CCR7-guided neutrophil redirection to skin-draining lymph nodes regulates cutaneous inflammation and infection.

Neutrophils are the first nonresident effector immune cells that migrate to a site of infection or inflammation; however, improper control of neutrophil responses can cause considerable tissue damage. Here, we found that neutrophil responses in inflamed or infected skin were regulated by CCR7-dependent migration and phagocytosis of neutrophils in draining lymph nodes (dLNs). In mouse models of Toll-like receptor-induced skin inflammation and cutaneous Staphylococcus aureus infection, neutrophils migrated from the skin to the dLNs via lymphatic vessels in a CCR7-mediated manner. In the dLNs, these neutrophils were phagocytosed by lymph node-resident type 1 and type 2 conventional dendritic cells. CCR7 up-regulation on neutrophils was a conserved mechanism across different tissues and was induced by a broad range of microbial stimuli. In the context of cutaneous immune responses, disruption of CCR7 interactions by selective CCR7 deficiency of neutrophils resulted in increased antistaphylococcal immunity and aggravated skin inflammation. Thus, neutrophil homing to and clearance in skin-dLNs affects cutaneous immunity versus pathology.

Regulatory T cells with superior immunosuppressive capacity emigrate from the inflamed colon to draining lymph nodes.

Foxp3+ Regulatory T cells (Tregs) play a critical role in the maintenance of colon homeostasis. Here we utilized photoconvertible KikGR mice to track immune cells from the caecum and ascending (proximal) colon in the steady state and DSS-induced colitis. We found that Tregs from the proximal colon (colonic migratory Tregs) migrated exclusively to the distal part of mesenteric lymph nodes (dMLN) in an S1PR1-dependent process. In the steady state, colonic migratory CD25+ Tregs expressed higher levels of CD103, ICOS, LAG3 and CTLA-4 in comparison with pre-existing LN Tregs. Intestinal inflammation led to accelerated Treg replacement in the colon, bidirectional Treg migration from the colon to dMLN and vice versa, as well as increases in Treg number, proliferation and expression of immunosuppressive molecules. This was especially apparent for CD25 very high Tregs induced in colitis. Furthermore, colonic migratory Tregs from the inflamed colon included more interleukin (IL)-10 producing cells, and demonstrated greater inhibition of T-cell proliferation in comparison with pre-existing LN Tregs. Thus, our results suggest that Tregs with superior immunosuppressive capacity are increased both in the colon and dMLN upon inflammation. These Tregs recirculate between the colon and dMLN, and are likely to contribute to the downregulation of intestinal inflammation.

Dental pulp dendritic cells migrate to regional lymph nodes.

Dendritic cell (DC) migration to regional lymph nodes (RLNs) is an essential step in adaptive immunity, and cell-surface antigens on migrating DCs greatly affect the quality and quantity of subsequent immune responses. Although MHC class II(+) DC-like cells exist in the dental pulp, the lineage and function of these cells remain unknown. Here, we identified migratory DCs from the dental pulp after cusp trimming and acid etching in KikGR mice, in which the photoconvertible fluorescent protein changed from green to red upon violet light exposure. Two major cell fractions from the dental pulp had migrated to the RLNs at 16 hrs after cusp treatment, which showed the following lineage markers in the main and second fractions: CD11c(high)CD11b(++)Ly6C(low) Ly6G(low) F4/80(+) and CD11c(med)CD11b(+++)Ly6C(++)Ly6G(+++)F4/80(-), respectively. These lineage markers indicate that the former cells were DCs that had migrated through afferent lymphoid vessels, and the latter were granulocytes recruited via blood circulation. Migratory dental pulp DCs were mature, expressing the highest levels of CD273 (B7-DC) and CD86 co-stimulators and MHC class II. Our results suggest that cariogenic-bacteria-exposed dental pulp DCs migrate to RLNs and there trigger adaptive immune responses.

A mandatory role for STAT4 in IL-12 induction of mouse T cell CCR5.

