Poxvirus Infection-Associated Downregulation of C-Type Lectin

of April 10, 2019.This information is current as

Protein-1BPrevents NK Cell Inhibition by NK ReceptorDownregulation of C-Type Lectin-Related-b Poxvirus Infection-Associated

BurshtynOscar A. Aguilar, Li Fu, James R. Carlyle and Deborah N. Kinola J. N. Williams, Evan Wilson, Chelsea L. Davidson,

ol.1103425http://www.jimmunol.org/content/early/2012/04/09/jimmun

published online 9 April 2012J Immunol 

MaterialSupplementary

5.DC1http://www.jimmunol.org/content/suppl/2012/04/10/jimmunol.110342

        average*  

4 weeks from acceptance to publicationFast Publication! •    

Every submission reviewed by practicing scientistsNo Triage! •    

from submission to initial decisionRapid Reviews! 30 days* •    

Submit online. ?The JIWhy

Subscriptionhttp://jimmunol.org/subscription

is online at: The Journal of ImmunologyInformation about subscribing to

Permissionshttp://www.aai.org/About/Publications/JI/copyright.htmlSubmit copyright permission requests at:

Email Alertshttp://jimmunol.org/alertsReceive 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 © 2012 by The American Association of1451 Rockville Pike, Suite 650, Rockville, MD 20852The American Association of Immunologists, Inc.,

is published twice each month byThe Journal of Immunology

by guest on April 10, 2019

http://ww

w.jim

munol.org/

Dow

nloaded from

by guest on April 10, 2019

http://ww

w.jim

munol.org/

Dow

nloaded from

The Journal of Immunology

Poxvirus Infection-Associated Downregulation of C-TypeLectin-Related-b Prevents NK Cell Inhibition by NKReceptor Protein-1B

Kinola J. N. Williams,* Evan Wilson,* Chelsea L. Davidson,* Oscar A. Aguilar,† Li Fu,*

James R. Carlyle,† and Deborah N. Burshtyn*

Innate immune recognition of virus-infected cells includes NK cell detection of changes to endogenous cell-surface proteins through

inhibitory receptors. One such receptor system is the NK cell receptor protein-1B (NKR-P1B) and its ligand C-type lectin-related-

b (Clr-b). NKR-P1B and Clr-b are encoded within the NK cell gene complex, a locus that has been linked to strain-dependent

differences in susceptibility to infection by poxviruses. In this study, we report the impact of vaccinia virus (VV) and ectromelia

virus infection on expression of Clr-b and Clr-b–mediated protection from NK cells. We observed a loss of Clr-b cell-surface

protein upon VVand ectromelia virus infection of murine cell lines and bone marrow-derived macrophages. The reduction of Clr-

b is more rapid than MHC class I, the prototypic ligand of NK cell inhibitory receptors. Reduction of Clr-b requires active viral

infection but not expression of late viral genes, and loss of mRNA appears to lag behind loss of Clr-b surface protein. Clr-b–

mediated protection from NK cells is lost following VV infection. Together, these results provide the second example of Clr-

b modulation during viral infection and suggest reductions of Clr-b may be involved in sensitizing poxvirus-infected cells to NK

cells. The Journal of Immunology, 2012, 188: 000–000.

Natural killer cells are innate lymphocytes that play animportant role in the defense against viruses by directlylysing infected cells, secreting antiviral cytokines, and

modulating the adaptive immune response (1). NK cells use a va-riety of activating receptors that interact with markers of cellularstress and inhibitory receptors that interact with endogenousligands to discriminate potential target cells from healthy cells (2).NK cells can be triggered by the increase of stimulatory ligands,the decrease of inhibitory ligands, or a combination of bothmechanisms. The first and best described of these receptor systemsis the inhibition of NK cells by classical and nonclassical MHCclass I (MHC-I) proteins. This recognition system that sensesmissing self is believed to have evolved to allow NK cells to detectvirus-infected cells, because many viruses interfere with normal

expression of MHC-I proteins to evade T cell responses (3, 4). Thereceptor systems for MHC-I have undergone considerable selectivepressure and have diverged significantly between species, and thisdivergence is exemplified by the differences between humans androdents. The receptors that bind to classical MHC-I in humansbelong to the Ig superfamily and are named killer cell Ig-likereceptors (5); in contrast, the parallel receptors in mice belong tothe C-type lectin family of receptors and are called Ly49 receptors(6). Mice and humans share the CD94/NKG2 heterodimeric C-typelectin-like receptors that detect nonclassical MHC-I (HLA-E andQa1) (2). HLA-E and Qa1 are not typically involved in T cellactivation but serve as surrogates of production of classicalMHC-I proteins, as they depend on MHC-I to provide their pep-tide. The NKG2 family encodes both inhibitory and activatingreceptors that pair with the invariant CD94 chain. Viruses haveevolved several mechanisms to evade these systems of detectionby NK cells, including expression of MHC-I mimics and pro-viding peptides for HLA-E (7–9).Humans and mice also share the NKR-P1 family of receptors that

recognize endogenous proteins, but the ligands for the NKR-P1receptors are unrelated to MHC-I and are not directly involvedin Ag presentation (10). Instead, NKR-P1 receptors interact withC-type lectin-related proteins encoded within the NK gene com-plex (NKC) (11, 12). The mouse NKC also encodes the NKR-P1receptors, allowing for coinheritance within haplotypes along withthe CD94/NKG2 and Ly49 families of receptors (6, 13). There areseveral NKR-P1 receptors and putative ligands within the NKC,and the region exhibits polymorphism between mouse strains thatincludes protein variation and relative expansion or contractionof the gene family (6, 13). The best understood receptor is NKRprotein-1B (NKR-P1B; also known as NKR-P1D in B6 mice),which recognizes C-type lectin-related-b (Clr-b). Clr-b is broadlyexpressed on normal cells, but frequently reduced on transformedcell lines that are susceptible to NK killing (11). Clr-b is alsoknown as osteoclast inhibitory lectin, and disruption of osteoclast

*Department of Medical Microbiology and Immunology, University of Alberta,Edmonton, Alberta T6G 2S2, Canada; and †Department of Immunology, SunnybrookResearch Institute, University of Toronto, Toronto, Ontario M4N 3M5, Canada

Received for publication December 2, 2011. Accepted for publication March 3, 2012.

This work was supported by Operating Grants from the Canadian Institutes forHealth Research (to D.N.B. [MOP 36344] and J.R.C. [MOP 74754]). D.N.B. is anAlberta Innovates Health Solutions-Alberta Heritage Foundation for Medical Re-search Senior Scholar. J.R.C. is supported by an Ontario Ministry of Research andInnovation Early Researcher Award, a Canadian Institutes for Health Research NewInvestigator Award, and an Investigator in the Pathogenesis of Infectious DiseaseAward from the Burroughs Wellcome Fund. K.J.N.W. is supported by a studentshipfrom the Canadian Liver Foundation.

Address correspondence and reprint requests to Dr. Deborah Burshtyn, Departmentof Medical Microbiology and Immunology, 659 Heritage Medical Research Centre,University of Alberta, Edmonton, AB T6G 2S2, Canada. E-mail address: [email protected]

The online version of this article contains supplemental material.

