of July 18, 2022.This information is current as

Inhibitory Killer Ig-Like ReceptorsIs Restricted by Their Limited Expression of The Lytic Potential of Human Liver NK Cells

DeMatteoBamboat, Hoang M. Nguyen, Bo Dupont and Ronald P. Bryan M. Burt, George Plitas, Zeguo Zhao, Zubin M.

http://www.jimmunol.org/content/183/3/1789doi: 10.4049/jimmunol.09005412009;

2009; 183:1789-1796; Prepublished online 8 JulyJ Immunol 

MaterialSupplementary

1.DC1http://www.jimmunol.org/content/suppl/2009/07/07/jimmunol.090054

Referenceshttp://www.jimmunol.org/content/183/3/1789.full#ref-list-1

, 23 of which you can access for free at: cites 50 articlesThis article

        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 © 2009 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 July 18, 2022http://w

ww

.jimm

unol.org/D

ownloaded from

by guest on July 18, 2022

http://ww

w.jim

munol.org/

Dow

nloaded from

by guest on July 18, 2022http://w

ww

.jimm

unol.org/D

ownloaded from

by guest on July 18, 2022

http://ww

w.jim

munol.org/

Dow

nloaded from

The Lytic Potential of Human Liver NK Cells Is Restricted byTheir Limited Expression of Inhibitory Killer Ig-Like Receptors1

Bryan M. Burt,* George Plitas,* Zeguo Zhao,† Zubin M. Bamboat,* Hoang M. Nguyen,*Bo Dupont,† and Ronald P. DeMatteo2*

The human liver is enriched in NK cells which are potent effectors of the innate immune system. We have determined that liverNK cells freshly isolated from surgical specimens from patients with hepatic malignancy have less cytolytic activity than autol-ogous blood NK cells. This difference was due to a higher proportion of CD16� NK cells in the liver and reduced cytotoxicity byCD16� liver NK cells compared with their blood counterparts. CD16� liver NK cells had similar expression of activating NKreceptors and had similar intracellular granzyme B and perforin content compared with CD16� blood NK cells. CD16� liver NKcells contained a reduced fraction of cells with inhibitory killer Ig-like receptors specific for self-MHC class I (self-killer Ig-relatedreceptor (KIR)) and an increased fraction of self-KIRnegNKG2Apos and self-KIRnegNKG2Aneg cells. Using single-cell analysis ofintracellular IFN-� production and cytotoxicity assays, we determined that CD16� liver NK cells expressing self-KIR were moreresponsive to target cells than those cells that did not express self-KIR molecules. CD16� liver NK cells gained cytolytic functionwhen stimulated with IL-2 or cultured with LPS or poly(I:C)-activated autologous liver Kupffer cells. Thus, the human livercontains NK cell subsets which have reduced effector function, but under appropriate inflammatory conditions become potentkillers. The Journal of Immunology, 2009, 183: 1789–1796.

T he liver is continuously confronted with a large antigenicload that includes toxins, harmless dietary proteins, andantigenic elements from commensal intestinal organisms.

The liver is also a common site of chronic viral and parasitic in-fection and is the most common location for distant metastaticdisease in cancer. Therefore, the hepatic immune system must dif-ferentiate harmless foreign Ags from invasive threats to deliverappropriate immune responses. Although the local immune mech-anisms that are required to direct immunogenic or tolerogenic re-sponses to these varied antigenic challenges are largely unknown,the overall balance in hepatic immunity tends to favor tolerance.For example, HLA matching between donor and recipient does notaffect the outcome of liver transplantation, whereas HLA-disparateheart, lung, and kidney transplants are frequently rejected (1). Fur-thermore, compared with other solid organ transplants, T cell-me-diated rejection of liver allografts is uncommon (2). Additionally,the liver is believed to play a central role in oral tolerance, a phe-nomenon in which systemic immune responses to specific Ags aresuppressed by prior administration of the Ag via the oral route orportal vein (3).

To cope with potentially toxic insults without launching harmfulsystemic immune responses, the liver may rely on its innate im-

mune system. The lymphoid system of the liver is uniquelyweighted with innate cells. Although peripheral blood lympho-cytes are dominated by T and B cells that possess clonotypic Ag-specific receptors and are capable of mediating adaptive responses,the lymphoid system of the liver is comprised chiefly of innateimmune cells such as NK cells, NKT cells, and �� T cells whichuse invariant receptors and are capable of detecting and respondingrapidly to pathogens and malignant cells (4).

NK cells are bone marrow-derived lymphocytes that are able torecognize and kill transformed and virus-infected cells without theneed for prior sensitization. NK cells are a predominant lympho-cyte population in the murine and human liver. Whereas NK cellsaccount for �20% of all circulating lymphocytes, they may com-prise up to 50% of the total lymphocyte pool in the human liver(4). In humans, two main functional subsets of NK cells existbased on their surface density of CD56 and CD16 (5). CD56dim

CD16pos NK cells (referred to hereafter as CD16� NK cells) com-prise upward of 90% of NK cells in the periphery, contain anabundance of lytic granules, and have high cytotoxic potential.CD56brightCD16neg NK cells (referred to hereafter as CD16� NKcells) constitute a minor population of circulating NK cells andhave decreased cytotoxic ability, but are capable of producing co-pious amounts of cytokines (6).

The distribution of human NK cell subsets varies within solidorgans, possibly reflecting a tissue-specific division of labor. Al-though rare in the blood, CD16� NK cells comprise �75% of NKcells in lymph nodes and 50% of NK cells in the spleen (7, 8). NKcells in lymphoid tissues can be found in close proximity to den-dritic cells (DCs)3 (9). Through production of IFN-�, NK cellsmay act to polarize adaptive T cell responses (10, 11) and can limitEBV-triggered B cell transformation (12). In the uterus and pla-centa, a unique subset of NK cells with the CD16� phenotype

*Hepatopancreatobiliary Service, Memorial Sloan-Kettering Cancer Center, NewYork, NY 10065; and †Immunology Program, Sloan-Kettering Institute, New York,NY 10065

Received for publication February 23, 2009. Accepted for publication May 31, 2009.

The costs of publication of this article were defrayed in part by the payment of pagecharges. This article must therefore be hereby marked advertisement in accordancewith 18 U.S.C. Section 1734 solely to indicate this fact.1 This work was supported by National Institutes of Health Grants DK068346 andAI70658 (to R.P.D.), CA23766 (to Z.Z. and B.D.), and CA123938 (to B.M.B.), theWilliam H. Goodwin and Alice Godman Fund and Commonwealth Cancer Founda-tion for Research/Experimental Therapeutics Center of Memorial Sloan-KetteringCancer Center, and the Swim Across America Foundation.2 Address correspondence and reprint requests to Dr. Ronald P. DeMatteo, MemorialSloan-Kettering Cancer Center, 1275 York Avenue, Box 203, New York, NY 10065.E-mail address: dematter@mskcc.org

3 Abbreviations used in this paper: DC, dendritic cell; MHC-I, MHC class I; KIR,killer Ig-like receptor; LMNC, liver mononuclear cell; KC, Kupffer cell.