IL-12 was recently shown to induce CCR5 on TCR-triggered mouse T cells. Considering that STAT4 is the most critical of IL-12 signaling molecules, this study investigated the role for STAT4 in the induction of CCR5 expression. IL-12R was induced by stimulation with anti-CD3 plus anti-CD28 mAb similarly on T cells from wild-type (WT) and STAT4-deficient (STAT4(-/-)) mice, but the levels of IL-12R induced on IFN-gamma-deficient (IFN-gamma(-/-)) T cells were lower compared with WT T cells. Exposure of TCR-triggered WT T cells to IL-12 induced CCR5 expression. In contrast, TCR-triggered STAT4(-/-) T cells failed to express CCR5 in response to IL-12. IL-12 stimulation induced detectable albeit reduced levels of CCR5 expression on IFN-gamma(-/-) T cells. Addition of rIFN-gamma to cultures of IFN-gamma(-/-) T cells, particularly to cultures during TCR triggering resulted in restoration of CCR5 expression. However, CCR5 expression was not induced in STAT4(-/-) T cells by supplementation of rIFN-gamma. These results indicate that for the induction of CCR5 on T cells, 1) STAT4 plays an indispensable role; 2) such a role is not substituted by simply supplementing rIFN-gamma; and 3) IFN-gamma amplifies CCR5 induction depending on the presence of STAT4.

Crystal engineering using anilic acids and dipyridyl compounds through a new supramolecular synthon.

The anilic acids, 2,5-dihydroxy-1,4-benzoquinone (1a), 2,5-dibromo-3,6-dihydroxy-1,4-benzoquinone (bromanilic acid; 1b), 2,5-dichloro-3,6-dihydroxy-1,4-benzoquinone (chloranilic acid; 1c), and 2,5-dicyano-3,6-dihydroxy-1,4-benzoquinone (cyananilic acid; 1d), were cocrystallized with rigid organic ligands containing two pyridine rings, 2,4-bipyridine (2a), 4,4'-bipyridine (2b), 1,2-bis(2-pyridyl)ethylene (3a), 1,2-bis(4-pyridyl)ethylene (3b), 2,2'-dipyridylacetylene (4a), 3,3'-dipyridylacetylene (4b), and 4,4'-dipyridylacetylene (4c). Fourteen complexes 5-18 were obtained as single crystals, and their crystal structures were successfully determined by X-ray analysis. All complexes except those with 2a are 1:1 and are composed of an infinite linear or zigzag tape structure, the formation of which is ascribed to intermolecular O-H...N, N(+)-H...O, or N(+)-H...O(-) hydrogen bonds or a combination of these between the anilic acids and the dipyridyl compounds. In the complexes 5 and 6, no infinite tape structure is observed although the molecular units connected by a similar hydrogen-bonding pattern are formed. For the 1:1 complexes, we have found two types of stacking arrangements, segregated stacks (7, 9, 12-15, 18) and alternated ones (8, 10, 11, 16, 17). In the complexes of 1c with the series of dipyridylacetylenes 4 (14, 15, 17), the neutral, dication, and monocaction states are formed depending on the nitrogen positions, which can be attributed to the different basicity of the pyridyl groups.

IL-12 plays a pivotal role in LFA-1-mediated T cell adhesiveness by up-regulation of CCR5 expression.

The chemokine receptor CCR5 has been implicated in the recruitment of T cells to inflammatory sites. However, the regulation of CCR5 induction on T cells and its contribution to T cell adhesiveness are poorly understood. Using a Th1 clone, 2D6, that can be maintained with interleukin (IL)-12 or IL-2 alone (designated 2D6(IL-12) or 2D6(IL-2), respectively), we investigated how CCR5 is induced on T cells and whether CCR5 is responsible for up-regulating the function of adhesion molecules. 2D6(IL-12) grew, forming cell aggregates, in culture containing IL-12. This was due to lymphocyte function-associated antigen (LFA)-1-intercellular adhesion molecule (ICAM)-1 interaction, because 2D6(IL-12) expressed both LFA-1 and ICAM-1 and cell aggregation was inhibited by anti-ICAM-1 monoclonal antibody. Despite comparable levels of LFA-1 and ICAM-1 expression, 2D6(IL-2) cells did not aggregate in culture with IL-2. It is important that there was a critical difference in CCR5 expression between 2D6(IL-12) and 2D6(IL-2); the former expressed high levels of CCR5, and the latter expressed only marginal levels. Both types of cells expressed detectable albeit low levels of RANTES (regulated on activation, normal T expressed and secreted) mRNA. Unlike IL-12 or IL-2, IL-18 induced high levels of RANTES mRNA expression without modulating CCR5 expression. Therefore, combined stimulation with IL-12 and IL-18 strikingly up-regulated 2D6 cell aggregation. Notably, LFA-1-mediated aggregation of 2D6(IL-12) cells was suppressed by anti-CCR5 antibody. These results indicate that IL-12 plays a critical role in CCR5 expression on Th1 cells and consequently contributes to CCR5-mediated activation of LFA-1 molecules.