Abbreviations used in this article: AraC, cytosine b-D-arabinofuranoside; BMMø,bone marrow-derived macrophage; Clr-b, C-type lectin-related-b; ECTV, ectromeliavirus; EGFP, enhanced GFP; MHC-I, MHC class I; MOI, multiplicity of infection;NKC, NK gene complex; NKR-P1B, NKR protein-1B; RCMV, rat CMV; SA, strep-tavidin; VV, vaccinia virus.

Copyright� 2012 by The American Association of Immunologists, Inc. 0022-1767/12/$16.00

www.jimmunol.org/cgi/doi/10.4049/jimmunol.1103425

Published April 9, 2012, doi:10.4049/jimmunol.1103425 by guest on A

pril 10, 2019http://w

ww

.jimm

unol.org/D

ownloaded from

inhibitory lectin/Clr-b in mice produces mild osteopenia but noapparent gross abnormalities in development of the immune sys-tem (14). Understanding of the role of the NKR-P1 receptor sys-tem in defense against viral infection is still emerging; however,a rat CMV (RCMV)-English isolate provides a clear example ofits importance. RCMV infection causes downregulation of Clr-bon rat cells, and consequently, RCMV possesses a Clr-b decoyhomolog (RCTL) that engages rat NKR-P1B to protect RCMV-infected cells from NK attack (15). The mechanism of Clr-bdownregulation during RCMV infection remains unknown, butother studies have shown that Clr-b is downregulated followingcellular stress, suggesting cells may have a host pathway to alertNK cells by deliberately downregulating Clr-b during infection(16).Poxviruses are interesting candidates to potentially regulate the

NKR-P1 and Clr families during NK cell-mediated detection ofinfection. Poxviruses are large DNAviruses that replicate within thecytoplasm and cause large perturbations in many cellular functions,including host protein synthesis. Poxviruses use a wide variety ofimmune evasion strategies, including interference with cytokinesthat impact the NK cell response (17, 18). The immune response inmice to both the mouse pathogen ectromelia virus (ECTV) and theprototypic poxvirus vaccinia virus (VV) involves NK cells (19–22). A very recent study indicates that NK cells even developmemory for VV (23). Various activating receptors on NK cellssuch as CD94/NKG2E and NKG2D are required for NK cell de-fense against ECTV in mouse (20, 24), and recognition of VVinfection by human NK cells involves NKG2D (25) and perhapsloss of HLA-E and classical MHC-I molecules (26, 27). Moreover,certain strains of mice, particularly C57BL/6, are resistant to VVand ECTV infection (22, 28–30), and depletion of NK cells con-verts resistant mice to a susceptible phenotype (22). The strain-dependent resistance to ECTV maps to the NKC (19), whichincludes the Ly49, CD94/NKG2, and NKR-P1 families of recep-tors. Although VV causes some reduction of mouse classicalMHC-I molecules (31), no evidence has emerged to suggestclassical MHC-I and Ly49 receptors are involved in resistance. Infact, NK cells with an activating Ly49 (Ly49H) are actually det-rimental to the response to ECTV in C57BL/6 mice (32). Recently,it was reported that the lack of CD94 in DBA/2J mice explainstheir high susceptibility to ECTV (24, 33). This study elegantlyshowed that CD94 in combination with NKG2E is essential forresistance to ECTVand synergizes with NKG2D to trigger the NKcells. However, it does not fully explain the differences in allstrains, as DBA/2J and BALB/c mice that express CD94 are alsoquite susceptible. Moreover, CD94 is not required for the responseto VV on a C57BL/6 background, suggesting that the dominantresistance trait in C57BL/6 for VV is mediated through anothermechanism (34).Given the potential for the NKR-P1 and Clr system of recep-

tors and ligands to be involved in the response to poxvirus infectionand the ability of poxviruses to interfere with protein expressionthrough destabilization of cellular transcripts (35–37), we testedthe hypothesis that Clr-b is modulated during poxvirus infection.We found that infection of mouse cells with either VV or ECTVreduces Clr-b on the cell surface. The loss of Clr-b requires activevirus infection, but not progression of the infection to the latephase and does not appear to require the loss of Clr-b mRNA. Clr-b–mediated protection from NK cell lysis is also gone followingVV infection in a time frame compatible with limiting viral rep-lication. These results suggest that murine NK cells may sensepoxviruses through the loss of Clr-b, and the proportion of NKcells sensitive to this pathway may contribute to the variation ofthe response by NK cells between various strains of mice.

Materials and MethodsCells and mice

NIH 3T3 mouse fibroblast cells were obtained from American Type CultureCollection (Manassas, VA) and cultured in DMEM with 10% calf serum.C1498, a murine myeloid leukemia cell line, was cultured in DMEMmediumwith 10% FBS. HuTK-143B cells (American Type Culture Collection) usedfor production of virus were cultured in DMEM with 10% FBS. The Uni-versity of Alberta AnimalWelfare and Policy Committee approved all animalprocedures. C57BL/6 mice were purchased from The Jackson Laboratory(Bar Harbor, ME). CD-1 mice were purchased from Charles River Labo-ratories (Wilmington, MA). Bone marrow-derived macrophages (BMMø)were generated from C57BL/6 mice. Briefly, bone marrow cells wereflushed out of the femur and tibia bones with PBS. These cells were washedthree times with PBS and cultured for 7 to 8 d in RPMI 1640 supplementedwith 10% FBS, 2 mM L-glutamine, 0.05 mM 2-ME, and 20% supernatantsfrom Chinese hamster ovary cells (kindly provided by Dr. H. Ostergaard,University of Alberta, Edmonton, AB, Canada) that produce GM-CSF. Thepurity of the macrophages was determined by staining for F4/80. NK cellswere derived from CD-1 splenocytes by harvesting and homogenizing thespleen. Splenocytes were incubated on a nylon wool column for 1 h to re-move adherent cells. The flowthrough containing the NK cells was culturedin 1000 U/ml IL-2 (Tecin) and the nonadherent fraction removed on day 3.The purity of the NK cells was $90% as determined by staining for CD3and NKR-P1B with PK136, and they were used in assays on day 7 or 8.

Abs and reagents

The Clr-b–specific Ab 4A6 (IgM) was purified and labeled with biotinas previously described (11). IgM isotype Ab control was purchasedfrom Cedarlane Laboratories (Burlington, ON, Canada). Streptavidin(SA)-allophycocyanin was purchased from eBioscience (San Diego, CA)and SA-PE from Cedarlane Laboratories. The anti–MHC-I Ab M142 (ratIgG2a) was kindly provided by Dr. K. Kane (University of Alberta). TheIgG2a (51.1) isotype Ab was produced in house from the hybridoma(American Type Culture Collection). Anti-mouse transferrin receptorR.17/217.1.3 Ab (rat IgG2a) was kindly provided by Dr. H. Ostergaard.Rabbit anti-I5L antiserum was provided by Dr. Michele Barry (Universityof Alberta) (38). Secondary goat anti-rabbit PE was purchased fromCedarlane Laboratories. Anti-mouse NK1.1-PerCP-Cy5.5 Ab and its iso-type (IgG2ak) were purchased from eBioscience. Anti-CD3 and anti-F4/80were from eBioscience. Cytosine b-D-arabinofuranoside (Sigma-Aldrich,Oakville, ON, Canada) stocks were prepared in water and used at a finalconcentration of 50 mg/ml. Cycloheximide (MP Biomedicals, Solon, OH)was used at a final concentration of 50 mg/ml. Actinomycin D (Sigma-Aldrich) was dissolved at 10 mg/ml and used at a final concentration of20 mg/ml.