Copyright © 2009 by The American Association of Immunologists, Inc. 0022-1767/09/$2.00

The Journal of Immunology

www.jimmunol.org/cgi/doi/10.4049/jimmunol.0900541

by guest on July 18, 2022http://w

ww

.jimm

unol.org/D

ownloaded from

predominates (13, 14). Although these NK cells do contain anabundance of cytolytic granules, they do not lyse the HLA-A- andHLA-B-negative extravillous trophoblasts that invade the deciduaduring pregnancy and, when freshly isolated, demonstrate weakcytolytic function (13). In the placenta, these tissue-resident NKcells appear to be specialized for regulating specific developmentalprocesses at the fetal-maternal interface (15).

The effector functions of NK cells are regulated in part by theirinteraction with MHC class I (MHC-I) molecules. Ligation of NKcell inhibitory receptors by their specific MHC-I ligands on targetcells results in inhibition of NK cell function. Therefore, targetswith normal MHC-I expression are more resistant to killing thanthose lacking MHC-I and target cells lacking MHC-I are sensitiveto NK cytotoxicity. Through a variety of inhibitory and activatingreceptors and their ligands on target cells, NK cells identify cellsthat have lost or altered MHC-I expression, as occurs in viral in-fection and malignant transformation. Recently, the “missing-self”paradigm (16) has been modified to accommodate the observationthat a substantial subset of NK cells are lacking all MHC-I-specificreceptors or inhibitory receptors for “self-” MHC-I. NK cells re-quire “licensing” or “education” by an inhibitory receptor that caninteract with a self-MHC-I ligand to elicit strong responses to tar-gets without self-MHC-I molecules, while NK cells without suchreceptors are hyporesponsive (17–23). The functional NK cell rep-ertoire is therefore controlled by the inhibitory receptors that rec-ognize self-MHC-I ligands. In humans, inhibitory NK receptorsthat recognize MHC-I include the polymorphic killer Ig-like re-ceptors (KIRs) and the lectin-like CD94-NKG2A heterodimer(24). The inhibitory KIRs consist of KIR2DL1 which recognizesthe group of HLA-C molecules with a Lys80 residue (HLA-C2group), KIR2DL2/2DL3 which recognize the group of HLA-Cwith an Asn80 residue (HLA-C1 group), and KIR3DL1 which rec-ognize Bw4-containing HLA-B alleles (HLA-Bw4 group). TheCD94-NKG2A heterodimer recognizes a complex of the nonclas-sical MHC-I molecule HLA-E bound to peptides from the leadersequences of other MHC-I molecules. Using fresh surgical speci-mens from patients with malignancy involving the liver, we showthat the human liver NK cell repertoire as based on inhibitoryreceptor status is substantially different from its circulating coun-terpart. We have characterized the phenotype and function of NKcells in paired autologous liver and blood samples. We have foundthat human liver NK CD16� NK cells display a decreased func-tional response toward MHC-I-deficient target cells and that this iscaused by a decrease in the proportion of NK cells with a licensedphenotype. Our findings emphasize the multitude of ways in whichimmune responses can be regulated in an organ-specific manner.

Materials and MethodsCell isolation

Blood and matched liver samples were collected from 46 patients under-going elective hepatic resection at Memorial Sloan-Kettering Cancer Cen-ter. The indications for operation included metastatic disease (n � 34),primary cancer (n � 10), and benign disease (n � 2). Detailed clinico-pathologic data are shown in supplemental Table I.4 Informed consent wasobtained according to an institutional review board-approved protocol.Blood was drawn intraoperatively and PBMCs were isolated by densitycentrifugation over Ficoll-Paque Plus (GE Healthcare). Liver tissue wasprocured more than 5 cm away from any visible tumor and processedimmediately after resection. Notably, we have not found any difference inexpression of the activation markers HLA-DR and CD69 on NK cells inthe peritumoral (within 2 cm from the tumor) vs distant liver (data notshown). To remove contaminating circulating blood cells, hepatic vesselswere flushed with a solution of HBSS containing type IV collagenase (1mg/ml; Sigma-Aldrich), DNase (50 ng/ml; Roche Diagnostics), and 2%

endotoxin-free human AB� serum (Invitrogen). The liver specimen wasthen morselized and digested in the collagenase solution at 37oC for 30min. The tissue slurry was quenched with cold HBSS containing DNase(25 ng/ml) and passed through a 100-�m filter. The resulting cell suspen-sion was centrifuged twice at 250 � g for 10 min to remove fat and debrisand then hepatocytes were removed with a low-speed spin (30 � g for 1min). The resulting nonparenchymal cells were layered over a Ficoll-Paquedensity gradient and centrifuged at 1000 � g for 20 min, at which time theinterface layer containing liver mononuclear cells (LMNCs) was harvested.

HLA genotyping

HLA typing was performed preoperatively on select patients using a com-bination of intermediate resolution sequence-specific typing for HLA-A,HLA-B, and HLA-Cw loci (Dynal Systems), followed by high-resolutiontyping by sequence-specific oligonucleotide probe analysis (Orchid Diag-nostics) as previously described (25). HLA-KIR ligand groups C1, C2, andBw4 were assigned to each donor and shown in supplemental Table I (26).

Flow cytometry

Immunophenotypic analysis of cells was performed with seven-colorflow cytometry on a FACSAria cytometer (BD Biosciences) usingmAbs conjugated to FITC, PE, PerCP-Cy5.5, PE-Cy7, allophycocyanin,allophycocyanin-Cy7, and Alexa Fluor 700 (AF700). Abs used for cellanalysis included FITC-CD16 (3G8), FITC-CD62L (Dreg56), FITC-CD94 (HP-3D9), PE-NKp30 (P30-15), PE-NKp44 (P44-81), PE-NKp46 (9E2), PE-TRAIL (RIK-2), PE-NKG2D (1D11), PerCP-Cy5.5-CD3 (SK7), PE-Cy7-CD16 (3G8), allophycocyanin-CD161 (DX12),allophycocyanin-Cy7-HLA-DR (L243), allophycocyanin–Cy7-CD69(FN50), AF700-CD56 (B159) (all from BD Biosciences), PE-CD56(AF12-7H3), allophycocyanin-KIR2DL1/2DS1 (11PB6), allophycocyanin-KIR2DL2/2DL3 (DX27), allophycocyanin-KIR3DL1 (DX9) (all fromMiltenyi Biotec), PE-NKG2A (Z199; Beckman Coulter), and PE-NKG2C (134591; R&D Systems). Intracellular perforin and granzymecontent was detected using FITC-perforin (dG9; eBioscience) and PE-granzyme B (eBioGrB; eBioscience) following cell permeabilization(Cytofix/Cytoperm; BD Biosciences). Ig isotype controls were usedwhere appropriate. Flow cytometry data were analyzed with FlowJosoftware (Tree Star).

