of September 2, 2015.This information is current as

NK CellsCardif (MAVS) Regulates the Maturation of

HedrickAriana Feuvrier, Chris A. Benedict and Catherine C.Wu, Robert Tacke, Heba N. Nowyhed, Jennifer Ekstein, LaTeira D. Haynes, Shilpi Verma, Bryan McDonald, Runpei

http://www.jimmunol.org/content/195/5/2157doi: 10.4049/jimmunol.1402060July 2015;

2015; 195:2157-2167; Prepublished online 31J Immunol 

MaterialSupplementary

0.DCSupplemental.htmlhttp://www.jimmunol.org/content/suppl/2015/07/30/jimmunol.140206

Referenceshttp://www.jimmunol.org/content/195/5/2157.full#ref-list-1

, 32 of which you can access for free at: cites 75 articlesThis article

Subscriptionshttp://jimmunol.org/subscriptions

is online at: The Journal of ImmunologyInformation about subscribing to

Permissionshttp://www.aai.org/ji/copyright.htmlSubmit copyright permission requests at:

Email Alertshttp://jimmunol.org/cgi/alerts/etocReceive free email-alerts when new articles cite this article. Sign up at:

Print ISSN: 0022-1767 Online ISSN: 1550-6606. Immunologists, Inc. All rights reserved.Copyright © 2015 by The American Association of9650 Rockville Pike, Bethesda, MD 20814-3994.The American Association of Immunologists, Inc.,

is published twice each month byThe Journal of Immunology

at Univ of C

alifornia-San Diego Serials/B

iomed L

ib 0699 on September 2, 2015

http://ww

w.jim

munol.org/

Dow

nloaded from

at Univ of C

alifornia-San Diego Serials/B

iomed L

ib 0699 on September 2, 2015

http://ww

w.jim

munol.org/

Dow

nloaded from

The Journal of Immunology

Cardif (MAVS) Regulates the Maturation of NK Cells

LaTeira D. Haynes,* Shilpi Verma,† Bryan McDonald,† Runpei Wu,* Robert Tacke,*

Heba N. Nowyhed,* Jennifer Ekstein,* Ariana Feuvrier,* Chris A. Benedict,† and

Catherine C. Hedrick*

Cardif, also known as IPS-1, VISA, and MAVS, is an intracellular adaptor protein that functions downstream of the retinoic acid–

inducible gene I family of pattern recognition receptors. Cardif is required for the production of type I IFNs and other inflam-

matory cytokines after retinoic acid–inducible gene I–like receptors recognize intracellular antigenic RNA. Studies have recently

shown that Cardif may have other roles in the immune system in addition to its role in viral immunity. In this study, we find that

the absence of Cardif alters normal NK cell development and maturation. Cardif2/2 mice have a 35% loss of mature CD272

CD11b+ NK cells in the periphery. In addition, Cardif2/2 NK cells have altered surface marker expression, lower cytotoxicity,

decreased intracellular STAT1 levels, increased apoptosis, and decreased proliferation compared with wild-type NK cells. Mixed

chimeric mice revealed that the defective maturation and increased apoptotic rate of peripheral Cardif2/2 NK cells is cell

intrinsic. However, Cardif2/2 mice showed enhanced control of mouse CMV (a DNA b-herpesvirus) by NK cells, commensurate

with increased activation and IFN-g production by these immature NK cell subsets. These results indicate that the skewed dif-

ferentiation and altered STAT expression of Cardif2/2 NK cells can result in their hyperresponsiveness in some settings and sup-

port recent findings that Cardif-dependent signaling can regulate aspects of immune cell development and/or function distinct from

its well-characterized role in mediating cell-intrinsic defense to RNA viruses. The Journal of Immunology, 2015, 195: 2157–2167.

Pattern recognition receptors recognize pathogen-associatedmolecular patterns (1, 2). Retinoic acid–inducible gene-I(RIG-I)–like receptors are a subset of pattern recognition

receptors that recognize intracellular viral nucleic acids and in-duce the production of type-I IFNs and NF-kB–regulated genes(3). Two members of the RIG-I–like receptor family, RIG-I andMDA5, have caspase recruitment and activation domains (CARDs)that allow for downstream signaling after activation. The adaptorprotein that interacts with RIG-I and MDA5 and allows downstreamsignaling was discovered by four different groups and is thus knownby four names: Cardif (CARD adaptor inducing IFN-b), MAVS(mitochondrial antiviral signaling), IPS-1 (IFN-b promoterstimulator-1), and VISA (virus-induced signaling adaptor) (4–7).We refer to this protein as Cardif.RIG-I and MDA5 initiate signaling through CARD–CARD

interactions with Cardif, which is a ubiquitously expressed proteinthat is located on the outer mitochondrial membrane of both im-mune and nonimmune cells (3, 8). The mitochondrial localizationof Cardif is essential to its signaling function. Once Cardif has

been engaged by RIG-I or MDA5, it aggregates with other Cardifmolecules. This aggregation is essential to propagation of down-stream signals (9). Cardif interacts with cytoplasmic adaptor mole-cules TRAF3, TRAF2, and TRAF6 to activate transcription factorsNF-kB, IRF3, and IRF7 to induce expression of type I IFN genesand IFN-induced genes (3, 10).Cardif is vital for signaling in response to viral pathogenic

nucleic acids sensed by MDA5 and RIG-I. However, there havebeen reports that suggest that both RIG-I and Cardif may play rolesin immune regulation that are separate from their roles in viraldefense (11–15). Wang et al. (12) report that RIG-I2/2 mice developcolitis and are more susceptible to dextran sulfate–induced colitis.Xu et al. (13) report that Cardif2/2 B cells have a cell-intrinsic defectin CD23 and TLR7 expression. In addition, Cardif2/2 mice developmore severe disease in the mouse model of multiple sclerosis, ex-perimental autoimmune encephalomyelitis (16). Cardif is also re-quired for the optimal activation of the NLRP3 inflammasome(14). RIG-I activity under the control of IRF1 has been implicatedin the progression of atherosclerosis. Wang et al. (15) propose that25-hydroxycholesterol induces IL-8 production in macrophages byinducing IRF1 and subsequent RIG-I expression and activation.These reports, as well as others, suggest that Cardif is active even inthe absence of pathogenic viral RNA.NK cells are innate cytotoxic lymphocytes that target virally

infected, stressed, or cancerous cells (17). NK cells primarily de-velop in the bone marrow, although some peripheral organs suchas the liver can house and develop NK cells (18–20). Mature NK(mNK) cells are the primary NK cells found in peripheral organssuch as the spleen, liver, and lymph nodes where they undergoadditional maturation (21). CD49b acquisition is the earliest stageof NK maturity. The acquisition of CD11b, CD43, and KLRG1occur after CD49b, and identify more advanced stages of NKmaturation. Surface markers CD27 and CD11b can be used tofurther delineate stages of maturation within immature NK (iNK)and CD49b+ NK (mNK) cells (22–24). Maturation using thesemarkers is divided into four stages that progress in the following

*Division of Inflammation Biology, La Jolla Institute for Allergy and Immunology,La Jolla, CA 92037; and †Division of Immune Regulation, La Jolla Institute forAllergy and Immunology, La Jolla, CA 92037

Received for publication August 13, 2014. Accepted for publication June 18, 2015.

This work was supported by the National Institutes of Health (Grant F31 HL110668 toL.D.H., Grant R01 HL097368 to C.C.H., Grant F32HL117533 to H.N.N., and Grant R01AI101423 to C.A.B.), American Heart Association Grant 13POST16990029 (to R.T.),and the American Diabetes Association (Award 1-13-MUI-02 to L.D.H. and C.C.H.).

Address correspondence and reprint requests to Prof. Catherine C. Hedrick, Divisionof Inflammation Biology, La Jolla Institute for Allergy and Immunology, 9420Athena Circle, La Jolla, CA 92037. E-mail address: hedrick@lji.org

The online version of this article contains supplemental material.

Abbreviations used in this article: CARD, caspase recruitment and activation domain;Cardif, CARD adaptor inducing IFN-b; DC, dendritic cell; iNK, immature NK;mNK, mature NK; MCMV, mouse CMV; RIG-I, retinoic acid–inducible gene I;SLE, systemic lupus erythematosus; WT, wild-type.

