Cellular Immunology xxx (2015) xxxxxx

Contents lists available at ScienceDirect

Cellular Immunology
journal homepage: www.elsevier.com/locate/ycimm

Research paper

Peripheral blood NK cells expressing HLA-G, IL-10 and TGF-b in healthy

donors and breast cancer patients
Yekaterina O. Ostapchuk a,, Esin Aktas Cetin b, Yuliya V. Perfilyeva a, Abdullah Yilmaz b, Yuriy A. Skiba d,
Alexandr P. Chirkin d, Nazgul A. Omarbaeva c, Shynar G. Talaeva c, Nikolai N. Belyaev a, Gunnur Deniz b
a

M.A.Aitkhozhins Institute of Molecular Biology and Biochemistry, Laboratory of Molecular Immunology and Immunobiotechnology, Almaty, Kazakhstan
Istanbul University, Institute of Experimental Medicine, Immunology Department, Istanbul, Turkey
Research Institute of Oncology and Radiology, Mammalogy Center, Almaty, Kazakhstan
d
M.A.Aitkhozhins Institute of Molecular Biology and Biochemistry, Laboratory of Genome, Almaty, Kazakhstan
b
c

a r t i c l e

i n f o

Article history:
Received 24 July 2015
Revised 25 August 2015
Accepted 1 September 2015
Available online xxxx
Keywords:
NK cells
IL-10
TGF-b
HLA-G
Breast cancer

a b s t r a c t
Human natural killer (NK) cells are not only professional cytotoxic cells integrated into effector branch of
innate immunity, but they are also regulatory cells, managing different immune processes.
Immunoregulatory NK cells, expressing HLA-G and IL-10, have been generated in vitro from human
hematopoietic progenitors and found in vivo among decidual NK cells of pregnant women. Human
peripheral blood NK cells have been shown to acquire suppressive properties after HLA-G uptake during
trogocytosis. Moreover, it has been shown that circulating NK cells contain a trace amount of cells producing TGF-b and IL-10, which exert a suppressive influence upon innate and adaptive immunity. In this
study, we report on a minor subset of peripheral blood HLA-G+ NK cells possessing suppressive activity
toward effector functions of NK cells. Further we demonstrate an increased number of circulating HLA-G+,
IL-10+, and TGF-b+ NK cells in breast cancer patients which might impair efficiency of anti-tumor
immunity.
2015 Elsevier Inc. All rights reserved.

1. Introduction
Human natural killer (NK) cells are considered professional
cytotoxic cells integrated into the effector branch of innate immunity during antiviral [1], antitumoral [2], antimicrobial [3], and
antiparasitic [4] responses and also exert regulatory effects on
innate and adaptive immunity through various cytokines and
chemokines or cell-to-cell-dependent contacts [5]. This
immunoregulatory function is attributed mostly to CD56bright NK
cells that are characterized by high CD56 expression [6], while
CD56dim NK cells have the most evident cytotoxic properties [7
9]. It has been shown that depending on cultivation conditions
in vitro (in the presence of IL-12 or IL-4), human NK cells differenAbbreviations: APC, antigen-presenting cells; CFSE, 5- (and 6-) carboxyfluorescein diacetate succinimidyl ester; CTL, cytotoxic T lymphocytes; FCS, fetal calf
serum; HLA-G, human leukocyte antigen G; ILT2, Ig-like transcript 2; mAb,
monoclonal antibody; PB, peripheral blood; PBMCs, peripheral blood mononuclear
cells; PHA, phytohemagglutinin; PI, propidium iodide; SD, standard deviation; i-IL10, intracellular IL-10; i-TGF-b, intracellular TGF; mb-IL-10, membrane-bound
IL-10.
Corresponding author at: 86 Dosmuhamedov Str., Almaty, Kazakhstan.
E-mail address: katyostapchuk@gmail.com (Y.O. Ostapchuk).

tiate into NK1 or NK2 cell subsets (analogous to Th1 or Th2 cells),
producing IFN-c or IL-5 with IL-13, respectively [10]. Phenotypic
analysis has shown that both subsets are present in CD56bright
and CD56dim fractions of peripheral blood PB NK cells in approximately equal proportions with absolute dominance of NK1 cells
[11]. It has also been shown that both fractions contain trace
amounts of regulatory NK3 and NKr1 cells expressing intracellular
TGF-b and IL-10 respectively [12]. IL-10-producing human regulatory NK cells obtained by immunomagnetic separation of activated
NK cells with bispecific mAb conjugates (anti-CD45 and anti-IL-10)
showed low levels of TGF-b and IL-13 but no IFN-c expression.
These cells inhibited antigen-induced proliferation and IFN-c production by T cells [13].
A suppressive subset of NK cells (NK-ireg) has been generated
in vitro from PB CD34+ hematopoietic progenitors expressing
membrane-bound IL-15 [14]. Such cells were found in vivo among
decidual NK cells of pregnant women. NK-ireg cells had a mature
cell phenotype, secreted suppressive molecules (e.g., HLA-G,
IL-10, IL-21), and did not exert cytolytic effects on target cells.
NK-ireg cells suppressed dendritic cells and the cytotoxic activity
of PB NK cells via membrane-bound HLA-G and secreted IL-10
[14]. Furthermore, NK cells have been shown to uptake HLA-G

http://dx.doi.org/10.1016/j.cellimm.2015.09.002
0008-8749/ 2015 Elsevier Inc. All rights reserved.

