IL-2 and IL-12 Alter NK Cell Responsiveness to IFN- Γ-Inducible Protein 10 by Down-Regulating CXCR3 Expression

IL-2 and IL-12 Alter NK Cell Responsiveness to IFN- γ-Inducible Protein 10 by Down-Regulating CXCR3 Expression

This information is current as Deborah L. Hodge, William B. Schill, Ji Ming Wang, Isaac of September 29, 2021. Blanca, Della A. Reynolds, John R. Ortaldo and Howard A. Young J Immunol 2002; 168:6090-6098; ; doi: 10.4049/jimmunol.168.12.6090

http://www.jimmunol.org/content/168/12/6090 Downloaded from

References This article cites 37 articles, 22 of which you can access for free at: http://www.jimmunol.org/content/168/12/6090.full#ref-list-1 http://www.jimmunol.org/

Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication by guest on September 29, 2021

*average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2002 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

IL-2 and IL-12 Alter NK Cell Responsiveness to IFN-␥-Inducible Protein 10 by Down-Regulating CXCR3 Expression1

Deborah L. Hodge,* William B. Schill,‡ Ji Ming Wang,† Isaac Blanca,* Della A. Reynolds,* John R. Ortaldo,* and Howard A. Young2*

Cytokine treatment of NK cells results in alterations in multiple cellular responses that include cytotoxicity, cytokine production, proliferation, and chemotaxis. To understand the molecular mechanisms underlying these responses, microarray analysis was performed and the resulting gene expression patterns were compared between unstimulated, IL-2, IL-2 plus IL-12, and IL-2 plus IL-18-stimulated NK92 cells. RNase protection assays and RT-PCR confirmed microarray predictions for changes in mRNA expression for nine genes involved in cell cycle progression, signal transduction, transcriptional activation, and chemotaxis. Multiprobe RNase protection assay also detected changes in the expression of CCR2 mRNA, a gene that was not imprinted on the microarray. We subsequently expanded our search for other chemokine receptor genes absent from the microarray and found an Downloaded from IL-2- and IL-12-dependent decrease in CXCR3 receptor mRNA expression in NK92 cells. A detailed analysis of CXCR3 expression in primary NK cells revealed that an IL-2 and an IL-12 together significantly decreased the CXCR3 receptor mRNA and receptor surface expression by 6 and 24 h of treatment, respectively. This decrease in receptor expression was associated with a significant reduction in chemotaxis in the presence of IFN-␥-inducible protein-10. The decline in CXCR3 mRNA was due to transcriptional and posttranscriptional mechanisms as the addition of actinomycin D to IL-2- and IL-12-treated NK92 slightly altered the half-life of the CXCR3 mRNA. Collectively, these data suggest that IL-2 and IL-12 directly affect NK cell migratory http://www.jimmunol.org/ ability by rapid and direct down-regulation of chemokine receptor mRNA expression. The Journal of Immunology, 2002, 168: 6090–6098.

atural killer (NK) cells are large granular lymphocytes predicting patterns of gene expression in tumor cells (7, 8). To that play an important role in the defense against virally examine gene expression in response to cytokine stimulation, a N infected or malignant cells (1). Their activity can be human NK cell line, NK92, was stimulated with IL-2 alone or in characterized as nonadaptive and independent of MHC restriction combination with IL-12 or IL-18. These cytokines were chosen

(1, 2). A variety of NK cell functions including cytotoxicity, pro- because of their ability to induce NK cell responses; however, little by guest on September 29, 2021 liferation, chemotaxis, and cytokine production are modulated by is known about the repertoire of genes that are activated by these regulatory cytokines including IFN-␣␤, IL-2, IL-12, IL-18, IL-10, cytokines. Microarray analysis of gene expression in NK92 cells and TNF (reviewed in Refs. 3 and 4). Because cytokines induce identified a variety of genes whose mRNA expression patterns such a broad range of effects in NK cells, the potential for alter- change in response to cytokine stimulation. The genes encoding ations in gene expression in stimulated cells is very great. To de- the mRNAs are not specific to any one pathway; however, changes termine which genes are regulated in response to cytokine stimu- in cytokine, chemokine, and chemokine receptor gene mRNAs lation, our laboratory has used cDNA microarray technology to were prevalent. Our mRNA studies on chemokine receptor gene examine gene expression in NK cells. Microarray technology is expression were extended to cell surface analysis of receptor den- very useful because it allows for large-scale examination of gene sities in cytokine-treated primary NK cells. Using FACS analysis, expression. Additionally, this technology has proved useful in we observed a significant decrease in CXCR3 receptor expression identifying physiologically relevant gene expression patterns in in NK cells treated for 24 h with IL-2 and IL-12 alone or in com- eukaryotic systems such as yeast (5) and fibroblasts (6) as well as bination. Recently, alterations in chemokine receptor expression were reported in IL-2-stimulated NK cells (9); however, the cells were cultured in IL-2 for 8Ð10 days. In contrast, our data demon- Laboratories of *Experimental Immunology and †Molecular Immunoregulation, Cen- strate that cytokines can modify chemokine receptor function ter for Cancer Research, National Cancer Institute, Frederick, MD 21702; and ‡Na- tional Fish Health Research Laboratory, U.S. Geological Survey-Leetown Science within hours, thus supporting a model whereby cytokines, in par- Center, Kearneysville, WV 25430 ticular IL-2 and IL-12, regulate chemokine receptor expression in Received for publication September 14, 2001. Accepted for publication April 5, 2002. a direct, rapid, and novel manner. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Materials and Methods 1 This project has been funded in whole or in part with Federal funds from the Na- NK cell preparation tional Cancer Institute, National Institutes of Health under Contract No. N01-CO- 12400. The publisher or recipient acknowledges the right of the U. S. Government to PBMC were isolated from buffy coats of healthy donors (obtained from the retain a nonexclusive, royalty-free license in and to copyright covering the article. The National Institutes of Health Blood Bank, Bethesda, MD) after centrifu- content of this publication does not necessarily reflect the views or policies of the gation on a lymphocyte separation medium. Cells were washed twice with Department of Health and Human Services, nor does mention of trade names, com- Dulbecco’s PBS and suspended in RPMI 1640 medium supplemented with mercial products, or organizations imply endorsement by the U. S. Government. 2mML-glutamine, 100 IU/ml penicillin, 50 ␮g/ml streptomycin, and 10% 2 Address correspondence and reprint requests to Dr. Howard A. Young, National FCS. Adherent cells were removed by incubation in plastic flasks for 1 h Cancer Institute, Building 560, Room 31-23, Frederick, MD 21702-1201. E-mail at 37¡C. Nonadherent cells were recovered by gently washing with warmed address: [email protected] medium and were further purified by incubating on nylon wool columns for

