatsuki R. Kataoka, Dae-Ki Kim, Yu-ichiro Koma, Young-Mi Lee, and Yukihiko Kitamura

The mouse mi locus encodes a basic-helix-loop-helix-leucine zipper-type transcriptionfactor, microphthalmia transcription factor(MITF). Mice of mi/mi genotype express amutant form of MITF ( mi-MITF), whereasmice of tg/tg genotype have a transgene inthe 5* flanking region of the mi gene and donot express MITF.Although the mi/mi mouseis deficient in natural killer (NK) activity, itwas found that the tg/tg mouse was normalin this respect. To know the cause, spleencells of both genotypes were compared.Although the proportion of spleen cells ex-pressing an NK cell marker, NK1.1, was

comparable in both mice, the proportion oflarge granular lymphocytes decreased onlyin mi/mi mice. The difference between mi/miand tg/tg mice was reproducible in the cul-ture supplemented with interleukin-2. More-over, the perforin gene expression was re-duced in mi/mi –cultured spleen cells. Wild-type (1 ) MITF transactivated, but mi-MITFsuppressed, the perforin gene promoterthrough the NF-P motif, a strong cis -actingelement. However, neither 1-MITF nor mi-MITF bound the NF-P motif. Instead, 2nuclear factors that bound the NF-P motifwere retained in the cytoplasm of mi/mi –

cultured spleen cells. In addition, overex-pression of mi-MITF resulted in cytoplasmicretention of the 2 NF-Pmotif–binding factorsin cytotoxic T lymphocytes. The presence ofmi-MITF rather than the absence of 1-MITFappeared to lead to poor transactivation ofthe NF-P motif by intercepting NF-P motif–binding factors. This inhibitory effect of mi-MITF may cause the deficient cytotoxicity ofNK cells in mi/mi mice. (Blood. 2001;97:2075-2083)

© 2001 by The American Society of Hematology

Introduction

The mousemi locus encodes a transcription factor belonging to thebasic-helix-loop-helix-leucine zipper family (microphthalmia [mi]transcription factor [MITF]).1,2 The mutantmi allele produces anabnormal MITF, in which 1 out of 4 consecutive arginines isdeleted in the basic domain (hereafter,mi-MITF).1,3,4Themi-MITFis defective in DNA binding, nuclear localization, and transactiva-tion of some target genes.5-10 Homozygous mutant mice ofmi/migenotype were found by Hertwig11,12 among the offspring of anX-irradiated male mouse. On the other hand, mice oftg/tggenotypeare null mutants at themi locus, a condition that was produced bytransgene insertion into the 59flanking region of themi gene.1,13

Both tg/tg and mi/mi mice share a lot of abnormal phenotypicfeatures, such as microphthalmia, white coat color, and thedecreased number of mast cells.13-18 However, they were clearlydistinguishable from each other at least in one respect:mi/mi miceare osteopetrotic whereastg/tgmice are not.

Another abnormality ofmi/mi mice is a deficiency in naturalkiller (NK) activity.17,19A decrease in the number of large granularlymphocytes (LGLs) has been reported inmi/mi mice.17 Since NKcells are morphologically identified as LGLs,20,21the decreased NKactivity is ascribed to the decreased number of LGLs.17 In thepresent study, we estimated the number of NK cells by anothermethod. When we examined the expression of NK1.1 surfaceantigen of spleen cells, the proportion of cells expressing theNK1.1 antigen did not decrease inmi/mi mice. Next, we examinedtg/tgmice according to the number and activity of their NK cells. Incontrast tomi/mi mice, tg/tg mice have a normal number of LGLsand NK1.1–expressing cells and a normal NK activity. Hence, we

attempted to investigate the mechanism that may explain thedifference betweenmi/miandtg/tgmice.

Granzyme (Gr) B and perforin are major effector proteins ofNK cells and cytotoxic T lymphocytes (CTLs).22-26 We alreadyfound that expression levels of the Gr B gene decreased in culturedmast cells derived from the spleen ofmi/mi mice.27 Sincemi/mi–cultured mast cells were deficient in killing activity, we consideredthat the poor killing activity ofmi/mi–cultured mast cells may beattributable to the deficient expression of the Gr B gene.28 Weexamined whethermi/mi NK cells were deficient in Gr B geneexpression. However, this was not the case, and NK1.11 cells ofbothmi/miandtg/tgmice were normal in Gr B gene expression. Onthe other hand, NK1.11 cells of tg/tg mice were normal, but thoseof mi/mi mice were deficient in perforin gene expression. Weexamined the transactivation effects ofmi-MITF on the perforingene promoter;mi-MITF appeared to disturb the nuclear transloca-tion of particular transcription factors that are primarily responsiblefor transactivation of the perforin gene.

Materials and methods

Mice

C57BL/6J-mi/1(mi/1) mice were purchased from the Jackson Laboratory(Bar Harbor, ME). VGA-9-tg/tgmice were kindly given by Dr H. Arnheiter(National Institutes of Health, Bethesda, MD) and carry the mousevasopressin–Escherichia coli–galactosidase transgene at the 59flankingregion of themi (MITF) gene.1 Both mice were maintained by consecutive

From the Department of Pathology, Osaka University Medical School, Suita,Osaka, Japan.

Submitted March 16, 2000; accepted December 7, 2000.

Reprints: Yukihiko Kitamura, Dept of Pathology, Osaka University MedicalSchool, Yamada-oka 2-2, Suita Osaka 565-0871, Japan; e-mail: kitamura

@patho.med.osaka-u.ac.jp.

The publication costs of this article were defrayed in part by page chargepayment. Therefore, and solely to indicate this fact, this article is herebymarked ‘‘advertisement’’ in accordance with 18 U.S.C. section 1734.

© 2001 by The American Society of Hematology

2075BLOOD, 1 APRIL 2001 z VOLUME 97, NUMBER 7

For personal use only. by guest on June 4, 2013. bloodjournal.hematologylibrary.orgFrom

backcrosses with our own inbred C57BL/6 colony (more than 12 genera-tions). Eithermi/mior tg/tgmice were produced by crosses between femaleand male heterozygotes of each genotype and were selected by their whitecoat color. To augment NK cells in vivo, 3-week-old mice received anintraperitoneal injection of polyinosinic-polycytidylic acid (poly I:C)(Sigma, St Louis, MO).29 We injected 100mg of poly I:C in 100mL ofphosphate-buffered saline (PBS) into each mouse.

Cell lines

The CTLL-2 and WEHI-3 cell lines were purchased from Riken (Tokyo,Japan). P-815 and YAC-1 cells were obtained from the American TypeCulture Collection (Bethesda, MD). CTLL-2 cells were cultured ina-MEMwith 10% fetal calf serum (FCS) and 1000 U/mL recombinant mouseinterleukin 2 (rmIL-2). The other cells were cultured ina-MEM with 10%FCS. The conditioned medium of WEHI-3 cells were obtained by culturingat 70% confluency for 2 days.

Spleen-cell culture

To prepare single-cell suspensions of spleen cells, spleens were asepticallyremoved frommi/mi or tg/tg mice and their normal (1/1) littermates at 3weeks of age and passed through the mesh. Spleen mononuclear cells wereobtained after centrifugation on Ficoll/Hypaque gradients (density5 1.077)(Sigma) at 300gfor 20 minutes. NK cells were augmented in the spleen-cellculture according to the procedure described.30 Briefly, 1.03 107 spleencells were suspended in 3 mL of the culture medium containing 70%a-MEM, 10% FCS, 20% WEHI-3–conditioned medium, and 1000 U/mLrmIL-2 (R&D Systems, Minneapolis, MN). Cells were transferred onto a25-cm2 tissue-culture flask (Corning Costar, Corning, NY) and incubated at37°C. Every 3 days, 3 mL ofa-MEM medium containing 10% FCS and1000 U/mL rmIL-2 was added to the culture. On the indicated days,nonadherent cells were harvested and used for further analysis.