Copyright� 2015 by The American Association of Immunologists, Inc. 0022-1767/15/$25.00

www.jimmunol.org/cgi/doi/10.4049/jimmunol.1402060

at Univ of C

alifornia-San Diego Serials/B

iomed L

ib 0699 on September 2, 2015

http://ww

w.jim

munol.org/

Dow

nloaded from

order: CD272CD11b2 (stage 1) → CD27+CD11b2 (stage 2) →CD27+CD11b+ (stage 3) → CD272CD11b+ (stage 4). Very fewCD272CD11b2 NK cells are found in the peripheral organs, andthe majority of CD272CD11b2 NK cells do not express CD49b.CD27+CD11b2 NK cells are highly proliferative and have theability to degranulate but are not as cytotoxic as CD27+CD11b+ orCD272CD11b+ NK cells. CD27+CD11b+ NK cells are highly re-sponsive to cytokine and dendritic cell (DC) stimulation in vitrocompared with CD272CD11b+ NK cells (22). CD272CD11b+ NKcells are terminally differentiated, long-lived, and express the in-hibitory NK receptor KLRG1 on their surface (25). CD27+CD11b+

NK cells are most prevalent in bone marrow and lymph nodes,whereas CD272CD11b+ NK cells are the predominant stage foundin spleen, liver, blood, and lung.NK cell numbers, maturation, and function are diminished in

the absence of type I IFN signaling. IFNAR2 /2 mice that lackthe ability to respond to type I IFN have fewer NK cells and areunable to control the growth of tumor cells that are normallysusceptible to NK killing (26–29). IFNAR-deficient NK cells arealso unable to kill some NK target cells in vitro. It has beensuggested that consistent and low levels of type I IFN are requiredto maintain NK cell numbers and functionality in vivo as well(30). Constitutive type I IFN signaling maintains STAT1 levels inNK cells. High STAT1 levels are required to maintain NK cytotoxicity,whereas STAT4 leads to IFN-g production (26, 31, 32). It has beensuggested that many of the effects of type I IFN deficiency in NK cellscan be attributed to the reduction of intracellular STAT1 levels.Considering the earlier studies and observations, we sought to

understand how the absence of Cardif affects the development,maturation, and function of NK cells. In this study, we found thatCardif expression is required for maintaining optimal NK cellnumbers in the periphery and for full NK cell maturation. As such,we found a marked reduction in the number of CD49b+ and CD272

CD11b+ NK cells with a concomitant increase in CD27+CD11b2

and CD27+CD11b+ NK cells in the periphery of mice lackingCardif. These NK cells had decreased cytotoxicity, although theyproduced comparable amounts of IFN-g after stimulation in vitro.Mixed bone marrow chimeras revealed that the maturation ofperipheral Cardif2/2 NK cells is cell intrinsic. Similar to NK cellsfrom IFNAR2/2 mice, Cardif2/2 NK cells showed decreased pro-liferation and decreased STAT1 activation. Intriguingly, despite theiraltered differentiation and effector function in these settings, in re-sponse to mouse CMV (MCMV) infection, NK cells from Cardif2/2

mice showed heightened IFN-g production and enhanced control ofthis dsDNA b-herpesvirus. Together, our results indicate that Cardifimposes intrinsic regulation of NK cell subset maintenance andfunction in diverse inflammatory settings.

Materials and MethodsMice

Cardif2/2 mice were a kind gift from the Shresta laboratory at La JollaInstitute for Allergy and Immunology. Cardif2/2 mice were generated asdescribed by Michallet et al. (33) and are on a C57BL/6 background.C57BL/6 mice purchased from The Jackson Laboratory (000664) or wild-type (WT) littermates were used as controls in experiments. Male andfemale mice were used at 6–12 wk of age. All experiments adhered to theguidelines outlined by the La Jolla Institute for Allergy and ImmunologyAnimal Care and Use Committee, according to criteria outlined in theGuide for the Care and Use of Laboratory Animals from the NationalInstitutes of Health. Mice were euthanized by CO2 inhalation.

Flow cytometry

Spleen and liver were harvested and pushed through a 40-mm strainer. Liverwas perfused with PBS before collection and placed in RPMI 1640 mediumsupplemented with 10 mm HEPES and 10% FBS. In addition, lymphocytes

were separated from hepatocytes via density centrifugation. RBCs were lysedwith RBC lysis buffer according to the manufacturer’s protocol (Biolegend).

Cells were resuspended in FACS buffer (1%BSA and 0.1% sodium azidein PBS), and 1–4 3 106 cells were incubated in 100 ml with anti-CD16/CD32 Ab (2.4G2) for 30 min on ice to block FCgRII/III binding. Sampleswere then incubated with a mixture of fluorochrome-conjugated Abs(Biolegend, eBioscience, BD Biosciences) for 30 min on ice in the dark.LIVE/DEAD Fixable Dead Cell Stain (Invitrogen) was used to determinecell viability. Intracellular staining of Abs was performed after cells werefixed and permeabilized using 0.55% paraformaldehyde and Permeabiliza-tion Buffer (BD Biosciences). Samples were analyzed for cellular fluores-cence on an LSR II (BD Biosciences), and data were analyzed withFlowJo software (Tree Star). A complete flow-cytometry gating strategy forperipheral NK cells is shown in Supplemental Fig. 1.

Ex vivo stimulation

Spleen was harvested and pushed through a 40-mm strainer. RBCs werelysed with RBC lysis buffer according to the manufacturer’s protocol(Biolegend). Splenocytes were cultured in 96-well plates for 4–5 h in thepresence of either PMA (50 ng/ml) and ionomycin (1 mg/ml) or IL-12(20 ng/ml) and IL-18 (20 ng/ml). When measuring degranulation, FITC-conjugated CD107a Ab (0.05 mg/ml) was added to the stimulation media.After stimulation, splenocytes were washed with PBS and then prepared forflow-cytometry analysis as described earlier.

In vitro culture

NK cells were enriched using EasySep Mouse NK Cell Enrichment kit(Stem Cell Technologies). Spleens from three or more mice were pooledtogether. Purity of NK cells was ∼80–85% as reported by supplier. Murinecytokines were used in the following concentrations: IL-15, 50 ng/ml(Peprotech); IL-18, 50 ng/ml (Peprotech); and IFN-b, 10 U/ml (Milli-pore). There were triplicates of each condition. Cells were stimulated withPMA (50 ng/ml) and ionomycin (1 mg/ml) for 4 h. After 4 h, cells wereharvested and processed for FACS analysis.

For type I IFN rescue experiments, NK cells were cultured in thepresence of IL-15 only or IL-15 with IFN-b for 5 d at a concentration of5e6 cells/ml in a 96-well round-bottom plate. IL-15 was vital for the survivalof NK cells in culture. After 5 d of cytokine treatment, NK cells werestimulated with PMA and ionomycin at the aforementioned concentrationsand analyzed by flow cytometry.

Cytotoxicity assay

Enriched NK cells were cultured at various ratios together with calcein-labeled YAC-1 cells for 4 h. YAC-1 cells were labeled with 0.5 mM cal-cein. Specific lysis was determined by measuring percent specific releaseof calcein using the following formula: Percent specific release = (exper-imental release 2 spontaneous release)/(maximum release 2 spontaneousrelease) 3 100.

Bone marrow transplantation

Bone marrow transplantation studies were performed as previously de-scribed (34). CD54.1/0.2 mice were irradiated with two doses of 550 radeach, 4 h apart. Bone marrow cells were isolated from CD45.1 and Cardif2/2

mice and processed under sterile conditions. A single-cell suspension inPBS was obtained with a 1:1 ratio of CD45.1 and Cardif2/2 bone marrowcells. Approximately 5e6 cells were retro-orbitally injected into recipientCD54.1/0.2 mice in a volume of 200 ml. Mice received autoclaved watertreated with antibiotics (trimethoprim-sulfamethoxazole) 1 wk before andcontinued until 1 wk after injection. Mice were euthanized for experiments8–10 wk after injection.

Apoptosis and proliferation analysis

Annexin V (Invitrogen) and propidium iodide (Invitrogen) were used toidentify apoptotic and dead cells, respectively. Cells were stained accordingto the protocol provided by the manufacturer. FITC-conjugated caspase-3(BD Biosciences) was also used to identify apoptotic cells. Cells werestained for intracellular caspase-3 according to the manufacturer’s protocol.For proliferation, Ki-67 (eBioscience) was used to identify proliferatingcells ex vivo. Splenocytes were stained for intracellular Ki-67 using theFoxp3/Transcription factor staining buffer set (eBioscience), and cellswere stained according to the protocol provided by the manufacturer.

Western blot analysis

NK cells were isolated using EasySep Mouse CD49b positive selection kit(STEMCELLTechnologies). Spleens from seven or more mice were pooled

2158 CARDIF AND NK CELL MATURATION

at Univ of C

alifornia-San Diego Serials/B

iomed L

ib 0699 on September 2, 2015

http://ww

w.jim

munol.org/

Dow

nloaded from

together. Purity of NK cells was ∼90% as reported by supplier. Isolated NKcells were then divided and either left untreated or treated with 100 U/ml

IFN-b (Millipore) for 15 min at 37˚C. Cells were then lysed for protein

collection using radioimmunoprecipitation assay buffer (50 mM Tris, pH

7.4, 150 mM NaCl, 1 mM EDTA, 1 mM EGTA, 1 mM NaVO4, 1 mM NaF,

0.5% NP-40, 0.1% Brij35, 0.1% deoxycholic acid). Total was quanti-

fied using BCA Protein Assay Reagent (Thermo Scientific). Afterward,

30 mg of each protein sample was loaded into SDS-PAGE. The fol-

lowing Abs were used at specified concentrations for immunoblots: STAT1

(1:1000; no. 9172; Cell Signaling), pSTAT1 (Tyr701; 1:1000; no. 9171; Cell

Signaling), STAT4 (1:1000; no. 2653; Cell Signaling), p-STAT4 (1:1000;

no. 5267; Cell Signaling), and b-Actin (1:2000; no. 9774; Cell Signaling).

Western blots were quantified using ImageJ software (National Institutes

of Health).

MCMV infection

MCMV WT and Δm157 salivary gland stocks were derived from thebacterial artificial chromosome–cloned MCMV K181 strain (kind gift

from Alec Redwood, Murdoch University) after in vivo passage of MEF-

derived virus twice in 3-wk-old BALB/c mice for 12 d, as previously

described (35). Cardif2/2 and WT mice were infected i.p. with 1 3 105

PFU, and viral replication in organs was measured by plaque assay in NIH

3T3 cells as described previously (36). NK cells were depleted by injecting

200 mg anti-NK1.1 Ab (clone PK136) in 200 ml PBS 24 h preinfection.