1 h at 37¡C. The nylon-nonadherent cells (mostly T cells and NK cells) Ϫ70¡C for 30 min, and subjected to centrifugation at 14,000 rpm for 15 were eluted with prewarmed RPMI-640 medium and fractionated on a min in a room temperature microcentrifuge. The supernatants were de- seven-step Percoll gradient as previously described (10). The NK cell- canted, a sterile cotton swap was used to remove excess liquid, and the enriched low-density fraction-2 (40Ð60% NK cells) was further depleted pellet was resuspended in 3 ␮l of BD PharMingen sample buffer. Each of remaining T lymphocytes and monocytes by negative selection with multiprobe RNase assay included probes specific for ribosomal L32 and anti-CD3 and anti-CD14 mAbs. Briefly, the cells were labeled for 30 min GAPDH mRNAs to assure equal amounts of input mRNA in each assay on ice with biotinylated anti-CD3 and anti-CD14 Abs. After removing the and to control for lane to lane variation during PAGE. unbound Abs by washing with cold PBS plus 1% BSA, the cells were For guanylate binding protein 1 (GBP-1)-1 and Src homology 2 domain- incubated 15 min with streptavidin microbeads (Miltenyi Biotec, Oslo, containing leukocyte protein of 76 kD (SLP-76) mRNA analysis, RPA Norway) and the positive cells (CD3ϩ and CD14ϩ) were removed with a probes were synthesized from plasmid DNA templates using T7 RNA magnetic column (MACS; Miltenyi Biotec). Purified NK cell populations polymerase and [␣-33P]UTP in an in vitro transcription reaction. The newly were Ͼ95% CD56ϩ/CD5Ϫ cells as determined by two-color flow cytom- synthesized riboprobes were loaded onto a 6% denaturing polyacrylamide etry analysis (FACSort; BD Biosciences, San Jose, CA) with anti-CD56 PE gel and full-length probes were excised and eluted from the gel by over- and anti-CD5 FITC (BD Biosciences). night incubation at 37¡C in gel elution buffer (supplied in RPA II kit; Ambion). For each RPA, ϳ100 ϫ 104 cpm and 50 ϫ 104 cpm of the Cell culture gene-specific and 18S rRNA probes were used, respectively. RPAs for GBP-1 and SLP-76 were performed according to the manufacturer’s pro- NK92 cells were maintained in RPMI 1640 medium (BioWhittaker, Walk- tocol using a RPA II RNase protection assay kit (Ambion) with 10 ␮gof ersville, MD), supplemented with 10% FCS, 2 mM L-glutamine, 100 U/ml input RNA per reaction. For all multiprobe and single-probe RPAs, pro- penicillin, 100 ␮g/ml streptomycin, 200 U/ml recombinant human IL-2, tected RNA products were separated by size on a 6% denaturing poly- and 10 ng/ml recombinant human IL-15. Cells were cultured at a density ϳ ϫ 5 acrylamide gel. Gels were dried under vacuum at 80¡C for 2 h and of 5 10 /ml in a 37¡C incubator with 5% CO2. For all experiments, cells ϫ 6 exposed to either x-ray film or a PhosphorImager screen for 16 to 24 h. were grown at a density of 1 10 /ml in medium lacking IL-2 and IL-15 Downloaded from Film images were scanned and signal intensities were quantitated using for 12 h before cytokine stimulation. TotalLab image analysis software (Phoretix, Newcastle Upon Tyne, U.K.). Freshly isolated primary NK cells were placed into warm RPMI 1640 Phosphorimages were generated and quantitated using PhosphorImager SI medium, supplemented with 10% FCS, 2 mM L-glutamine, 100 U/ml pen- analysis and ImageQuant software (Molecular Dynamics, Sunnyvale, CA). icillin, and 100 ␮g/ml streptomycin and immediately treated with 100 U/ml IL-2, 10 U/ml IL-12, or 10 ng/ml IL-18, alone or in combination. mRNA half-life measurement ϩ Poly(A) RNA preparation, microarrays, and data analysis Rested NK92 cells were treated with IL-2 (100 U/ml) and IL-12 (10 U/ml) alone or in combination in the presence of actinomycin D (5 ␮g/ml). Total http://www.jimmunol.org/ Approximately 100 ϫ 106 NK92 cells were stimulated for 3 h with 100 RNA was isolated at multiple times and CXCR3 mRNA expression was U/ml IL-2 alone or in combination with 10 U/ml IL-12 or 10 ng/ml IL-18. 33 ϩ measured by multiprobe RPA using [␣- P]UTP-labeled riboprobes tran- Poly(A) RNA was isolated according to the manufacturer’s protocol us- scribed from a hCR-6 template (BD PharMingen). The protected mRNA ing a FastTrack 2.0 mRNA isolation kit (Invitrogen, Carlsbad, CA). Incyte fragments were size separated on a 6% polyacrylamide gel under denatur- Genomics (St. Louis, MO) performed the cDNA generation, hybridization, ing conditions. The RPA gels were dried and exposed to a PhosphorImager and data collection. Alterations in gene expression were evaluated by re- ϩ screen for 12Ð24 h. Images were visualized and quantitated using Phos- verse transcription of poly(A) RNAs in the presence of Cy3 or Cy5 flu- phorImager SI analysis and ImageQuant software (Molecular Dynamics). orescent labeling dyes followed by hybridization to UniGEM human V microarray chips (Incyte Genomics). Each chip contains a total of 7075 Flow cytometry analysis elements of which 6794 are unique genes/clusters. These unique genes/ clusters can be further defined as 4610 annotated and 2184 unannotated To examine CXCR expression, primary NK cells were stained with FITC- by guest on September 29, 2021 sequences. Subsets of genes were selected based on differential Cy3/Cy5 conjugated CXCR3 or CXCR4 mAbs (R&D Systems, Minneapolis, MN) expression ratios that were Ն͉2͉ in response to any treatment. Differential and PE-conjugated CD56 mAb (BD Biosciences). For CCR expression, expression of representative selected genes was confirmed by RT-PCR NK cells were stained with PE-labeled CCR1 or CCR2 mAbs (R&D Sys- and/or RNase protection assay (RPA).3 tems) and FITC-conjugated CD16 mAb (BD Biosciences). Cells were also stained using isotype controls FITC-conjugated IgG1 and PE-conjugated RNA isolation, relative quantitative RT-PCR, RPA IgG2A Abs (BD Biosciences). All cells were analyzed using a FACSort To confirm microarray predictions, ϳ10 ϫ 106 NK92 cells were rested flow cytometer (BD Biosciences). overnight without IL-2 and IL-15. The cells were either nonstimulated or Chemotaxis assay stimulated with IL-2 (100 U/ml), IL-12 (10 U/ml), and IL-18 (10 ng/ml) alone or in combination for 3 h. Total RNA was isolated from the untreated NK cell migration was assessed with 48-well microchemotaxis chambers and cytokine-treated cells using the TRIzol (Life Technologies, Bethesda, (NeuroProbe, Cabin John, MD). The chemoattractants were placed in the MD) extraction procedure. For relative quantitative RT-PCR, ϳ5 ␮gof wells of the lower compartment. The cells (at 3 ϫ 106/ml) were placed in RNA was reverse transcribed using a Thermoscript RT-PCR system (Life the wells of the upper compartment. A 5-␮m pore-sized polycarbonate Technologies). The cDNA template was generated according to the man- filter precoated with fibronectin separated the two compartments. After ufacturer’s protocol with random hexamers as primers for initiation of incubation at 37¡Cfor3h,thefilter was removed, stained with DiffQuik, reverse transcription. Multiplex PCR was performed using Platinum PCR and the cells migrated across the filter were counted under microscopy with Supermix (Life Technologies) and primers specific for the human 18S the samples coded. The cell migration was tested in triplicates and the rRNA gene and also the gene of interest. The amplification efficiency of the chemotaxis index represents the fold-increase of cell migration in response 18S rRNA gene was modulated using an 18S Competitimer technology to chemokines over medium control. The statistical significance was ex- (Ambion, Austin, TX). This allowed the 18S gene to be expressed in the amined with the Student t test. same linear range as the gene of interest when amplified. The amount of PCR product for the gene of interest could then be compared with the amount of 18S product to estimate the amount of variation between Results samples. Cytokine-induced alterations in gene expression The multiprobe RPAs were performed according to the manufacturer’s Microarray experiments were performed to examine changes in directions (BD PharMingen, San Diego, CA) with the following modifi- cations: RNase inactivation and precipitation was performed using a master gene expression patterns in NK cells in response to cytokine ac- mixture containing 200 ␮l Ambion RNase inactivation reagent, 50 ␮l eth- tivation. In one experiment, changes in mRNA expression in rest- anol, 5 ␮g yeast tRNA, and 1 ␮l Ambion GycoBlue coprecipitate per RNA ing cells were compared with those observed in IL-2-activated NK sample. After adding the individual RNase-treated samples to 250 ␮lofthe cells. In a second experiment, gene expression patterns were com- inactivation/precipitation mixture, the samples were mixed well, placed at pared in NK cells under two different stimulatory conditions. For this, mRNA from IL-2 plus IL-12-stimulated cells was compared 3 Abbreviations used in this paper: RPA, RNase protection assay; GBP-1, guanylate with mRNA from IL-2 plus IL-18-stimulated cells. In all experi- binding protein 1; SLP-76, Src homology 2 domain-containing leukocyte protein of ϩ 76 kD; MIP, macrophage-inflammatory protein; MX1, myxovirus resistance 1; PI3K, ments, cells were stimulated for 3 h and poly(A) mRNA was phosphoinositide-3-kinase; IP-10, IFN-␥-inducible protein-10. isolated and evaluated by microarray analysis. 6092 CYTOKINE-INDUCED CHANGES IN NK CELL GENE EXPRESSION