Please cite this article in press as: Y.O. Ostapchuk et al., Peripheral blood NK cells expressing HLA-G, IL-10 and TGF-b in healthy donors and breast cancer
patients, Cell. Immunol. (2015), http://dx.doi.org/10.1016/j.cellimm.2015.09.002

Y.O. Ostapchuk et al. / Cellular Immunology xxx (2015) xxxxxx

during trogocytosis [15]. Short-term cell-to-cell contact with HLAG1-transfected tumor cell lines led to acquisition of HLA-G1 membrane fragments and loss of cytotoxic activity by IL-2-activated NK
cell lines and human PB NK cells. Blockade or loss of HLA-G1 fully
restored effector functions of NK cells toward tumor cells. NK cells
acquiring HLA-G1 showed suppressive properties toward other NK
cells. Inhibition of NK cell cytotoxicity was shown to directly
depend on the interaction between HLA-G1 and its receptor,
ILT2, on effector NK cells [15]. In addition, HLA-G-expressing
CD4+ and CD8+ T cells in human PB and sites of inflammation have
been recently revealed. These cells were hypo-proliferative and
exhibited immunosuppressive properties that were mediated by
HLA-G and IL-10 [1618].
HLA-G is an important immunoregulatory molecule that protects trophoblasts from maternal allorecognition. HLA-G was also
detected in pathological processes, such as viral infections, allotransplantation, and inflammatory diseases, and was found in various types of malignant cells [16]. Under alloreactive conditions,
such as allograft transplantation and mixed lymphocyte reactions,
T cells were shown to suppress proliferative responses by soluble
HLA-G. Expression of membrane-bound HLA-G in T cells was
described after transplantation and in HIV-infected patients [17].
Aberrant expression of HLA-G by tumor cells has been suggested
to be a major mechanism of escape from immunosurveillance.
HLA-G induces development of iDC, expansion of MDSC, and
induction of T suppressive cells and tumor-associated macrophages [19]. Participation of HLA-G in immune regulation has been
assessed for inhibition of allogeneic T cell proliferation, as well as
CD8+ T and NK cell cytotoxicity [20]. Ligation of NK cell receptors
(KIR2DL4, NKG2A, ILT2, and CD160) is responsible for both soluble
and membrane-bound HLA-G recognition and results in inactivation of NK cell effector functions. The ability of HLA-G to inhibit
NK cell-mediated lysis and IFN-c production was recently shown
[21,22]. Moreover, HLA-G has been shown to modulate proinflammatory or immunosuppressive cytokine secretion profiles
of PB and decidual NK cells [2327]. Thus, increasing evidence suggests that NK cells expressing IL-10, TGF-b, and HLA-G are important in regulation of the immune response and may contribute to
suppression of antitumor immunity.
For women worldwide, breast cancer is the most common cancer diagnosed and has the highest death toll [26]. Immunoloregulatory cell investigation is urgent and attracts a lot of attention in
the last decades and it is generally accepted that these subsets
may play a role in immuno-editing in cancer. Multiple studies have
demonstrated that effector activity of NK cells is decreased in
peripheral blood and tumor microenvironment of breast cancer
patients [2730], but little is known about the prevalence of
immunoregulatory NK cells in various types of cancer. Thus, the
study of regulatory NK cell subsets in healthy donors and their
alterations and suppressor role in breast cancer patients may have
direct clinical implications for design and implementation of future
immunotherapy-based clinical trials.

2. Materials and methods

2.1. Patients and healthy donors
PB samples were obtained from healthy volunteers and patients
with primary operable breast cancer in stages I, II, and III (Table 1).
None of the patients received chemotherapy or any other anticancer treatment prior to surgery, nor did they have metastases
or malignancies other than breast cancer. Recurrent breast cancer
cases were also excluded. Tumor staging was stratified according
to the sixth edition of the tumor-node metastasis classification
by the International Union Against Cancer. Healthy subjects had

Table 1
Donor characteristics.
Study subjects (n)

Male
Female
Average age SD (range)
Male
Female
Stage I
Stage II
Stage III
Stage IV

Normal

Breast cancer patients

6
21

22

44.2 12.4 (3060)

40.4 15.3 (2263)

52.4 14.1 (2369)

3
15
4

no clinical evidence of breast cancer or any other overt diseases.

Only females were analyzed in assays where breast cancer patients
and healthy donors were compared. The relevant institutional
review boards approved this study and all volunteers gave their
written informed consent according to the Declaration of Helsinki.
The research protocol was also approved by the Research Institute
of Oncology and Radiology Committee (Almaty, Kazakhstan).