FIGURE 1. Cardif is required for optimal NK

cell numbers in the periphery and for the ter-

minal differentiation of NK cells. (A) The ab-

solute number of NK1.1+CD32CD192 cells was

determined in the spleen. (B) Representative plots

(left and middle panels) and bar graphs (right

panels) show the percentage of CD49b2NK cells

(NK1.1+CD32CD192) in the spleen (upper pan-

els) and liver (lower panels). (C) Representative

plots and bar graphs plots (left and middle panels)

and bar graphs (right panels) show the distri-

bution of mNK (NK1.1+CD32CD192 CD49b+)

cells based on CD27 and CD11b expression in the

spleen (upper panels) and liver (lower panels).

Stages 1–4 are as follows: stage 1(CD272CD11b2),

stage 2 (CD27+CD11b2), stage 3 (CD27+CD11b2),

and stage 4 (CD272CD11b+). Experiments were

repeated at least three times. n = 3–5 mice/group/

experiment. Data are means 6 SEM. **p , 0.01,

***p , 0.001.

The Journal of Immunology 2159

at Univ of C

alifornia-San Diego Serials/B

iomed L

ib 0699 on September 2, 2015

http://ww

w.jim

munol.org/

Dow

nloaded from

Statistical analysis

All data are presented with mean 6 SEM as determined by Prism software(GraphPad). Unpaired t tests were used to compare WT and Cardif2/2

samples. The p values#0.05 were considered significant for all experiments.

ResultsNK maturation is impaired in Cardif2/2 mice

Cardif is reportedly active in the absence of a viral infection and isa potent inducer of type I IFN (3, 13, 16). Studies have shown thattype I IFN regulates NK cell maturation and function (27, 30).With this information, we hypothesized that Cardif may influenceNK cell maturation and function. To test this hypothesis, we char-acterized NK cells in the bone marrow and in the periphery ofWT and Cardif2/2 mice. We found that, compared with WT mice,Cardif2/2 mice have fewer numbers of CD192CD32NK1.1+ NKcells in spleen, but not in bone marrow (Fig. 1A).The acquisition of CD49b marks the developmental step from

iNK to mNK cells (21, 37). We found that the frequency ofCD49b2 (immature) NK cells in Cardif2/2 spleen and liver issignificantly higher than in WT NK cells (Fig. 1B). This 2-fold in-crease of CD49b2 NK cells in the spleens and livers of Cardif2/2

mice suggests a delay in mNK development and maintenance in theperiphery.In addition to a reduction in mNK cells (NK1.1+CD32CD192

CD49b+), we observed a significant reduction in terminally dif-ferentiated CD272CD11b+ NK cells in the spleen and liver ofCardif2/2 mice (Fig. 1C). This reduction of CD272CD11b+ NK

cells is accompanied by a concomitant increase in frequencies ofCD27+CD11b2 and CD27+CD11b+ NK cells. Cardif2/2 micehave a 35–40% loss of terminally differentiated mature CD272

CD11b+ NK cells with a concomitant increase in less mature(CD27+CD11b2 and CD27+CD11b+) NK cells in the spleen. Inaddition, there is a .50% decrease in terminally differentiatedCD272CD11b+ NK cells and a 50% increase in CD27+CD11b+

NK cells in liver of Cardif2/2 mice (Fig. 1C). NK cells in theblood and lymph nodes of Cardif2/2 mice displayed similar al-terations in NK maturation (data not shown). However, we foundno differences in NK maturation in the bone marrow (SupplementalFig. 2). Taken together, these data suggest that Cardif is impor-tant for NK maturation but does not impair NK development inbone marrow.

Inhibitory receptor expression is reduced in Cardif2/2 mice

We next wanted to examine homeostatic NK function in the ab-sence of Cardif signaling. Ly49 receptor expression on NK cells isrequired for NK cell “licensing,” the process that allows NK cellsto become active cytolytic cells after their encounter with selfMHC-1–expressing cells (38–40). The expression of the inhibitoryreceptors such as Ly49C/I and KLRG1 on NK cells is associatedwith NK cell maturity (23, 25). We investigated the expression ofKLRG1 and select Ly49 receptors in Cardif2/2 mice. We foundthat significantly fewer Cardif2/2 mNK (NK1.1+CD32CD192

CD49b+) cells express KLRG1 (Fig. 2A, 2B) (21). Moreover, thereis a decrease in the surface expression of KLRG1 on Cardif2/2

mNK cells (Fig. 2B). We found that WT and Cardif2/2 mNK cells

FIGURE 2. Cardif2/2 NK cells are less mature. (A) Representative histograms show the percentages of KLRG1+ cells, and cells expressing select

activating and inhibitory Ly49 receptors in WT and Cardif2/2 mNK populations (NK1.1+CD32CD192 CD49b+). (B) Bar graphs depict the percentage (left

panel) and amount of expression (right panel) of NK maturation markers and Ly49 receptors. Amount of expression was determined by flow-cytometric

analysis of mean fluorescence intensity (MFI). Experiments were repeated twice. n = 3 mice/group/experiment. Data are means 6 SEM. *p , 0.05, **p ,0.01.

2160 CARDIF AND NK CELL MATURATION

at Univ of C

alifornia-San Diego Serials/B

iomed L

ib 0699 on September 2, 2015

http://ww

w.jim

munol.org/

Dow

nloaded from

express similar amounts of the activating Ly49H receptor and theinhibitory Ly49G2 receptor (Fig. 2A, 2B). However, the numberof Ly49C/I+ mNK cells is markedly decreased in Cardif2/2

mice. Likewise, the surface expression of Ly49C/I on Cardif2/2

mNK cells is significantly lower than WT mNK cells. The de-crease in the frequency and cellular expression of KLRG1 andLy49C/I, which are associated with NK cell activation andmaturation, suggest that Cardif is involved in late mNK cellmaturation (18, 41).

Cardif2/2 NK cells are more apoptotic and less proliferative

To identify mechanistically why there are fewer mNK cells inCardif2/2 mice, we investigated the survival and proliferation ofNK cells in Cardif2/2 mice. We found that more Cardif2/2 NKcells were caspase-3+, suggesting that they were more apoptotic(Fig. 3A). We confirmed the increase in apoptosis using AnnexinV, because we found more Annexin V+ NK cells in Cardif2/2

mice (Fig. 3A). The greatest increase in apoptotic mNK cells wasobserved in terminally differentiated CD272CD11b+ NK cells,where 2-fold more CD272CD11b+ Cardif2/2 NK cells wereapoptotic as noted by caspase-3 expression (Fig. 3B). This in-crease of caspase-3+ CD272CD11b+ NK cells in Cardif2/2 miceaccounts for the increase in frequencies of total caspase-3+ NKcells. These data suggest that Cardif2/2 NK cells, particularlyterminally differentiated CD272CD11b+ NK cells, are more ap-optotic in the absence of tonic Cardif signaling. We used Ki-67 toidentify proliferating NK cells in Cardif2/2 and WT mice. Wefound an ∼15% decrease in proliferation of total splenic Cardif2/2

NK cells compared with WT NK cells (Fig. 3C). However, uponfurther analysis, we found that CD27+CD11b2 and CD27+

CD11b+ Cardif2/2 NK cells have an ∼25 and ∼30% reduction inproliferation, respectively. These two stages of NK cells areknown to mature into CD272CD11b+ NK cells (22, 23). Thus, inthe absence of Cardif, early-stage NK cells proliferate less whereasterminally differentiated CD272CD11b+ NK cells are prone toapoptosis. These data support our findings that Cardif2/2 micehave fewer total NK cells and increased numbers of early-stageNK cells.

Cardif deficiency alters NK cell function

Because we found that NK cells in Cardif2/2 mice are less matureand fewer are Ly49C/I+, we hypothesized that Cardif-deficient NKcells may not function as well as WT NK cells. Fernandez et al.(42) have previously shown that unlicensed Ly49C/I2 NK cellsare less cytotoxic than Ly49C/I+ NK cells. We stimulated WT andNK Cardif2/2 NK cells with either IL-12 and IL-18 or PMA andionomycin ex vivo. We found that granzyme B and CD107a,a marker of degranulation, were significantly reduced in Cardif2/2

NK cells compared with WT NK cells, suggesting that Cardif2/2

NK cells are likely impaired in their ability to kill target cells(Fig. 4A). The production of IFN-g, however, was comparablebetween NK cells from both WT and Cardif2/2 mice. Next, wetested the ability of Cardif2/2 NK cells to directly kill target cellsin vitro. YAC-1 cells served as target cells in the assay. Cardif2/2

NK cells had ∼20–25% less cytotoxic activity than WT NK cellswhen cultured with target B cells (Fig. 4B). These data indicate

FIGURE 3. Cardif2/2 NK cells are more apoptotic and less proliferative. (A) Bar graphs show the percentage of Caspase-3+ (upper panel) and Annexin

V+ (lower panel) NK cells (NK1.1+CD32CD192) from WT and Cardif2/2 mice. (B) Representative plots and bar graph of the percentage of WT and

Cardif2/2 caspase-3+ CD272CD11b+ NK cells. (C) Bar graphs depict the percentage of Ki-67+ total NK cells (left panel), as well as stage 1–4 NK cells

(middle and right panels). Experiments were repeated twice. n = 4–5 mice/group/experiment. Data are means 6 SEM. *p , 0.05, **p, 0.01, ***p , 0.001.

