Epub ahead of print April 22, 2016 - doi:10.1189/jlb.

2A1115-506RR

Article

Listeria monocytogenes infection

differentially affects expression of ligands for
NK cells and NK cell responses, depending
on the cell type infected
Hamid Shegarfi,*,,,1 Bent Rolstad,*,2 Kevin P. Kane, and Janne Nestvold*
*Department of Anatomy, Institute of Basic Medical Sciences, University of Oslo, Norway; Institute for Surgical Research, Oslo
University Hospital, Oslo, Norway; Atlantis Medical University College, Kolbotn, Norway; and Medical Microbiology and
Immunology, University of Alberta, Edmonton, Alberta, Canada
RECEIVED NOVEMBER 13, 2015; REVISED APRIL 1, 2016; ACCEPTED APRIL 6, 2016. DOI: 10.1189/jlb.2A1115-506RR

ABSTRACT

Introduction

The pivotal role of NK cells in viral infection is extensively

studied, whereas the role of NK cells in bacterial
infection has been poorly investigated. Here, we have
examined how Listeria monocytogenes (LM) affects
expression of ligands for NK cell receptors and subsequent NK cell responses, depending on the type of
cell infected. LM infected rat cell lines derived from
different tissues were coincubated with splenic NK cells,
and NK cell proliferation and IFN-g production were
measured. In addition, expression of ligands for the NK
cell receptors Ly49 and NK cell receptor protein 1 (NKRP1), MHC class I and C-type lectin-related molecules,
respectively, was assessed. Infected pleural R2 cells,
but not epithelium-derived colon carcinoma cell line
CC531 cells, induced proliferation of NK cells. Reporter
cells expressing the inhibitory NKR-P1G receptor or the
activating NKR-P1F receptor were less stimulated under
incubation with infected CC531 cells versus uninfected
CC531 controls, suggesting that the ligand(s) in question
were down-regulated by infection. Conversely, LM infection of R2 cells did not affect reporter cell stimulation
compared with uninfected R2 controls. We characterized a rat monocyte cell line, termed RmW cells. In
contrast to LM infected R2 cells that up-regulate MHC
class I molecules, RmW cells displayed unchanged
MHC class I expression following infection. In line with
MHC class I expression, more NK cells produced a
higher amount of IFN-g against infected R2 cells compared with RmW cells. Together, L. monocytogenes
infection may variously regulate cellular ligands for NK
cells, depending on the cell type infected, affecting the
outcome of NK cell responses. J. Leukoc. Biol.
100: 000000; 2016.

NK cells are a subset of circulating immune cells capable of

recognizing and destroying a wide variety of target cells,
including transformed, pathogen-infected, transplanted,
antibody-coated, and stressed cells [1]. The activities of NK cells
are regulated by a repertoire of activating and inhibitory
receptors interacting with adequate cell surface ligands. NK cell
stimulation may result in effector responses as release of
proinammatory cytokines or lysis of target cells [2]. In
mammals, there are 2 main classes of NK cell receptors: the Ig
superfamily receptors, including the KIRs, and the structurally
unrelated killer lectin-like receptors, which consists of the Ly49
and NKR-P1 families [3]. Akin to KIRs in humans recognizing
MHC-I, rat NK cells express functionally analogous Ly49
receptors [1, 4]. In contrast to a single NKR-P1 receptor in
humans, the rat NKR-P1 family has 4 members, including the
activating NKR-P1A, NKR-P1F and the inhibitory NKR-P1B,
NKR-P1G receptor interacting with Clr (C-type lectin-related or
osteoclast inhibitory lectin) molecules [5]. Rat NK cells are
phenotypically dened as NKR-P1A+CD32 cells [4].
NK cells use mechanisms to determine whether a cell is healthy
or pathologically altered. The missing self hypothesis suggests
that NK cells monitor cells for normal expression of MHC-I by
inhibitory KIR/Ly49 receptors [6]. Viruses down-regulate surface
MHC-I molecules to avoid recognition by CD8+ T cells, which can
render the cells susceptible to NK cell lysis [1]. To escape NK
cell-mediated control, viruses, such as cytomegalovirus, can
encode surrogate ligands for inhibitory NK cell receptors and/or
are able to regulate MHC-I molecules differentially [3]. The
impact of bacterial infection on ligands for NK cell receptors is
not well investigated.
LM is a Gram-positive intracellular bacterial pathogen that
generally causes acute infection, known as listeriosis, in

Abbreviations: BM = bone marrow, CC531 = colon carcinoma cell line

CC531, CD = cluster of differentiation, CHO = Chinese hamster ovary, CLN =
cervical lymph node, Clr = C-type lectin related, cRPMI = complete RPMI,
EGFP = enhanced GFP, KIR = killer cell Ig-like receptor, LM = Listeria
monocytogenes, MHC-I = MHC class I, NKR-P1 = NK cell receptor protein 1,
RCMV = rat CMV, RmW = rat myeloid cell line, RT1 = rat MHC class I

1. Correspondence: Institute for Surgical Research, Oslo University Hospital

HF, Sognsvannsveien N-0316 Oslo, Norway. E-mail: hamid.shegar@gmail.
com
2. Deceased.