2.2. Antibodies and reagents

The following anti-human mAbs were used for flow cytometry
and cell sorting: anti-CD3-FITC, anti-CD57-APC, anti-PerforinAPC, anti-CD107a-APC, anti-Granzyme B-FITC, Mouse IgG1-PE
and PerCP-Cy5.5 Isotypic mAbs (Miltenyi Biotec, Germany); antiCD56-PerCP-Cy5.5, anti-CD14-PE, BD Simultest IMK Plus CD3/
CD19, CD4/CD8, Simultest LeucoGATE CD45/CD14 (BD Pharmingen, USA); anti-CD3-FITC, anti-CD56-PerCP-Cy5.5, anti-CD56PerCP, anti-HLA-G-PE and anti-HLA-G-APC (clone 87G), and antiCD14-PerCP (BioLegend, USA); anti-TGF-b-FITC, anti-TGF-b-PE,
anti-IL-10-FITC, anti-IL-10-PE (R&D Systems, USA), and anti-IFNc-FITC (Beckman Coulter, USA). For ELISPOT assay, human IL-10
ELISPOT set and Abs from human TGF-b1 ELISA set (BD Biosciences,
USA) were used. The following reagents were used for cell culture
and cytotoxicity test: human IL-2 (R&D Systems, USA), Brefeldin A
(BioLegend, USA), propidium iodide (PI) (Miltenyi Biotec, Germany), RPMI-1640, Hystopaque-1077, L-glutamine, penicillin/
streptomycin, fetal calf serum (FCS), 5- (and 6-) carboxyfluorescein
diacetate succinimidyl ester (CFSE), and phytohemagglutinin M
(PHA) (SigmaAldrich, USA). For PCR assay the following reagents
were used: TRI Reagent (SigmaAldrich, USA), DNase I, Maxima
Reverse Transcriptase, Taq polymerase (Thermo Scientific, USA).

2.3. Cell separation

Peripheral blood mononuclear cells (PBMC) were isolated from
venous PB by density gradient centrifugation on a Hystopaque1077. NK cells were purified on a FACSAria II Sorter (BD Biosciences, USA) using anti-CD3-FITC and anti-CD56-PerCP-Cy5.5
(purity of CD3
CD56+ population was >98%). For further separation
into HLA-G+ and HLA-G
fractions, NK cells were activated overnight by incubating 2  106 cells/ml in 24-well plates in RPMI1640 complete medium supplemented with PHA-M (10 lg/ml).
Then, cells were sorted using anti-HLA-G-PE or anti-HLA-G-APC
Abs. The purity of HLA-G+ cells was greater than 85%.
For PCR assay, NK cells were negatively isolated by immunomagnetic cell separation using a NK isolation kit and VarioMACS
cell separator (Miltenyi Biotech, Germany) according to the manufacturers protocol or sorted using CD3-FITC and CD56-PerCP-Cy5.5
mAbs. The purity of the obtained NK cell fraction was greater than
97% as assessed by flow cytometry. PB B lymphocytes, CD8+ T cells,

Please cite this article in press as: Y.O. Ostapchuk et al., Peripheral blood NK cells expressing HLA-G, IL-10 and TGF-b in healthy donors and breast cancer
patients, Cell. Immunol. (2015), http://dx.doi.org/10.1016/j.cellimm.2015.09.002

Y.O. Ostapchuk et al. / Cellular Immunology xxx (2015) xxxxxx

and monocytes were purified on a FACSAria II Sorter using BD

Simultest kits (BD Bioscience).
2.4. Flow cytometry and ELISPOT assay
For surface and intracellular staining, cells were subsequently
incubated with mAbs specific for surface markers according to
the manufacturers protocols, then fixed and permeabilized with
Fixation/Permeabilization solution (BD Pharmingen, USA) and
incubated for 20 min in the dark at room temperature. Cells were
washed with Perm/Wash Buffer (BD Biosciences, USA) and stained
with mAbs specific for intracellular cytokines. Afterwards, cells
were washed with PBS, re-suspended in flow solution, and immediately analyzed by flow cytometry on a FACSCalibur (BD Biosciences) using CellQuest Pro software (BD Biosciences).
Lymphocytes were gated based on FSC and SSC properties. From
this gate, CD3
CD56+ NK cells were selected and analyzed for
expression of IL-10, TGF-b, HLA-G, CD57. Unstained cells, single
fluorochrome stained cells, and cells stained as fluorescenceminus-one (FMO) controls were used to set-up the machine.
For ELISPOT assay, 106 cells/ml NK cells from healthy donors
and breast cancer patients in 96-well flat-bottom ELISPOT plates
were stimulated with 5 lg/ml PHA in 200 ll of medium for 18 h
(BD Pharmingen, USA); secreted IL-10 and TGF-b were analyzed.
Locally produced cytokines were captured by specific mAbs. After
culture and cell lysis, the plates were washed and then incubated
with secondary biotinylated detection mAbs according to the protocol. Spots were developed using streptavidin-horseradish peroxidase and freshly prepared substrate (3-amino-9-ethylcarbazole).
Spots were counted microscopically.
2.5. Cell culture
In all experiments, cells were incubated at 2  106 cells/ml in
plates for 18 h in RPMI-1640 supplemented with 10% FCS, 100 U/
ml penicillin, 100 mg/ml streptomycin, and 2 mM L-glutamine in
a humidified CO2 incubator. Membrane-associated HLA-G and
IL-10 were detected on cell surfaces of freshly purified NK cells
and after their incubation without stimulation or in the presence
of PHA (10 lg/ml) and IL-2 (20 ng/ml) alone or in combination
for 18 h. For assessment of HLA-G+ NK cell cytokine profiles, levels
of intracellular IL-10, TGF-b, and IFN-c were assessed in freshly
purified and activated NK cells (18 h) with PHA. In some experimental comparisons of intracellular IL-10, TGF-b, and surface
HLA-G, PBMCs were used and gated on CD3
CD56+ cells. For
analysis of intracellular cytokines, Brefeldin A (1 ll/ml) was added
after 1 h of cultivation to predict secretion of cytokines.