The Journal of Immunology 2161

at Univ of C

alifornia-San Diego Serials/B

iomed L

ib 0699 on September 2, 2015

http://ww

w.jim

munol.org/

Dow

nloaded from

that Cardif plays an important role in manipulating the functionsof NK cells.

STAT1 protein is reduced in Cardif2/2 NK cells

Reports have shown that the maintenance of STAT1 levels in NKcells is important for NK cell cytotoxicity upon activation, whereasthe levels of STAT4 are important for IFN-g production (31). Inearlier experiments, we found that Cardif2/2 NK cells are lesscytotoxic than WT NK cells, whereas IFN-g production wasthe same (Fig. 4A, 4B). With these observations, we decided tomeasure STAT1 and STAT4 phosphorylation in NK cells andfound that there were dramatic reductions in both the phosphor-ylation and the levels of STAT1 protein in Cardif2/2 NK cells(Fig. 5A). STAT4 protein and phosphorylation seemed similarbetween Cardif2/2 and WT NK cells, but we found that thep-STAT4/STAT4 ratio was higher in Cardif2/2 NK cells (Fig. 5B).Based on a known role for STAT1 in regulating NK cytolyticfunctions, these data indicate that the reduced cytolytic functionof Cardif2/2 NK cells is most likely due to the low intracellularamount of STAT1 and phosphorylated. The effects of an increasein the p-STAT4/STAT4 ratio in Cardif2/2 NK cells was mostlikely masked because of the already high amounts of STAT4present in WT NK cells.

Impaired NK cell phenotype in Cardif2/2 mice is cell intrinsic

To determinewhether the impact of Cardif on NK cell maturation iscell intrinsic, we used a mixed chimera approach. WT CD45.1+

bone marrow and CD45.2+Cardif2/2 bone marrow were mixed1:1 and transplanted into irradiated CD45.1+/CD45.2+ recipientmice (Fig. 6A). After 10 wk of reconstitution, we assessed thephenotype of NK cells in various organs. We found that CD45.1+

and Cardif2/2 cells contributed equally to the total number ofCD45+ cells in the reconstituted bone marrow (Fig. 6B). However,there were slight but consistently lower percentages of Cardif2/2

origin NK (CD45.12CD45.2+NK1.1+CD32CD192) cells in thebone marrow of chimeric mice compared with WT CD45.1 originNK cells (data not shown). We found that this NK cell–specificphenotype was more pronounced in the periphery (Fig. 6C). Wefound that Cardif2/2 NK cells reconstituted only ∼40%, whereasCD45.1+ NK cells reconstituted ∼60% of total NK cells in thespleen and liver (Fig. 6D). We also observed a slight decrease ofthe CD272CD11b+ subset within Cardif2/2 NK cells in the spleenand liver, similar to what we observed in the global knockout mice(Figs. 1C, 6E). The decrease in Cardif2/2 CD272CD11b+ NKcells was accompanied by an increase in CD27+CD11b2 Cardif2/2

NK cells in both the spleen and the liver of the mixed bone marrowchimeric mice (Fig. 6E).We also investigated the function of Cardif2/2 NK cells in the

chimeric mice. We found that although Cardif2/2 NK cells andWT NK cells displayed a similar amount of degranulation, gran-zyme B production was lower in Cardif2/2 NK cells (Fig. 6F). Inaddition, we found that more Cardif2/2 NK cells are caspase-3+

compared with WT NK cells in chimeric mice (Fig. 6G). Similarto global knockout mice, CD272CD11b+ Cardif2/2 NK cells in

FIGURE 4. Cardif2/2 NK cells are less cytotoxic than WT mNK cells. (A) Representative plots and bar graphs show the percentage of CD107a+ (upper

panels), granzyme B+ (middle panels), and IFN-g+ (lower panels) mNK (NK1.1+CD32CD192 CD49b+) cells in WT and Cardif2/2 samples. (B) Rep-

resentative line graph depicting the cytotoxic activity of WT and Cardif2/2 NK cells against target YAC-1 cells at various ratios. Ratios are NK/target cells.

Experiments were repeated at least twice. n = 3–5 mice/group/experiment. Data are means 6 SEM. *p , 0.05, **p , 0.01.

2162 CARDIF AND NK CELL MATURATION

at Univ of C

alifornia-San Diego Serials/B

iomed L

ib 0699 on September 2, 2015

http://ww

w.jim

munol.org/

Dow

nloaded from

chimeric mice have the greatest increase in apoptosis (Fig. 6G).We examined proliferation of WT and Cardif2/2 NK cells inchimeric mice using Ki-67 and found that both WT and Cardif2/2

NK cells proliferated at the same rate (Fig. 6H).Signaling via Cardif leads to the production of type I IFN. We,

along with other groups, have found that similar to Cardif2/2mice,IFNAR12/2 mice have fewer terminally differentiated NK cellsand abrogated functionality in the absence of type I IFN signaling(27, 28, 43, 44) (Supplemental Fig. 3). Gough et al. (30) havetheorized that low homeostatic levels of type I IFN are required inthe absence of infection to sustain levels of STAT proteins in thecytoplasm of various immune cells and consequently maintainnormal cell function upon activation. Thus, we hypothesized thatIFN-b signaling was likely impaired in response to Cardif dele-tion because Cardif2/2 NK cells have low intracellular levels ofSTAT1 (Fig. 3). In an attempt to rescue NK maturation, we treatedWT and Cardif2/2 NK cells with IFN-b in vitro. Similar to Bradyet al. (45), we noticed drastically reduced levels of CD11b ex-pression on NK cells in culture. This made it difficult to confi-dently identify the maturation stages of NK cells. We did, however,

measure total expression of CD11b and KLRG1 in the cultured NKcells. Both CD11b and KLRG1 expression significantly increasedwith IFN-b treatment in WTand Cardif2/2 NK cells (SupplementalFig. 3). Still, CD11b and KLRG1 levels were not restored to thoseof WT with IFN-b stimulation.

Cardif2/2 mice show enhanced NK-mediated control ofMCMV infection

NK cells are a key component of the host innate response requiredto control MCMV (46), which is a commonly used viral model forassessing NK effector function. MCMV replication in Cardif2/2

mice was controlled ∼10-fold better than that seen in WT mice atday 4 postinfection (Fig. 7A). However, infection with an MCMVmutant lacking the viral m157 protein (detected by Ly49h-expressing NK cells) resulted in no difference in viral controlbetween Cardif2/2 and WT mice (Fig. 7A) (36, 47–49). In ad-dition, depletion of NK cells before MCMV infection alsonormalized viral replication to WT levels in Cardif2 /2 mice(Supplemental Fig. 4). These data indicate the better control ofMCMV seen in Cardif2/2 mice is due to enhanced NK-mediatedeffector function(s). We have previously shown that a proportionof splenic NK cells express IFN-g at ∼12 h post MCMV infectionin response to the first wave of type I IFN produced, and this isa key time point regulating antiviral control in subsequent days(36, 50). Maslowski et al. (51) have also described an early pro-duction of IFN-g by NK cells that is STAT4- and type I IFN–dependent during LCMV infection. Consequently, we specificallychose to measure NK IFN-g production at 12 h after MCMV in-fection because type I IFN produced by infected splenic stromalcells is the primary cytokine driving this at this early time point, asopposed to plasmacytoid DC–produced IL-12, which is the pri-mary inducer of NK IFN-g at 36 h (36). We hypothesized that theSTAT1/STAT4 ratio present in Cardif2/2 NK cells might result inenhanced IFN-g production by NK cells at 12 h after MCMVinfection, potentially accounting for the observed increase in an-tiviral defense. Indeed, at 12 h postinfection, ∼4-fold highernumbers of NK cells in Cardif2/2 mice produced IFN-g than wasseen in WT mice (Fig. 7B). All Cardif2/2 NK cell subsets pro-duced more IFN-g+ when compared with their WT counterparts;however, the highest percentage of IFN-g+cells was seen in theless mature CD27+CD11b+ subset as compared with the moremature CD272CD11b+ NK cells. In addition to the CD27+

CD11b+ NK subset containing the highest percentage of IFN-g+

cells in MCMV-infected WT and Cardif2/2 spleens, Cardif2/2

spleens also contained nearly twice the normal proportion of thesecells (Fig. 7B). These data support a model where increasednumbers of CD27+CD11b+ NK cells in Cardif2/2 mice, as well asenhanced STAT1/STAT4 ratios, are largely responsible for theincreased IFN-g production observed in their NK cell compart-ment in response to initial MCMV infection, and likely underliethe enhanced control of MCMV infection in spleens of Cardif-deficient mice.