0741-5400/16/0100-0001 Society for Leukocyte Biology

Volume 100, November 2016

Journal of Leukocyte Biology 1

Copyright 2016 by The Society for Leukocyte Biology.

immunocompetent individuals or pregnant women. Naturally

acquired listeriosis in humans is food borne, and hence, invasion
proceeds via the intestine [7]. After oral uptake, LM cross the
intestinal mucosa, disseminate via the bloodstream, and
replicate in different cell types, such as epithelial cells and
macrophages [8]. NK cells are part of the rst line of defense
against invading pathogens [1]. Innate production of IFN-g is
critical for resistance to infection and for control of an
ongoing LM infection [9]. NK cells and NKT cells are known to
secrete IFN-g, which consecutively leads to activation of
macrophages necessary for limiting the infection [1012]. NK
cell responses to LM-infected cells involve a complex network of
priming, proinammatory cytokine stimulation, and cell-to-cell
contact [13]. As NK cell activation and function may be
differently regulated in an organ-specic manner [14], LMinfected cells of different tissue origin may induce distinct NK
cell responses.
In a preclinical model, we have previously reported that LM
inuences the 2 main rat NK cell subsets, expressing Ly49
receptors or NKR-P1 receptors. Whereas LM infection expanded
the Ly49+ subset, the NKR-P1+ subset was reduced, indicating
that expression of ligands for these NK cell subsets might be
altered or regulated under LM infection [15]. In contrast to
down-regulation of MHC-I molecules, often observed in virus
infections, LM infection can up-regulate classical (Ia) and
nonclassical (Ib) MHC-I molecules on several rat cell types
[1618]. The interaction of activating Ly49 receptors with
increased MHC-Ib expression may be an explanation for
expansion of the Ly49+ subset detected in our in vivo studies
of LM-infected rats. Subsequently, in vitro coincubation of
these infected cell types with Ly49-expressing reporter cells
induced stronger signals. Whether other ligands for NK cell
receptors are inuenced by LM infection, or the up-regulation
of MHC-I occurs in all types of infected cells remains largely
unknown.
NK cell functions are strongly affected by the tissue microenvironment in which they reside [14]. In vitro infection of
different tissue-derived cell lines in coincubation with NK cells
may shed a light on the mechanisms that lead to NK cellmediated antibacterial immunity and regulation of the ensuing
NK cell response in various tissues. Here, we use LM infection of
myeloid monocyte-like RmW cells, pleural cavity-derived R2
macrophages, and colon epithelium-derived CC531 cells to
investigate expression of the ligands MHC-I and Clr for the
corresponding Ly49 receptor and NKR-P1 receptors. Accordingly, immune responses, including production of IFN-g and NK
cell proliferation, were monitored. These ndings revealed that
LM has different impacts on the expression of ligands for NK
cells that differ between cell types, which in turn may inuence
local NK cell responses.

MATERIALS AND METHODS

Animals

Bacteria and infection

Freeze-dried LM (strain L 242/73, type 4b; originally a gift from Arja de Klerk,
Department of Toxicology, Pathology and Genetics, National Institute of
Public Health and the Environment, RIVM, Bilthoven, the Netherlands [19])
was grown in tryptic soy broth until midlog phase. After addition of glycerol
(15% v/v), 2 ml aliquots were stored frozen at 270C until use. We have
previously established optimal conditions for infection of cells with LM to
develop a robust and reproducible in vitro infection model [16, 17]. In short,
cells were infected at a multiplicity of infection of 1:5 and incubated at 37C
for 1 h, followed by washing with RPMI and resuspension in 10 mg/ml
gentamicin (Sigma, St. Louis, MO, USA) containing cRPMI (RPMI 1640
supplemented with 10% FCS, 5 3 1025 M 2-ME, L-glutamine; all from Thermo
Fisher Scientic, Waltham, MA, USA) to kill extracellular bacteria. After 4872 h,
cells were heavily infected but still viable, as routinely monitored by Giemsa-stained
(Sigma) cytospins and FACS analysis.

Cell lines and IFN-g stimulation of cells

The R2 macrophage cell line (RT1d haplotype) originates from pleural
macrophages induced by silica injection in the pleural cavity of Wistar rats
[17]. The rat CC531 (RT1u haplotype) is a dimethyl hydrazine-induced
adenocarcinoma from the colon of WAG rats (a gift from Dr. Peter Kuppen,
Department of Surgery, Leiden University Medical Center, Leiden, The
Netherlands) [16]. These cell lines and stably transfected CHO cells
expressing Clr2 and Clr10 [5] were grown in cRPMI. For cell IFN-g
stimulation, cells were treated with 100 U/ml rat IFN-g (a gift from Peter H.
van der Meide, University Medical Centre Utrecht, the Netherlands).

Flow cytometry and mAb

Cells (50 ml; 25 3 105) were incubated with primary mAb (listed in Table 1)
for 30 min on ice. After 3 washes, labeled cells were incubated with a FITC-,
PE-, or PerCP-labeled secondary anti-rat or anti-mouse IgG (BD PharMingen,
San Diego, CA, USA) for 30 min and washed. Dead cells were excluded by
propidium iodide and analysis performed on FACSCalibur (BD Biosciences,
San Jose, CA, USA).

Phenotyping of rat RmW cell line

The myeloid cell line RmW originates from PVG (RT1c haplotype) rat spleen
as a side product during the generation of an in vitro passageable variant of
the in vivo-restricted Roser leukemia (a rat T cell leukemic cell line) [20, 21].
The phenotype of this cell line was determined by ow cytometry using a
panel of mAb specic for the rat (listed in Table 1).