To generate CTLs in mixed cell culture, spleen mononuclear cells werecultured as described previously.31 Briefly, 1.03 107 spleen cells wereobtained from DBA/2(H-2d) mice, and C57BL/6(H-2b)1/1, C5DBL/6(H-2b) mi/mi, and C57BL/6(H-2b) tg/tg mice. DBA/2 spleen cells wereirradiated with 20 Gy. Spleen cells of C57BL/6 mice of either genotypewere cocultured with irradiated DBA/2 spleen cells for 4 days. Cells wereharvested and used as anti–H-2d CTLs.

Fluorescence-activated cell sorting

Cells were incubated with biotin-conjugated anti-NK1.1 antibody (Pharm-ingen, San Diego, CA) at 4°C for 30 minutes, and stained with fluoresceinisothiocyanate–conjugated streptoavidin (Pharmingen) for 15 minutes at4°C in PBS containing 1.0% bovine serum albumin and 0.1% sodium azide.Cells were analyzed on a FACScan (Becton Dickinson, Los Angeles, CA).

Cytology

We centrifuged 1.03 105 cultured cells in 0.1 mLa-MEM at 600 rpm for 5minutes onto microscope slides using a Cytospin 2 centrifuge (Shandon,Pittsburgh, PA). Air-dried preparations were fixed in methanol and stainedwith 10% Giemsa solution (Merck, Darmstadt, Germany) diluted inTris-buffered saline (pH 6.4). LGLs were identified as being larger thansmall and medium-sized lymphocytes.20,21 They have a relatively highcytoplasmic-to-nuclear ratio and weakly basophilic cytoplasm with a lot ofazurophilic granules. Macrophages were distinguished from LGLs on thebasis of their larger size, vacuolar cytoplasm, and indented nucleus. At least1000 cells were analyzed per slide.

Northern blot analysis

Northern blot analysis was performed according to the standard method.27

Template complementary DNA (cDNA) for the mouse perforin probe wasobtained by reverse-transcription polymerase chain reaction (RT-PCR) withthe use of the following primers: 59-TGCCACTCGGTCAGAATG-CAAGC-39 and 59-CTTCCAGTAATGTGTGCAGGGGC-39. Relative sig-

nal intensity was calculated with the BAS 2000 system (Fuji Photo Film,Tokyo, Japan).

Cytotoxicity assay

NK-cell cytotoxicity to YAC-1 cells and H-2d–specific cytotoxicity toP-815 cells (H-2d) were measured by a51Cr release assay as describedpreviously.32 Briefly, cultured spleen cells ina-MEM with 10% FCS weredistributed at different cell numbers (1.0, 4.0, and 7.53 105 cells for YAC-1cells; 0.5, 1.0, and 2.53 105 cells for P-815 cells) in triplicate into 96-wellmicrotiter plates. Some wells were pre-incubated with 100 nM concanamy-cin A (CMA) (Wako, Osaka, Japan) or diluent (0.1% dimethyl sulfoxide[DMSO]) alone for 1 hour at 37°C. YAC-1 and P-815 cells were labeledwith [51Cr] Na2CrO4 (Amersham, Arlington Heights, IL). We mixed1.03 104 of either YAC-1 or P-815 cells with various numbers of spleencells in a total volume of 200mL a-MEM with 10% FCS. After incubatingplates at 37°C for 4 hours, the radioactivity was determined in 100-mLsamples of cell-free supernatants. The radioactivity released in the wellscontaining YAC-1 or P-815 cells alone with and without adding 0.01%Triton X-100 was designated total release (TR) and spontaneous release(SR), respectively. The percentage of specific51Cr release was calculated bymeans of the following formula: (cpm in the presence of spleen cells2SR)/(TR2 SR)3 100.

Transient transfection

The pEF-BOS plasmid33 containing 1-MITF or mi-MITF cDNA wasconstructed as described previously7,8 and used as an effector. ThepEF-BOS containing theb-galactosidase gene was used as an internalcontrol.7,8 To prepare reporter constructs, the promoter region of theperforin gene (nucleotide [nt]2822 to1173 [11 is the transcription startsite])34 was obtained with PCR and subcloned into the upstream region ofthe luciferase gene in pSPLuc plasmid.7,8 Transient transfection intoCTLL-2 cells were done as described previously.28 Luciferase andb-galac-tosidase activities were measured 24 hours after cotransfection with theeffector and reporter constructs.35 To express1-MITF and mi-MITF inCTLL-2 cells, the cells were cultured in 5 mLa-MEM with 10% FCS and1000 U/mL rmIL-2 for 5 days after transfection with the effector constructs.

Extraction and blotting of proteins

Nuclear and cytoplasmic fractions were prepared as described previously.36

We separated 10mg cell lysate proteins by 12% sodium dodecyl sulfate–polyacrylamide gel electrophoresis and transferred it onto an Immobilonmembrane (Millipore, Bedford, MA). The blot was incubated with themouse monoclonal antibody to pan-actin (Boehringer Mannheim, Ger-many) or proliferating cell nuclear antigen (PCNA) (PC10) (Dako, Kyoto,Japan) and then with horseradish peroxidase–conjugated anti–mouse IgG1antibody (Pharmingen). The blot was reacted with Renaissance chemilumi-nescence reagents (NEN, Boston, MA) before exposure.

Electrophoretic gel mobility shift assay

Electrophoretic gel mobility shift assay (EGMSA) was done according tothe procedure described previously.27 The nuclear fraction was used afterdialysis against 103 cell-pellet volume of a buffer (20 mM Hepes, pH 7.9;20% glycerol; 100 mM KCl; 0.2 mM EDTA; 0.5 mM phenylmethylsulfonylfluoride; and 0.5 mM dithiothreitol) for 1 hour.

Results

Number of NK cells

We assessed the proportion of NK cells in spleen mononuclear cellsby estimating LGLs and NK1.11 cells. The proportion of LGLsdecreased inmi/mi mice but not intg/tg mice (Table 1). On theother hand, the proportion of NK1.11 cells was comparable among1/1, mi/mi, andtg/tgmice (Table 1). This inconsistency suggested

2076 ITO et al BLOOD, 1 APRIL 2001 z VOLUME 97, NUMBER 7

For personal use only. by guest on June 4, 2013. bloodjournal.hematologylibrary.orgFrom

a failure in mi/mi mice to form cytoplasmic granules duringNK-cell development.

Expression of Gr B and perforin genes in spleen

We reported a reduction of the Gr B gene expression inmi/mi–cultured mast cells.27,28If this was also the case inmi/mi NK cells,the reduction of Gr B would be responsible for the deficient NKactivity. We examined the Gr B gene expression in the spleentissues of nontreated 3-week-oldmi/mi, tg/tg, and1/1 mice. TheGr B gene expression was below the limit of detection, and theperforin gene expression was just faintly detectable regardless ofthe genotype (Figure 1). Low levels of the Gr B and perforin geneexpression were consistent with the fact that significant NK activitywas not detectable in the spleen cells of 3-week-old mice.37,38 Toaugment the number of NK cells in the spleen, poly I:C wasinjected intraperitoneally. At 1 day after the injection, a markedincrease in the Gr B mRNA expression was detected in the spleentissues (Figure 1). The degree of the induction was comparableamong1/1, mi/mi, andtg/tgmice. On the other hand, the increasein the perforin gene expression was much smaller in mice of allgenotypes. Especially in the spleen ofmi/mi mice, the induction ofthe perforin gene was too small to be noticeable (Figure 1).