DiscussionWe have discovered a novel intrinsic role for Cardif in the mat-uration and function of NK cells. It is widely accepted that Cardif isvital for antiviral signaling in response to RIG-I and MDA5 ac-tivation; however, several studies indicate that Cardif and othercomponents in this pathway have relatively unexplored nonviralfunctions (13–16). Cardif has been shown to reduce the severity ofexperimental autoimmune encephalomyelitis in mice, activate theNLRP3 inflammasome, and regulate TLR7 expression in B cells.However, currently, no reports describe the requirement for, orinvolvement of, Cardif in NK cell development. In this report, we

FIGURE 5. Cardif2/2 mNK cells have decreased levels of STAT1 and

p-STAT1. (A) Western blot shows the amount of total STAT1 (top panel),

p-STAT1 (second panel), STAT4 (fourth panel), and p-STAT4 (bottom

panel) protein in WT (left) and Cardif2/2 (right) mNK cells. Cells were

treated with PBS or 100 U/ml IFN-b. Afterward, 30 mg protein from cell

lysates was loaded per well. b-Actin was used to normalize protein data.

(B) pSTAT1/STAT1 and p-STAT4/STAT4 ratios derived Western blot

quantification. Experiments were repeated twice. n = 4 mice/group/ex-

periment. Data are means 6 SEM.

The Journal of Immunology 2163

at Univ of C

alifornia-San Diego Serials/B

iomed L

ib 0699 on September 2, 2015

http://ww

w.jim

munol.org/

Dow

nloaded from

describe a novel and cell-intrinsic role for Cardif in the homeo-static maturation and function of NK cells.We discovered that Cardif mice have fewer splenic NK cells when

compared with WT mice, and Cardif2/2 NK cells are ∼40–45%more likely to lack CD49b, a marker of mNK cells. Splenic andliver Cardif2/2 NK cells populations also have ∼30 and ∼60%fewer terminally differentiated NK cells, respectively. Markers as-sociated with NK maturation, Ly49C/I and KLRG1, are also lowerin Cardif2/2 NK cells. Functionally, we found that Cardif2/2 NKcells had lower cytotoxic activity, because Cardif2/2 NK cells didnot degranulate nor produce as much granzyme B, nor kill targetcells and WT NK cells. However, we observed no changes in IFN-gproduction in Cardif-deficient NK cells. We did not measure pro-duction of other cytokines by Cardif NK cells in this study. BecauseSTAT1 has been shown to regulate the cytotoxicity functions of NKcells, whereas STAT4 governs IFN-g production by NK cells, wemeasured total STAT protein and protein phosphorylation. STAT1protein levels and phosphorylation were considerably reduced inCardif2/2 NK cells. In contrast, p-STAT4/STAT4 protein levelswere higher in Cardif2/2 NK cells. These data suggest that thediminished cytotoxicity of Cardif2/2 NK cells is due to the reducedintracellular levels of STAT1. Although there is an increase ofp-STAT4/STAT4 in Cardif2/2 NK cells, this increase may not havetranslated into increased levels of IFN-g in our ex vivo stimulationassays because the already high overall STAT4 levels in WT NKcells is saturating. However, the altered STAT1/STAT4 ratio could

have more dramatic consequences in the context of NK effectorfunction in vivo, as appears to be the case from our studies withMCMV.Cardif likely plays a role in stabilizing NK cells and preventing

their apoptosis. A recent report indicated that Cardif associateswith caspase-8 to induce apoptosis via caspase-3 in response toviral infection (52). Our data show that Cardif deficiency in-creases apoptosis in NK cells, particularly in terminally differenti-ated CD272CD11b+ NK cells. Moreover, our data suggest that thepresence of Cardif decreases NK cell proliferation, predominantlyin CD27+CD11b2 and CD27+CD11b+ NK cells. Cardif’s apparentrole in apoptosis and proliferation in NK cells explains much ofthe immature phenotype seen in NK cells from Cardif2/2 mice(Fig. 1). The lack of proliferation in CD27+CD11b2 and CD27+

CD11b+ NK cells suggests a lack of differentiation into terminallydifferentiated stage 4 NK cells. Together our data suggest thatCardif2/2 mice have fewer total NK cells, with a lower percentageof terminally differentiated CD272CD11b+ NK cells, and this islikely caused by an increase in CD272CD11b+ NK cell deathand a decrease in proliferation of early-stage CD27+CD11b2 andCD27+CD11b+ NK cells. NF-kB is a major downstream target ofCardif signaling along with type I IFN production (5, 6). NF-kB isknown to be involved in cell survival in multiple cell types, in-cluding B and T cells (53). Yet, little is known about NF-kB ac-tivity in NK cells. A few reports suggest that changes in NF-kBsignaling regulate NK cell proliferation, Ly49 expression, and

FIGURE 6. Cardif2/2 NK cells show maturation defect in 1:1 bone marrow chimeric mice. (A) Schematic diagram of method used to create 1:1 mixed

bone marrow chimeric mice. Cardif2/2 (CD45.2) and CD45.1 bone marrow were i.v. injected into CD45.1/0.2 recipient mice in a 1:1 ratio. Mice were

harvested 10 wk after injection. (B) Representative plot depicting equal amounts of WT and Cardif2/2 CD45+ cells in the bone marrow. (C) Representative

plots that show the ratios of CD45.1 and Cardif2/2 cells in the liver in specific populations: live cells → CD32CD192 → NK1.1+CD32CD192 → NK1.1+

CD32CD192 → NK1.1+CD32CD192CD272CD11b+. (D) Representative bar graphs show the contribution of cells from CD45.1 and Cardif2/2 origins to

total NK1.1+CD32CD192 cells in spleen and liver. Bar graphs (right panels) depict the distribution of WT (CD45.1) and Cardif2/2 NK cells based on the

four maturation stages, denoted by CD27 and CD11b in the spleen and liver of 1:1 mixed bone marrow chimeric mice. (E) Bar graphs depict the production

of CD107a (left panel) and granzyme B (right panel) by NK cells. (F) Bar graphs depict the percentage of Caspase-3+ NK cells. Representative plots and

bar graph depict the percentage of Caspase-3+ CD272CD11b+ NK cells (middle and right panels). (G) Bar graphs depict the percentage of Ki-67+ NK cells.

Experiments were performed twice (A–E) or once (F and G). n = 6 mice/group/experiment. Data are means 6 SEM. *p , 0.05, **p , 0.01, ***p , 0.001.

2164 CARDIF AND NK CELL MATURATION

at Univ of C

alifornia-San Diego Serials/B

iomed L

ib 0699 on September 2, 2015

http://ww

w.jim

munol.org/

Dow

nloaded from

increased apoptosis (53–55). NF-kB signaling is tightly regulatedin NK cells and changes in NF-kB activity are likely to perturbNK cell survival, and we surmise that the absence of a Cardif–NF-kB signaling axis may contribute to the differences observed inNK cell survival.Ourmixed bonemarrow chimera experiments revealed that Cardif is

intrinsically required for optimal terminal differentiation of NK cells.Moreover, Cardif is intrinsically required for the fitness and survival ofNK cells in peripheral organs, because Cardif2/2 NK cell numberswere lower than WT in the periphery of the mixed chimeric mice, andCD272CD11b+ terminally differentiated NK cells were nearly twiceas likely to be derived from WT mice in this mixed chimera scenario.We also found that increased apoptosis and decreased granzymeB levels in Cardif2/2 NK cells are due to cell-intrinsic deficiencies.The decreased proliferation and increased degranulation that wasobserved in Cardif2/2 mice, however, is not cell intrinsic.We initially thought that the cell-extrinsic factor acting upon

Cardif2/2 NK cells was IFN-b; however, we found that treatmentwith IFN-b in vitro, although somewhat effective, is not able tocompletely rescue the maturation phenotype that is present inCardif2/2 NK cells. Type I IFN is known to indirectly affect thedifferentiation of NK cells by upregulating the amount of IL-15and IL-15Ra present on DCs (56, 57). IL-15 transpresentation byDCs has been shown to have a dose-dependent effect on the de-velopment, maturation, and proliferation of NK and NKT cells(58–60). IL-15 transpresentation is required for the expressionof NK1.1 on NKT cells in the thymus. We have found that fewerNKT cells in the thymus express NK1.1 in young Cardif2/2 mice,which leads us to believe that there may be a perturbation of“basal” type I IFN production in Cardif2/2 mice that leads tolower amounts of IL-15 transpresentation (data not shown). Wealso found that Cardif2/2 DCs express lower amounts of IL-15/

IL-15Ra complexes when stimulated with TLR9 agonists (datanot shown). The involvement of IL-15 transpresentation in the NKphenotype of Cardif2/2 mice is also supported by results that theproliferation and cytotoxicity of Cardif2/2 NK cells is normal inWT hosts, two effector functions that are mediated by IL-15transpresentation. Thus, Cardif appears to regulate the matura-tion and function of NK cells by both direct and indirect mech-anisms. Further studies to delineate the role of type I IFNs in theterminal differentiation of NK cells in the absence of Cardif andother RIG-I molecules will be of interest, particularly in thecontext of viral immunity and autoimmune diseases.MCMV infection was used as another model to test the function of

Cardif2/2 NK cells, because these cells are key for innate defenseto this virus. Somewhat surprisingly, we found that Cardif2/2 micecontrolled MCMV replication better than WT mice at times of peakacute infection, and this enhanced control in the absence of Cardifwas shown to be NK dependent by both depleting NK cells and usingan MCMVΔm157 mutant virus (47–49). This amplified innate anti-viral defense was consistent with an increased percentage of IFN-g+

NK cells in the spleen seen at 12 h of infection, a time point wheninnate defense to MCMV is “kick-started” and the baseline forMCMV replication is being established (61). This initial IFN-gproduction by NK cells occurs in response to type I IFN produced byMCMV-infected marginal zone stromal cells in the spleen, a cell typefor which MCMV shows specific tropism because of its differentia-tion by constitutive lymphotoxin-ab signaling (36, 50). Data fromMack et al. (32) also show that very early after LCMV infection,a short burst of IFN-g is produced by NK cells via the type I IFNpathway, independent of IL-12. In turn, Doring et al. (62) have shownthat Cardif2/2 plasmacytoid DCs, and macrophages produce similarlevels of IFN-a in response to MCMV, cells that produce the secondwave of innate cytokines in the spleen in response to MCMV gen-

FIGURE 7. Cardif2/2 mice show

heightened NK activation and en-

hanced control of MCMV infection.