CFSE proliferation assay

Spleens harvested from PVG.1U rats were crushed through a cell strainer.
Lymphocytes were separated by using Lymphoprep (Thermo Fisher Scientic) density gradient centrifugation, according to the standard protocol [22].
Nylon wool nonadherent lymphocytes were labeled with 5 mM CFSE (Thermo
Fisher Scientic) at 1 3 107 cells/ml in PBS with 2% FBS for 10 min at 37C.
Isolated lymphocytes were washed and incubated with infected or uninfected
cells for 5 d. Lymphocytes cultured in the presence of IL-2 (125 U/ml) were
used as positive controls. Harvested cells were surface stained with the
following mAb: FITC-conjugated 3.2.3 or 10/78 (anti-NKR-P1A), in combination with PE-conjugated G4.18 (anti-CD3; all mAb from BD PharMingen),
and analyzed by ow cytometry.

BWN reporter assay

u

Male 8- to 10-wk-old PVG.1U [RT1 or u; i.e., RT1-A , -B/D , -CE/N/M (class

Ia, class II, class Ib), abbreviated u-u-u] and PVG.7b (RT1c-c-c) rats were bred in
our animal house or purchased from Harlan UK (Bicester, United Kingdom).
Rats were regularly screened for common pathogens and housed in

compliance with guidelines set by the Experimental Animal Board under the
Ministry of Agriculture of Norway. Euthanasia of rats was performed with CO2.

Journal of Leukocyte Biology

Volume 100, November 2016

The BWN cell line is derived from BW5147 T, stably transfected with EGFP
under the control of NFAT-1. These cells were stably transfected with NKRP1A, -B, -D, and -E receptors, using a fusion protein consisting of the
intracellular part of the z-chain of the TCR complex and the extracellular

www.jleukbio.org

Shegarfi et al. LM infection affects ligands and subsequent response for NK cells

TABLE 1. Panel of antibodies used in the study, including phenotyping of RmW cells
mAb
AAS5
AAS6
ED1
G4.18
LOV8
OX1
OX6
OX12
OX17
OX18
OX27
OX41
OX42
OX62
R73
W3/13
W3/25
3.2.3

Specicity

Fluorescence intensity modes

RT1 (MHC-Ib)
RT1 (MHC-Ia,b)
CD68
CD3
CD93/C1qRp
CD45
RT1C/D (MHC-II)
Ig
RT1D (MHC-II)
RT1 (MHC-Ia,b)
RT1 (MHC-Ia,b)
Signal-regulatory protein a (CD47)
CD11b/c
CD103
TCR (a,b)
CD43
CD4
NKR-P1A

+
++
+
2
++
++
2/+
2/+
2/+
++
++
2
++
2
2
+
+
2

RmW cells were labeled with the indicated mAb, followed by secondary PE-conjugated anti-rat/mouse Ig
reagents. The specicity of the antibodies is shown. The uorescence-intensity modes include the
following: ++, 2.53 log10; +, 1.52 log10; 2/+, 11.5 log10; 2, ,1 log10.
ligand-binding part of the NKR-P1 receptors [5]. The BWN cells were
maintained in cRPMI, supplemented with 0.5 mg/ml Hygromycin B and
1 mg/ml G418 sulfate (Thermo Fisher Scientic). PMA (3 ng/ml), which
activates PKC, was added to the BWN-NKR-P1 reporter cells before mixing
with R2 or CC531 cells (with or without LM) in 18 h and analyzed for EGFP
expression by FACS analysis. Ionomycin (3 mM), a potent and selective
calcium ionophore, was added to the cell medium to enhance the specic
reporter cell signal without increasing the nonspecic background signal as
positive control.

Analysis of IFN-g production by splenic NK cells and

T cells
Fresh splenic isolated lymphocytes (PVG.7b) were incubated with cells
infected with LM or uninfected cells in the presence of IL-2 for 18 h.
Stimulation with IL-12 (2 ng/ml) was used as a positive control. Brefeldin A
(10 mg/ml; Thermo Fisher Scientic) was added to the cells to block transport
of newly synthesized IFN-g molecules to the cell surface. Cells were incubated
further for an additional 34 h and then surface stained with mAb against
NKR-P1A and CD3 for ow cytometric analysis, followed by 2% paraformaldehyde xation. Cells were permeabilized with 0.5% saponin (Sigma),
and intracellular labeling with anti-rat IFN-g mAb (BD PharMingen) was
carried out, according to standard protocols.

Statistical analysis
Statistical analyses were performed by 1-way ANOVA with Bonferroni
correction or by Students t test, using Prism software (GraphPad Software, La
Jolla, CA, USA). Differences were considered signicant at P , 0.05.

explore whether LM infection of R2 or CC531 cells, originated

from pleural cavity and intestine, respectively, induces proliferation of NK cells. The proliferation of coincubated NK cells
was measured by ow cytometric analysis. Increased proliferative
activity of NK (NKR-P1A+CD32) cells upon coculturing with
LM-infected R2 cells (23.2%) reached the same level as NK cells
stimulated with IL-2 (25.1%) compared with uninfected control
cells (,1%; Fig. 1A, left). Interestingly, although R2 and
CC531 cells behave similarly in response to LM infection that
includes up-regulation of MHC-I molecules [16, 17] (summarized
in Table 2), splenic NK cells did not proliferate toward infected
CC531 cells (Fig. 1A, right). Infection did not elicit proliferation of
T (CD3+NKR-P1A2) cells, and a negligible amount of NKT (NKRP1A+CD3+) cells was present (Fig. 1B). The cell counts of
proliferated NK cells against R2 cells (uninfected, 98.0 6 37.3; LM,
321.3 6 79.7; P , 0.05) and CC531 cells (uninfected, 161.5 6
135.1; LM, 142.5 6 33.2; P = not signicant) are shown (Fig. 1C).
The proliferation experiments suggest a cell type-specic NK
cell response, as only LM-infected R2, but not epithelial CC531
cells, potentiate the proliferative activity of NK cells.