CFSE. In brief, sorted NK cells were initially activated with PHA

(10 lg/ml) for 18 h to ensure an optimal level of HLA-Gexpressing cells. Then NK cells were fractionated into HLA-G+
and HLA-G
NK cell subsets and used for cytotoxicity suppression
assay in two variants. First, for cellcell dependent contact, both
cell subsets were cocultured for 48 h together or separately with
K562 cells at suppressor:effector:target ratios of 25:5:1 in roundbottom plates with or without HLA-G blocking mAbs HLA-G (clone
87G, 10 lg/ml). Second, for cellcell independent contact, Corning
Transwell polycarbonate membrane inserts (SigmaAldrich, USA)
were used. HLA-G+ NK cells placed in the upper chamber were
cocultured with autologous HLA-G
NK cells in the lower chamber
at suppressor:effector ratios of 5:1 for 18 h in RPMI-1640 supplemented with 10% FCS. Then, HLA-G
NK cells were placed in the
round-bottom plates and incubated for an additional 48 h with
K562 cells at effector:target ratios of 5:1. Also, HLA-G+ NK cells
in the upper chamber were separated from HLA-G
NK and K562
cells by Transwell membrane and incubated for 48 h (25:5:1 suppressor:effector:target ratios). The percentage of specific NK cell
cytotoxicity toward gated green K562 cells was analyzed by counting of PI-positive cells. Suppressive activity of HLA-G+ NK cells was
also assessed in cellcell contact direct suppression tests by
expression of granzyme B, perforin, IFN-c, and CD107a in HLA-G

gated NK cells.
2.8. Statistical analysis
Statistical analysis was performed using Prizm 6 software
(Graph Pad) or STATISTICA 6.0. Data are presented as mean values standard deviation (SD) or dot and box plots. To explore
the existence of statistically significant differences, two-tailed,
two-sample Students t-test was used. The Wilcoxons signedrank test was used to determine pairwise differences between
CD57+ and CD57
NK cell subsets. p values 60.05 were considered
statistically significant.
3. Results
3.1. PB NK cells of breast cancer patients contain an increased number
of IL-10+ and TGF-b+ cells
The percentage of PB NK cells was increased in the breast cancer
group compared to healthy donors (Fig. 1).
Freshly isolated PBMCs from breast cancer patients contained
higher amounts of NK cells expressing intracellular IL-10 (i-IL-10)

2.6. PCR analysis

Total RNA was isolated from unstimulated PB NK cells using TRI
Reagent following the manufacturers protocol. Samples were
treated with DNase I. Total RNA was reverse transcribed using random primers and Maxima Reverse Transcriptase. PCR was
performed using Taq polymerase. Primers used to amplify the
HLA-G1 isoform were: 50 -GCTGCAGCGCGCGGAC-30 (forward) and
50 -TGGTGGGCAGGGAAGACTGCTT-30 (reverse) [33]. PB B lymphocytes were used as negative control [34], and CD8+ T cells [17],
and PB monocytes [35] were used as positive controls for HLA-G1
transcripts.
2.7. NK cell cytotoxicity assay
Suppressive activity of the HLA-G+ NK cell subset was determined using a cytotoxicity assay against HLA class I-deficient
K562 erythroleukemia cells, which were previously stained with

Fig. 1. Prevalence of NK cells is elevated in peripheral blood of patients with breast

cancer. PBMCs from 12 healthy individuals and 20 breast cancer patients were
labeled with anti-CD56, anti-CD3, and frequency of NK cells was analyzed by flow
cytometry. Representative flow cytometry and cumulative results for each group
are indicated. Each dot represents one individual. Bars represent the median.
Students t-test showed that there was a significant difference between the groups,
as indicated: *p 6 0.005.