FACS

LMNCs or PBMCs were stained with the following Abs for cell separation:FITC-CD16, PE-CD56, PE-Cy7-CD11b (ICR-F44; BD Biosciences),allophycocyanin-KIR2DL1/2DL2, allophycocyanin-KIR2DL2/2DL3, allo-phycocyanin-KIR3DL1, allophycocyanin-CD3 (SK7; BD Biosciences), al-lophycocyanin-Cy7-CD3 (SK7; BD Biosciences), and allophycocyanin-Cy7-CD14 (M�P9; BD Biosciences). Cell sorting was performed using aMoFlo (DakoCytomation) or BD FACSAria (BD Biosciences) cell sorterand post sort purities were routinely �98%. NK cells were selected fromthe lymphocyte gate as CD56posCD3neg cells and in some cases as CD56dim

CD16posCD3neg cells. Liver Kupffer cells (KC) were sorted as CD14pos

CD11bhigh cells contained within the monocyte gate.

Cytotoxicity assay

Standard cytotoxicity assays were performed using freshly isolated NKcells and 51Cr-labeled MHC-I-negative target cells. The K562 cell line wasobtained from the American Type Culture Collection and the LCL721.221cell line was provided by P. Parham (Stanford University, Palo Alto, CA).Cell lines were maintained in complete RPMI 1640 containing 10% heat-inactivated FCS, 2 mM L-glutamine, 0.1% 2-ME, 100 U/ml penicillin, and100 �g/ml streptomycin. Two million target cells were labeled with 100�Ci of Na51CrO4 (PerkinElmer Life and Analytical Sciences) at 37°C for90 min and then washed with medium. Lysis assays were performed intriplicate for 4 h at the indicated E:T ratio. In some experiments, NK cellswere treated with 100 ng/ml concanamycin A (ICN Biomedical) or aDMSO (Fischer Scientific) vehicle for 2 h at 37°C before the addition oftargets. Chromium release was measured with a TopCount NXT microplatescintillation and luminescence counter (PerkinElmer) and percent specificlysis was calculated as (cpm experimental � cpm spontaneous release) �100/(cpm maximum release � cpm spontaneous release). In other exper-iments, freshly isolated NK cells were cultured overnight in medium alone,in medium containing 200 IU/ml recombinant human IL-2 (Proleukin; No-vartis), or with autologous liver KC at an NK:KC ratio of 5:1. Beforecoculture, KC were pulsed for 1 h with LPS (1 �g/ml from Escherichia coli055:B5; Sigma-Aldrich) or poly(I:C) (50 ng/ml; InvivoGen) and thenwashed. Following the culture period, NK cells were recounted for use incytotoxicity assays.4 The online version of this article contains supplemental material.

1790 HUMAN LIVER NK CELLS

by guest on July 18, 2022http://w

ww

.jimm

unol.org/D

ownloaded from

Intracellular IFN-� detection

Detection of intracellular IFN-� was performed as previously described,with minor modifications (26). After the percentage of NK cells in LMNCsand PBMCs was determined by flow cytometry, 1 million LMNCs orPBMCs were cultured in 96-well round-bottom plates in complete mediumwith K562 cells at an NK cell:target cell ratio of 1:1. Brefeldin A (10�g/ml; BD Biosciences) was added after 1 h and the cells were cultured foran additional 12 h. For detection of intracellular IFN-� following CD16receptor cross-linking, similar experiments were performed using anti-CD16-coated plates prepared by labeling 96-well enzyme immunoassayhigh-binding plates (Fisher Scientific) with purified anti-CD16 Ab(3G8; BD Biosciences) in PBS overnight at 4°C. After culture, the cellswere stained with Abs to CD3, CD56, CD16, NKG2A, and inhibitoryself-KIR receptors before fixation and permeabilization and stainingwith a FITC-conjugated anti-IFN-� Ab (intracellular cytokine detectionkit; BD Biosciences).

Statistics

Data were analyzed for statistical significance using the paired Student ttest (GraphPad Prism statistical software). Values of p � 0.05 were con-sidered statistically significant.

ResultsHuman liver contains a lower percentage of CD16� NK cells

Using flow cytometry, we analyzed LMNCs and autologousPBMCs from 15 patients to determine the composition of NK cellsubsets in the liver (Fig. 1, A and B). There was considerablevariability in NK cell subset composition within the liver. WhereasCD16� NK cells accounted for a mean (SEM) of 95 1%(range, 89–98%) of NK cells in the blood, CD16� NK cells rep-resented 50 4% (range, 27–75%) of all liver NK cells (Fig. 1B).Furthermore, liver NK cells displayed an activated phenotypewhen compared with blood NK cells because they had higher ex-pression of CD69, HLA-DR, and NKp44 and lower CD62L(Fig. 1, C and D).

Human liver NK cells are poorly cytolytic

Because of their activated phenotype, we hypothesized that resi-dent liver NK cells would have greater cytolytic function thancirculating NK cells. On the contrary, we found that freshly iso-lated blood NK cells were consistently more potent than autolo-gous liver NK cells in their ability to kill K562 and MHC-I-defi-cient LCL721.221 targets (Fig. 2, A and B). To determine whethersubset composition was merely responsible for the difference inlysis between blood and liver bulk NK cells, we compared the lyticfunction of blood and liver CD16� NK cells. Strikingly, CD16�

liver NK cells demonstrated markedly less killing (Fig. 2, C andD). The low number of CD16� NK cells within our limited num-ber of paired blood and liver samples that were freshly isolated didnot allow for further analysis. However, in agreement with previ-ous reports (27), CD16� blood NK cells were less cytolytic thanCD16� blood NK cells (data not shown) and in two patients whereadequate numbers could be isolated to compare sorted liver andblood CD16� NK cells, lysis of class I-negative targets was sim-ilar (liver: 30.9 0.6% vs blood: 32.9 1.3% at E:T 15:1 forK562 and liver: 12.8 1.6% vs blood: 10.6 0.2% at E:T 15:1for LCL721.221). Therefore, human liver is enriched in NK cellswith decreased cytotoxic function because of both a higher per-centage of CD16� NK cells and overall decreased cytolytic activ-ity of liver CD16� NK cells.

CD16� liver NK cells express activating receptors and containintracellular perforin and granzyme B

We next sought to determine the reason for decreased killing byCD16� liver NK cells. It has become clear that activating re-ceptors contribute substantially to NK cell specificity and mayprovide the critical threshold of signals needed to override thecounterbalancing effects of inhibitory receptors to mount a

FIGURE 1. Human liver contains a variable percent-age of CD16� NK cells. A, Contour plots from onerepresentative patient demonstrate the gating strategyused for analysis by flow cytometry and cell sorting ofliver and blood mononuclear cell preparations. Numbersrepresent the percentage of cells within the indicatedgate. B, Scatter plots for 15 patients show the percent-ages of CD16� and CD16� NK cells in liver and bloodmononuclear cell preparations. C, Representative dotplots from one of eight patients are gated on CD56�

CD3� cells to demonstrate the phenotype of liver com-pared with blood NK cells. Numbers represent the per-centage of cells in their respective quadrants. D,Summary plots for activation marker expression onCD16� and CD16� NK cells in matched blood andliver samples from eight patients.