Modulation of Clr expression in LM-infected CC531

but not in infected R2 cells

We have previously reported that enrichment of the Ly49+ subset

occurs in the spleen of LM-infected rats, not observed in
corresponding blood or BM, suggesting tissue-specic NK cell
subset expansion [15, 18]. In the current study, we aimed to

Having previously demonstrated that LM infection of R2 and

CC531 cells up-regulates MHC-I moleculesligands for the
Ly49+ subset [16, 17]we investigated whether ligands for the
other main functional NK cell subset in the rat are inuenced by
LM infection. The rat NKR-P1 receptor family consists of 4
membersNKR-P1A, NKR-P1B, NKR-P1F, and NKRP1Grecognizing a large family of Clr molecules [23]. The
effect of LM infection on rat Clr expression is unknown. It has
been reported that R2 and CC531 cells express ligands for the
inhibitory NKR-P1G and activating NKR-P1F receptors [5].
Antibodies against Clr molecules in the rat are not yet available,

www.jleukbio.org

Volume 100, November 2016

RESULTS

LM-infected R2 macrophages, but not infected CC531

cells, induce proliferation of rat splenic NK cells

Journal of Leukocyte Biology 3

Figure 1. LM-infected R2 cells, but not infected

CC531 cells, induce proliferation of NK cells. (A)
Representative dot plots (upper) indicate gating
strategy of lymphocyte subsets of splenocytes in
coculture with LM-infected cells. NK cells in the
rat are phenotypically dened as NKR-P1A+CD32,
NKT cells as NKR-P1A+CD3+, and T cells as
CD3+ NKR-P1A2. CFSE-labeled, fresh PVG splenic
lymphocytes were cultured for 5 d with uninfected
R2 (left) or CC531 (right) cells or corresponding
cells infected with LM. IL-2-stimulated lymphocytes were included as controls. The CFSE dilution proles of lymphocytes were analyzed by
ow cytometry. Representative dot plots (CFSE
uorescence) of proliferation are shown for NK
cells and (B) T cells. Lower-right square of all plots
indicates percentage of positive CFSE cells. (C)
Data represent means 6 SEM of proliferated NK
cells cultured with R2 cells (left) or CC531 cells
(right), treated with IL-2 (black bars), uninfected
(white bars), or LM-infected (gray bars); n = 3.
All statistical analyses were performed by 1-way
ANOVA with Bonferroni.

i.e., direct examination of ligand expression is not achievable.

Therefore, by using reporter cells expressing GFP and a single
NKR-P1 receptor, we indirectly examined whether LM infection
of these cells may have an impact on Clr expression. We infected

CC531 and R2 cells with LM and further coincubated with BWN

reporter cells. Ionomycin, which strongly stimulates the reporter
cells, served as a positive control (Fig. 2AC, left). To prove
functionality of reporter cells, the cells were coincubated with

TABLE 2. Comparison of ligand expression on LM-infected R2 macrophage with infected

CC531 epithelial cell lines and subsequent impact on NK cell responses
Change in ligand expression or NK cell response

R2 macrophages

Epithelial CC531 cells

Modulation of MHC-I
Change in stimulation of reporter cells expressing
Ly49 receptors
Change in stimulation of reporter cells expressing
NKR-P1A and -B receptors
Change in stimulation of reporter cells expressing
NKR-P1G and -F receptors
NK cell IFN-g production
Proliferation of NK cells

Yes
Yes

Yes
Yes

No

No

No

Yes

Yes
Yes

Yes
No

The ligands include MHC-I molecules (for Ly49 receptors) and Clr molecules (for NKR-P1 receptors).
NK cell responses measured are as follows: stimulation of reporter cells expressing the NK cell receptors
Ly49 or NKR-P1B, production of IFN-g and NK cell proliferation. Arrows indicate the following: ,
increscent; , reduction; , unchanged response or regulation.

Journal of Leukocyte Biology

Volume 100, November 2016

www.jleukbio.org

Shegarfi et al. LM infection affects ligands and subsequent response for NK cells

Figure 2. Modulation of Clr expression by LM infection. (A) Representative dot blots of BWN.NKR-P1 reporter cells cocultured with medium
(negative control) or stimulated with ionomycin (positive control; left), in parallel with CHO cells expressing Clr ligands (Clr2 or Clr10; middle).
BWN.NKR-P1A reporter cells were incubated overnight with uninfected or LM-infected CC531 cells (upper right) or R2 cells (lower right). Similar
experiments were performed with (B) NKR-P1G and (C) NKR-P1F reporter cells. FSC, Forward-scatter. (D) Data represent means 6 SEM of
standardized reporter cell EGFP expression of NKR-P1G and -F reporter cells to LM-infected or uninfected CC531 or R2 cells (n . 3). Statistical
signicance was determined with a 2-tailed unpaired Students t test (P , 0.05).