Please cite this article in press as: Y.O. Ostapchuk et al., Peripheral blood NK cells expressing HLA-G, IL-10 and TGF-b in healthy donors and breast cancer
patients, Cell. Immunol. (2015), http://dx.doi.org/10.1016/j.cellimm.2015.09.002

Y.O. Ostapchuk et al. / Cellular Immunology xxx (2015) xxxxxx

and TGF-b (i-TGF-b) compared to healthy donors (Fig. 2A and B).

After overnight incubation with or without PHA, the percentage
of i-IL-10+ and i-TGF-b+ in NK cells were measured. Mitogen
stimulation significantly increased expression of i-IL-10 and iTGF-b in both healthy volunteers and breast cancer patients. The
number of i-IL-10+ and i-TGF-b+ NK cells increased after PHA
stimulation in breast cancer patients compared to healthy donors
(Fig. 2A and B). In addition, ELISPOT assay was used as an alternative method for cytokine production by NK cells and demonstrated
a similar increase in IL-10- and TGF-b-secreting NK cells after
activation with PHA in breast cancer patients (data not shown).
Earlier it has been shown that NK-ireg cells generated from
CD34+ PB hematopoietic progenitors express membrane-bound
IL-10 (mb-IL-10) together with HLA-G [14]. With respect to this,

we examined the assumption that NK cells could express

mb-IL-10 on their surface. After surface labeling, a small number
of mb-IL-10+ NK cells in healthy donors were observed
(Fig. 2A and C). IL-2 alone or in combination with PHA did not
increase the number of cells bearing mb-IL-10 beyond baseline,
while PHA enhanced the percentage of mb-IL-10+ NK cells compared to unstimulated NK cells (Fig. 2C).
Breast cancer patients showed an increased number of mb-IL10+ NK cells in both freshly purified and PHA-stimulated PBMCs
compared to healthy volunteers (Fig. 2A and B).
It is well known that cytokine-producing NK cells are traditionally attributed mostly to the CD56bright NK subset. CD56bright NK
cells have been characterized as a CD57
immature subset [36].
For accurate definition of IL-10+ and TGF-b+ NK cells, expression

Fig. 2. Prevalence of NK cell subsets expressing intracellular IL-10, intracellular TGF-b and membrane-bound IL-10 is elevated in peripheral blood of patients with breast
cancer. Freshly isolated (f.i.) and cultured without stimuli (u.s.) or with 10 lg/ml PHA (PHA) PBMCs from 10 healthy individuals and 20 breast cancer patients were labeled
with anti-CD3, anti-CD56, anti-IL-10, anti-TGF-b, anti-IL-10, and anti-CD57 and analyzed. (A) Representative flow cytometry results and (B) cumulative results of i-IL-10+,
i-TGF-b+, mb-IL-10+ cells in CD3
CD56+ gate for each group are shown. (C) Freshly isolated (f.i.), cultured without any stimulation (u.s.), with 10 lg/ml PHA (PHA), 20 ng/ml
IL-2 (IL-2), and its combination (IL-2 + PHA) PBMCs from 17 healthy donors were gated on CD3
CD56+ and assessed for mb-IL-10 expression. (D) Freshly isolated PBMCs were
gated on CD3
CD56+ and percentage of i-IL-10+, i-TGF-b+, and mb-IL-10+ cells was analyzed in CD57+ and CD57
gates. Box plots in (B) and (D) show the inter quartile range
(p25-p75), bars show the median. Error bars indicate the minimum and maximum values for the healthy donors and breast cancer patients groups. The Wilcoxons signedrank test showed pairwise differences in mb-IL-10 expression between CD57+ and CD57
NK cells (p < 0.01, p < 0.05). Significant differences between columns assessed by
parametric Students t-test are indicated: *p 6 0.0005, **p 6 0.005, ***p 6 0.05.

Please cite this article in press as: Y.O. Ostapchuk et al., Peripheral blood NK cells expressing HLA-G, IL-10 and TGF-b in healthy donors and breast cancer
patients, Cell. Immunol. (2015), http://dx.doi.org/10.1016/j.cellimm.2015.09.002

Y.O. Ostapchuk et al. / Cellular Immunology xxx (2015) xxxxxx

of CD57 was estimated. In healthy donors, most mb-IL-10+ NK cells

were found in the CD57
NK subset, while i-IL-10+ and i-TGF-b+ NK
cells were not associated with the CD57
immature phenotype.
The same distribution of cells was maintained in the breast cancer
group. Interestingly, there was a marked enhancement in the number of mb-IL-10+, i-IL-10+, and i-TGF-b+ NK cells among both CD57+
and CD57
NK cell subsets in cancer patients compared to healthy
women (Fig. 2D).

To verify the specificity of HLA-G immunostaining, PB monocytes

were used as a positive control (Fig. 3A), whereas their constitutive
expression of HLA-G has been previously demonstrated [18].
Moreover, results of PCR showed that PB NK cells express HLAG1 mRNA (Fig. 3B). PHA induced a twofold increase in the HLAG+ NK cell percentage compared to unstimulated NK cells, while
the effect of PHA in combination with IL-2 was slightly lower.
IL-2 alone did not affect expression of HLA-G (Fig. 3C).