1791The Journal of Immunology

by guest on July 18, 2022http://w

ww

.jimm

unol.org/D

ownloaded from

productive response (28). We did not find a difference in theexpression of activating receptors between blood and liverCD16� NK cells that could account for their disparate cytotox-icity (supplemental Table II). Using concanamycin A to inhibitthe granule exocytosis pathway (29), we found that both liverand blood CD16� NK cells killed K562 targets predominantlythrough granule release (Fig. 3A). Both liver and blood CD16�

NK cells contained intracellular perforin and granzyme B, al-though a trend toward decreased perforin content was observedin liver NK cells (Fig. 3, B–D). Therefore, CD16� liver NKcells are armed for killing but have overall decreased cytotoxicactivity.

A lower proportion of liver NK cells possesses inhibitory KIRreceptors for self-MHC-I

NK cells that possess inhibitory receptors with specificity for self-MHC-I Ags are endowed with functional competence (i.e., li-

censed) and are highly responsive to tumor targets (17–23). There-fore, we investigated whether the expression of these inhibitoryreceptors was different on liver and blood NK cells. HLA-KIR ligandgroups C1, C2, and Bw4 were assigned preoperatively to each patientby HLA genotyping and KIRs specific to self-MHC-I were identified(self-KIR). NK cell expression of self-KIR and NKG2A receptorswas determined for eight patients (Fig. 4 and Table I).

The phenotype distribution of CD16� NK cells was markedly dif-ferent between liver and blood. CD16� NK cells in the liver contained

FIGURE 2. Human liver NK cellshave weaker cytolytic function thanblood NK cells. Bulk NK cells weresorted from matched blood and livermononuclear cell preparations andtested for their ability to lyse chromi-um-labeled K562 (A; five patients) and721.221 target cells (B; three patients).Similarly, CD16� NK cells were sortedfrom matched blood and liver mononu-clear cell preparations and their lyticability was tested in chromium releaseassays using K562 (C; five patients)and 721.221 (D; four patients) as tar-gets. One representative patient isshown in the left panel and in the rightpanel, data from blood and liver sam-ples of individual patients are summa-rized by percent specific lysis at thehighest E:T ratio.

FIGURE 3. CD16� liver NK cells contain cytolytic granules and ex-press perforin and granzymes. A, Sort-purified blood and liver CD16� NKcells were assayed for their ability to lyse K562 target cells in the presenceor absence of concanamycin A, an inhibitor of perforin release, or a DMSOvehicle control. One of three representative experiments with similar re-sults is shown. B, Representative contour plots of perforin expression inliver and blood NK cells. C and D, The mean fluorescence index (MFI) ofintracellular perforin (11 patients) and granzyme B (6 patients) in bloodand liver CD16� NK cells and CD16� NK cells is shown.

FIGURE 4. Liver NK cells express less inhibitory receptors for self-MHC-I than blood NK cells. Inhibitory KIRs with affinity for self-MHC-I(self-KIR) were determined preoperatively as described in Materials andMethods. A, Dot plots from one representative patient show the distributionof self-KIR and NKG2A on CD16� liver and blood NK cells. In this C1/C2Bw4/Bw6 patient, self-KIR-expressing NK cells are represented byKIR2DL1�, KIR2DL2�, or KIR3DL1� NK cells. The percent self-KIR�

(B) or NKG2A� CD16� NK cells (C) is shown for eight patients.

1792 HUMAN LIVER NK CELLS

by guest on July 18, 2022http://w

ww

.jimm

unol.org/D

ownloaded from

a statistically significant decreased percentage of both self-KIRpos

NKG2Aneg and self-KIRposNKG2Apos NK cells and a higher per-centage of self-KIRnegNKG2Apos and self-KIRnegNKG2Aneg NKcells (Table I and Fig. 4, B and C). In contrast, the KIR and NKG2Aphenotype of CD16� NK cells in blood and liver were similar (TableI). These data demonstrate that the liver when compared with bloodcontains a decreased number of licensed NK cells that express inhib-itory receptors for self-MHC-I. These results would suggest that thisdifference in NK cell phenotype should be associated with decreasedeffector function against MHC-I-deficient target cells.

The relative distribution of NK cell subpopulations in the liverand blood accounts for the differences in effector function inCD16� NK cells

We investigated how the inhibitory receptor subset phenotype ofliver CD16� NK cells contributed to their functional capacity. Weobserved that liver CD16� NK cells had modestly decreased in-tracellular IFN-� responses to K562 target cells (Fig. 5, A and B)and CD16 cross-linking than their matched blood CD16� NK cells(Fig. 5, C and D). We next compared the four subsets of NK cellsdefined by self-KIR and NKG2A and their IFN-� production fol-

lowing stimulation with K562 (Fig. 5B) and CD16 cross-linking(Fig. 5D). There were no significant differences between the re-sponses for each subset of NK cells in blood and liver. For exam-ple, the two self-KIRpos subsets (i.e., self-KIRposNKG2Aneg andself-KIRposNKG2Apos) had higher responses than the two self-KIRneg subsets in both blood and liver and displayed a similarhierarchy of responsiveness (Fig. 5, B and D). Cytolytic effectorfunction of the self-KIR expressing liver CD16� NK cells wascompared with liver CD16� NK cells that did not possess self-KIRusing standard chromium-release assays. Consistent with the NKcell licensing hypothesis, self-KIRpos liver NK cells had highercytolytic activity than self-KIRneg cells (Fig. 5E). We have notfound the degree of CD107a membrane expression on liver NKcells to correlate reliably with classical chromium-based lysis as-says (data not shown). Similar to a previously published compar-ison between licensed and unlicensed blood NK cells (19), we didnot find a difference in perforin expression between self-KIRpos

and self-KIRneg subsets of liver CD16� NK cells that could ac-count for the observed difference in killing (Fig. 5F). Taken to-gether, the decreased frequency of self-KIR-expressing CD16�

FIGURE 5. Self-KIR and NKG2A endow similarfunction to blood and liver CD16� NK cells. Matchedliver and blood mononuclear cells were cultured withK562 cells at a NK:K562 ratio of 1:1 (A and B) or onanti-CD16-coated plates (C and D) in the presence ofbrefeldin A. The percentage of IFN-��CD16� NK cellsis shown for five individual patients in A and C. In eachblood and liver, the percentage of self-KIR and/orNKG2A-expressing CD16� NK cells that are IFN-�� isshown for five patients in B and D. Each color-coded setof points represents a single patient. E, Liver CD16�

NK cells from five individual patients were sorted intoself-KIR� and self-KIR� subsets and tested for theirability to lyse K562 targets. F, Intracellular perforin ex-pression in self-KIR� and self-KIR� CD16� liver NKcells was compared in four patients.

Table I. Inhibitory receptor expression on liver NK cellsa

Self-KIR�NKG2A� Self-KIR�NKG2A� Self-KIR�NKG2A� Self-KIR�NKG2A�

Bulk NK cellsLiver 10.4 2.1 9.2 1.9 52.3 2.4 30.8 4.0Blood 29.4 5.7 20.2 2.6 26.3 3.3 20.2 2.4p 0.009 0.002 �0.0001 0.016

CD16� NK cellsLiver 2.9 0.6 3.6 0.6 63.6 3.2 29.7 3.1Blood 3.7 1.1 4.9 1.3 64.1 5.0 24.5 2.8p 0.28 0.14 0.87 0.09

CD16� NK cellsLiver 22.8 4.2 15.5 3.0 37.2 4.8 23.8 3.1Blood 39.7 4.4 21.7 2.8 22.5 3.0 15.8 2.2p 0.002 0.02 0.001 0.007

a Matched liver and blood samples are shown for eight individual patients.