CHO cells expressing Clr2 or Clr10, ligands known to be

recognized by NKR-P1 receptors [5]. The response patterns to
the respective ligands, resulting from NKR-P1A recognition of
Clr10 and from NKR-P1G and -F recognition of Clr2, are
depicted (Fig. 2AC, middle). In parallel with the controls,
uninfected and LM-infected CC531 cells (upper right) and R2
cells (lower right) were coincubated with BWN-NKR-P1A (Fig.
2A), -G (Fig. 2B), -F (Fig. 2C), or -B (not shown) reporter cells.
The analyses revealed that stimulation of BWN-NKR-P1G (uninfected, 20%; LM, 7.7%)- and BWN-NKR-P1F (uninfected, 40%;
LM, 22%)-expressing reporter cells is reduced with LM-infected
CC531 cells compared with uninfected cells (Fig. 2B and C). By
contrast, there was a negligible change in the reporter cell
recognition of R2 cells upon infection (Fig 2AC). Neither
CC531 cells nor R2 cells were recognized by the BWN-NKR-P1A
(Fig. 2A) and NKR-P1B reporter cells (not shown). The
signicance of experiments with CC531 cells and NKR-P1G
reporter cells (uninfected, 32.3 6 6.2; LM, 18.9 6 6.4; P , 0.085,
indicating a weak but not statistically signicant reduction) and
NKR-P1F reporter cells (uninfected, 38.0 6 1.5; LM, 18.3 6 2.1;

www.jleukbio.org

P , 0.01) is shown (Fig. 2D). To exclude any unspecic NKR-P1

reporter cell stimulation, we applied BWN.Ly49 reporter cells as
additional control [16, 17]. Converse to less stimulation of
NKR-P1-reporters against CC531 cells, the Ly49 reporters induced a stronger response as a result of MHC-I overexpression
(not shown). Hence, with regards to the several controls
described above, the NKR-P1 reporter assays indicated a downward shift in the ligands expression upon LM infection of CC531
cells. Collectively, these experiments suggest that endogenous Clr
molecules, including Clr2 (or in addition to other unidentied
ligands), are down-regulated on LM-infected, epithelium-derived
CC531 cells but unchanged on infected R2 macrophages.

LM-infected RmW cells, unlike infected R2 cells,

displayed unchanged MHC-I expression and
concomitantly induced less IFN-g production by
NK cells
Both LM-infected CC531 and R2 cells overexpress MHC-I
molecules [16, 17]. During in vivo experiments, macrophages in
the CLNs in rats did not up-regulate MHC-I molecules in

Volume 100, November 2016

Journal of Leukocyte Biology 5

response to LM infection, contrary to other tissues, such as BM,

spleen, and blood [18]. In search for an appropriate cell line to
mimic this phenomenon, we screened a number of rat cells
retrieved from different sources. One of these cell lines was the
RmW, originally derived as a side-product of a T cell leukemia
cell line (Roser Leukaemia) [20, 21]. The RmW cell line has not
previously been fully described and therefore, was characterized
by using an array of rat mAb (Table 1). The ow cytometric
analysis revealed that RmW cells were negative for T cell markers
[CD3 (G4.18) and TCR a/b (R73)], B cell markers (Ig and
OX12), and a marker for NK cells [NKR-P1 (3.2.3)]. RmW cells
stained positive for MHC-I [OX18; broadly reactive with rat
classic (Ia) and nonclassical (Ib) MHC-I and AAS5; reactive with
the specic nonclassical RT1-CE region of MHC-I] and the
monocyte marker CD11b/c (OX42). The RmW cell line weakly
presents molecules associated with MHC-II (OX6). In addition,
RmW cells express the antigens CD4 (W3/25), CD43 (W3/13),
CD45 (OX1), and CD93 (LOV8), related to a variety of
hematopoietic cells, including monocytes. Based on the phenotypic analysis and morphologic examination, RmW cells were
evaluated to be monocyte-like cells (Figs. 3 and 4A, left).
Following LM infection, R2 and RmW cells were labeled with
mAb against both classic and nonclassical MHC-I molecules for
ow cytometry. We observed that unlike R2 cells (right), both
classic and nonclassical MHC-I expression were not signicantly
altered on LM-infected RmW cells (left; Fig. 4B). However,
stimulation of RmW cells with 100 U/ml IFN-g induced an upregulated expression of MHC-Ia and -Ib molecules. Hence, the

2 monocyte/macrophage cell lines, originating from different

tissues (spleen and pleural cavity, respectively), were similar in
infection susceptibility but different in MHC-I expression. Rat NK
cells produced large amount of IFN-g in coincubation with LMinfected CC531 cells [16]. To investigate the functional involvement of the differential MHC-I expression, we measured NK
cell IFN-g production in cultures with LM-infected R2 or RmW
cells. Stimulation of splenocytes with IL-2 (upper left) or IL-12
(upper right) developed a high IFN-g secretion and served as
negative and positive controls, respectively (Fig. 4B). Splenocytes
were additionally coincubated with target cells, uninfected or
infected RmW cells (lower left) in parallel to R2 cells (lower right).
We found that a minority of NK cells produced IFN-g against LMinfected RmW cells (uninfected, 4.1%; LM, 13.7%) compared with
infected R2 cells (uninfected, 4.7%; LM, 56.8%; Fig. 4B). The
signicance of IFN-g secretion against RmW cells (uninfected,
2.8 6 1.3; LM, 10.8 6 3.2; P = 0.13) and R2 cells (uninfected, 4.7 6
0.9; LM, 63.7 6 7.2; P , 0.01) is shown (Fig. 4C).
Collectively, and in line with our prior in vivo observations, our
results demonstrate that LM infection may alter the level of
surface MHC-I, depending on the cell type infected. Accordingly,
enhanced NK cell IFN-g production proportionally correlates
with the magnitude of MHC-I expression.

DISCUSSION
Our previous reports demonstrated a complex engagement of
NK cells with LM-infected cells through receptor/ligand

Figure 3. Phenotyping of the rat RmW cell line.