Development of NK cells in culture

We augmented the number of NK cells by culturing spleen cells inthe presence of rmIL-2. The number of NK1.11 cells reached apeak from day 7 to day 10 and declined toward day 14. Only asmall population of hematopoietic cells could survive longer thanday 14. On day 10, the proportion of NK1.11 cells in the culture

Figure 1. Messenger RNA (mRNA) induction of the Gr B and perforin genes inthe spleen tissue by poly I:C injection. Mice of 1/1, mi/mi, and tg/tg genotypesreceived an intraperitoneal injection of poly I:C. Before (0 days) and 24 hours after theinjection (1 day), spleens were excised and used for RNA extraction. Five micro-grams of total RNA were blotted and hybridized with the Gr B or perforin probe.Reprobing with b-actin probe verified an RNA equal loading. Similar results wereyielded in 3 independent experiments.

Figure 2. In vitro development of NK cells from the spleen-cell culturesupplemented with rmIL-2. Spleen cells derived from 1/1, mi/mi, and tg/tg micewere cultured in a-MEM continuously supplemented with rmIL-2. On days 7 and 10,nonadherent cells of the culture were harvested and examined by immunologic(panel A), morphological (panel B) and molecular biologic (panel C) methods. Threeindependent experiments yielded similar results. (A) FACScan analysis to detectNK1.1 marker expression in the cultured spleen cells on day 10. Gates indicate thepopulation of NK1.11 cells in the culture. (B) Morphological detection of NK cells.Cytospin preparation of the culture was done on days 7 and 10, and cells werestained with Giemsa solution. Through days 7 to 10, azurophilic cytoplasmic granuleswere well developed in large lymphocytes of both 1/1 and tg/tg mice, but not of mi/mimice. (C) Northern blot analysis to detect the Gr B, perforin, and MITF geneexpression in the spleen-cell culture. Total RNA (5.0 mg) was extracted from thealiquots of the cultured cells on day 10, blotted, and hybridized with the Gr B, perforin,or MITF probe. Reprobing with b-actin verified an RNA equal loading.

Table 1. Proportion of large granular lymphocytes and NK1.1 1 cells in spleenmononuclear cells of 3-week-old mice

Genotype LGLs NK1.11 cells

1/1 0.9 6 0.2 1.8 6 0.4

mi/mi ,0.1* 1.8 6 0.5

tg/tg 0.8 6 0.1 1.9 6 0.3

Data are expressed as the mean 6 SE (%) of 3 experiments.LGLs indicate large granular lymphocytes; NK, natural killer.*P , .01 by t test when compared with the values of 1/1 and tg/tg spleen cells.

GENE EXPRESSION IN NATURAL KILLER CELLS 2077BLOOD, 1 APRIL 2001 z VOLUME 97, NUMBER 7

For personal use only. by guest on June 4, 2013. bloodjournal.hematologylibrary.orgFrom

was examined by FACScan. A considerable proportion of NK1.11

cells was observed in the culture ofmi/mi, tg/tg, and1/1 mouseorigin (Figure 2A). Giemsa staining revealed that the major part ofthe cultured spleen cells consisted of large lymphocytes with wideand pale cytoplasm (Figure 2B). The proportion of large lympho-cytes was consistent with the proportion of NK1.11 cells in all1/1, mi/mi, and tg/tg spleen-cell cultures (Table 2). Smalllymphocytes, granulocytes, and macrophages belonged to a minorpopulation of the culture (Table 2). A striking difference wasobserved in the content of azurophilic granules. Most largelymphocytes of1/1 and tg/tg mice contained cytoplasmic gran-ules and are considered to be LGLs (Figure 2B; Table 3). Thegranules developed through days 7 to 10 of the culture (Figure 2B).In contrast, most large lymphocytes ofmi/mi mice did not containcytoplasmic granules and remained agranular even on day 10 of theculture (Figure 2B; Table 3).

The cultured spleen cells were harvested on day 10 andexamined for NK activity. Collected cells were cultured togetherwith 51Cr-labeled YAC-1 cells at various effector/target (E/T)ratios, and51Cr release from YAC-1 cells was measured after 4hours. A dose-dependent killing activity was detected in thecultured spleen cells derived from1/1 mice (Table 4). Compa-rable killing activity was also detected in the cultured spleen cellsfrom tg/tg mice. In contrast, the cultured spleen cells frommi/mimice showed markedly reduced levels of killing activity (Table 4).We examined the extent to which the killing activity dependedupon perforin in the present assay. Pretreatment of effector cellswith CMA has been shown to exclude the perforin-mediatedcytotoxicity.39,40 Spleen cells were pre-incubated with CMA andthen cocultured with YAC-1 cells at an E/T ratio of 75:1. Killingactivities were significantly reduced (Table 4). Pretreatment withthe control medium containing 0.1% DMSO alone did not reducethe activity (data not shown).

Gene expression for Gr B and perforin was examined in aliquotsof the cultured spleen cells. There was no difference in the Gr Bgene expression among the cultured spleen cells of the 3 genotypes(Figure 2C). Cultured spleen cells obtained from1/1 and fromtg/tg mice contained the perforin message abundantly. However,the perforin gene expression was reduced to approximately onetenth in themi/mi–cultured spleen cells (Figure 2C). These resultsindicated thatmi/mi NK cells were deficient in killing activityowing to the defect of the perforin gene expression.

Cytotoxicity of CTLs

Although decreased NK activity has been reported inmi/mimice, itremains unknown whether the mice are also deficient in cytotoxic-ity of CTLs. We generated anti–H-2d CTLs from C57BL/6 mice(H-2b) of 1/1, mi/mi, and tg/tg genotypes and examined theircytotoxic activities to P-815 cells (H-2d). A dose-dependentcytotoxic activity was detected in CTLs derived from all 3 mice(Table 5). At any E/T ratio, cytotoxic values ofmi/mi CTLs weresmaller than those of1/1 CTLs, but there was no significantdifference. CTLs oftg/tg mice showed a cytotoxicity comparableto that of1/1 mice.

Expression levels of Gr B and perforin genes were examined inaliquots of CTLs. Both genes were expressed almost equallyamong the CTLs of the 3 genotypes (Figure 3). The results of geneexpression were in accordance with those of killing activity.

The above results indicated thatmi-MITF was involved in thedeficient transactivation of perforin gene in NK cells, but not inCTLs. We attempted to focus on NK cells, but the regulatorymechanisms for the perforin gene expression have been analyzedmore intensely in CTLs by using mouse CTL cell lines, such asCTLL-2.41,42 Moreover, there were no available mouse NK celllines. In the following experiments, we first examined how theperforin gene promoter was transactivated in CTLL-2 cells aftertransfection with1-MITF andmi-MITF.

Effects of MITF on transactivation of the perforin promoter

We examined the effect of1-MITF andmi-MITF on the perforingene transactivation by using the transient cotransfection assay.The 59-flanking sequence of the perforin gene (nt2822 to1173)34 was cloned upstream from the luciferase gene. We also obtained3 deletion constructs containing the promoter sequences from nt2609,2491, or2141 to1173. These luciferase constructs wereshown to function in CTLL-2 cells.41,42 The constructs werecotransfected into CTLL-2 cells with the expression plasmidpEF-BOS containing1-MITF cDNA, mi-MITF cDNA, or noinsert. A strong transactivation effect of1-MITF was detected onthe deletion construct containing2609 to1173 but not on either

Table 2. Proportion of hematopoietic cells on Day 10 of the spleen-cell culturesupplemented with recombinant mouse interleukin-2

GenotypeSmall

lymphocytesLarge

lymphocytes Granulocytes Macrophages

1/1 15.4 6 1.4 70.5 6 3.7 10.3 6 1.1 3.8 6 0.7

mi/mi 14.7 6 2.5 65.4 6 2.1 16.2 6 2.6 3.7 6 0.4

tg/tg 16.0 6 2.8 69.2 6 3.3 9.3 6 2.9 5.5 6 0.5

Data are expressed as the mean 6 SE (%) of 3 experiments.