1793The Journal of Immunology

by guest on July 18, 2022http://w

ww

.jimm

unol.org/D

ownloaded from

NK cells in the liver compared with the blood accounts for theoverall decreased cytolytic potential of liver CD16� NK cells.

The cytolytic restriction on liver CD16� NK cells is removed byIL-2 or an activated liver APC

IL-2 potentiates the cytotoxic capability of NK cells by up-regu-lating cytolytic machinery and increasing intracellular perforin(30). Moreover, human blood NK cells gain an equal level of cy-totoxicity when cultured in IL-15 regardless of the expression ofinhibitory receptors for MHC-I (20). To determine whether liverCD16� NK cells were anergic or whether their cytolytic restrictioncould be overcome, they were cultured overnight in IL-2. NK cellsbecame robust cytolytic effectors after activation with IL-2 withcomparable potency to IL-2-cultured blood CD16� NK cells (Fig.6A). Similar results were obtained using IL-15 (data not shown).Although a predominance of NK cells with weak resting cytolyticfunction may be advantageous to prevent hyperactivity in the po-tentially volatile, endotoxin-rich environment of the liver, theircytolytic function could be beneficial during an infectious or met-astatic insult. Therefore, we asked whether an activated residentliver immune cell could provide the necessary signals to boosttheir killing potential. When freshly isolated liver CD16� NK cellswere cultured overnight with autologous Kupffer cells activatedeither by LPS or poly(I:C), they became more potent killers ofK562 targets (Fig. 6B). Hence, although the liver contains an abun-dance of CD16� NK cells with cytolytic potential that is limited tosome degree by their inhibitory receptor expression, they are ca-pable of mounting effective lytic responses in the appropriate in-flammatory settings.

DiscussionThe threshold for activation of innate immune responses variesamong different organs. These activation profiles are controlled bydifferent organ-specific regulatory mechanisms that are required tosustain immune homeostasis yet can result in immunopathology if

disrupted (31). For example, the teeming microbial mass in thecolon does not induce overt inflammation in the intestinal mucosa,suggesting the existence of specialized regulatory mechanisms thatcurtail initiation of inflammatory and bactericidal programs againstthe luminal microflora. Recent studies have suggested that intes-tinal epithelial cells may have evolved polarized signaling mech-anisms to maintain colonic homeostasis and regulate tolerance andimmunity. Although basolateral TLR9 stimulation mobilizes aninflammatory immune cascade, apical (luminal) TLR9 stimulationresults in tolerogenic immune responses (32). Additionally, humanintestinal macrophages do not express innate response receptorsfor LPS (CD14) and do not produce proinflammatory cytokines inresponse to a range of inflammatory stimuli (33). Similarly, humanliver DCs promote CD4 T cell hyporesponsiveness and favor thegeneration of regulatory T cells (34) and human Kupffer cells re-spond to LPS paradoxically by secreting the anti-inflammatory cy-tokine IL-10 (35). In the lung, administration of TLR ligands in-duces high levels of the negative immunoregulator IDO that candecrease Th2-driven experimental asthma (36).

The immune environment of the liver is unique because it iscontinually exposed to commensal-derived endotoxin that ispresent in the portal circulation and therefore lends itself to con-stitutive TLR4 ligation in the absence of an infectious threat. LPSis a powerful activator of both innate and adaptive immunitythrough the triggering of inflammatory cascades, secretion ofproinflammatory cytokines, and maturation of specialized APCs.NK cells are present in abundant numbers in the liver and areresponsible for hepatocyte death and liver inflammation duringhepatitis B virus infection (37). Furthermore, hepatocytes make up60–80% of the cellular mass of the liver and the expression ofMHC-I on human hepatocytes has been well documented to be lowand even absent in vivo (38–40). This potentially MHC-I deficientenvironment lends itself to relief of the tonic inhibition on NKcells; however, NK cells do not appear to be hyperactive in thepotentially hostile immune environment of the liver. Therefore, toprevent liver injury and compromise liver function, it is crucial thatautoaggressive activation of NK cells be regulated.

It has been previously recognized that human liver NK cells areenriched in CD16� NK cells (41) and that this subset of NK cells,through cross-talk with activated resident liver macrophages, willproduce IFN-� (42). We have determined that the subset compo-sition of NK cells in the human liver is variable but on average isrepresented by equal percentages of CD16� and CD16� NK cells.Furthermore, we and others have found that NK cells in the humanliver display an activated phenotype (43) and we show here thatthis phenotype is accounted for predominantly by the CD16� sub-set of liver NK cells. Similar to CD16� lymph node NK cells, thissubset of human liver NK cells express high levels of HLA-DR,NKp44, and CD69 and low levels of CD62L (8, 44). Despite theiractivated phenotype, CD16� NK cells had poor cytotoxic capacitylike their respective blood NK cell equivalents. However, CD16�

NK cells that are regarded as the cytotoxic subset of NK cells inthe liver were also poorly cytotoxic, suggesting one potential reg-ulatory means by which liver NK cell autoreactivity could bequelled.

The isolation of immune cells from solid organs is more difficultthan isolating circulating lymphocytes and requires processing ofthe tissues. Using propidium iodide and annexin V staining, weinitially established that the method of isolation did not influencethe viability or killing ability of the cells (data not shown). Fur-thermore, by subjecting whole blood to enzyme treatment and amock isolation procedure identical to that used to isolate liver NK

FIGURE 6. Liver CD16� NK cell cytolytic function is increased byIL-2 or activated KCs. CD16� NK cells were purified from the blood orliver of the same patient and cultured overnight in either medium or IL-2(200 IU/ml). The following day the lytic capacity of the cultured NK cellswas assayed using chromium-labeled K562 targets. These data are repre-sentative of experiments from four individual patients. B, CD16� liver NKcells were cultured overnight with autologous sort-purified KCs that werefirst pulsed with poly(I:C), LPS, or medium. The following day their lyticability was tested using chromium-labeled K562 targets. These data arerepresentative of experiments from three individual patients.

1794 HUMAN LIVER NK CELLS

by guest on July 18, 2022http://w

ww

.jimm

unol.org/D

ownloaded from

cells, we clarified that KIR and NKG2A expression were unaf-fected by the isolation process and, importantly, did not find anydifference in lysis of K562 cells (data not shown).

To avoid inadvertent destruction of normal tissues, the potenteffector functions of NK cells are controlled through their inter-actions with ubiquitously expressed MHC-I molecules. KIRs andNKG2A are inhibitory NK cell receptors that recognize MHC-Iand play a major role in mediating the balance between self-tol-erance and appropriately targeted effector responses. For example,these inhibitory NK cell receptors fulfill the missing-self hypoth-esis by inhibiting NK cell lysis and cytokine production when theyare engaged by their cognate ligands on normal tissues. Con-versely, NK cell effector functions are released when these ligandsare missing from an infected or malignant target cell. It is nowevident that these same inhibitory receptors are required for a NKcell to become a functionally competent effector. In other words, ithas been shown that to be strongly reactive to missing-self-targets,a NK cell is required to have at least one inhibitory receptor thatcan interact with a self-MHC-I ligand (17–23). This model for NKcell licensing provides an explanation for the decreased lytic re-sponses of liver CD16� NK cells based upon their repertoire ofself-KIR-deficient NK cells.