RmW cells were labeled with a panel of rat mAb
against cell surface markers. The specicity of all
antibodies is listed in Table 1. Staining with
secondary anti-rat or anti-mouse antibody as
negative controls is presented (2 left upper).
Representative histograms show that the RmW
cell line, derived from lymphoblastic T cell
leukemia, presents the surface antigens CD45,
CD93, CD43, MHC-I molecules, CD4, and
CD11b, associated with a monocyte-like
phenotype.

Journal of Leukocyte Biology

Volume 100, November 2016

www.jleukbio.org

Shegarfi et al. LM infection affects ligands and subsequent response for NK cells

Figure 4. Differential MHC-I expression on LM-infected cell types lead to unequal NK cell IFN-g production. (A) Representative histograms of
RmW (left) and R2 (right) cells displaying CD11b/c (OX42), classic and nonclassical MHC-Ia,-Ib (OX18), and nonclassical MHC-I (AAS5).
Uninfected (gray lled areas), LM-infected (thick lines), IFN-g-stimulated cells (thin lines), and secondary anti-rat/mouse control mAb (broken
lines) are presented (n . 5). (B) NK cells were analyzed for IFN-g production by intracellular ow cytometric analysis. Freshly isolated spleen cells
cultured in IL-2 medium (negative control) and by addition of IL-12 (positive control) are depicted in representative dot plots (upper). Following
coincubation with RmW (left) or R2 (right) cells, uninfected or LM-infected, production of IFN-g by splenic NK cells was measured (n = 3). The
amount of IFN-g-secreting NK cells is denoted in the upper-right square of all plots. (C) Graphs showing the mean percentages 6 SEM of the IFNg-producing splenic NK cells, as described above. Statistical analyses were performed by 1-way ANOVA with Bonferroni.

interaction [1518]. The in vivo experiments revealed that the

size of the Ly49+ subset of splenic NK cells increased, whereas the
NKR-P1+ subset simultaneously decreased following LM infection. The change was not observed in other tissues, as the
percentage of the Ly49+ population concurrently decreased in
BM and remained unchanged in blood [18]. Yet, CLN
macrophages did not up-regulate the ligands, MHC-I molecules,
for Ly49 receptors following LM infection compared with

macrophages isolated from other tissues sources [18]. Based on

the in vivo data, we hypothesized that ligands for NK cell
receptors and the corresponding subsets are variously inuenced
by LM infection.
NK cell defense against pathogens is triggered by diverse
mechanisms requiring the presence of several accessory cell types
[24]. Proliferation of NK cells in response to LM-infected cells
indicates that LM induces expression of ligands for activating NK

www.jleukbio.org

Volume 100, November 2016

Journal of Leukocyte Biology 7

cell receptors. The advantages of Ly49+ expansion might be

dependent on activating receptors for pathogen-induced ligands
important for NK cell maintenance [25]. Furthermore, the
complex NK cell signaling results from multiple receptor types
contributing to NK cell stimulation, which in turn, is based on the
presence or absence of specic cellular ligands [26]. The
different NK cell proliferation response to LM-infected R2 versus
CC531 cells suggests that the expression of ligands for NK cells
may be variously affected depending on cell type. As CC531 and
R2 cells overexpress their MHC-I molecules in response to LM,
down-regulation of ligands for other NK cell receptors, e.g., Clr
molecules, may reect the lack of such ligands to stimulate NK
cell proliferation. This notion is supported by the data in the
current study (Fig. 2) and our in vivo experiments, in which the
NKR-P1 subset is reduced in LM-infected rats [15].
Interaction between inhibitory NK cell receptors and nonMHC-I has been proposed to play an alternative role in selftolerance [27]. The rat NKR-P1 receptor family recognizes Clr
molecules [23]. RCMV infection down-regulates Clr molecules,
whereas it up-regulates RCMV C-type lectin-like, a decoy ligand,
to protect infected cells from NK cell killing via direct interaction
with the NKR-P1B inhibitory receptor [28, 29]. It was postulated
that the NKR-P1/Clr receptor/ligand recognition system may
represent an alternative mechanism for NK cell discrimination
between normal and damaged cells [28]. Our data show that LM
infection down-regulates the Clr molecules on the cell surface of
epithelial cells. Compared with RCMV infection, it is intriguing to
speculate that LM may down-regulate Clr molecules in epithelial
cells [29] as a defense mechanism to avoid NK cell stimulation.
This assumption may be supported by the high expression of
NKR-P1F and NKR-P1G receptors in rat gut
mucosa lymphocytes [23].
A robust IFN-g production promotes bacterial clearance and is
thus critical in controlling an ongoing LM infection [12]. Release
of IFN-g by NK cells has been proposed to be the major
contribution factor against LM infection at an early stage [11].
We previously suggested a relation between up-regulation of
MHC-I on LM-infected cells and production of IFN-g by NK cells
[16, 18]. We have been looking for cell types to mimic the CLN
macrophages, which remained unaffected in MHC-I expression
upon LM infection [18]. Similar to CLN macrophages, infection
of RmW cells with LM did not result in modulation of MHC-Ia
and -Ib molecules. A comparison of RmW cells with R2 cells that
overexpress MHC-I in response to LM infection [17] indicated a
close correlation between enhanced MHC-I expression and
increased NK cell IFN-g production (Fig. 4). The mechanism
behind the unchanged expression of MHC-I in LM-infected
RmW cells remains under investigation. However, this RmW cell
property gave the opportunity to examine the involvement of
MHC-I molecules in stimulating NK cells without mAb-blocking
strategies that have been shown to be difcult to perform. In
other words, lower MHC-I expression on LM-infected cells may
lead to less IFN-g production by NK cells. This, in turn, supports
unequal NK cell responses to infected cell types, which depends
not only on ligand expression but is also related to the magnitude
of expression.
Macrophages have been used extensively in experimental in
vitro models of LM infection. However, the rst encounter with
8