Viruses and infection

VV strain WR and VV-EGFP were prepared as previously described (27).Briefly, the viruses were grown in HuTK-143B cells and the cell-associated virus released by sonicating the cells in 10 mM Tris (pH 9).The virus was concentrated by ultracentrifugation through a 36% sucrosecushion. After resuspension of the pellet in 10 mM Tris (pH 9), the pelletwas sonicated, and aggregated material was removed by centrifuging atlow speed. The virus was collected by ultracentrifugation and the pelletresuspended in a low volume of 10 mM Tris (pH 9) and sonicated. Theresulting viral suspension was stored at 280˚C and titered on HuTK-143B cells in six-well plates by crystal violet staining (WR) or X-gal(EGFP-WR). ECTV strain Moscow was provided by Dr. Michele Barry(University of Alberta), prepared in the same way as the VV prepara-tions, and titered on BGMK cells. Viruses were UV-inactivated for 30min as previously described (27) and the efficacy of the inactivationverified by titering on HuTK-143B cells. For infection of NIH 3T3 cells,the cells were plated at 5.5 3 105 cells/well in 12-well plates ∼24 h priorto infection. The medium was removed from the cells and the virusadded in 0.5 ml medium without additives. Following incubation at 37˚C/5% CO2 for 1 h, complete medium was added and the incubationcontinued for the duration of the experiment. For infection of C1498 andBMMø, the cells were pelleted in a 15-ml conical tube. The virus wasadded in 0.5 ml medium without additives and incubated with the cells at37˚C/5% CO2 for 1 h. Complete media was added, and the cells wereplated 106 cells per well in 12-well plates.

Flow cytometry

For surface staining, cells were harvested, washed with PBS supplementedwith2%FBS,andprimaryAbswereadded.Cellswere thenstained for30min

2 POXVIRAL INFECTION-MEDIATED LOSS OF Clr-b

by guest on April 10, 2019

http://ww

w.jim

munol.org/

Dow

nloaded from

at 4˚C and washed. If a secondary was needed, cells were washed again andstained for 30 min at 4˚C. For intracellular staining, cells were harvested andwashed with PBS/1% FBS and permeabilized using the Cytofix/Cytopermkit (BD Biosciences, Mississauga, ON, Canada) as per the manufacturer’sprotocol. Cells were then stained with primary Ab followed by secondary Abfor 30 min each at 4˚C. Where indicated, cells were also stained with theMolecular Probes LIVE/DEAD Fixable Dead Cell Stain Kit–violet (catalognumber L34955; Invitrogen, Burlington, ON, Canada) according to theinstructions supplied by the manufacturer. All samples were fixed with 4%PFA, acquired on an FACS Canto II (BD Biosciences), and analyzed usingFlowJo Software (version 7.2.5; Tree Star, Ashland, OR). Unless otherwiseindicated, live cellsweregated on for analysis based on their forward and sidescatter. Histogram overlays were generated using the percent of maximumoption that plots each histogram relative to the channel with maximum cellcount for that sample and is indicated on the y-axis as the relative cell number.

Analysis of mRNA

RNA was isolated from 2.5 3 106 cells using the RNeasy kit (Qiagen,Toronto, ON, Canada). Samples were treated on the columns with DNAseto remove any genomic or viral DNA. RT-PCR reactions were performedwith 10–100 ng RNA. Primers specific for Clr-b amplification were 59-CTC GGT TTT GAC AAC CAG GT-39 and 59-GAT CCC GTT GTT GTTCAG GGT-39; for b-actin, 59-TGT TAC CAA CTG GGA CGA CA-39 and59-GGG GTG TTG AAG GTC TCA A-39; and VV viral protein D10, 59-TTC CAG AGT GTT TAT CCA GGG A-39 and 59-CTC GTT AGA GATATT CTT CCG ACA A-39. The resulting PCR products were analyzed onan agarose gel stained with ethidium bromide.

Cytotoxicity assay

Cytotoxicity was measured in a standard [51Cr] release assay (39). C1498target cells were infected with VV (multiplicity of infection [MOI] of 10)for 18 h, and an aliquot was removed to assess the expression of Clr-b. Thecells (5 3 105) were labeled with ∼10 mCi [51Cr] (sodium chromate; NEN)for 1 h at 37˚C in 5% CO2. The target cells were washed and the viable cellcount determined using trypan blue exclusion and preincubated with either40 mg/ml 4A6 or isotype control Ab for 10 min prior to the addition of NKcells. Target cells were plated at 2500 live cells per well in V-bottommicrotiter plates in triplicate. The spontaneous release was determined inthe presence of the Abs. The NK cells were washed, diluted to the appro-priate concentration in assay medium, and added to the labeled target cells.The plates were incubated at 37˚C in 5% CO2 for 4 h. The released [51Cr]was quantified for 25 ml supernatant and analyzed in a Wallac 1450Microbeta Trilux (PerkinElmer). The percent lysis was calculated as the100 3 (specific [51Cr] release 2 spontaneous release)/(maximum release 2spontaneous release) for measured for each target. The spontaneous releasewas not increased in the infected cells relative to the uninfected controls.

Statistics

The statistical significance between conditions was calculated using aStudent t test, considered significant at a 95% confidence limit, and notedas *p , 0.05, **p , 0.005, and ***p , 0.0005.

ResultsInfection with VV reduces Clr-b cell-surface expression

To assess the effect of VV infection on Clr-b expression, we in-fected NIH 3T3 and C1498 cells with VVat an MOI of 10 for 12 h.Cells were then analyzed by flow cytometry for surface Clr-b usingbiotinylated 4A6 (11) and SA-allophycocyanin. We observeda substantial loss of Clr-b surface expression for both cell lines(Fig. 1A, 1B). In contrast, exposure of the cells to UV-treated virushad no effect on the level of Clr-b. To determine if the loss ofClr-b corresponded to infection of the cells and whether the losswas occurring on cells that were still alive and able to replicate thevirus, we performed a similar experiment with C1498 cells anda virus expressing enhanced GFP (VV-EGFP). In this case, thecells were also treated with a fixable live/dead stain. We observea good correlation between our live gate by forward and sidescatter analysis and the ability to exclude the live/dead cell stain(Fig. 1C). Moreover, the loss of Clr-b is evident within the livecells and within the infected cells expressing EGFP (Fig. 1D).We next assessed how the dose of virus affects the amount of

Clr-b that is lost using the VV-EGFP virus to simultaneously

monitor the infected and uninfected cell subsets. The comparisonof Clr-b loss with the amount of EGFP expression is illustrated inthe dot plots depicted in Fig. 2A for NIH 3T3 cells and 2C forC1498 cells. A dose-dependent response was observed for bothcell subsets, with significant reductions for both the EGFPhi andEGFPlo populations of cells, particularly at higher MOI (Fig. 2E,2F), corresponding to the doses in which most of the cells arealso EGFP positive (Fig. 2). Notably, for the NIH-3T3 cells, theEGFPlo subset often has low Clr-b expression similar to the EGFPhi

subset at each dose, whereas for the C1498 cells, a reduction inClr-b is more evident on the EGFPhi cells. However, by thesemethods, we cannot distinguish if the GFPlo cells with lower Clr-bare actually infected, are more resistant to infection by VV, or ifsoluble factors released from the infected cells may drive downClr-b expression on uninfected cells.