Table 3. Percentage of large granular lymphocytes in large lymphocytes ondays 7 and 10 of the spleen-cell culture supplementedwith recombinant mouse interleukin-2

Genotype Day 7 Day 10

1/1 50.4 6 2.4 92.8 6 4.7

mi/mi 7.7 6 1.5* 13.7 6 2.1*

tg/tg 42.0 6 2.8 90.5 6 3.3

Data are expressed as the mean 6 SE (%) of 3 experiments.*P , .01 by t test when compared with the values of 1/1 and tg/tg spleen cells.

Table 4. Cytotoxic activity to YAC-1 cells of spleen cells cultured withrecombinant mouse interleukin-2 for 10 days

E/Tratio

CMAtreatment

Specific 51Cr release (%)*

1/1 mi/mi tg/tg

10 No 15.6 6 1.1 7.8 6 0.7† 15.0 6 1.1

40 No 41.6 6 2.5 15.7 6 2.1† 45.2 6 3.2

75 No 52.5 6 2.8 18.5 6 2.2† 55.3 6 3.9

75 Yes 18.4 6 1.0‡ 10.5 6 1.2 18.5 6 1.8‡

E/T ratio indicates the ratio of cultured spleen cells to YAC-1 cells; CMA,concanamycin A.

*Mean 6 SE of 3 experiments.†P , .05 by t test when compared with the values of 1/1 and tg/tg spleen cells.‡P , .05 by t test when compared with the values of 1/1 and tg/tg spleen cells in

the absence of CMA.

Table 5. Cytotoxic activity of anti–H-2 d cytotoxic T lymphocytes to P-815 cells

E/Tratio

Specific 51Cr release (%)*

1/1 mi/mi tg/tg

5 9.9 6 1.1 4.5 6 1.3 5.6 6 1.6

10 19.7 6 1.5 17.7 6 2.2 16.0 6 1.0

25 23.7 6 0.8 21.8 6 1.0 24.8 6 1.7

E/T ratio is the ratio of anti–H-2d cytotoxic T lymphocytes to P-815 cells.*Mean 6 SE of 3 experiments.

2078 ITO et al BLOOD, 1 APRIL 2001 z VOLUME 97, NUMBER 7

For personal use only. by guest on June 4, 2013. bloodjournal.hematologylibrary.orgFrom

the longer (nt2822 to1173) or the 2 shorter (nt2491 or2141 to1173) constructs (Figure 4A). This indicated that the positivecis-actingelement for1-MITF was present between2609 and2491.

We reported previously that1-MITF directly transactivated anumber of genes through binding to CANNTG (E-box) mo-tifs.7-10,27 In the region between2609 and2491, 2 CANNTGmotifs were present: CAACTG (nt2567 to 2562; E1) andCAGCTG (nt2525 to2520; E2). The 2 motifs were mutated inthe deletion construct containing2609 to 1173: CAACTG toCTACAG, or CAGCTG to CTGCAG. Neither mutation reducedthe luciferase activity of the original nt2609 promoter construct(Figure 4B). Therefore,1-MITF did not appear to transactivate theperforin promoter through the direct binding with the E-box motifs.

Between nt2609 and2491 of the perforin promoter, there isan NF-P motif (nt2505 to 2493) that is known to be a strongcis-acting element for transactivation of the perforin gene.43,44Twonuclear factors, termed NF-P1 and NF-P2, have been shown to bindand transactivate the NF-P motif in cytolytic lymphocytes.43,44

Reporter constructs that contained the NF-P motif (nt2822 and nt2609 constructs) were transactivated 6-fold as strongly as theconstruct lacking the promoter (nt1173 construct), even when anempty pEF-BOS was cotransfected (Figure 4A). This transactiva-tion effect was remarkably diminished either when the NF-P motifwas deleted from the promoter, as seen in the case of nt2491 or2141 construct (Figure 4A), or when the NF-P motif was mutatedto ACATTCCTG (Figure 4B).43,44 Endogenous factors, such asNF-P1 and NF-P2, of CTLL-2 cells appeared to transactivate thepromoter containing the NF-P motif.

When compared with the endogenous transactivation observedin CTLL-2 cells, the coexpression of1-MITF transactivated the nt2609 reporter construct 3-fold further (Figure 4A). The mutationin the NF-P motif abrogated this effect (Figure 4B), suggesting that1-MITF transactivated the nt2609 reporter construct through theNF-P motif. In contrast, when the nt2609 reporter construct wascotransfected with cDNA encodingmi-MITF, the luciferase activitydecreased to one third of the value obtained by the cotransfection withempty pEF-BOS (Figure 4A). The enhancing effect of1-MITF and theinhibiting effect of mi-MITF were still observed when the 2 E-boxmotifs between2609 and2491 were mutated (Figure 4B).

Poor binding of MITF to E-box and NF-P motifs

In vitro binding of 1-MITF to the E-box and NF-P motifs wasexamined by EGMSA. As controls, we used 2 oligonucleotides

containing the CACATG (nt2128 to2123; oligonucleotide P6)and mutated (CTCAAG; oligonucleotide mP6) motif of the mousemast cell protease (MMCP)–6 gene promoter.5,8 In addition, 3oligonucleotides containing CAACTG (oligonucleotide E1),CAGCTG (oligonucleotide E2), and NF-P (oligonucleotide NF)motifs of the perforin gene promoter were synthesized (Figure 5A).The binding between1-MITF and oligonucleotide P6 was com-peted out completely by the cold oligonucleotide P6, but not at allby the cold oligonucleotide E1 or oligonucleotide mP6 (Figure 5B).The competitive effect of the oligonucleotide E2 was slight (Figure5B). Consistently,1-MITF yielded a little amount of the complexwith oligonucleotide E2, but no complex with oligonucleotide E1(Figure 5B). Neither1-MITF nor mi-MITF bound oligonucleotideNF (Figure 5B).

Figure 4. Positive effect of 1-MITF and negative effect of mi -MITF on transacti-vation of the mouse perforin gene. (A) The perforin gene promoter sequences from2822,2609,2491, and 2141 to 1173 were inserted into the upstream of theluciferase gene of pSPLuc. These reporter constructs and empty pSPLuc weretransfected into CTLL-2 cells with the pEF-BOS containing no insert (pEF-BOS) or1-MITF or mi-MITF cDNA by electroporation. The bars represent the mean 6 SE ofthe luciferase activities obtained by 3 independent experiments: (f) indicates1-MITF; (M), mi-MITF; (u), pEF-BOS. (B) The effect of mutations in the CANNTGand NF-P motifs on the luciferase activity. In the 2609 promoter construct, theCAACTG (E1, nt 2567 to 2562; upstream oval) and CAGCTG (E2, nt 2525 to 2520;downstream oval) motifs, and NF-P motif (nt 2505 to 2493; black square) weremutated in various combinations. X indicates mutated motifs: CACATG to CTACAG;CAGCTG to CTGCAG; ACAGGAAGT to ACATTCCTG. Only the mutation in theNF-P motif abrogated the transactivation effect of 1-MITF. *P , .05 by t test whencompared with the values when pEF-BOS containing no insert was cotransfected.