A critical role for IL-12 in CCR5 induction on T cell receptor-triggered mouse CD4(+) and CD8(+) T cells.

Despite increasing evidence for the role of the chemokine system in leukocyte trafficking, the mechanism underlying the induction of chemokine receptors is poorly understood. Here, we investigated how CCR5, a chemokine receptor implicated in T cell migration to inflammatory sites, is induced in the T cell. CCR5 mRNA was hardly detected in resting T cells and marginally induced following T cell receptor (TCR) stimulation. However, TCR-triggered T cells expressed IL-12 receptor, and stimulation with recombinant IL-12 resulted in high levels of CCR5 expression on both CD4(+) and CD8(+) T cells. In contrast, IL-2 failed to up-regulate CCR5 expression. The effect of IL-12 was selective to CCR5 because IL-12 did not up-regulate CXCR3 expression. Surface expression of CCR5 was shown by staining with anti-CCR5 monoclonal antibody. Stimulation of these CCR5-positive T cells with the relevant chemokine MIP-1 alpha elicited Ca(2+) influx, showing that IL-12-induced CCR5 is functional. These results indicate a critical role for IL-12 in the induction of CCR5 on TCR-triggered T cells.

An absolute requirement for STAT4 and a role for IFN-gamma as an amplifying factor in IL-12 induction of the functional IL-18 receptor complex.

IL-12 and IL-18 are both proinflammatory cytokines that contribute to promoting Th1 development and IFN-gamma expression. However, neither IL-12R nor IL-18R is expressed as a functional complex on most resting T cells. This study investigated the molecular mechanisms underlying the induction of an IL-18R complex in T cells. Resting T cells expressed IL-18Ralpha chains but did not exhibit IL-18 binding sites as detected by incubation with rIL-18 followed by anti-IL-18 Ab, suggesting a lack of IL-18Rbeta expression in resting T cells. Although they also failed to express IL-12R, stimulation with anti-CD3 plus anti-CD28 generated IL-12R. Exposure of these cells to IL-12 led not only to up-regulation of IL-18Ralpha expression but also to induction of IL-18R binding sites on both CD4(+) and CD8(+) T cells concomitant with IL-18Rbeta mRNA expression. The IL-18 binding site represented a functional IL-18R complex capable of exhibiting IL-18 responsiveness. IL-12 induction of an IL-18R complex and IL-18Rbeta mRNA expression was not observed in STAT4-deficient (STAT4(-/-)) T cells and was substantially decreased in IFN-gamma(-/-) T cells. However, the failure of STAT4(-/-) T cells to induce an IL-18R complex was not corrected by IFN-gamma. These results indicate that STAT4 and IFN-gamma play an indispensable role and a role as an amplifying factor, respectively, in IL-12 induction of the functional IL-18R complex.

CD28 costimulation is required not only to induce IL-12 receptor but also to render janus kinases/STAT4 responsive to IL-12 stimulation in TCR-triggered T cells.

The activation of resting T cells for the acquisition of various functions depends on whether CD28 costimulatory signals are provided upon T cell receptor stimulation. Here, we investigated how CD28 costimulation functions to allow TCR-triggered resting T cells to acquire IL-12 responsiveness. When T cells are stimulated with low doses of anti-CD3 mAb, CD28 costimulation was required for the optimal levels of IL-12 receptor (IL-12R) expression. However, stimulation of T cells with high doses of anti-CD3 alone induced comparable levels of IL-12R expression to those induced upon CD28 costimulation. Nevertheless, there was a substantial difference in IL-12 responsiveness between these two groups of T cells: compared to anti-CD28-costimulated T cells, T cells that were not costimulated with anti-CD28 exhibited decreased levels of Janus kinases (JAK) JAK2/TYK2 and STAT4 phosphorylation and IFN-y production following IL-12 stimulation. Importantly, STAT6 phosphorylation following IL-4 stimulation was not decreased in anti-CD28-uncostimulated T cells. These resutls indicate that CD28 costimulation not only contributes to up-regulating IL-12R expression but is also required to render JAKs/STAT4 responsive to IL-12 stimulation.