The regulation of NK cell responses by inhibitory receptor phe-notypes applies to the resting state and the requirement for self-MHC recognition in establishing functional tolerance can be over-come by cytokine stimulation (17, 20, 23). Along these lines, wehave found that CD16� liver NK cells like blood CD16� NK cellswould become potent effectors by short culture in IL-2 or IL-15.Similarly, others have found that 4 days of culture in IL-2 willactivate bulk human liver NK cells to become more potent cyto-lytic killers (45). Like human blood NK cells which can be acti-vated by DCs (46–48) and macrophages (49), liver CD16� NKcells can be activated by autologous liver KCs (42). We have sim-ilarly found that the lytic capacity of liver CD16� NK cells can beenhanced by activated liver DCs (data not shown) or KCs. Takentogether, this suggests that the lytic capacity of liver CD16� NKcells is regulated by their inhibitory receptor phenotype in the rest-ing state, but in an inflammatory state, they can become potenteffectors through cross-talk with liver APCs. Whether the tumor-bearing liver has an impact upon these results is difficult to deter-mine since most hepatectomy specimens that we can obtain arefrom cancer patients. For example, it has been suggested that com-pared with normal donors, patients with hepatic malignancy havesignificantly reduced proportions of NK and NKT cells expressingKIR receptors and CD94 (50). However, this study did not com-pare the proportions of CD16� NK cells and CD16� NK cells inthe diseased and normal livers as this ratio could potentially ex-plain a difference in overall KIR and NKG2A expression. Further-more, we have not found differences in CD94 expression betweenliver and blood NK cell subsets (supplemental Table II and datanot shown).

In conclusion, the liver contains a repertoire of NK cells that areless cytolytic than circulating NK cells. One of the predominantcontributing factors is that the subset of CD16� NK cells expressesless inhibitory KIR molecules with specificity for self-MHC-I li-gands and is less cytotoxic. We hypothesize that through someprocess of selection, the liver gains a repertoire of tolerant nonli-censed CD16� NK cells that may be beneficial in preventing liverinjury and autoimmunity in its potentially hostile local immuneenvironment.

AcknowledgmentsWe are grateful to Hebroon Obaid for technical support and Jan Hendrikx,Mark Kweens, and Patrick Anderson of the Memorial Sloan-Kettering

Cancer Center Flow Cytometry Core Facility for assistance with cellsorting.

DisclosuresThe authors have no financial conflict of interest.

References1. Opelz, G., T. Wujciak, B. Dohler, S. Scherer, and J. Mytilineos. 1999. HLA

compatibility and organ transplant survival: collaborative transplant study. Rev.Immunogenet. 1: 334–342.

2. Mazariegos, G., V. J. Reyes, I. R. Marino, A. J. Demetris, B. Flynn, W. Irish,J. McMichael, J. J. Fung, and T. E. Starzl. Weaning of immunosuppression inliver transplant recipients. 1997. Transplantation 63: 243–249.

3. Crispe, I. N., M. Giannandrea, I. Klein, B. John, B. Sampson, and S. Wuensch.2006. Cellular and molecular mechanisms of liver tolerance. Immunol. Rev. 213:101–118.

4. Doherty, D. G., and C. O’Farrelly. 2000. Innate and adaptive lymphoid cells inthe human liver. Immunol. Rev. 174: 5.

5. Caligiuri, M. A. 2008. Human natural killer cells. Blood 112: 461–469.6. Cooper, M. A., T. A. Fehniger, S. C. Turner, K. S. Chen, B. A. Ghaheri,

T. Ghayur, W. E. Carson, and M. A. Caligiuri. 2001. Human natural killer cells:a unique innate immunoregulatory role for the CD56bright subset. Blood 97:3146–3151.

7. Fehniger, T. A., M. A. Cooper, G. J. Nuovo, M. Cella, F. Facchetti, M. Colonna,and M. A. Caligiuri. 2003. CD56bright natural killer cells are present in humanlymph nodes and are activated by T cell-derived IL-2: a potential new link be-tween adaptive and innate immunity. Blood 101: 3052–3057.

8. Ferlazzo, G., D. Thomas, S. L. Lin, K. Goodman, B. Morandi, W. A. Muller,A. Moretta, and C. Munz. 2004. The abundant NK cells in human secondarylymphoid tissues require activation to express killer cell Ig-like receptors andbecome cytolytic. J. Immunol. 172: 1455–1462.

9. Bajenoff, M., B. Breart, A. Y. Huang, H. Qi, J. Cazareth, V. M. Braud,R. N. Germain, and N. Glaichenhaus. 2006. Natural killer cell behavior in lymphnodes revealed by static and real-time imaging. J. Exp. Med. 203: 619.

10. Martin-Fontecha, A., L. L. Thomsen, S. Brett, C. Gerard, M. Lipp,A. Lanzavecchia, and F. Sallusto. 2004. Induced recruitment of NK cells tolymph nodes provides IFN-� for TH1 priming. Nat. Immunol. 5: 1260–1265.

11. Morandi, B., G. Bougras, W. A. Muller, G. Ferlazzo, and C. Munz. 2006. NKcells of human secondary lymphoid tissues enhance T cell polarization via IFN-�secretion. Eur. J. Immunol. 36: 2394–2400.

12. Strowig, T., F. Brilot, F. Arrey, G. Bougras, D. Thomas, W. A. Muller, andC. Munz. 2008. Tonsilar NK cells restrict B cell transformation by the Epstein-Barr virus via IFN-�. PLoS Pathog. 4: e27.

13. Kopcow, H. D., D. S. Allan, X. Chen, B. Rybalov, M. M. Andzelm, B. Ge, andJ. L. Strominger. 2005. Human decidual NK cells form immature activating syn-apses and are not cytotoxic. Proc. Natl. Acad. Sci. USA 102: 15563–15568.

14. Manaster, I., S. Mizrahi, D. Goldman-Wohl, H. Y. Sela, N. Stern-Ginossar,D. Lankry, R. Gruda, A. Hurwitz, Y. Bdolah, R. Haimov-Kochman, et al. 2008.Endometrial NK cells are special immature cells that await pregnancy. J. Immu-nol. 181: 1869–1876.

15. Hanna, J., D. Goldman-Wohl, Y. Hamani, I. Avraham, C. Greenfield,S. Natanson-Yaron, D. Prus, L. Cohen-Daniel, T. I. Arnon, I. Manaster, et al.2006. Decidual NK cells regulate key developmental processes at the humanfetal-maternal interface. Nat. Med. 12: 1065–1074.

16. Ljunggren, H. G., and K Karre. 1990. In search of the “missing self ”: MHCmolecules and NK cell recognition. Immunol. Today 11: 237–244.