Figure 3. Gene expression for Gr B and perforin in anti–H-2 d CTLs derived from1/1, mi/mi , and tg/tg mice. Total RNA (3.0 mg) was extracted from the aliquots ofCTLs, blotted, and hybridized with the Gr B or perforin probe. Reprobing with b-actinprobe verified an RNA equal loading.

GENE EXPRESSION IN NATURAL KILLER CELLS 2079BLOOD, 1 APRIL 2001 z VOLUME 97, NUMBER 7

For personal use only. by guest on June 4, 2013. bloodjournal.hematologylibrary.orgFrom

Inhibitory effect of mi-MITF on nuclear translocationof NF-P motif–binding factors

The nuclear fraction was obtained from CTLL-2 cells and exam-ined with EGMSA by using oligonucleotide NF as a probe. Wedetected 2 retarded bands (Figure 6A; denoted by arrowheads 1 and2). Addition of cold oligonucleotide NF as a competitor erased bothbands completely (Figure 6A). This competitive effect disappearedwhen the NF-P motif was mutated to ACATTCCTG (oligonucleo-tide mNF) (Figure 6A), indicating that the 2 bands were NF-Pmotif–specific. The bands shown by arrowheads 1 and 2 wereconsidered to represent the NF-P motif–binding nuclear factors thathave been named NF-P1 and NF-P2, respectively.43,44

There was the possibility that NF-P1, NF-P2, or both mediatedthe enhancing effect of1-MITF and the inhibiting effect ofmi-MITF on the perforin promoter. When the CTLL-2 cell lysatewas divided into nuclear and cytoplasmic fractions, NF-P1 andNF-P2 were detected in the nuclear fraction (Figure 6B), indicatingan efficient translocation of both factors into the nucleus. After thetransfection with1-MITF, mi-MITF cDNA, or vector alone,CTLL-2 cells were fractionated and subjected to EGMSA accord-ing to the same procedure (Figure 6C). Transfection with vectoralone or1-MITF cDNA did not change the nuclear localization ofeither NF-P1 or NF-P2. NF-P2 was also detectable in the nuclear

fraction of CTLL-2 cells transfected withmi-MITF cDNA. Theband intensity for NF-P2 was slightly weaker. By contrast, NF-P1was scarcely detectable in this fraction. Instead, both NF-P1 andNF-P2 were detectable in the cytoplasmic fraction of CTLL-2 cellstransfected withmi-MITF, but not in the fraction of CTLL-2 cellstransfected with1-MITF or with vector alone.

Aliquots of the transfected CTLL-2 cells were examined forgene expression. High levels of mRNA expression for the exog-enous1-MITF and mi-MITF were detected in the correspondingCTLL-2 cells (Figure 6D). The RNA blot was rehybridized with theGr B and perforin probes. No significant difference was detected inthe expression of either gene among 3 types of the transfectedCTLL-2 cells (Figure 6D).

Figure 6. Cytoplasmic retention of NF-P motif–binding factors in CTLL-2 cellsoverexpressing mi-MITF. (A) Two nuclear factors that specifically bind the NF-Pmotif are present in the nucleus of CTLL-2 cells. The nuclear extract (NE) was preparedfrom CTLL-2 cells and incubated with 32P-labeled oligonucleotide NF in the presence orabsence (2) of the competitor. For competition, NE was incubated with an excess amount(100-fold) of unlabeled oligonucleotide NF or oligonucleotide mNF prior to the reaction withthe probe. The reaction mixture was separated by polyacrylamide gel.Arrowheads 1 and 2indicate the same retarded bands that specifically bind the NF-Pmotif. (B)The nuclear (NE)and cytoplasmic (CE) fractions of CTLL-2 cells were examined with EGMSA by usingoligonucleotide NF as a probe. Arrowheads 1 and 2 indicate the same retarded bands thatspecifically bind the NF-P motif. (C) EGMSA of CTLL-2 cells transfected with 1-MITF andmi-MITF. The nuclear and cytoplasmic fractions were extracted from CTLL-2 cells that hadbeen transfected with 1-MITF, mi-MITF, or vector alone (Vec) 5 days before. The extractswere incubated with 32P-labeled oligonucleotide NF, and separated on polyacrylamide gel.Arrowheads 1 and 2 indicate the same retarded bands that bind specifically the NF-P motif.(D) Gene expression in CTLL-2 cells transfected with cDNA encoding 1-MITF or mi-MITF.CTLL-2 cells were transfected with 1-MITF, mi-MITF, or vector alone. After 5 days, totalRNA(3.0 mg) was extracted from the cells, blotted, and hybridized with the MITF, Gr B, andperforin probes. Reprobing with b-actin probe verified an RNAequal loading.

Figure 5. EGMSA using oligonucleotides containing E-box or NF-P motif. (A)Sequences of oligonucleotides used in EGMSA. P6 contains the CACATG motif(boxed) of the MMCP-6 gene promoter, which is mutated (underlined) in mP6. E1 andE2 contain the CAACTG and CAGCTG motifs (boxed) of the perforin gene promoter,respectively. NF contains the NF-P motif (boxed) of the perforin gene promoter, whichis mutated (underlined) in mNF. Arrowheads indicate the position of nucleotides in thepromoters. (B) In vitro binding of MITF to the 2 CANNTG motifs (E1 and E2) and theNF-P motif in the perforin promoter. GST-fused 1-MITF or mi-MITF protein wasincubated with 32P-labeled oligonucleotides (probe oligonucleotide) in the presenceor absence (2) of the competitor. For competition, an excess amount (100-fold) ofindicated unlabeled oligonucleotides was added to the reaction prior to the probe.The reaction mixture was separated on polyacrylamide gel. An arrowhead indicatesthe position of protein-DNA complex.

2080 ITO et al BLOOD, 1 APRIL 2001 z VOLUME 97, NUMBER 7

For personal use only. by guest on June 4, 2013. bloodjournal.hematologylibrary.orgFrom

Cytoplasmic localization of the NF-P motif–binding factorin mi/mi –cultured spleen cells

Using the same procedure as described above, we examined1/1,mi/mi, andtg/tg spleen cells after culturing with rmIL-2 for 10days. In the nuclear fraction of thetg/tg spleen cells, 2 retardedbands were detected at positions comparable to those of NF-P1 andNF-P2 (Figure 7A). Addition of cold oligonucleotide NF, but not ofoligonucleotide mNF, as a competitor erased the 2 bands (Figure7A), indicating that the 2 bands were NF-P motif–specific. NK1.11

cells of tg/tg mice appeared to possess both NF-P1 and NF-P2 intheir nucleus.

Similarly, NF-P1 and NF-P2 were detected in the nuclearfraction of the1/1 spleen cells (Figure 7B). In contrast, no bandswere detected at the position of either NF-P1 or NF-P2 in thenuclear fraction obtained from themi/mi spleen cells (Figure 7B).On the other hand, both NF-P1 and NF-P2 were weakly detected inthe cytoplasmic fractions of 3 types of the cultured cells. Theintensity of the bands was comparable in1/1 andtg/tg cells andwas slightly stronger inmi/mi cells (Figure 7B). Competitionexperiments showed that NF-P1 and NF-P2 detected in eachfraction were NF-P motif–specific (data not shown).

Aliquots of the nuclear and cytoplasmic fractions were blottedwith the PCNA and pan-actin antibodies (Figure 7B, lower panels).The immunoreactive signals for PCNA and actin were detectedprimarily in the nuclear and cytoplasmic fractions, respectively,indicating that the cell lysate of each genotype was dividedappropriately into nuclear and cytoplasmic fractions.