Stimulation of NK cells by IL-2 resulted in identification of a lation of the chemokine receptor CCR1 by IL-2 (Fig. 1). Addi- total of 65 genes with greater than 2-fold changes in expression tionally, RPA analysis demonstrated a slight up-regulation of (Table I). Of these, the majority of the genes (n ϭ 49) were up- CCR2 mRNA expression by IL-2 plus IL-12. This gene was not regulated by IL-2. The identified genes were involved in a variety imprinted on the microarray and was examined only as a result of of cellular processes that included transcriptional activation, inter- our usage of the multiprobe RPA. As a result of this observation, mediary metabolism, signal transduction, and cell cycle regulation. we extended our study to other chemokine receptor genes that Microarray analysis of mRNA from IL-2 plus IL-12 and IL-2 plus were absent on the microarray. Using a CXCR-specific RPA tem- IL-18-stimulated NK92 cells did not reveal a large of amount of plate, we found that CXCR3 mRNA expression was down-regu- differential regulation. Of the 17 genes identified, most (n ϭ 10) lated 13% by IL-2. The combination of IL-2 with IL-12 or IL-18 were up-regulated by IL-2 plus IL-12 as compared with IL-2 plus further enhanced this effect with decreases of 38 and 56%, respec- IL-18 (Table II). We were surprised at the overall lack of differ- tively (Fig. 1). In contrast, CXCR4 mRNA expression was up- ential gene expression between cells treated with IL-12 and IL-18. regulated ϳ1.2-fold by all cytokine treatments (Fig. 1), suggesting Many cytokines bind receptors that have common subunit chains that effects on this gene were IL-2-specific. that are shared with other receptors. The commonality between Although our microarray data did not contain information per- receptors may attribute to the activation of similar signaling path- taining to chemokines, changes in cytokine and chemokine recep- ways by different cytokines. IL-12 and IL-18 receptors, however, tor mRNA expression lead us to examine changes in these impor- are structurally very different and do not share common receptor tant chemotactic-inducing proteins. Consistent with a previous chains. Thus, their cellular target genes would potentially be very report (20), multiprobe RPA analysis demonstrated an IL-2 induc- different; however, the converse is true. These data suggest that tion of macrophage-inflammatory protein (MIP)-1␣ mRNA. The Downloaded from IL-12 and IL-18 signaling pathways converge at the nuclear level inability of IL-12 to further enhance the MIP-1␣ mRNA expres- to activate similar subsets of genes. sion indicated that IL-2 was the primary inducer of this mRNA. This was further supported by only a slight enhancement in Predicted activation of genes confirmed by relative quantitative MIP-1␣ mRNA expression in IL-2 plus IL-18-treated NK92 cells. RT-PCR and RPA In contrast to MIP-1␣, MIP-1␤ mRNA expression was not affected

Genes were selected from the microarray results for further anal- by IL-2. However, IL-18 in combination with IL-2 did up-regulate http://www.jimmunol.org/ ysis based on their predicted expression change or their potential to expression of this chemokine. affect NK cell biology. These genes were directly tested for We next examined mRNA expression of GBP-1 and myxovirus changes in mRNA expression by RPA or relative quantitative RT- resistance 1 (MX1) genes. Both the GBP-1 and MX1 genes code for PCR. Multiprobe and single-probe RPA were performed on RNA IFN-inducible GTP-binding proteins that protect cells against viral isolated from unstimulated NK92 cells as well as cells treated with infection (21Ð23). Because NK cells are primary defenders against IL-2 alone or in combination with IL-12 or IL-18. Predicted viral invaders, we speculated that these proteins might provide changes in gene expression in three primary families of genes were antiviral protection to NK cells during virus encounters. Our mi- confirmed by multiprobe RPA. These families represent genes that croarray results suggested that IL-2 should specifically up-regulate code for cytokine, chemokine, and chemokine receptors. mRNA expression of both the GBP-1 and MX1 genes. RPA probes by guest on September 29, 2021 To ensure that our cytokine treatments properly induced known were designed and used for RPA analysis of these mRNAs. Con- cellular functions, an IFN-␥-specific probe was used as a positive sistent with the microarray results, we found that IL-2 up-regulated control in RPA and RT-PCR analyses. Changes in IFN-␥ mRNA expression of the GBP-1 mRNA in NK92 cells (Fig. 2). This ex- expression are directly linked to cytokine stimulation (11Ð13) and pression was specific to IL-2 in that costimulation with IL-12 or thus, are useful measurements of cytokine-induced changes in IL-18 had little or no effect on GBP-1 mRNA expression. Unlike gene expression. Consistent with previous reports (11, 14, 15), the GBP-1, MX1 mRNA expression did not correlate with the mi- RPA result demonstrated cytokine-induced increases in IFN-␥ croarray results. RPA analysis revealed that NK92 cells did ex- mRNA by IL-2 alone (Fig. 1). Moreover, the expected synergy press MX1 mRNA; however, the expression was constitutive and between IL-2 and IL-12 or IL-18 on IFN-␥ mRNA expression was cytokine-independent (Fig. 2). The lack of inducibility of the MX1 observed. These data demonstrated that NK92 cells responded to gene demonstrates that microarray results, like those from any sin- cytokine treatment in an expected manner thus providing assur- gle method, can be incorrect, thus necessitating independent con- ance that the microarray results accurately predicted changes in formation of gene expression changes by alternative methods. mRNA expression of other genes. The microarray results also predicted that cytokine treatments of For examining changes in IFN-␥ mRNA, we used a multiprobe NK cells could alter mRNA expression of genes that code for template (hCK1) which also contains an IL-10-specific probe. This signal transduction proteins. Two such genes were the p85 subunit was useful because our microarray results predicted a greater than of phoshoinositide-3-kinase (PI3K) and the lymphocyte cytosolic 2-fold increase in IL-10 mRNA in response to IL-2. Consistent protein 2/SLP-76. RPA analysis of PI3K mRNA expression dem- with previous reports (16, 17, 18), RPA analysis demonstrated that onstrated an IL-2-dependent decrease in PI3K mRNA (Fig. 3). The IL-10 gene expression was altered by cytokine stimulation of NK IL-2 effect was enhanced by costimulation with IL-12 and IL-18, cells (Fig. 1). Interestingly, the patterns of IL-10 and IFN-␥ mRNA thus implicating these cytokines as coregulators of PI3K mRNA expression were parallel with respect to IL-2 alone and in combi- expression. In contrast, SLP-76 mRNA expression was signifi- nation with IL-12 or IL-18. We do not know why these two an- cantly increased by IL-2 (Fig. 2). IL-2 appeared to be the primary tagonistic cytokines are coexpressed in a similar fashion in NK regulator of SLP-76 mRNA in that no increase in mRNA expres- cells; however, it is possible that the IL-10 may act as a negative sion was observed in the IL-2 plus IL-12-treated NK cells. Addi- regulator to shutdown IFN-␥ expression. This is supported by a tionally, costimulation by IL-2 and IL-18 only slightly up-regu- report that demonstrates that IL-10 secretion from Th2 cells acts to lated the SLP-76 mRNA expression over that observed with IL-2 inhibit IFN-␥ production in both Th1 cells and NK cells (19). alone. We next used RPA analysis to examine chemokine receptor Relative-quantitative RT-PCR corroborated the predicted mRNA expression in NK92 cells. Using a CCR-specific multi- changes in the expression of five genes in response to IL-2 (Fig. 3). probe RPA, we confirmed the microarray prediction for up-regu- Of these, pim-1, cyclin D2, and c-myc genes code for proteins that The Journal of Immunology 6093