3.2. NK cells display membrane-bound HLA-G

3.3. HLA-G+ NK cell subset is enhanced in breast cancer patients

We found a minor subset of PB NK cells that constitutively

expressed surface HLA-G, averaging 4.5% (range, 2.86.6%; n = 11)
of the total NK cells (Fig. 3A). The frequency of HLA-G+ NK cells
did not statistically correlate with age and sex (data not shown).

Our analysis showed increased expression of membraneassociated HLA-G by PB NK cells obtained from breast cancer
patients compared to healthy women (Fig. 4A). HLA-G expression
was not associated with the age of patients or disease stage. Simi-

Fig. 3. A minor subset of NK cells expresses HLA-G cell surface molecule. Freshly isolated PBMCs from 11 healthy individuals were labeled with anti-CD56, anti-CD3, antiHLA-G, anti-CD14, and analyzed by flow cytometry. (A) Representative dot plots of isotype and FMO controls (upper panel) and HLA-G expression on CD14+ and CD3
CD56+
cells (lower panel) are depicted. (B) PCR analysis of HLA-G1 mRNA isoform in purified PB NK cells of healthy donors. Specific primers to amplify membrane-associated isoform
HLA-G1 (forward: 50 -GCTGCAGCGCGCGGAC-30 ; reverse: 50 -TGGTGGGCAGGGAAGACTGCTT-30 ) were used for PCR amplification. cDNA from PB monocytes and CD8+ T cells
was used as positive controls and B-cells were used as a negative control. Bands corresponding to HLA-G1 transcripts are indicated by arrow. (C) PBMCs from 10 healthy
donors were cultured without stimulation (u.s.) and in the presence of 10 lg/ml PHA (PHA), 20 ng/ml IL-2 (IL-2), and its combination (IL-2 + PHA), after that cells were gated
on CD3
CD56+ and frequency of HLA-G+ cells was estimated. There were significant differences in HLA-G expression between unstimulated and PHA-activated (*p = 0.0001)
cells and between unstimulated and IL-2 plus PHA-activated (**p < 0.05) cells.

Please cite this article in press as: Y.O. Ostapchuk et al., Peripheral blood NK cells expressing HLA-G, IL-10 and TGF-b in healthy donors and breast cancer
patients, Cell. Immunol. (2015), http://dx.doi.org/10.1016/j.cellimm.2015.09.002

Y.O. Ostapchuk et al. / Cellular Immunology xxx (2015) xxxxxx

lar to healthy donors, PHA activated HLA-G expression in NK cells

of breast cancer patients compared to unstimulated cells. Moreover, in the breast cancer group, the percentage of HLA-G+ NK cells
increased after PHA stimulation compared to healthy donors
(Fig. 4A).
The prevalence of HLA-G+ NK cells was higher among CD57

compared to CD57+ cells in freshly purified PBMCs in both breast
cancer patients and healthy donors (Fig. 4B). There was a marked
enhancement in the number of HLA-G+ NK cells among both
CD57+ and CD57
subsets in cancer patients compared to healthy
women (Fig. 4B). In addition, a PHA-induced increase of HLA-G+
NK cells was observed among CD57
but not mature CD57+ subsets
in both groups (data not shown). Notably, no correlation was seen
between HLA-G expression in freshly isolated or PHA-activated NK
cells and IL-10 and TGF-b levels in all groups (data not shown).
3.4. HLA-G-expressing NK cells of healthy donors suppress cytotoxicity
of HLA-G non-expressing NK cells in vitro
Next, cytokine profiles of HLA-G+ NK cells after PHA stimulation
were investigated. The HLA-G+ NK subset contained around 55% iIL-10+, 66% i-TGF-b+, and 34% i-IFN-c+ cells (Fig. 5A). HLA-G+ subset
contained significantly higher number of i-TGF-b+ and less i-IFN-c+
cells comparing to HLA-G
NK (Fig. 5A). Notably, analysis of cytokine profiles of freshly purified, unstimulated, or PHA-activated

HLA-G+ NK cells disclosed no significant difference in expression

of i-IL-10, i-TGF-b, or i-IFN-c between healthy donors and breast
cancer groups (data not shown).
Secretion of potentially tolerogenic/immunosuppressive factors
by HLA-G+ NK cells, as well as inhibitory effects of HLA-G on NK
cell effector functions demonstrated earlier, led us to suggest
involvement of this subset in paracrine regulation of NK cell cytotoxic activity. To investigate this possibility, we assessed sorted
HLA-G+ and HLA-G
NK cell subsets in a suppression assay
(Fig. 5B). HLA-G+ NK cells did not exert cytotoxic effects on K562
cells (Fig. 5C and D). To address whether expression of HLA-G
was associated with the regulatory NK cell subset, HLA-G+ NK cells
were added as suppressors to autologous HLA-G
NK cells
co-cultured with K562 cells (suppressor:effector:target ratio
25:5:1). In this case HLA-G+ NK cells inhibited HLA-G
NK cell
cytotoxic activity up to 50% (Fig. 5C and D). Moreover, HLA-G+
NK cells suppressed expression of CD107a (Fig. 5E), production of
perforin, and granzyme B, but not IFN-c (Fig. 5F) by autologous
HLA-G
NK cells compared to co-culture of HLA-G
NK cells with
K562 cells alone. Furthermore, analysis of the influence of possible
soluble factors secreted by HLA-G+NK cells using the Transwell
membrane showed there was no suppression of HLA-G
NK cell
cytotoxicity (data not shown). Therefore, we suggest that inhibition of HLA-G
NK cell cytotoxicity depends on cell-to-cell contact
with their HLA-G+ counterpart and is not mediated by HLA-G+