To examine whether the cytoplasmic NF-P1 and NF-P2 werederived from NK1.11 cells but not from NK1.12 cells, weextracted nuclear and cytoplasmic fractions frommi/mi–culturedspleen cells on various days of the culture. Each fraction wasexamined with EGMSA by using oligonucleotide NF as a probe(Figure 7C). In the nuclear fractions, NF-P2 was detected throughdays 2 to 7, but disappeared almost completely on day 10. NF-P1was faintly detectable only on day 2. In the cytoplasmic fractions,no retarded bands appeared on day 2. On day 5, NF-P2 becamedetectable and gradually increased through days 5 to 10. Inaddition, NF-P1 became detectable on day 7 and more intense onday 10. Since NK1.11 cells were only a minor population on day 2and day 5, but became a major population through days 7 to 10(Table 3, and data not shown), it was reasonable to consider thatcytoplasmic NF-P1 and NF-P2 were derived primarily fromNK1.11 cells. Together, the results of EGMSA suggested adeficient nuclear translocation of NF-P1 and NF-P2 inmi/miNK1.11 cells.

Discussion

The proportion of NK1.11 cells was normal in the spleen ofmi/mimice, though the proportion of LGLs decreased significantly. Boththe proportion of NK1.11 cells and that of LGLs were normal in thespleen oftg/tgmice. The difference betweenmi/miandtg/tgmutantanimals was reproducible in the culture of spleen cells continu-ously stimulated by rmIL-2. Through days 7 to 10, the populationof NK1.11 and large lymphocytic cells became predominant in theculture of1/1 spleen cells. Bothmi/mi and tg/tg cultured spleencells showed a population similar in this respect. However, most ofmi/mi lymphocytic cells remained agranular in their cytoplasm,whereas cytoplasmic granules were well developed in lymphocyticcells of1/1 andtg/tggenotypes. Mice ofmi/migenotype appeared

Figure 7. Detection of NF-P motif–binding factors in the cultured spleen cells.Arrowheads 1 and 2 indicate the same positions of protein-DNA complexes as shownin Figure 6. (A) Two nuclear factors that specifically bind with the NF-P motif arepresent in the nucleus of tg/tg–cultured spleen cells. The nuclear extract (NE) wasprepared from tg/tg spleen cells after culturing with rmIL-2 for 10 days and incubatedwith 32P-labeled oligonucleotide NF in the presence or absence (2) of the competitor.For competition, NE was incubated with an excess amount (100-fold) of unlabeledoligonucleotide NF or oligonucleotide mNF prior to the reaction with the probe. Thereaction mixture was separated by polyacrylamide gel. (B) Detection of the NF-Pmotif–binding factors in the nuclear and cytoplasmic (CE) extracts by EGMSA. Eachcell extract was prepared from CTLL-2 cells, and 1/1, mi/mi, and tg/tg spleen cellsafter culturing with rmIL-2 for 10 days. The extracts were incubated with 32P-labeledoligonucleotide NF and separated on polyacrylamide gel (upper panel). Aliquots (10mg per lane) of the extracts were blotted with the PCNA and pan-actin antibodies(lower panels). (C) Detection of the NF-P motif–binding factors during in vitrodevelopment of mi/mi NK cells. To produce NK cells, spleen cells from mi/mi micewere cultured as described in “Materials and methods.” The nuclear and cytoplasmicfractions were extracted from the cultured cells on the indicated days after theinitiation of the culture. The extracts were incubated with 32P-labeled oligonucleotideNF and separated on polyacrylamide gel.

GENE EXPRESSION IN NATURAL KILLER CELLS 2081BLOOD, 1 APRIL 2001 z VOLUME 97, NUMBER 7

For personal use only. by guest on June 4, 2013. bloodjournal.hematologylibrary.orgFrom

to be defective in cytoplasmic granule formation during NK-celldevelopment, but not in the entire process of NK-cell development.Although it is poorly understood how NK cells produce thegranules, heparin is required for the formation of normal cytoplas-mic granules in mast cells.45,46 We previously reported that theheparin content decreased in the skin mast cells ofmi/mi mice.47

However, it must be pointed out that the principal proteoglycan inNK-cell granules is chondroitin sulfate A.48 Reduced production ofchondroitin sulfate A could be a cause for the poor development ofcytoplasmic granules inmi/miNK1.11 cells.

Gr B and perforin are major effector proteins contained by NKcells. The expression of Gr B was normal but that of perforin wasreduced inmi/miNK1.11 cells. In proportion to the reduction of theperforin gene expression, NK activity was reduced inmi/miNK1.11 cells. These abnormalities exhibited bymi/mi mice werereminiscent of Chediak-Higashi syndrome (bg/bg) mice49 andperforin gene knock-out mice (perforin [2/2] mice).24-26 Incontrast tomi/mi NK1.11 cells, NK cells ofbg/bg mice containgiant granules in their cytoplasm. Butbg/bg NK cells are asdefective in NK activity asmi/mi NK cells.50 Perforin (2/2) miceshowed more profound reduction in NK activity thanmi/mi mice.The magnitude of the reduction observed inmi/mi mice wascomparable to that of perforin (1/2) mice.26 Normal developmentof azurophilic granules and full expression of perforin geneappeared necessary for normal NK activity.

Although deficient in NK activity,mi/mi mice were normal inCTL cytotoxic activity. Moreover,mi/mi CTLs expressed both GrB and perforin genes as abundantly as1/1 CTLs. Themi-MITFappeared to inhibit transactivation of the perforin gene in NK cells,but not in CTLs. In this sense, the effect ofmi-MITF on the perforingene transactivation should be analyzed in NK cells. However, weused CTLL-2 cells initially for this purpose, because positive andnegative regulatory regions of the perforin promoter have beendetermined in CTLs.41,42The positive regulatory region is active incytolytic lymphocytes, such as CTLL-2 cells, whereas the negativeregulatory region functions predominantly in other types of cells.The NF-P motif (nt2505 to2493) is one of the strongest positivecore elements.41,42 In CTLL-2 cells, 1-MITF enhanced andmi-MITF suppressed the transactivation of the perforin promoterthrough the NF-P motif, but not through the E-box motifs.However, neither1-MITF nor mi-MITF bound the NF-P motif.There seemed to be other factors that mediated the enhancingeffects of1M-MITF and suppressing effects ofmi-MITF on theNF-P motif.

NF-P1 and NF-P2 are defined as 2 nuclear factors thatspecifically bind the NF-P motif.43,44 Although both NF-P1 andNF-P2 are considered to belong to the Ets family,43 molecularcloning of them has not been reported yet. In CTLL-2 cells, bothfactors were detected in the nucleus, but not in the cytoplasm,indicating their efficient translocation into the nucleus. Overexpres-sion of mi-MITF resulted in the cytoplasmic retention of most ofNF-P1 and a part of NF-P2 (Figure 6C). The similar effect ofmi-MITF was found inmi/miNK1.11 cells, but its inhibitory effectwas more profound even thoughmi-MITF was expressed muchless abundantly than in CTLL-2 cells overexpressingmi-MITF. InNK1.11–cultured spleen cells ofmi/mi mice, both NF-P1 andNF-P2 were detected in the cytoplasm, but not in the nucleus

(Figure 7B). The results of EGMSA appeared to account for thedifferent levels of the perforin gene expression among 3 types ofcells expressingmi-MITF. Reduction of the perforin expressionwas undetectable inmi/mi CTLs. Similarly, there was little, if any,reduction of perforin expression in CTLL-2 cells even whenmi-MITF was overexpressed (Figure 6D). In contrast, the reductionwas much larger inmi/mi NK1.11–cultured spleen cells. Themi-MITF seemed to suppress the transactivation of the perforingene more efficiently in NK cells than in CTLs.