Table I. Analysis of gene expression in nonstimulated and IL-2Ðstimulated NK92 cells

Gene Name GenBank Accession Number Nonstimulated Signal IL-2ÐStimulated Signal Fold Change

Genes downÐregulated by IL-2 EST AI380052 17,963 6,861 Ϫ2.62 AHNAK nucleoprotein (desmoyokin) M80899 5,022 2,085 Ϫ2.41 Homo sapiens nuclear Ag H731-like protein U96628 4,810 2,287 Ϫ2.1 IL-9R M84747 3,902 1,415 Ϫ2.76 V-yes-1 Yamaguchi sarcoma viralÐrelated oncogene M16038 2,562 1,233 Ϫ2.08 homolog IL-16 (lymphocyte chemoattractant factor) AI652705 2,540 1,235 Ϫ2.06 DB1 AA195154 2,492 1,050 Ϫ2.37 Conserved gene ampliﬁed in osteosarcoma AF022231 2,162 1,043 Ϫ2.07 5Ј fragment, not mapped (Incyte PD: 194162) N/Aa 1,668 713 Ϫ2.34 PI3K, regulatory subunit, polypeptide 1 (p85 ␣) M61906 1,275 398 Ϫ3.2 Fucosidase, ␣-L-1 M80815 1,122 318 Ϫ3.53 Human EV12 protein gene M55267 800 394 Ϫ2.03 EST, highly similar to HMG box containing protein 1 H86395 729 364 Ϫ2 B cell translocation gene 1, anti-proliferative AI925293 654 234 Ϫ2.79 EST AI436055 370 159 Ϫ2.33 Sortilin-related receptor, L(DLR class) Y08110 265 126 Ϫ2.1 Genes up-regulated by IL-2 Chemokine (C-C motif) receptor 1 D10925 4,456 22,840 5.13 Downloaded from EST, moderately similar to heat shock protein HSP 90-␤ AL040289 7,070 15,715 2.22 Human pim-2 protooncogene homolog pim-2h AI634033 4,062 13,916 3.43 V-myc avian myelocytomatosis viral oncogene homolog V00568 4,229 13,642 3.23 Human serine protease gene J02907 5,694 12,758 2.24 Lactate dehydrogenase B AL044172 4,993 12,238 2.45 Solute carrier family 7 (cationic amino acid transporter, AJ130718 2,242 11,686 5.21 yϩ system), member 7

Differentiated embryo chondrocyte expressed gene 1 AB004066 3,309 9,317 2.82 http://www.jimmunol.org/ Lymphocyte cytosolic protein 2 U20158 4,184 9,299 2.22 IL-10 M57627 3,373 7,861 2.33 Pim-1 oncogene M54915 1,533 7,829 5.11 GBP 1, interferon-inducible M55542 3,229 7,274 2.25 Cyclin D2 D13639 1,918 6,180 3.22 CSF 1 (macrophage) M11296 2,691 5,753 2.14 Perforin 1 (preforming protein) AI076019 2,618 5,736 2.19 Tryptophanyl-tRNA synthetase X59892 2,247 5,730 2.55 Ornithine decarboxylase 1 M16650 1,806 5,380 2.98 Putative translation initiation factor (Incyte PD: 58399) N/A 2,478 5,344 2.16

MX1 (influenza), homolog of murine (IFN-inducible M30817 1,645 5,275 3.21 by guest on September 29, 2021 protein p78) IFN regulatory factor 1 X14454 2,385 5,254 2.2 FLN29 gene product AB007447 1,854 5,040 2.72 Dyskeratosis congenita 1, dyskerin U59151 2,068 4,873 2.36 Solute carrier family 1 (neutral amino acid transporter), U53347 2,031 4,818 2.37 member 5 5Ј not mapped fragment (Incyte PD: 2498667) N/A 1,456 4,625 3.18 EST AI127628 1,839 4,381 2.38 Dual specificity phosphatase 5 U15932 1,735 4,286 2.47 Paired basic amino acid cleaving enzyme (furin, X17094 1,450 3,341 2.3 membrane associated receptor protein) EST AI188513 1,165 3,335 2.86 EST AA559093 695 3,260 4.69 Nucleolar phosphoprotein p130 Z34289 1,116 2,929 2.62 Human ras-like protein AA459547 1,109 2,914 2.63 Pleckstrin homology, Sec7 and coiled/coil domains, AI538459 745 2,546 3.42 binding protein Topoisomerase (DNA) I U07806 911 2,138 2.35 Flotillin 1 AA618098 850 1,915 2.25 Protein tyrosine phosphatase type IVA, member 1 U48296 912 1,904 2.09 Carcinoembryonic Ag gene family member 6 X52378 654 1,847 2.82 B cell CLL/lymphoma 2 M14745 789 1,691 2.14 Serine/threonine kinase 17b (apoptosis-inducing) AA725600 783 1,580 2.02 Bystin-like L36720 665 1,437 2.16 H. sapiens mRNA expressed in osteoblast F12860 459 1,393 3.03 Pregnancy specific ␤-1-glycoprotein 7 (Incyte PD: N/A 551 1,389 2.52 64457) Methionine adenosyltransferase II, ␣ F07456 592 1,342 2.27 Glutamyl aminopeptidase (aminopeptidase A) L12468 535 1,231 2.3 Cytochrome P450, 51 (lanosterol 14-␣-demethylase) U23942 601 1,227 2.04 Low density lipoprotein receptor (familial L00352 559 1,121 2.01 hypercholesterolemia) Fatty-acid-coenzyme A ligase, very long-chain 1 D88308 460 1,012 2.2 H. sapiens DKFZp564L176 (from clone D64110 353 862 2.44 DKFZp564L176) Regulator of G-protein signaling 16 U70426 327 778 2.38 IL 6 signal transducer (gp 130, oncostatin M receptor) M57230 348 709 2.04

a N/A represents genes that do not have GenBank accession numbers. The Incyte PD identiﬁcation numbers are provided. Expression of genes listed in italics was evaluated by RPA or semi-quantitative RT-PCR analysis. 6094 CYTOKINE-INDUCED CHANGES IN NK CELL GENE EXPRESSION

Table II. Analysis of gene expression in IL-2 ϩ IL-12 and IL-2 ϩ IL-18Ðtreated NK92 cells

GenBank Accession IL-2 ϩ IL-12ÐStimulated IL-2 ϩ IL-18ÐStimulated Fold Gene Name Number Signal Signal Change