The rate of Clr-b loss from the cell surface is greater than thatof MHC-I

To further characterize the reduction of Clr-b expression, we per-formed a time-course analysis using C1498 cells to compare thechanges in Clr-b relative to other cell-surface proteins. We alsoevaluated the effects of these treatments on the expression of MHC-Isurface proteins, the ligands of the Ly49 receptors, and the highlystable transferrin receptor that recycles through the endosomalpathway. We infected the cells with VV-EGFP at an MOI of 10 toensure all of the cells were infected. For comparison with the normalrate ofClr-b cell-surface turnover, we also treated the cellswith 50mg/ml cycloheximide to block protein synthesis. Cycloheximide treat-ment caused a detectable loss of Clr-b at 4 h, illustrating that Clr-b isnormally quite rapidly turned over at the cell surface under steady-state conditions (Fig. 3A). The virus-induced downregulation wasslightly slower, but both treatments promoted a significant reductionof Clr-b expression by 12 h (Fig. 3A, 3B). We observed less VV-mediated downregulation of MHC-I, only obvious in the histogramsafter 12 h of infection (Fig. 3A), whereas the effects of cycloheximideon MHC-I expression were more obvious and more rapid. In someexperiments, VV had little effect on MHC-I levels at any time point(Fig. 3B). However, as expected, surface TfR expression remainedrelatively unaffected by VV infection or cycloheximide treatment.These results show that VV does not alter the expression of each cell-surface protein to the same extent, and, of particular interest in thisstudy, the rate of Clr-b loss is closer to treatment with a proteinsynthesis inhibitor than is the lossMHC-I, as depicted in Fig. 3B. Thissuggests the two are modulated by different mechanisms, and, per-haps, Clr-b loss could be detected by NK cells prior to loss of MHC-I.

Clr-b transcript is reduced during infection

Infection with VV could cause Clr-b to be reduced through a varietyof mechanisms including internalization, degradation, or decreasedproduction. It is well established that poxviruses can interfere withcellular transcripts, leading to reductions in host proteins (35, 36, 40,41). Therefore, we compared the rate of Clr-b protein loss to theamount of mRNA. As a positive control to detect loss of Clr-bmRNA, we treated the cells with 20 mg/ml of actinomycin D toinhibit transcription elongation. The rate of loss of Clr-b from thecell surface is quite similar on VV-infected cells compared with cellstreated with actinomycin D (Fig. 4A). To measure the amount ofClr-b message by conventional PCR, we first titrated the RNA fromuninfected cells to determine the point of sensitivity for the PCR (notshown). Using 50 ng of RNA, there is a clear loss of Clr-bmessage atthe 12-h time point and some loss by 8 h (Fig. 4C). In contrast, therewas little change in the detection of the abundant and stable messagefor b-actin. We could also detect a corresponding increase in mes-sage of the viral gene D10. We also performed a quantitative real-

The Journal of Immunology 3

by guest on April 10, 2019

http://ww

w.jim

munol.org/

Dow

nloaded from

time PCR analysis in which we observed a parallel loss in controlcellular transcripts as assessed by the raw threshold cycle scores, butlittle to no change in the relative amount of Clr-b compared with thetypical controls (b-actin or GAPDH) (data not shown). Therefore, asexpected, VV infection of C1498 cells causes a reduction in Clr-bmessage, but in comparison with treatment with actinomycin D, the

kinetics of the loss of Clr-b mRNA appear to be delayed relative tothe loss of Clr-b cell-surface protein.

Early infection is sufficient for Clr-b reduction

To better delineate the viral processes involved in Clr-b down-regulation, we repressed late gene expression by inhibiting viral

FIGURE 1. Clr-b downregulation by VV.

NIH 3T3 and C1498 cells were incubated in

medium or infected at an MOI of 10 with VV

or an equivalent amount of UV-inactivated

VV for 12 h. (A) Clr-b surface staining with

the Ab 4A6 shown relative to isotype control

Ab was determined by flow cytometry. (B)

The average mean fluorescence intensity

(MFI) for three experiments done in tripli-

cate is plotted for each condition. The error

bars represent the SE. (C and D) C1498 cells

were infected at an MOI of 10 for 12 h prior

to staining for Clr-b and with a fixable live/

dead cell stain. (C) illustrates the staining on

the uninfected control cells, whereas (D)

shows the gating strategy, EGFP expression,

and Clr-b staining for the infected live cells.

The p values were calculated using an un-

paired two-tailed t test. The results are an

average of the three experiments, each done

in triplicate. ***p , 0.0005. RCN, Relative

cell number (see Materials and Methods for

details).

4 POXVIRAL INFECTION-MEDIATED LOSS OF Clr-b

by guest on April 10, 2019

http://ww

w.jim

munol.org/

Dow

nloaded from

FIGURE 2. Clr-b loss is correlated with increasing dose of virus. NIH 3T3 cells (A, B, E) and C1498 cells (C, D, F) were infected at the indicated MOI

with VV-EGFP for 12 h and analyzed by flow cytometry for EGFP and Clr-b expression. (A) and (C) show the relative expression of Clr-b and EGFP in

representative samples from one experiment. The quadrants are set to the uninfected cells and isotype control for Clr-b (not shown). (B) and (D) depict the

histogram profiles for Clr-b expression on the gated subsets for EGFP expression as indicated by the legend below the panels. (E) and (F) are the average

MFI for triplicate samples on the same day. The numbers within the bars indicate the percentage of the live cells found within the gate. The error bars

indicate the SE. The p values shown above the panels were calculated using an unpaired two-tailed t test. The results are representative of three

experiments. *p , 0.05, **p , 0.005.

The Journal of Immunology 5

by guest on April 10, 2019

http://ww

w.jim

munol.org/

Dow

nloaded from

DNA replication using cytosine b-D-arabinofuranoside (AraC).AraC treatment eliminated the appearance of cells with high ex-pression of EGFP in both cell types (Fig. 5A). This is likely due toAraC’s effect on replication of the viral genome, preventing greaterproduction of EGFP that in this recombinant is under the control ofa synthetic early and late promoter (42). It is also interesting that

by this measure, the infection seems to progress better in the NIH

3T3 cells than in the C1498 cells, although we did not observe this

in all experiments (Fig. 2A, 2D). AraC treatment alone led to

a slight reduction of Clr-b at 12 h, but this was not statistically

significant (Fig. 5C). AraC did not prevent VV-EGFP–mediated

Clr-b downregulation (Fig. 5B, 5C). The results suggest that nei-

ther DNA replication nor late gene expression are required for the

observed VV-mediated reduction of Clr-b.

Clr-b reduction occurs with normal macrophages and ECTV

Clr-b is often reduced in transformed cells; therefore, we wanted toensure the effects we observed upon infection were not dependent on

FIGURE 3. Time-course analysis of Clr-b

loss following infection. (A) C1498 cells were

infected with a MOI 10 of VV-EGFP or treated

with 50 mg/ml cycloheximide (CHX). The

cells were surface stained for Clr-b, MHC-I, or

transferrin receptor (TfR) at the specified time

points and analyzed by flow cytometry. (B) The

relative loss of each protein shown by decay

curves that were plotted by calculating the

ratio of the MFI of infected to uninfected (UI)

at each time point for the treatments. The

results correspond to the average of three in-

dependent experiments. The error bars indicate

the SE.