4,7-bis(4-pyridylethynyl)-2,1,3-benzothiadiazole and its dipyridinium diperchlorate.

The title compound, C20H10N4S, and its dipyridinium salt, 4,4'-(2,1,3-benzodiazol-4,7-diyldiethynyl)dipyridinium diperchlorate, C20H12N4S2+*2ClO4-, display bond alternation in the 2,1,3-benzothiadiazole rings, which suggests their quinonoid character. The dipyridinium dication molecules stack along the a axis and form a dimer with short S.N interheteroatom contacts [3.146 (4) A] between the two 1,2,5-thiadiazole rings. The dimer is surrounded by the perchlorate anions with which it forms a large number of intermolecular N-H...O and C-H...O hydrogen bonds.

A molecule with a C1-homobasketane framework.

The title compound, 6-(1,3-benzodithiol-2-ylidene)-5,7-dimethyl-1,2-diphenylpentacyclo[5.4.0.0(2,5).0(3,11).0(4,8)]undecane, C(32)H(28)S(2), with a C(1)-homobasketane framework, crystallizes in the P-1 space group with one molecule in the asymmetric unit. The two cyclobutane rings in the cage are in a puckered conformation. Due to the enhanced through-bond interaction of the phenyl pi systems involving a strained sigma bond, the (Ph-)C-C(-Ph) bond length is significantly extended, to 1.610 (3) A.

Linear hydrogen-bonded molecular tapes in the cocrystals of squaric acid with 4,4'-dipyridylacetylene and 1,2-bis(4-pyridyl)ethylene.

The title compounds, 4,4'-(ethyne-1,2-diyl)dipyridinium bis(squarate), C(12)H(10)N(2)(2+).2C(4)HO(4)(-), and 4,4'-(ethene-1,2-diyl)dipyridinium bis(squarate), C(12)H(12)N(2)(2+).2C(4)HO(4)(-), are isomorphous and crystallize in space group P-1. The cocrystals contain linear hydrogen-bonded molecular tape structures along the [120] direction. The squarate monoanions form a ten-membered dimer linked by two intermolecular O-H.O hydrogen bonds. Each component molecule forms a segregated stack along the c axis. The bond lengths of the squarate monoanion indicate delocalization of the enolate anion.

IL-12 as well as IL-2 upregulates CCR5 expression on T cell receptor-triggered human CD4+ and CD8+ T cells.

The expression of chemokine receptors on leukocytes is related to their activation state. However, the exact mechanism underlying the induction of each chemokine receptor is poorly understood. Here, we investigated how CCR5, a chemokine receptor implicated in T cell trafficking and HIV infection, is induced in human T cells. CCR5 was marginally detected on a freshly prepared human peripheral blood mononuclear cell (PBMC) population. Long-term (8-day) stimulation of PBMC with IL-2 resulted in high levels of CCR5 expression on T cells. IL-12 failed to induce CCR5 on T cells in such a directly stimulated PBMC population. Stimulation of PBMC T cells with anti-CD3 plus anti-CD28 induced detectable albeit very low levels of CCR5 along with the induction of IL-12 receptor. However, these TCR-triggered T cells expressed much higher levels of CCR5 when stimulated with IL-12. Although IL-2 also induced CCR5 expression, CCR5 expression was more potent in IL-12 than IL-2 stimulation. These results indicate that, in addition to IL-2, IL-12 plays an important role in the induction of CCR5 expression on T cells, particularly TCR-triggered T cells.

A role of interferon-gamma (IFN-gamma) in tumor immunity: T cells with the capacity to reject tumor cells are generated but fail to migrate to tumor sites in IFN-gamma-deficient mice.