Fig. 4. Frequency of HLA-G+ NK cells is increased in breast cancer patients. Freshly isolated PBMCs from 11 healthy individuals and 15 breast cancer patients were labeled
with anti-CD3, anti-CD56, anti-CD57, anti-HLA-G, and analyzed by flow cytometry. (A) Mean percentage of HLA-G+ NK cells in freshly isolated (f.i.), cultured without
stimulation (u.s.), and with 10 lg/ml PHA (PHA) PBMCs is shown. (B) CD3
CD56+ NK cells were gated on CD57+ and CD57
cells and expression of HLA-G was analyzed.
Representative results and cumulative data for the groups of healthy donors and breast cancer patients are depicted. Each dot represents one individual. Bars represent the
median. The Wilcoxons signed-rank test showed pairwise differences in HLA-G expression between CD57+ and CD57
NK cells (p < 0.05) in both groups. Students t-test
showed that there were significant differences between columns, as indicated: *p 6 0.0005, **p 6 0.005.

Please cite this article in press as: Y.O. Ostapchuk et al., Peripheral blood NK cells expressing HLA-G, IL-10 and TGF-b in healthy donors and breast cancer
patients, Cell. Immunol. (2015), http://dx.doi.org/10.1016/j.cellimm.2015.09.002

Y.O. Ostapchuk et al. / Cellular Immunology xxx (2015) xxxxxx

Fig. 5. HLA-G+ NK subset produces immunoregulatory cytokines IL-10, TGF-b and IFN-c and suppresses cytotoxic activity of HLA-G
NK cells. (A) PBMCs obtained from 6
healthy donors were incubated with PHA, stained with anti-CD3, anti-CD56, anti-IL-10, anti-TGF-b, anti-IFN-c and analyzed by flow cytometry. Mean percentage of IL-10+,
TGF-b+, and IFN-c+ cells in CD3
CD56+HLA-G+ and CD3
CD56+HLA-G
gates is depicted. Isolated NK cells from healthy individuals were stimulated with PHA for 18 h, labeled
with HLA-G-PE and sorted into HLA-G+ and HLA-G
subsets. (B) Representative results of the cell sorting are shown. HLA-G
cells were co-cultured for 2 days with K562 cell
line that had been previously stained with CFSE (effector:target ratio of 5:1) in round-bottom plates. HLA-G+ NK cells were added as suppressors at a suppressor:effector:target ratio of 25:5:1. Anti-HLA-G blocking Ab (a-HLA-G) was added at the final concentration equal to 10 lg/ml, then cells were stained with PI and percentage of
CFSE+PI+ labeled cells was analyzed by flow cytometry (n = 8). (C) Representative results and (D) mean percentage are shown. Obtained cultures and HLA-G
NK cells were
also labeled with (E) anti-CD107a-APC, and (F) anti-perforin-APC, anti-GranzymeB-FITC, anti-IFN-c-FITC, and percentage of positive cells was analyzed in CD3
CD56+HLA-G

gate. (J) Sorted HLA-G+ NK cells were washed, cultured for 48 h without any stimuli, labeled with anti-CD56, and presence of HLA-G mAbs was assessed by flow cytometry
with or without secondary labeling of HLA-G. One of three representative analyses is depicted. p values calculated using the Students t-test and indicated as: *p 6 0.0005,
**
p 6 0.005, ***p 6 0.05.

Please cite this article in press as: Y.O. Ostapchuk et al., Peripheral blood NK cells expressing HLA-G, IL-10 and TGF-b in healthy donors and breast cancer
patients, Cell. Immunol. (2015), http://dx.doi.org/10.1016/j.cellimm.2015.09.002