Genes downÐregulated by IL-2 ϩ IL-18 as compared to IL-2 ϩ IL-12 Ornithine decarboxylase 1 M16650 2,372 1,134 Ϫ2.09 Peptidylprolyl isomerase E (cyclophilin E) AF042386 810 320 Ϫ2.53 GTP-binding protein overexpressed in skeletal U10550 719 134 Ϫ5.37 muscle PGER2 (subtype EP2) U19487 477 225 Ϫ2.12 EST highly similar to DNAJ protein homolog AI337322 236 115 Ϫ2.05 MTJ1 (M. musculus) EST AI890347 145 68 Ϫ2.13 Hyaluronan-mediated motility receptor (RHAMM) U29343 128 58 Ϫ2.21 EST R55801 122 55 Ϫ2.22 EST AA868888 113 37 Ϫ3.05 V-maf musculoaponeurotic ﬁbrosarcoma (avian) AF059195 108 46 Ϫ2.35 oncogene family, protein G Genes up-regulated by IL-2 ϩ IL-18 as compared to IL-2 ϩ IL-12 Apoptosis inhibitor 2 U37546 724 2,466 3.41 Downloaded from NF of ␬ light polypeptide gene enhancer in B cell M69043 546 1,752 3.21 inhibitor, ␣ Discoidin domain receptor family, member 1 Z29093 663 1,334 2.01 Carboxypeptidase D D85390 427 1,007 2.36 Regulator of G-protein signaling 16 U70426 327 674 2.06 Hypoxia-inducible factor 1, ␣ subunit (basic helix- U22431 239 536 2.24

loop-helix transcription factor) http://www.jimmunol.org/ Apoptosis inhibitor 1 U37547 181 477 2.64 are involved in cell cycle progression. Additionally, pim-1 and whether these changes were functional in primary NK cells, we c-myc have been reported to act synergistically to prevent apopto- examined the potential for cytokine-induced alterations in chemo- sis (24) and thus may play a novel role in protecting NK cells from kine receptor expression. Recent studies have demonstrated the stress-induced cellular death. presence of chemokine receptors on NK cells (9, 25Ð27); however, The NK92 cell line is a good model for studying changes in no studies have addressed alterations in these receptors in response cytokine gene expression; however, it is cytokine-dependent and to short-term activation by IL-2 alone and in combination with by guest on September 29, 2021 may not entirely reﬂect changes in primary NK cells. To address IL-12 or IL-18. this concern, we compared alterations in cell cycle gene expression To examine changes in chemokine receptor density, freshly iso- in primary and NK92 cells. Overnight depletion of IL-2 from lated NK cells that were greater than 95% CD56ϩ/CD5Ϫ by FACS NK92 cells might result in changes in cell cycle gene expression analysis, were placed into culture and stimulated with IL-2, IL-12, that could be dramatically different from primary NK cells that and IL-18 alone. Additionally, cells were stimulated with IL-2 in have not been IL-2-depleted in culture. combination with IL-12 or IL-18. After 24 h of cytokine stimula- The results from RT-PCR and RPA analysis of RNA from tion, alterations in CCR1, CCR2, CXCR3, and CXCR4 densities NK92 and primary NK cells, respectively, demonstrated that both were evaluated by FACS analysis. Expression of CCR1 and CCR2 cell types had large increases in cyclin D2 mRNA expression after was virtually absent in all treatments (Table III). The lack of CCR1 3 h of stimulation by all cytokines (Figs. 3 and 4A). Interestingly, expression was in direct contrast to CCR1 mRNA expression in the primary NK cells exhibited a more robust response to IL-2 and NK92 cells. We considered the likelihood that there were differ- IL-18 than did the NK92 cells. Moreover, cyclin D3 mRNA ex- ences in the expression of this receptor between the NK92 cell line pression was slightly up-regulated in primary NK cells. This effect and primary cells; however, it has been recently reported that pri- was not seen in NK92 cells (data not shown). mary NK cells express CCR1 mRNA although surface expression We continued our study of cell cycle-related genes by examin- of the receptor is undetectable (9). ing c-myc and pim-1 mRNA expression in primary NK cells. RT- Examination of CXCR3 surface expression revealed dramatic PCR analysis revealed that all cytokine treatments could induce decreases in receptor density in response to cytokine treatment. In c-myc and pim-1 mRNA levels in primary NK cells (Fig. 4B). cells treated with IL-2 and IL-12 alone, there was a 28 and 26% These observations were in good agreement with RT-PCR results reduction in CXCR3 expression, respectively (Table III and Fig. from NK92 cells. Collectively, these data indicate that regulation 5). The combination of these cytokines was synergistic with a 68% of NK92 cells is similar in many ways to primary NK cells with decrease in receptor density. IL-18 alone or in combination with regard to IL-2, IL-2 plus IL-12, and IL-2 plus IL-18 treatments. IL-2 had no effect on CXCR3 expression. There was general Thus, NK92 cells appear to be a reliable model to predict global agreement between CXCR3 expression in primary NK cells and changes in gene expression in primary NK cells. mRNA expression in NK92 cells with regard to IL-2 and IL-12 regulation of this receptor. In contrast, RPA analysis of CXCR3 Cytokine-induced changes in chemokine receptor density on mRNA expression in NK92 cells predicted that the maximum de- primary NK cells crease in receptor density should result from IL-2 plus IL-18 treat- RPA analysis demonstrated that IL-2 alone and in combination ment. This prediction was not substantiated in primary NK cells with IL-12 or IL-18 had a large impact on cytokine, chemokine, and may reﬂect differences in the manner by which primary NK and chemokine receptor expression in NK92 cells. To determine cells and NK92 respond to IL-2 plus IL-18 with respect to RNA The Journal of Immunology 6095 Downloaded from

FIGURE 1. Multiprobe RNase protection analysis of gene expression in NK92 cells. NK92 cells were rested overnight in RPMI medium without IL-2 and IL-15. Cells were either untreated (NS) or stimulated with IL-2 (100 U/ml) alone or in combination with IL-12 (10 U/ml) and IL-18 (10

ng/ml) for 3 h. Total RNA was isolated as described in Materials and http://www.jimmunol.org/ FIGURE 2. Single-probe RPA analysis of mRNA expression in NK Methods and ϳ10 ␮g of input RNA was used per hybridization. All RPAs cells. All probes were in vitro synthesized from gene specific sequences were performed by hybridizing RNA to hCK1, hCK5, hCR5, hCR6, or and gel purified before hybridization. The RPAs were performed according hAPO5 multiprobe templates (BD PharMingen). The exposure time to film to manufacturer specifications using a RPA II kit (Ambion). The gels were varied for individual RPAs and depended on the strength of the signal of exposed to film for a minimum of 24 h; however, exposure times varied for the target genes. Differences in exposure are reflected by differences in individual RPAs and depended on the strength of the signal of the target GAPDH signal intensities. The data are representative of two independent genes. The data are representative of two independent experiments. experiments.