Journal of Leukocyte Biology

Volume 100, November 2016

LM is usually intestinal epithelial cells in the gut [8]. Epithelial

CC531 and R2 macrophages illustrate that LM infection elicits
distinct NK cell responses that might be associated with the cell
type infected. The complexity of responses is revealed, as
infection of CC531 cells triggers NK cell IFN-g production [16]
yet negligible NK cell proliferation when compared with R2 cells
under the same infection and culture conditions. A comparison
of outcomes of NK cell proliferation and IFN-g production
between LM-infected CC531 and R2 cells, including previously
published results [16, 17], is illustrated in Table 2. NK cells
integrate signals received from multiple receptors to sense their
environment and respond appropriately [30] to control infections [31]. The synergistic engagement of combinations of NK
cell receptors may allow NK cells to have more exibility in
recognizing and responding quickly to changes in their
environment. LM has been shown to inuence other ligands for
NK cells, as the ligands for the activating receptor NKG2D also
have been reported to be regulated in response to LM infection
[32]. A broad evaluation of the alterations in ligand expression
for rat NK cell receptors following LM infection remains
incomplete, as a result of lack of suitable reagents.
In conclusion, in vitro infection of R2 macrophages and
epithelial CC531 cells may be a useful way to gather new
knowledge about ligands for NK cells and NK cell responses in
various tissues, i.e., spleen and intestine, the major routes of LM
infection. Our ndings demonstrate that LM differently affects
the expression of ligands for NK cell receptorshere MHC-I and
Clr moleculesdepending on the cell type infected. Modication of ligand expression may induce activation of NK cells in the
surrounding tissue that exert distinct effector functions related to
the type of ligand alteration in infected cells.

AUTHORSHIP
Experiments were conceived of and designed by H.S. and B.R.
and conducted by H.S. H.S., J.N., and K.P.K. contributed to data
analysis, interpreted data, and wrote the paper.

ACKNOWLEDGMENTS
This work was supported by grants from the Norwegian Cancer
Society, Research Council of Norway, Anders Jahres Foundation for
the Promotion of Science, and University of Oslo. The generous
technical support of the staff in the B.R. laboratory is gratefully
acknowledged. This article is dedicated to the memory of B.R.

DISCLOSURES

The authors have no nancial or commercial conicts of interests.

REFERENCES

1. Lanier, L. L. (2005) NK cell recognition. Annu. Rev. Immunol. 23,

225274.
2. Raulet, D. H., Vance, R. E. (2006) Self-tolerance of natural killer cells.
Nat. Rev. Immunol. 6, 520531.
3. Carlyle, J. R., Mesci, A., Fine, J. H., Chen, P., Belanger, S., Tai, L. H.,
Makrigiannis, A. P. (2008) Evolution of the Ly49 and Nkrp1 recognition
systems. Semin. Immunol. 20, 321330.

www.jleukbio.org

Shegarfi et al. LM infection affects ligands and subsequent response for NK cells
4. Rolstad, B., Naper, C., Lvik, G., Vaage, J. T., Ryan, J. C., BackmanPetersson, E., Kirsch, R. D., Butcher, G. W. (2001) Rat natural killer cell
receptor systems and recognition of MHC class I molecules. Immunol.
Rev. 181, 149157.
5. Kveberg, L., Dai, K. Z., Westgaard, I. H., Daws, M. R., Fossum, S., Naper,
C., Vaage, J. T. (2009) Two major groups of rat NKR-P1 receptors can be
distinguished based on chromosomal localization, phylogenetic analysis
and Clr ligand binding. Eur. J. Immunol. 39, 541551.
6. Karre, K., Ljunggren, H. G., Piontek, G., Kiessling, R. (1986) Selective
rejection of H-2-decient lymphoma variants suggests alternative
immune defence strategy. Nature 319, 675678.
7. Yamamoto, S., Russ, F., Teixeira, H. C., Conradt, P., Kaufmann, S. H. (1993)
Listeria monocytogenes-induced gamma interferon secretion by intestinal
intraepithelial gamma/delta T lymphocytes. Infect. Immun. 61, 21542161.
8. Pamer, E. G. (2004) Immune responses to Listeria monocytogenes. Nat. Rev.
Immunol. 4, 812823.
9. Bancroft, G. J., Schreiber, R. D., Bosma, G. C., Bosma, M. J., Unanue,
E. R. (1987) A T cell-independent mechanism of macrophage activation
by interferon-gamma. J. Immunol. 139, 11041107.
10. Kaufmann, S. H. (1993) Immunity to intracellular bacteria. Annu. Rev.
Immunol. 11, 129163.
11. Dunn, P. L., North, R. J. (1991) Early gamma interferon production by
natural killer cells is important in defense against murine listeriosis.
Infect. Immun. 59, 28922900.
12. Dai, W. J., Bartens, W., Kohler, G., Hufnagel, M., Kopf, M., Brombacher,
F. (1997) Impaired macrophage listericidal and cytokine activities are
responsible for the rapid death of Listeria monocytogenes-infected IFNgamma receptor-decient mice. J. Immunol. 158, 52975304.
13. Humann, J., Lenz, L. L. (2010) Activation of naive NK cells in response to
Listeria monocytogenes requires IL-18 and contact with infected dendritic
cells. J. Immunol. 184, 51725178.
14. Sojka, D. K., Plougastel-Douglas, B., Yang, L., Pak-Wittel, M. A., Artyomov,
M. N., Ivanova, Y., Zhong, C., Chase, J. M., Rothman, P. B., Yu, J., Riley,
J. K., Zhu, J., Tian, Z., Yokoyama, W. M. (2014) Tissue-resident natural
killer (NK) cells are cell lineages distinct from thymic and conventional
splenic NK cells. eLife 3, e01659.
15. Shegar, H., Sydnes, K., Lvik, M., Inngjerdingen, M., Rolstad, B., Naper,
C. (2009) The role of natural killer cells in resistance to the intracellular
bacterium Listeria monocytogenes in rats. Scand. J. Immunol. 70, 238244.
16. Shegar, H., Dai, K. Z., Inngjerdingen, M., Ryan, J. C., Vaage, J. T.,
Rolstad, B., Naper, C. (2010) The activating rat Ly49s5 receptor responds
to increased levels of MHC class Ib molecules on Listeria monocytogenesinfected enteric epithelial cells. Eur. J. Immunol. 40, 35353543.
17. Shegar, H., Dai, K. Z., Daws, M. R., Ryan, J. C., Vaage, J. T., Rolstad, B.,
Naper, C. (2011) The rat NK cell receptors Ly49s4 and Ly49i4 recognize
nonclassical MHC-I molecules on Listeria monocytogenes-infected
macrophages. J. Leukoc. Biol. 89, 617623.
18. Shegar, H., Naper, C., Rolstad, B., Inngjerdingen, M. (2010) Listeria
monocytogenes infection affects a subset of Ly49-expressing NK cells in the
rat. PLoS One 5, e15579.