(A) WT and Dm157 MCMV replica-

tion levels in spleens of day 4 infected

WT and Cardif2/2 mice. (B) Repre-

sentative plots of IFN-g production by

NK cells at 12 h in MCMV-infected

mice (upper panels), and bar graphs

quantifying this in multiple mice and

in NK cell subsets (lower panels). Bar

graph portraying the percentage of

IFN-g+ cells within NK cell subsets in

the spleens of mice 12 h postinfection

(lower right panel). n $ 3 mice/

group. Experiments were performed

twice with at least three mice/group.

(A) Combined data from two separate

experiments, with (B) representative of

a single experiment. Data are means6SEM. *p , 0.05, ***p , 0.001.

The Journal of Immunology 2165

at Univ of C

alifornia-San Diego Serials/B

iomed L

ib 0699 on September 2, 2015

http://ww

w.jim

munol.org/

Dow

nloaded from

erated from the infected marginal zone stroma (61). Together, thissuggests that instead of type I IFN being altered in MCMV-infectedCardif2/2 mice, which is perhaps not surprising because MCMV isa dsDNA virus, that Cardif2/2 NK cells are more highly subject toactivation by these key innate cytokines. Hayakawa et al. (22) foundthat CD27+CD11b+ NK cells produced more IFN-g than CD272

CD11b+ NK cells in response to cytokines and DC cross talk. We alsofound higher IFN-g production by CD27+CD11b+ NK cells in re-sponse to MCMV, accounting for a substantial overall increase in NKcell IFN-g seen in Cardif2/2 mice. This suggests that the CD272

CD11b+ NK subset may preferentially contribute to MCMV controlin some contexts, although this needs to be tested more directly. Inaddition, the prevalence of more unlicensed, Ly49C/I2 NK cellsin Cardif2/2 mice may also contribute to the observed enhancedcontrol of MCMV, because these cells are activated to a higherextent in the response to this virus because of fewer inhibitorymechanisms (63, 64). In contrast, these unlicensed Ly49C/I2 NKcells are not as functional as licensed NK cells in contexts otherthan viral infection, such as tumor immunosurveillance, contributingto the reduced cytotoxicity of Cardif2/2 NK cells seen in vitro (39).Also, lower STAT1/STAT4 ratios might prolong IFN-g productionby Cardif2/2 NK, because it could take longer to induce highenough STAT1 levels in response to type I IFN to downregulateSTAT4.Increased Cardif NK cell activation in the context of MCMV in-

fection may be indicative of a potentially broader role of Cardif inregulating autoimmunity. Indeed, Cardif has recently been associatedwith incidence of systemic lupus erythematosus (SLE). Pothlichetet al. (65) described a loss-of-function mitochondrial antiviral signal-ing (Cardif) variant in humans that is associated with SLE in AfricanAmericans. Patients with this variant were characterized by low type IIFN levels and a lack of autoantibodies specific for RNA-bindingprotein. Similarly, Molineros et al. (66) described a risk allele ofIFIH1 (MDA5) in African Americans that is associated with an in-creased risk for SLE with downregulation of type I IFN signaling.Interestingly, SLE patients have reduced absolute numbers and a re-duction in the cytolytic activity of NK cells, similar to the pheno-type observed in Cardif2/2 mice (67–71). Furthermore, just as inCardif2/2 mice, the NK cells of SLE patients have an increase inthe number of iNK cells and a concomitant decrease of maturecytotoxic NK cells. Future studies should investigate Cardif sig-naling in NK cells in the context of SLE to further delineate theeffect of Cardif on NK cells and SLE pathogenesis.Our data suggest that Cardif negatively regulates NK cell matu-

ration and function in the absence of viral infection but may result intheir hyperreactivity in the context of pathogen infection. Several otherpublications have suggested a similar “dual role” for Cardif; however,clear mechanisms have not been established (7, 15, 16, 65). Withregard to the role of Cardif in regulating immune cell homeostasis,tonic signaling through the Rig-I/MDA5–Cardif axis triggered byendogenous agonists may play a role. There have been reports ofendogenous agonists of Rig-I such as the small self-RNAs producedby RNase-L, IRE1a, and endogenous retroviral elements in the hu-man genome (72–77). Recently, Dupuis-Maurin et al. (78) discoveredthat overexpression of the transcription factor specificity protein1 activates the Rig-I–Cardif pathway in the absence of a pathogen bystimulating the 29-59-oligoadenylate synthetase 1–RNase-L pathwayto produce small self-RNAs. Tonic Rig-I/MDA5–Cardif signalingactivity stimulated by specificity protein 1 or other transcriptionfactors may lead to low-level production of IFN-b and NF-kB ac-tivation. It is also possible that Cardif is involved in other signalingcascades rather than the RIG-I/MDA5–Cardif axis, because Sub-ramanian et al. (14) recently reported the activity of Cardif in theNLRP3 inflammasome.

AcknowledgmentsWe thank Dr. Alec Redwood (Murdoch University) for providing the

MCMV strain. We thank Amy Blatchley and Deborah Yoakum at La Jolla

Institute for Allergy and Immunology for assistance with mice and the Im-

aging Facility at La Jolla Institute for Allergy and Immunology for excellent

technical assistance with flow cytometry.

DisclosuresThe authors have no financial conflicts of interest.

References1. Takeuchi, O., and S. Akira. 2010. Pattern recognition receptors and inflamma-

tion. Cell 140: 805–820.2. Ireton, R. C., and M. Gale, Jr. 2011. RIG-I like receptors in antiviral immunity

and therapeutic applications. Viruses 3: 906–919.3. Eisenacher, K., and A. Krug. 2012. Regulation of RLR-mediated innate immune

signaling–it is all about keeping the balance. Eur. J. Cell Biol. 91: 36–47.4. Meylan, E., J. Curran, K. Hofmann, D. Moradpour, M. Binder,

R. Bartenschlager, and J. Tschopp. 2005. Cardif is an adaptor protein in the RIG-I antiviral pathway and is targeted by hepatitis C virus. Nature 437: 1167–1172.

5. Seth, R. B., L. Sun, C.-K. Ea, and Z. J. Chen. 2005. Identification and charac-terization of MAVS, a mitochondrial antiviral signaling protein that activatesNF-kappaB and IRF 3. Cell 122: 669–682.

6. Kawai, T., K. Takahashi, S. Sato, C. Coban, H. Kumar, H. Kato, K. J. Ishii,O. Takeuchi, and S. Akira. 2005. IPS-1, an adaptor triggering RIG-I- and Mda5-mediated type I interferon induction. Nat. Immunol. 6: 981–988.

7. Xu, L.-G., Y.-Y. Wang, K.-J. Han, L.-Y. Li, Z. Zhai, and H.-B. Shu. 2005. VISAis an adapter protein required for virus-triggered IFN-beta signaling. Mol. Cell19: 727–740.

8. Kawai, T., and S. Akira. 2006. TLR signaling. Cell Death Differ. 13: 816–825.9. Takamatsu, S., K. Onoguchi, K. Onomoto, R. Narita, K. Takahasi, F. Ishidate,

T. K. Fujiwara, M. Yoneyama, H. Kato, and T. Fujita. 2013. Functional char-acterization of domains of IPS-1 using an inducible oligomerization system.PLoS One 8: e53578.

10. Loo, Y.-M., and M. Gale, Jr. 2011. Immune signaling by RIG-I-like receptors.Immunity 34: 680–692.

11. Li, X. D., Y. H. Chiu, A. S. Ismail, C. L. Behrendt, M. Wight-Carter,L. V. Hooper, and Z. J. Chen. 2011. Mitochondrial antiviral signaling protein(MAVS) monitors commensal bacteria and induces an immune response thatprevents experimental colitis. Proc. Natl. Acad. Sci. USA 108: 17390–17395.

12. Wang, Y., H.-X. Zhang, Y.-P. Sun, Z.-X. Liu, X.-S. Liu, L. Wang, S.-Y. Lu,H. Kong, Q.-L. Liu, X.-H. Li, et al. 2007. Rig-I-/- mice develop colitis associatedwith downregulation of G alpha i2. Cell Res. 17: 858–868.

13. Xu, L.-G., L. Jin, B.-C. Zhang, L. J. Akerlund, H.-B. Shu, and J. C. Cambier. 2012.VISA is required for B cell expression of TLR7. J. Immunol. 188: 248–258.

14. Subramanian, N., K. Natarajan, M. R. Clatworthy, Z. Wang, and R. N. Germain.2013. The adaptor MAVS promotes NLRP3 mitochondrial localization andinflammasome activation. Cell 153: 348–361.