human NK cells were isolated and changes in CXCR3 mRNA and protein expression of CXCR3. Finally, we examined changes were monitored for up to 12 h following IL-2 plus IL-12 addition. by guest on September 29, 2021 in CXCR4 receptor expression. Our RPA results suggested that In all donors tested, CXCR3 mRNA expression remained rela- there should be cytokine-independent expression of CXCR4 in NK tively unchanged for at least 3 h following cytokine treatment (Fig. cells. Flow cytometry analysis demonstrated that ϳ25% of pri- 7, A and B). However, by6hofIL-2 plus IL-12 treatment, there mary NK cells expressed CXCR4 (Table III). Moreover, receptor was a slight 20% reduction in CXCR3 mRNA accumulation. The density remained unchanged with respect to cytokine stimulation. CXCR3 mRNA continued to decrease rapidly with time until a total 80% drop in expression was seen after 12 h of cytokine treat- Alterations in chemotaxis in cytokine-treated primary NK cells ment. Interestingly, we found that IL-2 plus IL-12 did not nega- The decrease in CXCR3 density in IL-2- and IL-12-treated NK tively affect all chemokine receptors. The fractalkine receptor, cells predicted that these cells would have decreased migratory CX3CR1 is expressed on NK cells and is involved in NK cell- capacity in the presence of CXCR3-speciﬁc chemoattractants. To mediated endothelial cell injury (28). In contrast to CXCR3, IL-2 test this assumption, freshly isolated primary NK cells were treated plus IL-12 greatly enhanced CX3CR1 mRNA expression (Fig. 7A). with IL-2 and IL-12 alone or in combination for 24 h. Cells were By3hofIL-2 plus IL-12 treatment, CX3CR1 mRNA increased placed into microchemotaxis chambers and their migration poten- 3-fold. The increase in mRNA expression continued to rise with a tial was evaluated in the presence of CXCR3 ligand, IFN-␥-induc- 4-fold increase after6hofcytokine treatment. Moreover, this ible protein (IP)-10. At IP-10 concentrations of 100 ng/ml or increase in mRNA was sustained throughout the remainder of the greater, we observed a signiﬁcant reduction in chemotaxis in the time course. IL-2- and IL-12-treated cells (Fig. 6). Furthermore, the combina- To understand whether CXCR3 mRNA regulation was entirely tion of IL-2 and IL-12 resulted in a substantial decrease in migra- due to transcriptional changes, we determined the half-life of the tion that was greater than that observed in cells treated with the CXCR3 mRNA in the presence and absence of the transcriptional cytokines alone (Fig. 6). Collectively, these data demonstrate that inhibitor, actinomycin D. For these experiments, NK92 cells were IL-2 and IL-12 can directly alter the chemotactic function of NK treated with IL-2 plus IL-12 Ϯ actinomycin D and changes in cells through changes in chemokine receptor mRNA and protein CXCR3 mRNA were measured at various times posttreatment. As expression. IL-2 plus IL-12 reduces the quantity of CXCR3 mRNA, the amount of CXCR3 mRNA measured at all times was expressed as Down-regulation of CXCR3 mRNA expression a percentage of that observed at the initial time of cytokine addi- Decreases in CXCR3 density are evident by 24 h after IL-2 plus tion. In the absence of actinomycin D, NK92 cells responded to IL-12 treatment. Therefore, to understand the molecular mecha- IL-2 plus IL-12 with the expected reduction in CXCR3 mRNA nism responsible for these changes, we chose to examine the ki- (Fig. 8, A and B). Under these conditions, the CXCR3 mRNA netic pattern of CXCR3 mRNA expression at times immediately half-life was estimated to be 1.5 h (Fig. 8B). The addition of ac- following cytokine addition to NK cells. For these studies, primary tinomycin D slightly increased the half-life of the CXCR3 mRNA 6096 CYTOKINE-INDUCED CHANGES IN NK CELL GENE EXPRESSION

FIGURE 3. Relative-quantitative RT-PCR analysis of alterations in gene expression in NK92 cells. Approximately 5 ␮g of RNA was reverse transcribed using a Thermoscript RT-PCR system (Life Technol- ogies). The resulting cDNA template was used for multiplex PCR with gene-speciﬁc and human 18S rRNA gene primers. The ﬁnal PCR products were size-separated by agarose gel electrophoresis and visualized using ethidium bromide staining. The microarray values for the NS and IL-2-treated samples are listed to the left of each gene tested. No values are given for the IFN-␥ gene as it was not represented on the microarray chip. However, relative-quantitative RT-PCR analysis was performed on the IFN-␥ gene as a positive control for cytokine-induced changes in gene expression. Downloaded from

from 1.5 to 2 h (Fig. 8B). This slight increase in the mRNA half- mRNA half-life may be insigniﬁcant and other as yet unknown life was surprising in that mRNA stabilization, in the absence of factors may contribute to the down-regulation of CXCR3 http://www.jimmunol.org/ other factors, should lead to an overall increase in mRNA accu- expression. mulation. However, we found that IL-2 plus IL-12 act collectively to decrease CXCR3 expression. Although this small increase in Discussion CXCR3 mRNA half-life suggests that posttranscriptional effects NK cells play a pivotal role in protecting the body against infec- may contribute to the regulation of CXCR3 expression, it should tious agents and cancers (1, 29, 30). Thus, cytokine-induced vari- be noted that the overall increase in mRNA half-life is small and ations in gene expression in NK cells may have profound effects on that the rate of CXCR3 mRNA decay is similar in the absence or NK cell biology and immune response. In this study, we have presence of actinomycin D. Thus, the overall increase in CXCR3

examined global changes in gene expression in response to IL-2 by guest on September 29, 2021 alone and in combination with IL-12 or IL-18 and found that 82 genes changed at least 2-fold in response to cytokine treatment. Collectively, alterations in mRNA expression were associated with genes whose protein products are involved with multiple cellular functions such as, cell cycle progression, signal transduction, transcriptional activation, and mRNA processing. However, the majority of genes that responded to cytokine treatment code for chemokines, chemokine receptors, and cytokines. Of the altered chemokine receptor genes identiﬁed, we chose to further investigate the effect of IL-2 and IL-12 on CXCR3 expression. Cellular distribution of CXCR3 is mainly localized to activated T cells, eosinophils, basophils, and NK cells but control of this expression is poorly understood. Alterations in CXCR3 receptor and mRNA expression have been studied in IL-2-treated T and NK cells (9, 31); however, cells were incubated in the presence of IL-2 for many days before mRNA analysis was made. This demonstrated the long-term effects of IL-2, but did not address rapid short-term effects of cytokines on CXCR3 expression.

Table III. FACS analysis of cytokine-induced changes in surface chemokine receptor expression in primary NK cellsa FIGURE 4. RPA and RT-PCR analysis of cell cycle-associated genes in primary human NK cells. Primary NK cells were stimulated with IL-2 (100 IL-2 IL-2 ϩ ϩ U/ml) alone or in combination with IL-12 (10 U/ml) or IL-18 (10 ng/ml) Chemokine Receptor NS IL-2 IL-12 IL-18 IL-12 IL-18 for 3 h. Total mRNA was isolated from the cells and mRNA expression was evaluated. A, mRNA expression of cyclins D2 and D3 were examined CCR1 0 0 0000 by RPA using a hCYC-1 multiprobe. GAPDH was included as a control for CCR2 0 0 0000 sample-to-sample variation. B, Relative-quantitative RT-PCR analysis c- CXCR3 81 58 60 83 26 55 myc and Pim-1 expression. 18S ribosomal RNA was used as an internal CXCR4 25 28 26 27 26 29 control in this assay. These results are representative of assays performed a These data represent one of two separate experiments. For both experiments, on RNA from two independent donors. freshly isolated NK cells were from individual donors. The Journal of Immunology 6097