6 POXVIRAL INFECTION-MEDIATED LOSS OF Clr-b

by guest on April 10, 2019

http://ww

w.jim

munol.org/

Dow

nloaded from

a transformed state of C1498 cells or a high passage number of NIH3T3 cells. To do this, we generatedGM-CSF–stimulated BMMø fromC57BL/6 mice. Clr-b expression on these cells peaks at about day 7(not shown), and although these cells do not have uniform expres-sion of Clr-b, the majority can be infected by VV-EGFP. The ex-pression is clearly downregulated within the infected (i.e., EGFP+)cells (Fig. 6). Curiously, it appears that the remaining uninfectedcells have low Clr-b expression as well, suggesting the Clr-b highcell subset or cell type may be preferentially infected.We next asked whether a more physiologically relevant poxvirus

caused a similar change in Clr-b expression by using the bona fidemouse pathogen ECTV. We optimized infection of our cells withECTVusing intracellular staining for the viral protein I5L.We foundgood expression of I5L in the cell lines and weak but detectableexpression in the BMMø at 18 h using anMOI of 5 (Fig. 7A, 7B).Wealso observe a pronounced loss of Clr-b expression in all three celltypes upon ECTV infection (Fig. 7C, 7D). Together, these resultsindicate that both VV and ECTV infection can promote Clr-bdownmodulation on primary cells.

VV infection leads to loss of functional recognition of Clr-b

To determine the functional consequences of VV-mediated Clr-bdownregulation, we assessed the effect of VV infection on Clr-b–mediated protection from IL-2–activated NK cells (LAK). To per-form these experiments, we used NK cells derived from CD-1 micethat have high expression of NKR-P1B, as shown in this study by

staining with the NK1.1 Ab (Fig. 8A). We used C1498 cells as thetarget cells because their expression of Clr-b is sufficient to protectcells from NK cells with NKR-P1B (Fig. 8B, 8C). As expected,masking Clr-b on uninfected targets with the anti–Clr-b Ab 4A6(IgM) caused an increase in lysis compared with target cells incu-bated with an isotype control Ab (Fig. 8C), demonstrating Clr-b–mediated protection of the target cells. The protection through Clr-binteraction with NKR-P1B was completely abrogated by 18 h in-fection with VV, as there is no difference in the lysis betweenisotype control and anti–Clr-b mAb treatment of the target cells(Fig. 8C). At this time point, the reduction in Clr-b was quitepronounced (Fig. 8B), even though the cells were still intact (∼6%dead by live/dead stain; data not shown). However, the lysis of theinfected targets was slightly less than mock-treated cells, suggestingthat the virus might also interfere with NK-mediated cytotoxicitythrough another mechanism. Notably, similar results were also ob-served after 12 h VV infection; however, at this intermediate timepoint, we only observed a partial reduction in Clr-b–mediatedprotection and a blunting of the overall lysis (Supplemental Fig. 1).

DiscussionWe have demonstrated in this study that Clr-b expression isdownregulated from the surface of mouse cells following infection

with two different orthopoxviruses. We observed that the loss was

relatively rapid in comparison with MHC-I. The loss was most

pronounced within the VV-EGFPhi cells, but it is also worth noting

FIGURE 4. Kinetics of changes in Clr-b surface protein expression and steady-state mRNA levels. (A) C1498 cells were incubated in medium alone or

20 mg/ml actinomycin D or infected with VV-EGFP at an MOI of 10 for the indicated periods. At each time point, C1498 cells were harvested and stained

with anti–Clr-b. (B) The expression of EGFP is depicted for the same samples as in (A). (C) RT-PCR analysis was performed using 50 ng of RNA/sample.

(D) The average MFI of Clr-b expression for three independent experiments is shown. The error bars represent the SE. UI, Uninfected.

The Journal of Immunology 7

by guest on April 10, 2019

http://ww

w.jim

munol.org/

Dow

nloaded from

that in some experiments, a subset of Clr-b low cells remained

uninfected or less productively infected (negative/low for EGFP),

particularly within the BMMø. As fixation tends to blunt the EGFP

signal (D. N. Burshtyn, unpublished observations), these cells may

well be infected but not resolved from the background. It is also

possible that these cells are responding as bystanders to the infected

cells or that the Clr-b low cells are more refractory to infection.In terms of delineating the mechanism that leads to the loss of

surface Clr-b, the precise mechanism remains unknown, but wehave established that it requires an active infection and not ex-pression of late viral genes. The mRNA for Clr-b became onlyweakly detectable in our assay by 12 h, suggesting that viral in-terference with host cell mRNA likely contributes to a proportion ofthe Clr-b loss, particularly at later time points. However, the re-duction of Clr-b message at early time points was less than ob-served for treatment with actinomycin D, an inhibitor of trans-cription (and DNA replication), whereas the loss of surface proteinbetween the two treatments was remarkably similar. This disparitysuggests that mechanisms in addition to virus-mediated suppres-sion of Clr-b mRNA are involved in the observed reduction ofClr-b surface protein. There are at least two possible scenarios bywhich virus infection could lead to decreased Clr-b: one possibilityis that the host cells innately sense infection and/or respond bydownregulating Clr-b surface protein (either specifically or byproxy) to alert NK cells of an infection, and a second is that thevirus may encode proteins that target Clr-b (and/or Clr/Clec2)function. Cellular stress pathways in response to chemical or

physical insults have already been shown to cause downregula-tion of Clr-b surface protein that is dependent on the ubiquitin/proteasomal pathway or at least a ubiquitin-dependent process

FIGURE 5. Clr-b loss occurs in the presence

of an inhibitor of late gene expression. NIH

3T3 or C1498 cells were incubated with media

or infected with an MOI of 10 of VV-EGFP for

12 h in the presence or absence of 50 mg/ml

AraC. The expression of EGFP (A) or Clr-b (B)

is shown for representative infected and unin-

fected cells relative to isotype control. (C)

shows average MFI for three experiments each

done with triplicate samples. The error bars

indicate the SE. The p values shown above the

panels were calculated using an unpaired two-

tailed t test. **p , 0.005, ***p , 0.0005. UI,

Uninfected.

FIGURE 6. Clr-b downregulation on primary BMMø. Day 7 BMMø were

infected with VV-EGFP at an MOI of 10 for 12 h and analyzed by flow

cytometry. (A) The expression ofClr-b andEGFP.The quadrantswere set using

theuninfectedcells (EGFP)stainedwith controlAb (not shown). (B)Theprofile

for Clr-b staining is shown in the EGFP+ gate for the infected cells relative to

uninfected (UI) cells. The staining is indicated in the legend. The results are

representative of three experiments. (C) The average MFI for Clr-b expression

within the EGFP+ cells for three experiments donewith triplicate samples. The

background staining is subtracted from the values. The error bars represent the

SE, and the p values were calculated using an unpaired two-tailed t test.