15. Wang, F., W. Xia, F. Liu, J. Li, G. Wang, and J. Gu. 2012. Interferon regulatorfactor 1/retinoic inducible gene I (IRF1/RIG-I) axis mediates 25-hydroxycholesterol-induced interleukin-8 production in atherosclerosis.Cardiovasc. Res. 93: 190–199.

16. Dann, A., H. Poeck, A. L. Croxford, S. Gaupp, K. Kierdorf, M. Knust, D. Pfeifer,C. Maihoefer, S. Endres, U. Kalinke, et al. 2012. Cytosolic RIG-I-like helicases actas negative regulators of sterile inflammation in the CNS.Nat. Neurosci. 15: 98–106.

17. Vivier, E., E. Tomasello, M. Baratin, T. Walzer, and S. Ugolini. 2008. Functionsof natural killer cells. Nat. Immunol. 9: 503–510.

18. Gordon, S. M., J. Chaix, L. J. Rupp, J. Wu, S. Madera, J. C. Sun, T. Lindsten, andS. L. Reiner. 2012. The transcription factors T-bet and Eomes control keycheckpoints of natural killer cell maturation. Immunity 36: 55–67.

19. Daussy, C., F. Faure, K. Mayol, S. Viel, G. Gasteiger, E. Charrier, J. Bienvenu,T. Henry, E. Debien, U. A. Hasan, et al. 2014. T-bet and Eomes instruct thedevelopment of two distinct natural killer cell lineages in the liver and in thebone marrow. J. Exp. Med. 211: 563–577.

20. Fathman, J. W., D. Bhattacharya, M. A. Inlay, J. Seita, H. Karsunky, andI. L. Weissman. 2011. Identification of the earliest natural killer cell-committedprogenitor in murine bone marrow. Blood 118: 5439–5447.

21. De Colvenaer, V., S. Taveirne, M. Delforche, M. De Smedt, B. Vandekerckhove,T. Taghon, L. Boon, J. Plum, and G. Leclercq. 2011. CD27-deficient mice shownormal NK-cell differentiation but impaired function upon stimulation. Immunol.Cell Biol. 89: 803–811.

22. Hayakawa, Y., and M. J. Smyth. 2006. CD27 dissects mature NK cells into two subsetswith distinct responsiveness and migratory capacity. J. Immunol. 176: 1517–1524.

23. Chiossone, L., J. Chaix, N. Fuseri, C. Roth, E. Vivier, and T. Walzer. 2009. Maturationof mouse NK cells is a 4-stage developmental program. Blood 113: 5488–5496.

24. Fu, B., F. Wang, R. Sun, B. Ling, Z. Tian, and H. Wei. 2011. CD11b and CD27reflect distinct population and functional specialization in human natural killercells. Immunology 133: 350–359.

25. Huntington, N. D., H. Tabarias, K. Fairfax, J. Brady, Y. Hayakawa, M. A. Degli-Esposti, M. J. Smyth, D. M. Tarlinton, and S. L. Nutt. 2007. NK cell maturationand peripheral homeostasis is associated with KLRG1 up-regulation. J. Immunol.178: 4764–4770.

2166 CARDIF AND NK CELL MATURATION

at Univ of C

alifornia-San Diego Serials/B

iomed L

ib 0699 on September 2, 2015

http://ww

w.jim

munol.org/

Dow

nloaded from

26. Nguyen, K. B., T. P. Salazar-Mather, M. Y. Dalod, J. B. Van Deusen, X. Q. Wei,F. Y. Liew, M. A. Caligiuri, J. E. Durbin, and C. A. Biron. 2002. Coordinated anddistinct roles for IFN-alpha beta, IL-12, and IL-15 regulation of NK cell re-sponses to viral infection. J. Immunol. 169: 4279–4287.

27. Swann, J. B., Y. Hayakawa, N. Zerafa, K. C. F. Sheehan, B. Scott, R. D. Schreiber,P. Hertzog, and M. J. Smyth. 2007. Type I IFN contributes to NK cell homeostasis,activation, and antitumor function. J. Immunol. 178: 7540–7549.

28. Mizutani, T., N. Neugebauer, E. M. Putz, N. Moritz, O. Simma, E. Zebedin-Brandl,D. Gotthardt, W. Warsch, E. Eckelhart, H. P. Kantner, et al. 2012. ConditionalIFNAR1 ablation reveals distinct requirements of Type I IFN signaling for NK cellmaturation and tumor surveillance. OncoImmunology 1: 1027–1037.

29. Lee, C. K., D. T. Rao, R. Gertner, R. Gimeno, A. B. Frey, and D. E. Levy. 2000. Distinctrequirements for IFNs and STAT1 in NK cell function. J. Immunol. 165: 3571–3577.

30. Gough, D. J., N. L. Messina, C. J. P. Clarke, R. W. Johnstone, and D. E. Levy.2012. Constitutive type I interferon modulates homeostatic balance through tonicsignaling. Immunity 36: 166–174.

31. Miyagi, T., M. P. Gil, X. Wang, J. Louten, W.-M. Chu, and C. A. Biron. 2007.High basal STAT4 balanced by STAT1 induction to control type 1 interferoneffects in natural killer cells. J. Exp. Med. 204: 2383–2396.

32. Mack, E. A., L. E. Kallal, D. A. Demers, and C. A. Biron. 2011. Type 1 inter-feron induction of natural killer cell gamma interferon production for defenseduring lymphocytic choriomeningitis virus infection. MBio 2: e00169–e11.

33. Michallet, M.-C., E. Meylan, M. A. Ermolaeva, J. Vazquez, M. Rebsamen,J. Curran, H. Poeck, M. Bscheider, G. Hartmann, M. Konig, et al. 2008. TRADDprotein is an essential component of the RIG-like helicase antiviral pathway.Immunity 28: 651–661.

34. Hanna, R. N., L. M. Carlin, H. G. Hubbeling, D. Nackiewicz, A. M. Green,J. A. Punt, F. Geissmann, and C. C. Hedrick. 2011. The transcription factorNR4A1 (Nur77) controls bone marrow differentiation and the survival of Ly6C-monocytes. Nat. Immunol. 12: 778–785.

35. McWhorter, A. R., L. M. Smith, L. L. Masters, B. Chan, G. R. Shellam, andA. J. Redwood. 2013. Natural killer cell dependent within-host competitionarises during multiple MCMV infection: consequences for viral transmission andevolution. PLoS Pathog. 9: e1003111.

36. Verma, S., Q. Wang, G. Chodaczek, and C. A. Benedict. 2013. Lymphoid-tissuestromal cells coordinate innate defense to cytomegalovirus. J. Virol. 87: 6201–6210.

37. Seillet, C., N. D. Huntington, P. Gangatirkar, E. Axelsson, M. Minnich, H. J. M. Brady,M. Busslinger, M. J. Smyth, G. T. Belz, and S. Carotta. 2014. Differential requirementfor Nfil3 during NK cell development. J. Immunol. 192: 2667–2676.

38. Sun, J. C. 2010. Re-educating natural killer cells. J. Exp. Med. 207: 2049–2052.39. Kim, S., J. Poursine-Laurent, S. M. Truscott, L. Lybarger, Y.-J. Song, L. Yang,

A. R. French, J. B. Sunwoo, S. Lemieux, T. H. Hansen, and W. M. Yokoyama.2005. Licensing of natural killer cells by host major histocompatibility complexclass I molecules. Nature 436: 709–713.

40. Yokoyama, W. M., and S. Kim. 2006. Licensing of natural killer cells by self-major histocompatibility complex class I. Immunol. Rev. 214: 143–154.

41. Robbins, S. H., M. S. Tessmer, L. Van Kaer, and L. Brossay. 2005. Direct effectsof T-bet and MHC class I expression, but not STAT1, on peripheral NK cellmaturation. Eur. J. Immunol. 35: 757–765.

42. Fernandez, N. C., E. Treiner, R. E. Vance, A. M. Jamieson, S. Lemieux, andD. H. Raulet. 2005. A subset of natural killer cells achieves self-tolerancewithout expressing inhibitory receptors specific for self-MHC molecules.Blood 105: 4416–4423.

43. Beuneu, H., J. Deguine, I. Bouvier, J. P. Di Santo, M. L. Albert, and P. Bousso.2011. Cutting edge: a dual role for type I IFNs during polyinosinic-polycytidylicacid-induced NK cell activation. J. Immunol. 187: 2084–2088.

44. Marcais, A., S. Viel, M. Grau, T. Henry, J. Marvel, and T. Walzer. 2013. Reg-ulation of mouse NK cell development and function by cytokines. Front.Immunol. 4: 450.

45. Brady, J., S. Carotta, R. P. L. Thong, C. J. Chan, Y. Hayakawa, M. J. Smyth, andS. L. Nutt. 2010. The interactions of multiple cytokines control NK cell matu-ration. J. Immunol. 185: 6679–6688.

46. Biron, C. A., K. B. Nguyen, G. C. Pien, L. P. Cousens, and T. P. Salazar-Mather.1999. Natural killer cells in antiviral defense: function and regulation by innatecytokines. Annu. Rev. Immunol. 17: 189–220.

47. Bubic, I., M. Wagner, A. Krmpotic, T. Saulig, S. Kim, W. M. Yokoyama,S. Jonjic, and U. H. Koszinowski. 2004. Gain of virulence caused by loss ofa gene in murine cytomegalovirus. J. Virol. 78: 7536–7544.