FIGURE 5. Inhibition of CXCR3 expression by IL-2 and IL-12. Freshly isolated primary NK cells were stimulated with 100 U/ml IL-2, 10 U/ml IL-12, or 10 ng/ml IL-18, alone or in combination for 24 h and CXCR3 expression was compared with that observed in nontreated (NT) cells by FIGURE 7. Kinetics of CXCR3 mRNA accumulation in primary hu- ﬂow cytometry analysis. A, Receptor expression in cells treated with IL-2 man NK cells. A, Freshly isolated NK cells were treated with IL-2 (100 and IL-12 alone and in combination. B, Receptor expression in cells treated U/ml) plus IL-12 (10 U/ml) for the times indicated. Total mRNA was with IL-2 and IL-18 alone and in combination. In both A and B, cntl rep- isolated and CXCR3 and CX3CR1 mRNAs were measured by RPA anal- resents isotype control Ig staining. The data are from one representative ysis using a hCR6 multiprobe. GAPDH was included as a control for sam- experiment of two performed. Individual donors were used as the source of ple-to-sample variation. A 24-h exposure is shown for all mRNAs. A 12-h NK cells for each experiment. exposure (exp) of GAPDH mRNA is included to demonstrate sample and gel loading variations. B, Graphical representation of the percent change in CXCR3 mRNA shown in A. The amount of CXCR3 mRNA at time 0 was Our data show that IL-12 and IL-2 alone and in combination arbitrarily assigned as 100%. This experiment is representative of three Downloaded from effectively down-regulate CXCR3 cell surface expression in NK independent assays performed on three separate donors. cells. Furthermore, alterations in CXCR3 mRNA precede changes in protein expression suggesting that changes in CXCR3 surface rates between NK92 and primary NK cells did not affect the over- expression are due to rapid changes in mRNA accumulation and all outcome of CXCR3 down-regulation and may reﬂect differ- are not due to receptor internalization. Initially, we observed an

ences in cultured and primary cells. http://www.jimmunol.org/ IL-2 plus IL-12 mediated down-regulation of CXCR3 mRNA ex- To closely examine the mechanism of CXCR3 mRNA regula- pression in NK92 cells. This observation was extended to primary tion, we conducted CXCR3 mRNA half-life studies with actino- human NK cells. The kinetics of IL-2 plus IL-12 down-regulation mycin D to determine whether regulation was entirely due to tran- of CXCR3 mRNA in primary NK cells demonstrated that CXCR3 scriptional effects. We expected that the transcriptional blocker, mRNA expression was relatively constant after3hofcytokine actinomycin D, would either have no effect or would shorten the treatment; however, by 6 h, an ϳ25% decrease in mRNA accu- half-life of CXCR3 mRNA as an increase in CXCR3 mRNA half- mulation was seen. At times following, CXCR3 mRNA continued life would reﬂect a mRNA stabilization event that should ulti- to decline. The decline in CXCR3 mRNA was also observed in mately lead to an increase in mRNA accumulation. Surprisingly, NK92 cells but was more rapid with a signiﬁcant decline visual- we found that concurrent addition of actinomycin D, IL-2, and by guest on September 29, 2021 ized by3hofIL-2 plus IL-12 treatment. The difference in decay IL-12 to NK92 cells slightly increased the half-life of the CXCR3