17. 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.

18. 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-tolerance withoutexpressing inhibitory receptors specific for self-MHC molecules. Blood 105:4416–4423.

19. Anfossi, N., P. Andre, S. Guia, C. S. Falk, S. Roetynck, C. A. Stewart, V. Breso,C. Frassati, D. Reviron, D. Middleton, et al. 2006. Human NK cell education byinhibitory receptors for MHC class I. Immunity 25: 331–342.

20. Cooley, S., F. Xiao, M. Pitt, M. Gleason, V. McCullar, T. L. Bergemann,K. L. McQueen, L. A. Guethlein, P. Parham, and J. S. Miller. 2007. A subpopu-lation of human peripheral blood NK cells that lacks inhibitory receptors forself-MHC is developmentally immature. Blood 110: 578–586.

21. Yu, J., G. Heller, J. Chewning, S. Kim, W. M. Yokoyama, and K. C. Hsu. 2007.Hierarchy of the human natural killer cell response is determined by class andquantity of inhibitory receptors for self-HLA-B and HLA-C ligands. J. Immunol.179: 5977–5989.

22. Kim, S., J. B. Sunwoo, L. Yang, T. Choi, Y. J. Song, A. R. French, A. Vlahiotis,J. F. Piccirillo, M. Cella, M. Colonna, et al. 2008. HLA alleles determine differ-ences in human natural killer cell responsiveness and potency. Proc. Natl. Acad.Sci. USA 105: 3053–3058.

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

24. Hsu, K. C., and B. Dupont. 2005. Natural killer cell receptors: regulating innateimmune responses to hematologic malignancy. Semin. Hematol. 42: 91–103.

1795The Journal of Immunology

by guest on July 18, 2022http://w

ww

.jimm

unol.org/D

ownloaded from

25. Hsu, K. C., C. A. Keever-Taylor, A. Wilton, C. Pinto, G. Heller, K. Arkun, R. J.O’Reilly, M. M. Horowitz, and B. Dupont. 2005. Improved outcome in HLA-identical sibling hematopoietic stem-cell transplantation for acute myelogenousleukemia predicted by KIR and HLA genotypes. Blood 105: 4878–4884.

26. Chewning, J. H., C. N. Gudme, K. C. Hsu, A. Selvakumar, and B. Dupont. 2007.KIR2DS1-positive NK cells mediate alloresponse against the C2 HLA-KIR li-gand group in vitro. J. Immunol. 179: 854–868.

27. Nagler, A., L. L. Lanier, S. Cwirla, and J. H. Phillips. 1989. Comparative studiesof human FcRIII-positive and negative natural killer cells. J. Immunol. 143:3183–3191.

28. Lanier, L. L. 2008. Up on the tightrope: natural killer cell activation and inhibi-tion. Nat. Immunol. 9: 495–502.

29. Kataoka, T., N. Shinohara, H. Takayama, K. Takaku, S. Kondo, S. Yonehara, andK. Nagai. 1996. Concanamycin A, a powerful tool for characterization and es-timation of contribution of perforin- and Fas-based lytic pathways in cell-medi-ated cytotoxicity. J. Immunol. 156: 3678–3686.

30. Zhang, J., I. Scordi, M. J. Smyth, and M. G. Lichtenheld. 1999. Interleukin 2receptor signaling regulates the perforin gene through signal transducer and ac-tivator of transcription (Stat)5 activation of two enhancers. J. Exp. Med. 190:1297–1308.

31. Raz, E. 2007. Organ-specific regulation of innate immunity. Nat. Immunol.8: 3–4.

32. Lee, J., J. H. Mo, K. Katakura, I. Alkalay, A. N. Rucker, Y. T. Liu, H. K. Lee,C. Shen, G. Cojocaru, S. Shenouda, et al. 2006. Maintenance of colonic ho-meostasis by distinctive apical TLR9 signalling in intestinal epithelial cells. Nat.Cell Biol. 8: 1327–1336.

33. Smythies, L. E., M. Sellers, R. H. Clements, M. Mosteller-Barnum, G. Meng,W. H. Benjamin, J. M. Orenstein, and P. D. Smith. 2005. Human intestinal mac-rophages display profound inflammatory anergy despite avid phagocytic and bac-tericidal activity. J. Clin. Invest. 115: 66–75.

34. Bamboat, Z. M., J. A. Stableford, G. Plitas, B. M. Burt, H. M. Nguyen,A. P. Welles, M. Gonen, J. W. Young, and R. P. DeMatteo. 2009. J. Immunol.182: 1901–1911.

35. Knolle, P., J. Schlaak, A. Uhrig, P. Kempf, K. H. Meyer zum Buschenfelde, andG. Gerken. 1995. Human Kupffer cells secrete IL-10 in response to lipopolysac-charide (LPS) challenge. J. Hepatol. 22: 226–229.

36. Hayashi, T., L. Beck, C. Rossetto, X. Gong, O. Takikawa, K. Takabayashi,D. H. Broide, D. A. Carson, and E. Raz. 2004. Inhibition of experimental asthmaby indoleamine 2,3-dioxygenase. J. Clin. Invest. 114: 270–279.

37. Dunn, C., M. Brunetto, G. Reynolds, T. Christophides, P. T. Kennedy,P. Lampertico, A. Das, A. R. Lopes, P. Borrow, K. Williams, et al. 2007. Cy-tokines induced during chronic hepatitis B virus infection promote a pathway forNK cell-mediated liver damage. J. Exp. Med. 204: 667–680.

38. Daar, A. S., S. V. Fuggle, J. W. Fabre, A. Ting, and P. J. Morris. 1984. Thedetailed distribution of HLA-A, B, C antigens in normal human organs. Trans-plantation 38: 287.

39. Lau, J. Y., G. L. Bird, N. V. Naoumov, and R. Williams. 1993. Hepatic HLAantigen display in chronic hepatitis B virus infection. Relation to hepatic expres-sion of HBV genome/gene products and liver histology. Dig. Dis. Sci. 38:888–895.

40. So, S. K., D. C. Snover, N. L. Ascher, and J. L. Platt. 1987. Class I majorhistocompatibility complex antigens are induced on hepatocytes in rejecting hu-man liver grafts. Transplant. Proc. 19: 2464–2465.

41. Hata, K., X. R. Zhang, S. Iwatsuki, D. H. Van Thiel, R. B. Herberman, andT. L. Whiteside. 1990. Isolation, phenotyping, and functional analysis of lym-phocytes from human liver. Clin. Immunol. Immunopathol. 56: 401–419.

42. Tu, Z., A. Bozorgzadeh, R. H. Pierce, J. Kurtis, I. N. Crispe, and M. S. Orloff.2008. TLR-dependent cross talk between human Kupffer cells and NK cells.J. Exp. Med. 205: 233–244.

43. Tu, Z., A. Bozorgzadeh, I. N. Crispe, and M. S. Orloff. 2007. The activation stateof human intrahepatic lymphocytes. Clin. Exp. Immunol. 149: 186–193.