48. Orr, M. T., J. C. Sun, D. G. T. Hesslein, H. Arase, J. H. Phillips, T. Takai, andL. L. Lanier. 2009. Ly49H signaling through DAP10 is essential for optimal naturalkiller cell responses to mouse cytomegalovirus infection. J. Exp. Med. 206: 807–817.

49. Brown, M. G., A. O. Dokun, J. W. Heusel, H. R. Smith, D. L. Beckman,E. A. Blattenberger, C. E. Dubbelde, L. R. Stone, A. A. Scalzo, andW. M. Yokoyama.2001. Vital involvement of a natural killer cell activation receptor in resistance to viralinfection. Science 292: 934–937.

50. Schneider, K., A. Loewendorf, C. De Trez, J. Fulton, A. Rhode, H. Shumway,S. Ha, G. Patterson, K. Pfeffer, S. A. Nedospasov, et al. 2008. Lymphotoxin-mediated crosstalk between B cells and splenic stroma promotes the initial type Iinterferon response to cytomegalovirus. Cell Host Microbe 3: 67–76.

51. Maslowski, K. M., A. T. Vieira, A. Ng, J. Kranich, F. Sierro, D. Yu, H. C. Schilter,M. S. Rolph, F. Mackay, D. Artis, et al. 2009. Regulation of inflammatory responsesby gut microbiota and chemoattractant receptor GPR43. Nature 461: 1282–1286.

52. El Maadidi, S., L. Faletti, B. Berg, C. Wenzl, K. Wieland, Z. J. Chen, U. Maurer,and C. Borner. 2014. A novel mitochondrial MAVS/Caspase-8 platform linksRNA virus-induced innate antiviral signaling to Bax/Bak-independent apoptosis.J. Immunol. 192: 1171–1183.

53. Samson, S. I., S. Memet, C. A. Vosshenrich, F. Colucci, O. Richard, D. Ndiaye,A. Israel, and J. P. Di Santo. 2004. Combined deficiency in IkappaBalpha andIkappaBepsilon reveals a critical window of NF-kappaB activity in natural killercell differentiation. Blood 103: 4573–4580.

54. Pascal, V., N. R. Nathan, E. Claudio, U. Siebenlist, and S. K. Anderson. 2007.NF-kappa B p50/p65 affects the frequency of Ly49 gene expression by NK cells.J. Immunol. 179: 1751–1759.

55. Tato, C. M., N. Mason, D. Artis, S. Shapira, J. C. Caamano, J. H. Bream, H.-C. Liou,and C. A. Hunter. 2006. Opposing roles of NF-kappaB family members in the reg-ulation of NK cell proliferation and production of IFN-gamma. Int. Immunol. 18:505–513.

56. Lucas, M., W. Schachterle, K. Oberle, P. Aichele, and A. Diefenbach. 2007.Dendritic cells prime natural killer cells by trans-presenting interleukin 15.Immunity 26: 503–517.

57. Baranek, T., T.-P. V. Manh, Y. Alexandre, M. A. Maqbool, J. Z. Cabeza,E. Tomasello, K. Crozat, G. Bessou, N. Zucchini, S. H. Robbins, et al. 2012.Differential responses of immune cells to type I interferon contribute to hostresistance to viral infection. [Published erratum appears in 2013 Cell Host Mi-crobe 13: 372.] Cell Host Microbe 12: 571–584.

58. Lee, G. A., Y.-H. Liou, S.-W. Wang, K.-L. Ko, S.-T. Jiang, and N.-S. Liao. 2011.Different NK cell developmental events require different levels of IL-15 trans-presentation. J. Immunol. 187: 1212–1221.

59. Castillo, E. F., and K. S. Schluns. 2012. Regulating the immune system via IL-15transpresentation. Cytokine 59: 479–490.

60. Gordy, L. E., J. S. Bezbradica, A. I. Flyak, C. T. Spencer, A. Dunkle, J. Sun,A. K. Stanic, M. R. Boothby, Y.-W. He, Z. Zhao, et al. 2011. IL-15 regulateshomeostasis and terminal maturation of NKT cells. J. Immunol. 187: 6335–6345.

61. Verma, S., and C. A. Benedict. 2011. Sources and signals regulating type I in-terferon production: lessons learned from cytomegalovirus. J. Interferon Cyto-kine Res. 31: 211–218.

62. Doring, M., I. Lessin, T. Frenz, J. Spanier, A. Kessler, P. Tegtmeyer, F. Da�g,N. Thiel, M. Trilling, S. Lienenklaus, et al. 2014. M27 expressed by cytomeg-alovirus counteracts effective type I interferon induction of myeloid cells but notof plasmacytoid dendritic cells. J. Virol. 88: 13638–13650.

63. Orr, M. T., W. J. Murphy, and L. L. Lanier. 2010. ‘Unlicensed’ natural killer cellsdominate the response to cytomegalovirus infection. Nat. Immunol. 11: 321–327.

64. Orr, M. T., and L. L. Lanier. 2011. Natural killer cell licensing during viralinfection. In Crossroads between Innate and Adaptive Immunity III. Advances inExperimental Medicine and Biology, Vol. 780. B. Pulendran, P. D. Katsikis, andS. P. Schoenberger, eds. New York: Springer, p. 37–44.

65. Pothlichet, J., T. B. Niewold, D. Vitour, B. Solhonne, M. K. Crow, and M. Si-Tahar. 2011. A loss-of-function variant of the antiviral molecule MAVS is as-sociated with a subset of systemic lupus patients. EMBO Mol. Med. 3: 142–152.

66. Molineros, J. E., A. K. Maiti, C. Sun, L. L. Looger, S. Han, X. Kim-Howard,S. Glenn, A. Adler, J. A. Kelly, T. B. Niewold, et al; BIOLUPUS Network. 2013.Admixture mapping in lupus identifies multiple functional variants within IFIH1associated with apoptosis, inflammation, and autoantibody production. PLoSGenet. 9: e1003222.

67. Poggi, A., and M. R. Zocchi. 2014. NK cell autoreactivity and autoimmunediseases. Front. Immunol. 5: 27.

68. Park, Y.-W., S.-J. Kee, Y.-N. Cho, E.-H. Lee, H.-Y. Lee, E.-M. Kim, M.-H. Shin,J.-J. Park, T.-J. Kim, S.-S. Lee, et al. 2009. Impaired differentiation and cyto-toxicity of natural killer cells in systemic lupus erythematosus. Arthritis Rheum.60: 1753–1763.

69. Henriques, A., L. Teixeira, L. Ines, T. Carvalheiro, A. Goncalves, A. Martinho,M. L. Pais, J. A. P. da Silva, and A. Paiva. 2013. NK cells dysfunction in sys-temic lupus erythematosus: relation to disease activity. Clin. Rheumatol. 32:805–813.

70. Hervier, B., V. Beziat, J. Haroche, A. Mathian, P. Lebon, P. Ghillani-Dalbin,L. Musset, P. Debre, Z. Amoura, and V. Vieillard. 2011. Phenotype and functionof natural killer cells in systemic lupus erythematosus: excess interferon-gproduction in patients with active disease. Arthritis Rheum. 63: 1698–1706.

71. Schepis, D., I. Gunnarsson, M.-L. Eloranta, J. Lampa, S. H. Jacobson, K. Karre,and L. Berg. 2009. Increased proportion of CD56bright natural killer cells inactive and inactive systemic lupus erythematosus. Immunology 126: 140–146.

72. Malathi, K., B. Dong, M. Gale, Jr., and R. H. Silverman. 2007. Small self-RNAgenerated by RNase L amplifies antiviral innate immunity. Nature 448: 816–819.

73. Ezelle, H. J., and B. A. Hassel. 2012. Pathologic effects of RNase-L dysregu-lation in immunity and proliferative control. Front. Biosci. (Schol Ed). 4: 767–786.

74. Cho, J. A., A. H. Lee, B. Platzer, B. C. Cross, B. M. Gardner, H. De Luca,P. Luong, H. P. Harding, L. H. Glimcher, P. Walter, et al. 2013. The unfoldedprotein response element IRE1a senses bacterial proteins invading the ER toactivate RIG-I and innate immune signaling. Cell Host Microbe 13: 558–569.

75. Malathi, K., T. Saito, N. Crochet, D. J. Barton, M. Gale, Jr., and R. H. Silverman.2010. RNase L releases a small RNA from HCV RNA that refolds into a potentPAMP. RNA 16: 2108–2119.

76. Zeng, M., Z. Hu, X. Shi, X. Li, X. Zhan, X. D. Li, J. Wang, J. H. Choi,K. W. Wang, T. Purrington, et al. 2014. MAVS, cGAS, and endogenous retro-viruses in T-independent B cell responses. Science 346: 1486–1492.

77. Grasset, E. K., and A. Cerutti. 2014. Immunology. Retroviral help for B cells.Science 346: 1454–1455.

78. Dupuis-Maurin, V., L. Brinza, J. Baguet, E. Plantamura, S. Schicklin,S. Chambion, C. Macari, M. Tomkowiak, E. Deniaud, Y. Leverrier, et al. 2015.Overexpression of the transcription factor Sp1 activates the OAS-RNAse L-RIG-I pathway. PLoS One 10: e0118551.

The Journal of Immunology 2167

at Univ of C

alifornia-San Diego Serials/B

iomed L

ib 0699 on September 2, 2015

http://ww

w.jim

munol.org/

Dow

nloaded from