FIGURE 8. Half-life of CXCR3 mRNA in IL-2 plus IL-12-treated FIGURE 6. Effect of IL-2 and IL-12 on chemotaxis of primary NK NK92 cells. A, NK92 cells were rested overnight without IL-2 and IL-15 cells. Freshly isolated primary NK cells were stimulated with 100 U/ml and then treated with IL-2 (100 U/ml), IL-12 (10 U/ml) Ϯ actinomycin D IL-2 and 10 U/ml IL-12, alone or in combination for 24 h. The NK cells (5 ␮g/ml) for the times indicated. Total mRNA was isolated from the cells were placed into the upper compartment of a microchemotaxis chamber and a multiprobe RPA analysis was performed with a hCR-6 probe set. while the chemoattractant, IP-10, was placed into the lower compartment GAPDH was included as an internal control for sample-to-sample varia- at the various concentrations indicated. The amount of chemotactic activity tion. The gel was exposed to a PhosphorImager screen for 24 h. B, Graph- was evaluated following a 4-h incubation. All results are expressed as the ical representation of the quantitation performed by ImageQuant analysis mean Ϯ SD for triplicate determinations. Significant changes in migration of the image shown in A. The amount of CXCR3 mRNA at time 0 was p Ͻ 0.05). arbitrarily set to 100%. The experiment is one of three separate assays, all ,ء) as compared with medium alone are indicated by an asterisk The data are from one representative experiment of two performed. with similar results. 6098 CYTOKINE-INDUCED CHANGES IN NK CELL GENE EXPRESSION mRNA suggesting that these cytokines have a stabilizing effect on 13. Ahn, H. J., S. Maruo, M. Tomura, J. Mu, T. Hamaoka, K. Nakanishi, S. Clark, the CXCR3 mRNA. These data demonstrate that IL-2 plus IL-12 M. Kurimoto, H. Okamura, and H. Fujiwara. 1997. A mechanism underlying synergy between IL-12 and IFN-␥-inducing factor in enhanced production of control of CXCR3 mRNA expression is not entirely at the tran- IFN-␥. J. Immunol. 159:2125. scriptional level and that as yet unknown posttranscriptional ef- 14. Ye, J., J. R. Ortaldo, K. Conlon, R. Winkler-Pickett, and H. A. Young. 1995. Cellular and molecular mechanisms of IFN-␥ production induced by IL- 2 and fectors must negate any stabilization effect of IL-2 plus IL-12 on IL-12 in a human NK cell line. J. Leukocyte Biol. 58:225. CXCR3 mRNA so that accumulation is ultimately reduced. 15. Bohn, E., A. Sing, R. Zumbihl, C. Bielfeldt, H. Okamura, M. Kurimoto, J. Heese- Overall, our data suggest that IL-2 and IL-12 alone or in com- mann, and I. B. Autenrieth. 1998. IL-18 (IFN-␥-inducing factor) regulates early cytokine production in, and promotes resolution of, bacterial infection in mice. bination have the ability to modulate immune function by altering J. Immunol. 160:299. CXCR3 receptor expression on NK cells. This is particularly im- 16. Mehrotra, P. T., R. P. Donnelly, S. Wong, H. Kanegane, A. Geremew, portant because NK cells are primary cellular defenders against H. S. Mostowski, K. Furuke, J. P. Siegel, and E. T. Bloom. 1998. Production of IL-10 by human natural killer cells stimulated with IL-2 and/or IL-12. J. Immu- viral infections and viral infections induce cellular release of po- nol. 160:2637. tent cytokines and chemokines. The chemokine MIP-1␣ is a potent 17. Lauwerys, B. R., N. Garot, J. C. Renauld, and F. A. Houssiau. 2000. Cytokine production and killer activity of NK/T-NK cells derived with IL-2, IL-15, or the inducer of NK cell chemotaxis (32, 33) and has been shown to combination of IL-12 and IL-18. J. Immunol. 165:1847. induce NK cell migration to livers of murine CMV-infected mice 18. D’Andrea, A., M. Aste-Amezaga, N. M. Valiante, X. Ma, M. Kubin, and (34). Once in the liver, NK cells produce IFN-␥ in an IL-12-de- G. Trinchieri. 1993. Interleukin 10 (IL-10) inhibits human lymphocyte interferon ␥-production by suppressing natural killer cell stimulatory factor/IL-12 synthesis pendent fashion that in turn up-regulates the expression of CXCR3 in accessory cells. J. Exp. Med. 178:1041. receptor ligands monocyte interferon-␥-inducible protein, IFN-␥- 19. Oliva, A., A. L. Kinter, M. Vaccarezza, A. Rubbert, A. Catanzaro, S. Moir, inducible T-cell ␣ chemoattractant, and IP-10 (34Ð37). Concurrent J. Monaco, L. Ehler, S. Mizell, R. Jackson, et al. 1998. Natural killer cells from human immunodeficiency virus (HIV)-infected individuals are an important with the IL-12 up-regulation of IFN-␥, it is possible that IL-12 source of CC-chemokines and suppress HIV-1 entry and replication in vitro. Downloaded from alone or in combination with IL-2 down-regulates CXCR3 expres- J. Clin. Invest. 102:223. 20. Chieux, V., W. Chehadeh, J. Harvey, O. Haller, P. Wattre, and D. Hober. 2001. sion on NK cells thus reducing NK cell chemotactic responsive- Inhibition of coxsackievirus B4 replication in stably transfected cells expressing ness to CXCR3 ligands. As monocyte interferon-␥-inducible pro- human MxA protein. Virology 283:84. tein, IFN-␥-inducible T-cell ␣ chemoattractant, and IP-10 can 21. Staeheli, P., O. Haller, W. Boll, J. Lindenmann, and C. Weissmann. 1986. Mx protein: constitutive expression in 3T3 cells transformed with cloned Mx cDNA promote the migration of activated Th1 lymphocytes to inflamma- confers selective resistance to influenza virus. Cell 44:147. tory sites, these chemokines may then initiate a secondary T cell- 22. Anderson, S. L., J. M. Carton, J. Lou, L. Xing, and B. Y. Rubin. 1999. Interferon- http://www.jimmunol.org/ mediated immune response. This model illustrates how IL-12 induced guanylate binding protein-1 (GBP-1) mediates an antiviral effect against vesicular stomatitis virus and encephalomyocarditis virus. Virology 256:8. alone and in combination with IL-2 may trigger both primary and 23. Shirogane, T., T. Fukada, J. M. Muller, D. T. Shima, M. Hibi, and T. Hirano. secondary immune responses, thus profoundly affecting NK cell 1999. Synergistic roles for Pim-1 and c-Myc in STAT3-mediated cell cycle progression and antiapoptosis. Immunity 11:709. biology and immune responses. 24. Qin, S., J. B. Rottman, P. Myers, N. Kassam, M. Weinblatt, M. Loetscher, A. E. Koch, B. Moser, and C. R. Mackay. 1998. The chemokine receptors CXCR3 and CCR5 mark subsets of T cells associated with certain inflammatory Acknowledgments reactions. J. Clin. Invest. 101:746. We thank William Bere and Stephanie Krebs for their assistance with pri- 25. Inngjerdingen, M., B. Damaj, and A. A. Maghazachi. 2000. Human NK cells mary NK cell isolation and RPAs. We are also indebted to Wanghua Gong express CC chemokine receptors 4 and 8 and respond to thymus and activation- regulated chemokine, macrophage-derived chemokine, and I-309. J. Immunol. by guest on September 29, 2021 for her assistance with the chemotaxis assays. 164:4048. 26. Campbell, J. J., S. Qin, D. Unutmaz, D. Soler, K. E. Murphy, M. R. Hodge, References L. Wu, and E. C. Butcher. 2001. Unique subpopulations of CD56ϩ NK and NK-T peripheral blood lymphocytes identified by chemokine receptor expression rep- 1. Trinchieri, G. 1989. Biology of natural killer cells. Adv. Immunol. 47:187. ertoire. J. Immunol. 166:6477. 2. Herberman, R. B., and J. R. Ortaldo. 1981. Natural killer cells: their roles in 27. Yoneda, O., T. Imai, S. Goda, H. Inoue, A. Yamauchi, T. Okazaki, H. Imai, defenses against disease. Science 214:24. O. Yoshie, E. T. Bloom, N. Domae, and H. Umehara. 2000. Fractalkine-mediated 3. Biron, C. A., K. B. Nguyen, G. C. Pien, L. P. Cousens, and T. P. Salazar-Mather. endothelial cell injury by NK cells. J. Immunol. 164:4055. 1999. Natural killer cells in antiviral defense: function and regulation by innate 28. Biron, C. A. 1997. Activation and function of natural killer cell responses during cytokines. Annu. Rev. Immunol. 17:189. viral infections. Curr. Opin. Immunol. 9:24. 4. Maghazachi, A. A. 2000. Chemokines, G proteins and natural killer cells. Arch. 29. Scott, P., and G. Trinchieri. 1995. The role of natural killer cells in host-parasite Immunol. Ther. Exp. 48:65. interactions. Curr. Opin. Immunol. 7:34. 5. DeRisi, J. L., V. R. Iyer, and P. O. Brown. 1997. Exploring the metabolic and 30. Loetscher, M., P. Loetscher, N. Brass, E. Meese, and B. Moser. 1998. Lympho- genetic control of gene expression on a genomic scale. Science 278:680. cyte-specific chemokine receptor CXCR3: regulation, chemokine binding and 6. Iyer, V. R., M. B. Eisen, D. T. Ross, G. Schuler, T. Moore, J. C. Lee, J. M. Trent, gene localization. Eur. J. Immunol. 28:3696. L. M. Staudt, J. Hudson, Jr., M. S. Boguski, et al. 1999. The transcriptional 31. Maghazachi, A. A., A. al Aoukaty, and T. J. Schall. 1994. C-C chemokines program in the response of human fibroblasts to serum. Science 283:83. induce the chemotaxis of NK and IL-2-activated NK cells: role for G proteins. 7. Ross, D. T., U. Scherf, M. B. Eisen, C. M. Perou, C. Rees, P. Spellman, V. Iyer, J. Immunol. 153:4969. S. S. Jeffrey, M. Van de Rijn, M. Waltham. 2000. Systematic variation in gene 32. Taub, D. D., T. J. Sayers, C. R. Carter, and J. R. Ortaldo. 1995. ␣ and ␤ che- expression patterns in human cancer cell lines. Nat. Genet. 24:227. mokines induce NK cell migration and enhance NK-mediated cytolysis. J. Im- 8. Perou, C. M., S. S. Jeffrey, M. Van de Rijn, C. A. Rees, M. B. Eisen, D. T. Ross, munol. 155:3877. A. Pergamenschikov, C. F. Williams, S. X. Zhu, J. C. Lee, et al. 1999. Distinctive 33. Salazar-Mather, T. P., J. S. Orange, and C. A. Biron. 1998. Early murine cyto- gene expression patterns in human mammary epithelial cells and breast cancers. megalovirus (MCMV) infection induces liver natural killer (NK) cell inflamma- Proc. Natl. Acad. Sci. USA 96:9212. tion and protection through macrophage inflammatory protein 1␣ (MIP-1␣)-de- 9. Inngjerdingen, M., B. Damaj, and A. A. Maghazachi. 2001. Expression and reg- pendent pathways. J. Exp. Med. 187:1. ulation of chemokine receptors in human natural killer cells. Blood 97:367. 34. Wright, T. M., and J. M. Farber. 1991. 5Ј regulatory region of a novel cytokine 10. Timonen, T., J. R. Ortaldo, and R. B. Herberman. 1981. Characteristics of human gene mediates selective activation by interferon ␥. J. Exp. Med. 173:417. large granular lymphocytes and relationship to natural killer and K cells. J. Exp. 35. Farber, J. M. 1990. A macrophage mRNA selectively induced by ␥-interferon Med. 153:569. encodes a member of the platelet factor 4 family of cytokines. Proc. Natl. Acad. 11. Chan, S. H., M. Kobayashi, D. Santoli, B. Perussia, and G. Trinchieri. 1992. Sci. USA 87:5238. Mechanisms of IFN-␥ induction by natural killer cell stimulatory factor (NKSF/ 36. Cole, K. E., C. A. Strick, T. J. Paradis, K. T. Ogborne, M. Loetscher, IL-12): role of transcription and mRNA stability in the synergistic interaction R. P. Gladue, W. Lin, J. G. Boyd, B. Moser, D. E. Wood, et al. 1998. Interferon- between NKSF and IL-2. J. Immunol. 148:92. inducible T cell ␣ chemoattractant (I-TAC): a novel non-ELR CXC chemokine 12. Chan, S. H., B. Perussia, J. W. Gupta, M. Kobayashi, M. Pospisil, H. A. Young, with potent activity on activated T cells through selective high affinity binding to S. F. Wolf, D. Young, S. C. Clark, and G. Trinchieri. 1991. Induction of inter- CXCR3. J. Exp. Med. 187:2009. feron ␥ production by natural killer cell stimulatory factor: characterization of the 37. Farber, J. M. 1997. Mig and IP-10: CXC chemokines that target lymphocytes. responder cells and synergy with other inducers. J. Exp. Med. 173:869. J. Leukocyte Biol. 61:246.