IFN-gamma-deficient (IFN-gamma-/-) mice induce potent in vitro immune responses such as anti-allo mixed lymphocyte reaction and CTL responses, whereas they often fail to exhibit in vivo immunity. Here, we investigated whether there exists a defect in tumor rejection responses and if so, which process of responses is impaired. IFN-gamma-/- and wild-type (WT) BALB/c mice were immunized with attenuated syngeneic CSA1M tumor cells. The capacity of T cells to mediate tumor protection was examined in Winn assays to assess the growth of tumor cells admixed with tumor-sensitized T cells. Splenic T cells from both groups of mice exhibited comparable levels of tumor-neutralizing activity. When portions of immunized mice were directly challenged with viable tumor cells, tumor rejection was induced only in WT mice. CD4(+) and CD8(+) T-cell infiltration were observed at the site of tumor challenge in WT mice, whereas such a T-cell infiltration did not occur in IFN-gamma-/- mice. Similarly, splenic T cells from interleukin 12-treated CSA1M-bearing IFN-gamma-/- and WT mice neutralized tumor cells at comparable efficacies in Winn assays. However, the migration of these T cells to tumor masses and the resultant interleukin 12-induced tumor regression took place in WT mice, but neither intratumoral T-cell infiltration nor tumor regression occurred in IFN-gamma-/- mice. These results indicate a critical requirement for IFN-gamma in the process of inducing T-cell migration to tumor sites rather than of generating antitumor protective T cells.

Transforming growth factor-beta up-regulates CD40-engaged IL-12 production of mouse Langerhans cells.

Transforming growth factor (TGF)-beta is an immunosuppressive agent that is efficacious in suppressing a wide variety of cell-mediated immune responses. However, the direct effect of this cytokine on Langerhans cells (LC) has not been clarified. In this study, we examined its modulatory effects on the expression of co-stimulatory molecules and LC IL-12 production. A highly purified population of LC (>95%) was prepared from BALB/c mouse skin by the panning method using anti-I-Ad mAb. Semiquantitative reverse transcription-PCR analysis showed that LC express TGF-beta receptor II mRNA. Interestingly, TGF-beta1 enhanced IL-12 p40 production of anti-CD40/IFN-gamma-stimulated LC, despite its down-regulatory effect on CD40 expression. A bioassay using an IL-12-dependent T cell line demonstrated the correlation of the IL-12 p40 level with the bioactivity of IL-12. More importantly, it was found that in contrast to TGF-beta, granulocyte/macrophage colony-stimulating factor (GM-CSF) strikingly inhibits IL-12 production of anti-CD40/IFN--stimulated LC and that the level of LC IL-12 production is determined by the relative amounts of TGF-beta1 and GM-CSF. Taken together, these results suggest that the two cytokines produced in the skin microenvironment, namely TGF-beta and GM-CSF, exert their important effects on LC function by regulating the secretion of IL-12, a cytokine influencing the Th1-Th2 balance.

IFN-alpha acts on T-cell receptor-triggered human peripheral leukocytes to up-regulate CCR5 expression on CD4+ and CD8+ T cells.

Interleukin-12 (IL-12) as well as IL-2 was recently shown to up-regulate CCR5 expression on T-cell receptor (TCR)-triggered human T cells. Because of the functional similarity between interferon-alpha (IFN-alpha) and IL-12, the present study investigated whether IFN-alpha also up-regulates T cell CCR5 expression. CCR5 was marginally detected on T cells from unstimulated human peripheral blood leukocytes (PBLs) and only slightly induced on PBL T cells following stimulation with anti-CD3 plus anti-CD28 monoclonal antibodies (mAbs). When anti-CD3/anti-CD28-triggered PBLs were exposed to IFN-alpha, T cells expressed high levels of CCR5. The levels of CCR5 expression were comparable to those induced by IL-12. However, when purified T cells instead of unfractionated PBL were stimulated with anti-CD3/CD28 and then exposed to IL-12 or IFN-alpha, CCR5 expression was induced by IL-12 but not by IFN-alpha. IFN-alpha was found to act on anti-CD3/anti-CD28-stimulated PBL to promote their IL-12 production. Moreover, addition of anti-IL-12 mAb to IFN-alpha-stimulated cultures of anti-CD3/CD28-pretreated PBL resulted in considerable inhibition of CCR5 expression. Together, these results indicate that IFN-alpha as well as IL-12 up-regulates CCR5 expression on TCR-triggered T cells and that IFN-alpha functions not by acting directly on T cells but via enhancing IL-12 production by PBL.

Requirement for IFN-gamma in IL-12 production induced by collaboration between v(alpha)14(+) NKT cells and antigen-presenting cells.