8 POXVIRAL INFECTION-MEDIATED LOSS OF Clr-b

by guest on April 10, 2019

http://ww

w.jim

munol.org/

Dow

nloaded from

affected by the proteasomal inhibitors MG132 or lactacystin (16).Therefore, it seems highly possible that in response to sensingviral DNA in the cytoplasm or the cellular stress imposed by in-fection (infection stress), cells internalize and/or degrade Clr-bprotein faster through cell-intrinsic mechanisms. In contrast, itremains possible that the virus (or host cell) encodes proteinsto interfere specifically with Clr/Clec2 function, perhaps throughdirect association and trafficking by the C-type lectin-relatedproteins or through virally encoded proteins that drive ubiq-uitination (43). Poxviruses are known to employ several strategiesto block expression of host proteins that are involved in activationof the immune response (44–46), and there are indications thatClr-b might be involved in immune activation, making it a targetfor the virus. For example, exposure of human APCs to inactivatedviruses or pathogen mimetics produces a substantial increase of

LLT1, a human Clr-b homolog, which may protect APCs from NKcell attack (47). For mouse cells, the loss of Clr-b appears todominate when an APC is actively infected, as we observe a de-crease of Clr-b expression on infected macrophages. Poxvirusesmight interfere with Clr-b protein trafficking and turnover, as theyencode a wealth of proteins that manipulate the ubiquitin path-way (43).In order for NK cells to defend against poxviruses through

cytolysis or cytokine production, it requires they respond during theinnate immune response, ideally within the window of time it takesthe virus to produce progeny virus. Therefore, it was of interest todelineate how the loss of Clr-b related to sensitivity to lysis by NKcells. We observed a reduction of Clr-b within the time frame ittakes for VV to complete its replication in culture (31), and thissuggests the ability of NK cells to detect the loss would be in time

FIGURE 7. Clr-b downregulation by ECTV. Cells were infected with ECTV at an MOI of 5 for 18 h for NIH 3T3 cells and BMMø and 24 h for C1498

cells. (A) The cells were analyzed for expression of I5L by intracellular staining. The background was established by staining with secondary alone and

staining on uninfected (UI) cells. (B) The background-corrected average MFI is plotted for I5L expression of triplicate samples. (C) The cells were analyzed

for surface expression of Clr-b as before. (D) The background-corrected average MFI is plotted for Clr-b expression of triplicate samples. For (B) and (D),

the p values were calculated using an unpaired two-tailed t test, and the error bars represent the SE. The results represent the average of three experiments.

**p , 0.005, ***p , 0.0005.

The Journal of Immunology 9

by guest on April 10, 2019

http://ww

w.jim

munol.org/

Dow

nloaded from

for NK-mediated antiviral effects to contain the virus and/or re-cruit other immune effector cells. In contrast, the reduction ofclassical MHC-I molecules we observe is quite modest and similarto that previously reported (31). The relatively slow and modestloss of MHC-I may explain why NK cells do not appear to sensepoxvirus-infected cells through the Ly49 receptor system. In termsof NK cells sensing the loss of Clr-b, we found that infection oftarget cells does prevent Clr-b from mediating protection ofinfected cells. However, we did not observe any significant in-crease in the lysis of the infected target cells compared with un-infected controls. In single experiments performed at an earliertime point, we observed an intermediate effect in which Clr-b wasreduced but not gone, and there remained some protection throughClr-b, albeit to a lesser extent than the uninfected targets (Sup-plemental Fig. 1). It is possible that at the rather late time pointshown in this study, the virus is also interfering with the expres-sion of molecules that would otherwise specifically enhance NKcell lysis of C1498 cells (e.g., ICAM-1, RAE-1, Qa-1, other Clr,etc.) and/or interfering with the lytic mechanism itself by pre-venting the induction of apoptosis through expression of anti-apoptotic proteins (48). In line with our observation, a modestreduction in lysis was also observed after 24-h infection using NKcells from a different mouse strain (C3H) and different target cellline (L929) (22). Previously, we observed that human NK cellslysed human target cells better following overnight infection withVV (27). However, it was recently found that the hemagglutininglycoproteins encoded by both VV and ECTV bind to the acti-vating receptors NKp30 and NKp46 on human NK cells, whichcan result in an overall reduction in target cell lysis (49). Coin-cidentally, the VV protein N1 limits mouse NK cell responsesin vivo (50), and the SPI-2 protein encoded by vaccinia and

ectromelia can prevent Fas-mediated apoptosis and NK cell acti-vation in vivo, respectively (51, 52). Another possible way thatVV might interfere with NK function by mouse NK cells wouldbe to encode decoy proteins as seen in RCMV-English (15). Al-though this seems more remote given that VV is not a naturalpathogen of mice, VV does encode at least two proteins that havestructural similarity to the C-type lectin family that are expressedon the surface of infected cells, namely A33 and A40R (53, 54).Additionally, A40R is a virulence factor with effects restricted tothe dermal route of inoculation (55), where NK cells are recruited(56).The reduction of Clr-b during infection by poxviruses opens the

possibility that NKR-P1 receptors influence the NK response topoxviruses in different strains of mice. Although many strains ofmice exhibit variegated expression of the NKR-P1 receptors, weexamined NK cells from CD-1 mice because they uniformly ex-press NKR-P1B; however, there are no reports, to our knowledge,regarding the relative susceptibility of CD-1 mice to ectromelia orVV. It is also worth noting that C57BL/6 mice, which are resistantto these viruses, express an activating NKR-P1 receptor on all NKcells. In contrast to rodents, humans appear to have only one ho-molog in this family, NKR-P1A, which is inhibitory on NK cellsand binds LLT-1. Nonetheless, there are also two more distantlyrelated receptors (KLRF1/2) that also have genetically linkedligands within the same region (CLEC2A/B), some of which mayfunction in a tissue-restricted fashion such as the skin (57, 58).Therefore, whether human NK cells can sense target cells throughmodulation of the NKR-P1 ligand following poxvirus infectionand whether this might vary between individuals are interestingand relevant points to consider, because poxviruses are beingdeveloped as vectors for vaccines and as oncolytic viruses to treat

FIGURE 8. Functional recognition of

Clr-b following VV infection. (A) Day 7

NK cells from CD-1 mice analyzed for

NKR-P1B expression by staining with

NK1.1. (B) C1498 cells were infected with

VV at an MOI 10 for 18 h and analyzed for

loss of Clr-b (left panel) and GFP expres-

sion (right panel) by flow cytometry. (C)

Mock-treated and infected C1498 cells

shown in (B) were labeled with [51Cr] and

used as targets in a cytotoxicity assay with

the NK cells shown in (A). The assay was

performed in the presence of 4A6 (40 mg/

ml) or isotype control Ab (40 mg/ml). The

spontaneous lysis was determined in the

presence of the Abs. The error bars repre-

sent the SE. The results are representative

of three experiments. UI, Uninfected.

10 POXVIRAL INFECTION-MEDIATED LOSS OF Clr-b

by guest on April 10, 2019

http://ww

w.jim

munol.org/

Dow

nloaded from

tumors. In conclusion, the findings presented in this study extendNK-mediated missing-self recognition of Clr-b to poxviruses, butfuture studies are needed to elucidate the mechanism(s) of Clr(Clec2) regulation and the relative role of the NKR-P1 (Klrb1)receptor system in the NK cell response to poxviruses in vivo.

AcknowledgmentsWe thank Dr. Michele Barry and Nick Van Buuren for helpful discussions,

providing the ECTV, and the gift of the anti-I5L Ab. We also thank Drs.

Hanne Ostergaard and Kevin Kane for providing reagents.

DisclosuresThe authors have no financial conflicts of interest.

References1. 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.

2. Lanier, L. L. 2005. NK cell recognition. Annu. Rev. Immunol. 23: 225–274.3. Ljunggren, H. G., and K. Karre. 1990. In search of the ‘missing self’: MHC

molecules and NK cell recognition. Immunol. Today 11: 237–244.4. Fruh, K., A. Gruhler, R. M. Krishna, and G. J. Schoenhals. 1999. A comparison

of viral immune escape strategies targeting the MHC class I assembly pathway.Immunol. Rev. 168: 157–166.