On the other hand,1-MITF did not appear essential for nucleartranslocation of NF-P1 and NF-P2, because both factors werelocated in the nucleus of NK1.11 cells of tg/tg mice lacking anyMITFs. Concomitantly, no reduction of the perforin expression wasfound in tg/tg NK1.11 cells or tg/tg CTLs. This result alsosuggested that the enhancing effect of1-MITF on the perforingene promoter was undetectable at the physiologic concentration of1-MITF. The presence ofmi-MITF, rather than the absence of1-MITF, appeared to be a cause of decreased expression of theperforin gene through poor transactivation of the NF-P motif.

We reported previously thatmi-MITF is mutated in its sequencefor the nuclear localization signal and thus deficient in entering thenucleus.6 This deficiency ofmi-MITF might cause the cytoplasmicretention of NF-P1 and NF-P2. The simplest explanation for thisresult may be as follows: NF-P1 and NF-P2 interact withmi-MITFin the cytoplasm, and the resulting complexes fail to enter thenucleus because of the defect inmi-MITF. In our previous report,we demonstrated the similar mechanisms that causemi-MITFs tointerfere with other transcription factors. Whenmi-MITF wasoverexpressed, c-Jun51 and PU.1,52 a transcription factor of the Etsfamily, were retained in the cytoplasm. Furthermore, the cytoplas-mic retention of c-Jun resulted in a decreased expression of theMMCP-7 gene inmi/mi–cultured mast cells.51 In P-815 mastocy-toma cells, mi-MITF interfered with endogenous transcriptionfactors, such as AP-1 and PEBP2, that are primarily responsible fortranscription of the Gr B gene.28 However, in our preliminaryexperiments,mi-MITF did not prefer to interact with NF-P1 orNF-P2 under in vitro conditions (A.I., unpublished observations,2000). Further study is necessary on the molecular mechanismsunderlying the inhibitory effect ofmi-MITF.

Besides NK cells and CTLs, mast cells are another cell typepossessing cytotoxicity. As reported previously,mi/mi–culturedmast cells were deficient in Gr B gene expression and cytotoxic-ity.28 However, NK cells and CTLs ofmi/migenotype were normalin Gr B gene expression. On the other hand,mi/mi NK cells weredeficient in perforin gene expression, whilemi/mi CTLs werenormal in this regard. Mast cells are lacking in perforin geneexpression even if the cells are derived from1/1 mice. This maybe the reason cultured mast cells are modest in their cytotoxicitywhen compared with NK cells and CTLs (A.I., unpublishedobservations). With respect to expression levels of Gr B andperforin, mi/mi NK1.11 cells resemble1/1 cultured mast cells.Thus, we considered that the deficient cytotoxicity ofmi/miNK1.11 cells was attributable to the poor expression of the perforingene. The inhibitory effect ofmi-MITF on the transactivation of theperforin gene appeared to explain the deficient cytotoxicity of NKcells inmi/mimice.

References

1. Hodgkinson CA, Moore KJ, Nakayama A, et al.Mutations at the mouse microphthalmia locus areassociated with defects in a gene encoding a

novel basic-helix-loop-helix-zipper protein. Cell.1993;74:395-404.

2. Hughes JJ, Lingrel JB, Krakowsky JM, Anderson

KP. A helix-loop-helix transcription factor-likegene is located at the mi locus. J Biol Chem.1993;268:20687-20690.

2082 ITO et al BLOOD, 1 APRIL 2001 z VOLUME 97, NUMBER 7

For personal use only. by guest on June 4, 2013. bloodjournal.hematologylibrary.orgFrom

3. Hemesath TJ, Streingrimsson E, McGill G, et al.Microphthalmia, a critical factor in melanocytedevelopment, defines a discrete transcription fac-tor family. Genes Dev. 1994;8:2770-2780.

4. Steingrimsson E, Moore KJ, Lamoreux ML, et al.Molecular basis of mouse microphthalmia (mi)mutations helps explain their developmental andphenotypic consequences. Nat Genet. 1994;8:256-263.

5. Morii E, Takebayashi K, Motohashi H, YamamotoM, Nomura S, Kitamura Y. Loss of DNA bindingability of the transcription factor encoded by themutant mi locus. Biochem Biophys Res Comm.1994;205:1299-1304.

6. Takebayashi K, Chida K, Tsukamoto I, et al. Re-cessive phenotype displayed by a dominantnegative microphthalmia-associated transcriptionfactor mutant is a result of impaired nuclear local-ization potential. Mol Cell Biol. 1996;16:1203-1211.

7. Tsujimura T, Morii E, Nozaki M, et al. Involvementof transcription factor encoded by the mi locus inthe expression of c-kit receptor tyrosine kinase incultured mast cells of mice. Blood. 1996;88:1225-1233.

8. Morii E, Tsujimura T, Jippo T, et al. Regulation ofmouse mast cell protease 6 gene expression bytranscription factor encoded by the mi locus.Blood. 1996;88:2488-2494.

9. Morii E, Jippo T, Tsujimura T, et al. Abnormal ex-pression of mouse mast cell protease 5 gene incultured mast cells derived from mutant mi/mimice. Blood. 1997;90:3057-3066.

10. Jippo T, Morii E, Tsujino K, et al. Involvement oftranscription factor encoded by the mouse mi lo-cus (MITF) in expression of p75 receptor of nervegrowth factor in cultured mast cells. Blood. 1997;90:2601-2608.

11. Hertwig P. Neue Mutationen und Koppelungs-gruppen bei der Hausmaus. Z Indukt Abstamm-uVererbLehre. 1942;80:220-246.

12. Hertwig P. Sechs neue Mutationen bei der Haus-maus in ihrer bedeutung fur allgemeine Ver-erbungsfragen. Z Menschl Vererbungs-u KonstL.1942;26:1-21.

13. Tachibana M, Hara Y, Vyas D, et al. Cochlear dis-order associated with melanocyte anomaly inmice with a transgenic insertional mutation. MolCell Neurosci. 1992;3:433-445.

14. Silvers WK. The Coat Colors of Mice: A Model forMammalian Gene Action and Interaction. NewYork, NY: Springer-Verlag; 1979.

15. Green MC. Catalog of mutant genes and poly-morphic loci. In: Lyon MF, Searle AG, eds. Ge-netic Variants and Strains of the LaboratoryMouse. Stuttgart, Germany: Gustav Fischer Ver-lag; 1981:158-160.

16. Stevens J, Loutit JF. Mast cells in spotted mutantmice (W, Ph, mi). Proc R Soc Lond. 1982;215:405-409.

17. Stechschulte DJR, Sharma KN, Dileepan KM, etal. Effect of the mi allele on mast cells, basophils,natural killer cells, and osteoclasts in C57BL/6Jmice. J Cell Physiol. 1987;132:565-570.

18. Ebi Y, Kasugai T, Seino Y, Onoue H, Kanemoto T,Kitamura Y. Mechanism of mast cell deficiency inmutant mice of mi/mi genotype: an analysis byco-culture of mast cells and fibroblasts. Blood.1990;75:1247-1251.

19. Seaman WE, Gindhart TD, Greenspan JS, Black-man MA, Talal N. Natural killer cells, bone, andthe bone marrow: studies in estrogen-treated

mice and in congenitally osteopetrotic (mi/mi)mice. J Immunol. 1979;122:2541-2547.

20. Kumagai K, Itoh K, Suzuki R, Hinuma S, Saitoh F.Studies of murine large granular lymphocytes, I:identification as effector cells in NK and K cyto-toxicities. J Immunol. 1982;129:388-394.