Two cytokines IL-4 and IL-12 are known to determine the balance between T(h)1 and T(h)2 development. In addition to IL-4 production of V(alpha)14(+) NKT cells, they have recently been demonstrated to have the capacity to stimulate IL-12 production by antigen-presenting cells (APC). This study demonstrates that IFN-gamma is absolutely required for the NKT cell-stimulated IL-12 production. Culture of B cell-depleted spleen cells from C57BL/6 mice with alpha-galactosylceramide (alpha-GalCer) capable of selectively stimulating V(alpha)14/J(alpha)281(+) NKT cells resulted in the production of IL-12 together with IL-4. Whereas IL-4 production occurred in culture of IFN-gamma(-/-) C57BL/6 splenocytes, the same culture failed to generate IL-12 production. While IL-12 production induced during culture of V(alpha)14(+) NKT cells and APC depended on the interaction between CD40 ligand on NKT cells and CD40 on APC, the expression levels of these key molecules were comparable in cells from wild-type and IFN-gamma(-/-) mice. Addition of rIFN-gamma to alpha-GalCer stimulated IFN-gamma(-/-) splenocyte culture, and administration of rIFN-gamma to alpha-GalCer-injected IFN-gamma(-/-) mice resulted in the restoration of IL-12 production in vitro and in vivo. These results illustrate a mandatory role for IFN-gamma in V(alpha)14(+) NKT cell-stimulated IL-12 production by APC.

Granulocyte/macrophage colony-stimulating factor inhibits IL-12 production of mouse Langerhans cells.

We investigated the capacity of mouse Langerhans cells (LC) to produce IL-12, a central cytokine in a Th1 type of immune responses. We prepared purified LC (>95%) from BALB/c mouse skin by the panning method using anti-I-Ad mAb. An ELISA showed that purified LC spontaneously produced IL-12 p40, and that its production was up-regulated following simultaneous stimulation with anti-CD40 mAb and IFN-gamma. Surprisingly, GM-CSF strikingly inhibited IL-12 p40 production by anti-CD40/IFN-gamma-stimulated LC (% inhibition = 97.0 +/- 0.9% at 1 ng/ml GM-CSF). Supernatants of 48-h cultured keratinocytes (KC) also caused the inhibition of LC IL-12 p40 secretion, and this effect was neutralized by anti-GM-CSF mAb. IL-1alpha (1 ng/ml)-stimulated KC produced much more GM-CSF than unstimulated KC (60.9 +/- 0.2 pg/ml vs 20.9 +/- 1.7 pg/ml), and IL-1alpha-stimulated KC supernatants strongly inhibited IL-12 p40 production by anti-CD40/IFN-gamma-stimulated LC (% inhibition = 89.4 +/- 1.4%). A bioassay using an IL-12-dependent T cell line demonstrated the correlation of the level of IL-12 p40 with the bioactivity of IL-12. These results provide important implications for the pathogenesis of atopic dermatitis, which involves the participation of LC and KC with the capacity to produce IL-12 and GM-CSF, respectively.

Interleukin-18 in combination with IL-2 enhances natural killer cell activity without inducing large amounts of IFN-gamma in vivo.

Interleukin-18 (IL-18) is known to synergistically enhance murine natural killer (NK) cell activity in vitro when combined with either IL-12 or IL-2. However, it has also been demonstrated that simultaneous administration of IL-18 and IL-12 to mice induces extraordinarily large amounts of interferon-gamma (IFN-gamma) in the serum. In this study, we examined the effects of a combination of IL-18 and IL-2 on in vivo NK cell activation in parallel with the induction of toxicity. In contrast to the IL-18 and IL-12 combination, the combined administration of IL-18 and IL-2 to BALB/c mice for 3 days induced neither high levels of IFN-gamma production nor other visible side effects. When compared with treatment with IL-18 or IL-2 alone, the combined treatment resulted in a significant increase in the number of DX-5 (pan-NK cells marker)-positive cells in spleens and a marked enhancement of splenic NK activity, as determined by standard cytotoxicity assays. Enhanced splenic cytotoxicity generated in the mice treated with both IL-18 and IL-2 was also observed against syngeneic Colon 26 adenocarcinoma cells. Consistent with this in vitro observation, combined treatment produced a significantly stronger inhibitory effect on the pulmonary metastases following i.v. injection of Colon 26 tumor cells than treatment with either cytokine alone. These results suggest that IL-18 combined with IL-2 potentiates in vivo NK cell activity and contributes to inhibition of tumor metastasis without inducing significant toxicity.