5. Vilches, C., and P. Parham. 2002. KIR: diverse, rapidly evolving receptors ofinnate and adaptive immunity. Annu. Rev. Immunol. 20: 217–251.

6. Carlyle, J. R., A. Mesci, J. H. Fine, P. Chen, S. Belanger, L. H. Tai, andA. P. Makrigiannis. 2008. Evolution of the Ly49 and Nkrp1 recognition systems.Semin. Immunol. 20: 321–330.

7. Revilleza, M. J., R. Wang, J. Mans, M. Hong, K. Natarajan, and D. H. Margulies.2011. How the virus outsmarts the host: function and structure of cytomegalo-virus MHC-I-like molecules in the evasion of natural killer cell surveillance. J.Biomed. Biotechnol. 2011: 724607–724619.

8. Senkevich, T. G., and B. Moss. 1998. Domain structure, intracellular trafficking,and beta2-microglobulin binding of a major histocompatibility complex class Ihomolog encoded by molluscum contagiosum virus. Virology 250: 397–407.

9. Wang, E. C., B. McSharry, C. Retiere, P. Tomasec, S. Williams,L. K. Borysiewicz, V. M. Braud, and G. W. Wilkinson. 2002. UL40-mediatedNK evasion during productive infection with human cytomegalovirus. Proc.Natl. Acad. Sci. USA 99: 7570–7575.

10. Mesci, A., B. Ljutic, A. P. Makrigiannis, and J. R. Carlyle. 2006. NKR-P1 bi-ology: from prototype to missing self. Immunol. Res. 35: 13–26.

11. Carlyle, J. R., A. M. Jamieson, S. Gasser, C. S. Clingan, H. Arase, andD. H. Raulet. 2004. Missing self-recognition of Ocil/Clr-b by inhibitory NKR-P1natural killer cell receptors. Proc. Natl. Acad. Sci. USA 101: 3527–3532.

12. Iizuka, K., O. V. Naidenko, B. F. Plougastel, D. H. Fremont, andW. M. Yokoyama. 2003. Genetically linked C-type lectin-related ligands for theNKRP1 family of natural killer cell receptors. Nat. Immunol. 4: 801–807.

13. Higuchi, D. A., P. Cahan, J. Gao, S. T. Ferris, J. Poursine-Laurent,T. A. Graubert, and W. M. Yokoyama. 2010. Structural variation of the mousenatural killer gene complex. Genes Immun. 11: 637–648.

14. Kartsogiannis, V., N. A. Sims, J. M. Quinn, C. Ly, M. Cipetic, I. J. Poulton,E. C. Walker, H. Saleh, N. E. McGregor, M. E. Wallace, et al. 2008. Osteoclastinhibitory lectin, an immune cell product that is required for normal bonephysiology in vivo. J. Biol. Chem. 283: 30850–30860.

15. Voigt, S., A. Mesci, J. Ettinger, J. H. Fine, P. Chen, W. Chou, and J. R. Carlyle.2007. Cytomegalovirus evasion of innate immunity by subversion of the NKR-P1B:Clr-b missing-self axis. Immunity 26: 617–627.

16. Fine, J. H., P. Chen, A. Mesci, D. S. Allan, S. Gasser, D. H. Raulet, andJ. R. Carlyle. 2010. Chemotherapy-induced genotoxic stress promotes sensitivityto natural killer cell cytotoxicity by enabling missing-self recognition. CancerRes. 70: 7102–7113.

17. Reading, P. C., and G. L. Smith. 2003. Vaccinia virus interleukin-18-bindingprotein promotes virulence by reducing gamma interferon production and naturalkiller and T-cell activity. J. Virol. 77: 9960–9968.

18. Born, T. L., L. A. Morrison, D. J. Esteban, T. VandenBos, L. G. Thebeau,N. Chen, M. K. Spriggs, J. E. Sims, and R. M. Buller. 2000. A poxvirus proteinthat binds to and inactivates IL-18, and inhibits NK cell response. J. Immunol.164: 3246–3254.

19. Delano, M. L., and D. G. Brownstein. 1995. Innate resistance to lethal mousepoxis genetically linked to the NK gene complex on chromosome 6 and correlateswith early restriction of virus replication by cells with an NK phenotype. J. Virol.69: 5875–5877.

20. Fang, M., L. L. Lanier, and L. J. Sigal. 2008. A role for NKG2D in NK cell-mediated resistance to poxvirus disease. PLoS Pathog. 4: e30.

21. Parker, A. K., S. Parker, W. M. Yokoyama, J. A. Corbett, and R. M. Buller. 2007.Induction of natural killer cell responses by ectromelia virus controls infection.J. Virol. 81: 4070–4079.

22. Bukowski, J. F., B. A. Woda, S. Habu, K. Okumura, and R. M. Welsh. 1983.Natural killer cell depletion enhances virus synthesis and virus-induced hepatitisin vivo. J. Immunol. 131: 1531–1538.

23. Gillard, G. O., M. Bivas-Benita, A. H. Hovav, L. E. Grandpre, M. W. Panas,M. S. Seaman, B. F. Haynes, and N. L. Letvin. 2011. Thy1+ NK [corrected] cellsfrom vaccinia virus-primed mice confer protection against vaccinia virus chal-lenge in the absence of adaptive lymphocytes. PLoS Pathog. 7: e1002141.

24. Fang, M., M. T. Orr, P. Spee, T. Egebjerg, L. L. Lanier, and L. J. Sigal. 2011.CD94 is essential for NK cell-mediated resistance to a lethal viral disease. Im-munity 34: 579–589.

25. Chisholm, S. E., and H. T. Reyburn. 2006. Recognition of vaccinia virus-infectedcells by human natural killer cells depends on natural cytotoxicity receptors. J.Virol. 80: 2225–2233.

26. Brooks, C. R., T. Elliott, P. Parham, and S. I. Khakoo. 2006. The inhibitoryreceptor NKG2A determines lysis of vaccinia virus-infected autologous targetsby NK cells. J. Immunol. 176: 1141–1147.

27. Kirwan, S., D. Merriam, N. Barsby, A. McKinnon, and D. N. Burshtyn. 2006.Vaccinia virus modulation of natural killer cell function by direct infection.Virology 347: 75–87.

28. Schell, K. 1960. Studies on the innate resistance of mice to infection withmousepox. II. Route of inoculation and resistance; and some observations on theinheritance of resistance. Aust. J. Exp. Biol. Med. Sci. 38: 289–299.

29. Schell, K. 1960. Studies on the innate resistance of mice to infection withmousepox. I. Resistance and antibody production. Aust. J. Exp. Biol. Med. Sci.38: 271–288.

30. Wallace, G. D., R. M. Buller, and H. C. Morse, III. 1985. Genetic determinants ofresistance to ectromelia (mousepox) virus-induced mortality. J. Virol. 55: 890–891.

31. Brutkiewicz, R. R., S. J. Klaus, and R. M. Welsh. 1992. Window of vulnerabilityof vaccinia virus-infected cells to natural killer (NK) cell-mediated cytolysiscorrelates with enhanced NK cell triggering and is concomitant with a decreasein H-2 class I antigen expression. Nat. Immun. 11: 203–214.