44. Mizrahi, S., E. Yefenof, M. Gross, P. Attal, A. Ben Yaakov, D. Goldman-Wohl,B. Maly, N. Stern, G. Katz, R. Gazit, R. V. Sionov, et al. 2007. A phenotypic andfunctional characterization of NK cells in adenoids. J. Leukocyte Biol. 82:1095–1105.

45. Ishiyama, K., H. Ohdan, M. Ohira, H. Mitsuta, K. Arihiro, and T. Asahara. 2006.Difference in cytotoxicity against hepatocellular carcinoma between liver andperiphery natural killer cells in humans. Hepatology 43: 362–372.

46. Ferlazzo, G., M. L. Tsang, L. Moretta, G. Melioli, R. M. Steinman, and C. Munz.2002. Human dendritic cells activate resting natural killer (NK) cells and arerecognized via the NKp30 receptor by activated NK cells. J. Exp. Med. 195:343–351.

47. Gerosa, F., B. Baldani-Guerra, C. Nisii, V. Marchesini, G. Carra, andG. Trinchieri. 2002. Reciprocal activating interaction between natural killer cellsand dendritic cells. J. Exp. Med. 195: 327–333.

48. Piccioli, D., S. Sbrana, E. Melandri, and N. M. Valiante. 2002. Contact-depen-dent stimulation and inhibition of dendritic cells by natural killer cells. J. Exp.Med. 195: 335–341.

49. Nedvetzki, S., S. Sowinski, R. A. Eagle, J. Harris, F. Vely, D. Pende,J. Trowsdale, E. Vivier, S. Gordon, and D. M. Davis. 2007. Reciprocal regulationof human natural killer cells and macrophages associated with distinct immunesynapses. Blood 109: 3776–3785.

50. Norris, S., D. G. Doherty, M. Curry, G. McEntee, O. Traynor, J. E. Hegarty, andC. O’Farrelly. 2003. Selective reduction of natural killer cells and T cells ex-pressing inhibitory receptors for MHC class I in the livers of patients with hepaticmalignancy. Cancer Immunol. Immunother. 52: 53–58.

1796 HUMAN LIVER NK CELLS

by guest on July 18, 2022http://w

ww

.jimm

unol.org/D

ownloaded from

Supplemental Table 1. Patient Characteristics.

Chemo (Chemotherapy within 3 months of surgery), Embo (embolization within 3 months of surgery), Cholangio (Cholangiocarcinoma), CRC (Colorectal cancer), GB (Gallbladder cancer), HCC (Hepatocellular carcinoma), GIST (Gastrointestinal stromal tumor), FNH (Focal nodular hyperplasia), M (Metastatic disease), P (Primary liver tumor), PV (Portal vein embolization), HA (Hepatic artery embolization).

Patient Age Gender Pathology Primary/Mets Chemo Embo Figure HLA Haplotype Self-KIR 1 ≥60 F Cholangio P N N 5A-D Bw4/Bw4 C1/C2 3DL1, 2DL2, 2DL1 2 ≥60 M CRC M Y N 3C-D, 5F Bw4/Bw6 C1/C1 3DL1, 2DL2 3 ≥60 F GB P N N 5E Bw4/Bw6 C1/C1 3DL1, 2DL2 4 <60 M CRC M Y N 4 Bw4/Bw6 C1/C1 3DL1, 2DL2 5 ≥60 M CRC M Y N 4 Bw4/Bw6 C1/C1 3DL1, 2DL2 6 ≥60 M HCC P N N 5E Bw4/Bw6 C1/C2 3DL1, 2DL2, 2DL1 7 <60 F GB M N N 5E Bw4/Bw6 C1/C2 3DL1, 2DL2, 2DL1 8 <60 M CRC M Y N 4, 5A-D Bw4/Bw6 C1/C2 3DL1, 2DL2, 2DL1 9 <60 F CRC M Y N 5A-D Bw4/Bw6 C1/C2 3DL1, 2DL2, 2DL1

10 ≥60 F GB P N N 5A-D, 6A Bw4/Bw6 C1/C2 3DL1, 2DL2, 2DL1 11 <60 F GB P N N 2, 4 Bw4/Bw6 C1/C2 3DL1, 2DL2, 2DL1 12 ≥60 M CRC M N N 4 Bw4/Bw6 C1/C2 3DL1, 2DL2, 2DL1 13 ≥60 F CRC M Y N 4 Bw4/Bw6 C1/C2 3DL1, 2DL2, 2DL1 14 <60 F CRC M N N 4 Bw4/Bw6 C1/C2 3DL1, 2DL2, 2DL1 15 ≥60 M CRC M Y N 5E Bw4/Bw6 C2/C2 3DL1, 2DL1 16 ≥60 F Adenoma M N N 5A-D Bw4/Bw6 C2/C2 3DL1, 2DL1 17 ≥60 M CRC M Y N 3C-D, 5F Bw4/Bw6 C2/C2 3DL1, 2DL1 18 <60 F HCC P N N 5E Bw6/Bw6 C1/C1 2DL2 19 <60 M CRC M Y N 3C-D, 5F Bw6/Bw6 C1/C1 2DL2 20 ≥60 F CRC M Y N 3C-D, 5F Bw6/Bw6 C2/C2 2DL1 21 ≥60 F Breast M Y N 3A, 6A 22 <60 M CRC M N N 6B 23 ≥60 M CRC M N N 4, 6B 24 <60 F CRC M Y N 3C-D, 6B 25 ≥60 F GIST M N N 1, 3C-D 26 ≥60 M HCC P N N 3C-D 27 <60 M GIST M N N 3C-D 28 ≥60 M GB P N N 3C-D 29 <60 M CRC M Y N 1, 3C-D 30 <60 M GIST M N N 3C-D 31 <60 F FNH P N N 1,2 32 ≥60 M CRC M N N 1 33 ≥60 M CRC M N N 2 34 ≥60 F CRC M Y PV 1 35 ≥60 M CRC M Y N 1 36 <60 F CRC M Y PV 1 37 <60 M CRC M Y N 1 38 <60 M CRC M Y N 1, 6A 39 ≥60 M CRC M Y N 2 40 ≥60 M CRC M Y N 2 41 <60 M GIST M N HA 2 42 <60 M HCC P N N 1 43 <60 F GB P N N 2 44 <60 F CRC M Y N 2 45 ≥60 M Melanoma M N N 2, 3A 46 ≥60 M Angiosarcoma M N N 2, 3A

Supplemental Table 2. Activating receptors on CD16+ liver and blood NK cells

Percent positivity of various surface markers on CD16+ NK cells (mean ± SD) from matched liver and blood samples from 6 patients.

TRAIL CD94 NKG2C NKG2D CD161 NKp30 NKp46 Liver 3.0 ± 1.4 50.2 ± 7.2 47.5 ± 2.5 90.8 ± 3.2 73.2 ± 8.5 54.3 ± 7.8 90.2 ± 2.9 Blood 1.9 ± 0.8 54.0 ± 5.7 46.5 ± 2.5 90.8 ± 4.4 77.8 ± 9.9 55.3 ± 8.5 93.5 ± 1.4 p value 0.2 0.3 0.8 1.0 0.1 0.6 0.3