21. Hackett J, Tutt M, Lipscomb M, Bennett M, KooG, Kumar V. Origin and differentiation of naturalkiller cells, II: functional and morphological stud-ies of purified NK-1.11 cells. J Immunol. 1986;136:3124-3131.

22. Heusel JW, Wesselschmidt RL, Shresta S, Rus-sell JH, Ley TJ. Cytotoxic lymphocytes requiregranzyme B for the rapid induction of DNA frag-mentation and apoptosis in allogenic target cells.Cell. 1994;76:977-987.

23. Shresta S, MacIvor DM, Heusel JW, Russell JH,Ley TJ. Natural killer and lymphokine-activatedkiller cells require granzyme B for the rapid induc-tion of apoptosis in susceptible target cells. ProcNatl Acad Sci U S A. 1995;92:5679-5683.

24. Kagi D, Ledermann B, Burki K, et al. Cytotoxicitymediated by T cells and natural killer cells isgreatly impaired in perforin-deficient mice. Na-ture. 1994;369:31-37.

25. Walsh GM, Matloubian M, Liu CC, et al. Immunefunction in mice lacking the perforin gene. ProcNatl Acad Sci U S A. 1994;91:10854-10858.

26. Lowin B, Beerman F, Schmidt A, Tschopp J. A nullmutation in the perforin gene impairs cytolytic Tlymphocyte- and natural killer cell-mediated cyto-toxicity. Proc Natl Acad Sci U S A. 1994;91:11571-11575.

27. Ito A, Morii E, Maeyama K, et al. Systematicmethod to obtain novel genes that are regulatedby mi transcription factor: impaired expression ofgranzyme B and tryptophan hydroxylase in mi/micultured mast cells. Blood. 1998;91:3210-3221.

28. Ito A, Morii E, Kim D-K, et al. Inhibitory effect ofthe transcription factor encoded by the mi mutantallele in cultured mast cells of mice. Blood. 1999;93:1189-1196.

29. Djeu JY, Heinbaugh JA, Holden HT, HerbermanRB. Augmentation of mouse natural killer cell ac-tivity by interferon and interferon inducers. J Im-munol. 1979;122:175-181.

30. Hattori M, Suda T, Iizuka M, et al. Generation ofcontinuous large granular lymphocyte lines byinterleukin 2 from the spleen cells of mice in-fected with Moloney leukemia virus: involvementof interleukin 3. J Exp Med. 1987;166:833-849.

31. Russel JH, Hale AH, Inbar D, Eisen HN. Loss ofreactivity of a BALB/c myeloma tumor with alloge-neic and syngeneic cytotoxic T lymphocytes. EurJ Immunol. 1978;8:640-645.

32. Bissonnette EY, Befus AD. Inhibition of mast cell-mediated cytotoxicity by IFN-a/b and -g. J Immu-nol. 1990;145:3385-3390.

33. Mizushima S, Nagata S. pEF-BOS, a powerfulmammalian expression vector. Nucleic AcidsRes. 1990;18:5322.

34. Youn B-S, Liu C-C, Kim K-K, Young JD, KwonMH, Kwon BS. Structure of the mouse pore-form-ing protein (perforin) gene: analysis of transcrip-tion initiation site, 59 flanking sequence, and alter-native splicing of 59 untranslated regions. J ExpMed. 1991;173:813-822.

35. Yasumoto K, Yokoyama K, Shibata K, Tomita T,Shibahara S. Microphthalmia-associated tran-scription factor as a regulator for melanocyte-spe-cific transcription of the human tyrosinase gene.Mol Cell Biol. 1994;14:8058-8070.

36. Schreiber E, Matthias P, Muller MM, Schaffner W.

Rapid detection of octamer binding proteins with’mini-extracts’, prepared from a small number ofcells. Nucleic Acids Res. 1989;17:6419.

37. Cudkowicz G, Hochman PS. Do natural killercells engage in regulated reactions against self toensure homeostasis? Immunol Rev. 1979;44:13-41.

38. Koo GC, Peppard JR, Hatzfeld A. Ontogeny ofNk-11 natural killer cells, I: proportion of Nk-11cells in fetal, baby, and old mice. J Immunol.1982;129:867-871.

39. Kataoka T, Takaku K, Magae J, et al. Acidificationis essential for maintaining the structure andfunction of lytic granules of CTL. J Immunol.1994;153:3938-3947.

40. Togashi K, Kataoka T, Nagai K. Characterizationof a series of vacuolar type H1-ATPase inhibitorson CTL-mediated cytotoxicity. Immunol Lett.1997;55:139-144.

41. Lichtenheld MG, Podack ER. Structure and func-tion of the perforin promoter and upstream re-gion: reciprocal gene activation or silencing per-forin positive and negative cells. J Immunol.1992;149:2619-2626.

42. Zhang Y, Lichtenheld MG. Non-killer cell-specifictranscription factors silence the perforin promoter.J Immunol. 1997;158:1734-1741.

43. Koizumi H, Horta MF, Youn B-S, et al. Identifica-tion of a killer cell-specific regulatory element ofthe mouse perforin gene: an Ets-binding site-homologous motif that interacts with Ets-relatedproteins. Mol Cell Biol. 1993;13:6690-6701.

44. Horta MF, Fu K-C, Koizumi H, Young JD, Liu C-C.Cell-free conversion of a ubiquitous nuclear pro-tein into a killer-cell-specific form that binds to theNF-P enhancer element of the mouse perforingene. Eur J Biochem. 1996;238:639-646.

45. Humphries DE, Wong GW, Friend DS, et al. Hep-arin is essential for the storage of specific granuleproteases in mast cells. Nature. 1999;400:769-772.

46. Forsberg E, Pejler, Ringvall M, et al. Abnormalmast cells in mice deficient in a heparin-synthe-sizing enzyme. Nature. 1999;400:773-776.

47. Kasugai T, Oguri K, Jippo-Kanemoto T, et al. Defi-cient differentiation of mast cells in the skin ofmi/mi mice: usefulness of in situ hybridization forevaluation of mast cell phenotype. Am J Pathol.1993;143:1337-1347.

48. MacDermott RP, Schmidt RE, Caulfield JP, et al.Proteoglycans in cell-mediated cytotoxicity: iden-tification, localization, and exocytosis of a chon-droitin sulfate proteoglycan from human clonednatural killer cells during target cell lysis. J ExpMed. 1985;162:1771-1787.

49. Roder J, Duwe A. The beige mutation in themouse selectively impairs natural killer cell func-tion. Nature. 1979;278:451-453.

50. Faigle W, Raposo G, Tenza D, et al. Deficientpeptide loading MHC class II endosomal sortingin a human genetic immunodeficiency disease:the Chediak-Higashi syndrome. J Cell Biol. 1998;141:1121-1134.

51. Ogihara H, Morii E, Kim D-K, Oboki K, KitamuraY. Inhibitory effect of the transcription factor en-coded by the mutant mi microphthalmia allele ontransactivaton of mouse mast cell protease 7gene. Blood. In press.

52. Sato M, Morii E, Takebayashi-Suzuki K, Yasui N,Kitamura Y, Nomura S. Microphthalmia-associ-ated transcription factor interacts with PU.1 andc-Fos: determination of their subcellular localiza-tion. Biochem Biophys Res Commun. 1999;254:384-387.

GENE EXPRESSION IN NATURAL KILLER CELLS 2083BLOOD, 1 APRIL 2001 z VOLUME 97, NUMBER 7

For personal use only. by guest on June 4, 2013. bloodjournal.hematologylibrary.orgFrom