32. Fodil-Cornu, N., S. H. Lee, S. Belanger, A. P. Makrigiannis, C. A. Biron,R. M. Buller, and S. M. Vidal. 2008. Ly49h-deficient C57BL/6 mice: a newmouse cytomegalovirus-susceptible model remains resistant to unrelatedpathogens controlled by the NK gene complex. J. Immunol. 181: 6394–6405.

33. Vance, R. E., A. M. Jamieson, D. Cado, and D. H. Raulet. 2002. Implications ofCD94 deficiency and monoallelic NKG2A expression for natural killer celldevelopment and repertoire formation. Proc. Natl. Acad. Sci. USA 99: 868–873.

34. Orr, M. T., J. Wu, M. Fang, L. J. Sigal, P. Spee, T. Egebjerg, E. Dissen,S. Fossum, J. H. Phillips, and L. L. Lanier. 2010. Development and function ofCD94-deficient natural killer cells. PLoS ONE 5: e15184.

35. Brum, L. M., M. C. Lopez, J. C. Varela, H. V. Baker, and R. W. Moyer. 2003.Microarray analysis of A549 cells infected with rabbitpox virus (RPV): a com-parison of wild-type RPV and RPV deleted for the host range gene, SPI-1. Vi-rology 315: 322–334.

36. Guerra, S., L. A. Lopez-Fernandez, A. Pascual-Montano, M. Munoz,K. Harshman, and M. Esteban. 2003. Cellular gene expression survey of vacciniavirus infection of human HeLa cells. J. Virol. 77: 6493–6506.

37. Yang, Z., D. P. Bruno, C. A. Martens, S. F. Porcella, and B. Moss. 2010. Si-multaneous high-resolution analysis of vaccinia virus and host cell tran-scriptomes by deep RNA sequencing. Proc. Natl. Acad. Sci. USA 107: 11513–11518.

38. van Buuren, N., B. Couturier, Y. Xiong, and M. Barry. 2008. Ectromelia virusencodes a novel family of F-box proteins that interact with the SCF complex. J.Virol. 82: 9917–9927.

39. Miller, R. G., and M. Dunkley. 1974. Quantitative analysis of the 51Cr releasecytotoxicity assay for cytotoxic lymphocytes. Cell. Immunol. 14: 284–302.

40. Becker, Y., and W. K. Joklik. 1964. Messenger Rna in Cells Infected withVaccinia Virus. Proc. Natl. Acad. Sci. USA 51: 577–585.

41. Pedley, S., and R. J. Cooper. 1984. The inhibition of HeLa cell RNA synthesisfollowing infection with vaccinia virus. J. Gen. Virol. 65: 1687–1697.

42. Hammond, J. M., P. G. Oke, and B. E. Coupar. 1997. A synthetic vaccinia viruspromoter with enhanced early and late activity. J. Virol. Methods 66: 135–138.

43. Shchelkunov, S. N. 2010. Interaction of orthopoxviruses with the cellularubiquitin-ligase system. Virus Genes 41: 309–318.

44. Alzhanova, D., D. M. Edwards, E. Hammarlund, I. G. Scholz, D. Horst,M. J. Wagner, C. Upton, E. J. Wiertz, M. K. Slifka, and K. Fruh. 2009. Cowpoxvirus inhibits the transporter associated with antigen processing to evade T cellrecognition. Cell Host Microbe 6: 433–445.

45. Dasgupta, A., E. Hammarlund, M. K. Slifka, and K. Fruh. 2007. Cowpox virusevades CTL recognition and inhibits the intracellular transport of MHC class Imolecules. J. Immunol. 178: 1654–1661.

46. Rehm, K. E., R. F. Connor, G. J. Jones, K. Yimbu, M. D. Mannie, andR. L. Roper. 2009. Vaccinia virus decreases major histocompatibility complex(MHC) class II antigen presentation, T-cell priming, and peptide association withMHC class II. Immunology 128: 381–392.

47. Rosen, D. B., W. Cao, D. T. Avery, S. G. Tangye, Y. J. Liu, J. P. Houchins, andL. L. Lanier. 2008. Functional consequences of interactions between humanNKR-P1A and its ligand LLT1 expressed on activated dendritic cells and B cells.J. Immunol. 180: 6508–6517.

48. Taylor, J. M., and M. Barry. 2006. Near death experiences: poxvirus regulationof apoptotic death. Virology 344: 139–150.

49. Jarahian, M., M. Fiedler, A. Cohnen, D. Djandji, G. J. Hammerling, C. Gati,A. Cerwenka, P. C. Turner, R. W. Moyer, C. Watzl, et al. 2011. Modulation ofNKp30- and NKp46-mediated natural killer cell responses by poxviral hemag-glutinin. PLoS Pathog. 7: e1002195.

50. Jacobs, N., N. W. Bartlett, R. H. Clark, and G. L. Smith. 2008. Vaccinia viruslacking the Bcl-2-like protein N1 induces a stronger natural killer cell responseto infection. J. Gen. Virol. 89: 2877–2881.

The Journal of Immunology 11

by guest on April 10, 2019

http://ww

w.jim

munol.org/

Dow

nloaded from

51. Kettle, S., A. Alcamı, A. Khanna, R. Ehret, C. Jassoy, and G. L. Smith. 1997.Vaccinia virus serpin B13R (SPI-2) inhibits interleukin-1beta-converting enzymeand protects virus-infected cells from TNF- and Fas-mediated apoptosis, butdoes not prevent IL-1beta-induced fever. J. Gen. Virol. 78: 677–685.

52. Melo-Silva, C. R., D. C. Tscharke,M. Lobigs, A. Koskinen, Y. C.Wong, R.M.Buller,A.Mullbacher, andM. Regner. 2011. The ectromelia virus SPI-2 protein causes lethalmousepox by preventing NK cell responses. J. Virol. 85: 11170–11182.

53. Wilcock, D., S. A. Duncan, P. Traktman, W. H. Zhang, and G. L. Smith. 1999.The vaccinia virus A4OR gene product is a nonstructural, type II membraneglycoprotein that is expressed at the cell surface. J. Gen. Virol. 80: 2137–2148.

54. Su, H. P., K. Singh, A. G. Gittis, and D. N. Garboczi. 2010. The structure of thepoxvirus A33 protein reveals a dimer of unique C-type lectin-like domains. J.Virol. 84: 2502–2510.

55. Tscharke, D. C., P. C. Reading, and G. L. Smith. 2002. Dermal infection withvaccinia virus reveals roles for virus proteins not seen using other inoculationroutes. J. Gen. Virol. 83: 1977–1986.

56. Jacobs, N., R. A. Chen, C. Gubser, P. Najarro, and G. L. Smith. 2006. Intra-dermal immune response after infection with Vaccinia virus. J. Gen. Virol. 87:1157–1161.

57. Dennehy, K. M., S. N. Klimosch, and A. Steinle. 2011. Cutting edge: NKp80uses an atypical hemi-ITAM to trigger NK cytotoxicity. J. Immunol. 186: 657–661.

58. Spreu, J., S. Kuttruff, V. Stejfova, K. M. Dennehy, B. Schittek, and A. Steinle.2010. Interaction of C-type lectin-like receptors NKp65 and KACL facilitatesdedicated immune recognition of human keratinocytes. Proc. Natl. Acad. Sci.USA 107: 5100–5105.

12 POXVIRAL INFECTION-MEDIATED LOSS OF Clr-b

by guest on April 10, 2019

http://ww

w.jim

munol.org/

Dow

nloaded from