19. Ruitenberg, E. J., Van Noorle Jansen, L. M., Kruizinga, W., Steerenberg,
P. A. (1976) Effect of pretreatment with bacillus Calmette-Guerin on the
course of a Listeria monocytogenes infection in normal and congenitally
athymic (nude) mice. Br. J. Exp. Pathol. 57, 310315.
20. Nestvold, J., Stokland, A., Naper, C., Rolstad, B. (2004) Phenotype and
natural killer cell sensitivity of a radiation-induced acute T-cell leukaemia
(Roser leukaemia) in PVG rats. Scand. J. Immunol. 60, 153158.
21. Lobato-Pascual, A., Saether, P. C., Dahle, M. K., Gaustad, P., Dissen, E.,
Fossum, S., Daws, M. R. (2013) Rat macrophage C-type lectin is an
activating receptor expressed by phagocytic cells. PLoS One 8, e57406.
22. Boyum, A. (1968) Separation of leukocytes from blood and bone
marrow. Introduction. Scand. J. Clin. Lab. Invest. Suppl. 97, 7.
23. Kveberg, L., Sudworth, A., Todros-Dawda, I., Inngjerdingen, M., Vaage,
J. T. (2015) Functional characterization of a conserved pair of NKR-P1
receptors expressed by NK cells and T lymphocytes in liver and gut. Eur.
J. Immunol. 45, 501512.
24. Newman, K. C., Riley, E. M. (2007) Whatever turns you on: accessory-celldependent activation of NK cells by pathogens. Nat. Rev. Immunol. 7,
279291.
25. Lee, S.-H., Biron, C. A. (2010) Here todaynot gone tomorrow: roles for
activating receptors in sustaining NK cells during viral infections. Eur.
J. Immunol. 40, 923932.
26. Bryceson, Y. T., March, M. E., Ljunggren, H. G., Long, E. O. (2006)
Synergy among receptors on resting NK cells for the activation of natural
cytotoxicity and cytokine secretion. Blood 107, 159166.
27. Kumar, V., McNerney, M. E. (2005) A new self: MHC-class-I-independent
natural-killer-cell self-tolerance. Nat. Rev. Immunol. 5, 363374.
28. Voigt, S., Mesci, A., Ettinger, J., Fine, J. H., Chen, P., Chou, W., Carlyle,
J. R. (2007) Cytomegalovirus evasion of innate immunity by subversion of
the NKR-P1B:Clr-b missing-self axis. Immunity 26, 617627.
29. Aguilar, O. A., Mesci, A., Ma, J., Chen, P., Kirkham, C. L., Hundrieser, J.,
Voigt, S., Allan, D. S., Carlyle, J. R. (2015) Modulation of Clr ligand
expression and NKR-P1 receptor function during murine
cytomegalovirus infection. J. Innate Immun. 7, 584600.
30. Long, E.O., Kim, H. S., Liu, D., Peterson, M. E., Rajagopalan, S. (2013)
Controlling NK cell responses: integration of signals for activation and
inhibition. Ann. Rev. Immunol. 31, 227258.
31. Bjorkstrom, N. K., Lindgren, T., Stoltz, M., Fauriat, C., Braun, M.,
Evander, M., Michaelsson, J., Malmberg, K. J., Klingstrom, J., Ahlm, C.,
Ljunggren, H. G. (2011) Rapid expansion and long-term persistence of
elevated NK cell numbers in humans infected with hantavirus. J. Exp.
Med. 208, 1321.
32. Hamerman, J. A., Ogasawara, K., Lanier, L. L. (2004) Cutting edge: Tolllike receptor signaling in macrophages induces ligands for the NKG2D
receptor. J. Immunol. 172, 20012005.

www.jleukbio.org

Volume 100, November 2016

KEY WORDS:
MHC-I Clr Ly49 NKR-P1 receptors rat cells

Journal of Leukocyte Biology 9