Y.O. Ostapchuk et al. / Cellular Immunology xxx (2015) xxxxxx

NK-secreted soluble factors/cytokines. Moreover, HLA-G neutralization by blocking Abs antagonized the suppressive effect,
leading to restoration of cytotoxic capacity of HLA-G
NK cells
(Fig. 5C and D). Therefore, we concluded that the suppressive effect
of HLA-G+ NK cells was mediated by the HLA-G molecule.
In note, 48 h was chosen for the cytotoxicity assay as far as
mAbs derived from the same clone (87G) were used for HLA-G+ cell
sorting and further HLA-G neutralization. We showed that sorted
HLA-G+ NK cells lost HLA-G mAbs (up to 80%) during following
two days of cultivation (Fig. 5J). We assumed that sorted HLA-G+
NK cells lost HLA-G mAbs through internalization during 48 h
incubation and further mAbs intracellular degradation, after that
the protein is restored on the cell surface. Furthermore, Davis
et al. showed that B cell line 721.221 transfected with gene encoding HLA-G lost HLA-G protein from the cell surface at a rate of
almost 6% per hour. They also found weak staining of HLA-Gtailed protein in the cytosol, which could represent endocytosed
protein [37]. Precise mechanism for the loss of HLA-G mAbs from
the cell surface is unknown, but the epitope for 87G clone mAbs
seemed to be restored during 2 days of incubation and can be
bound by blocking Abs used in our experimental settings.

4. Discussion
Immunoregulatory properties of NK cells have recently been
demonstrated. It has been shown that PB IL-10-producing
CD56bright and CD56dim NK cell subsets are increased in early pregnancy compared to spontaneous abortion cases [12] and in the follicular phase of the ovarian cycle [38]. Also, a portion of decidual
NK cells has been shown to secrete IL-10 [13]. Recent findings suggest the involvement of IL-10-producing NK cells in regulation of
maternal tolerance. Furthermore, IL-10-producing NK cells
obtained from PB of healthy donors have been shown to express
TGF-b and inhibit Ag-induced T cell proliferation [13]. Implication
of TGF-b and IL-10 in tumor growth via suppression of anti-cancer
immunity has been reported by numerous studies [3942]. However, the blood level of immunoregulatory NK cells in cancer,
including breast cancer, has remained elusive. Thus, the goal of
the present study was to evaluate immunoregulatory NK cells producing TGF-b and IL-10 in breast cancer patients. Our analyses
showed a significant increase in the proportion of these cells in
the circulation of breast cancer patients compared to healthy
donors. In breast cancer patients, the percentage of i-IL-10+ and
i-TGF-b+ NK cells was increased in both CD57+ and CD57
NK cell
subsets. It is well known that CD57 expression identifies the final
stages of NK cell maturation and defines a subset with a higher
cytotoxic capacity, greater responsiveness to signaling via CD16
and natural cytotoxicity receptors, and decreased responsiveness
to cytokines and proliferative activity [40]. It is possible that not
only immature CD57
NK cells, but also a minor part of mature
CD57+ NK cells acquire cytokine-producing potential, and the pool
of these cells is increased in breast cancer.
Expression of membrane-associated forms of IL-10 has been
shown in ex vivo-generated suppressive NK cells [14]. In the present study, we could barely detect expression of IL-10 on the surface of freshly purified and PHA-activated PB NK cells of healthy
donors. However, PB NK cells of breast cancer patients either
freshly isolated or PHA-stimulated showed an increased level of
membrane-associated IL-10. If this cytokine is anchored directly
to the cell membrane or bound to its receptor on the cell surface,
it is a matter of special interest.
The biological significance of IL-10- and TGF-b-expressing NK
cells in cancer is unclear. The dual role of IL-10 and TGF-b in
tumorigenesis has been demonstrated. IL-10 may exhibit antitumor activities by promotion of NK cell cytotoxicity in preclinical

models. It has been shown that malignant cells exposed to IL-10

down-regulate HLA class I molecules on their surface, which
reduces their sensitivity to CTL but increases one to NK cell cytotoxicity [43]. At the same time, IL-10 exhibits pro-tumoral activity,
inhibiting classical APC maturation, which preserves their antigenuptake function and promotes polarization towards the M2 phenotype [44]. In addition, IL-10 inhibits IFN-c and TNF-a production by
NK cells [4549], which play a big role in NK cytotoxic activity
towards target cells [48]. TGF-b acts as a tumor suppressor affecting cell cycle inhibition, induction of apoptosis, and prevention of
cell immortalization. Expression of the TGF-b type II receptor in
breast cancer cells prevents tumor formation [49]. On the other
hand, TGF-b plays a major role in promoting breast cancer migration, invasion, and metastasis by acting on the stroma, neighboring
cells surrounding the tumor, and directly on cancer cells. TGF-b
also inhibits expression of NKG2D and NKp30 in NK cells [50].
Our data do not allow us to exclude paracrine or autocrine roles
of IL-10 and TGF-b produced by NK cells and their negative or positive impact on anti-cancer immunity. Further studies are needed
to explore the role of these cells in cancer pathogenesis.
A second aspect of this study was HLA-G1 expression in PB NK
cells. It has been reported that around 4% of decidual NK cells
express HLA-G [14], but neither HLA-G+ NK cells in PB of pregnant
women, nor soluble or membrane-bound forms of HLA-G of PB NK
cells in the control group using Western blot were found. Our findings support evidence that a small subset of NK cells expressing
HLA-G are constitutively presented in PB of healthy individuals.
We assume that these cells can migrate to decidua and promote
immunosuppression [51], determining fetomaternal tolerance
[52] during pregnancy which would explain the absence of HLAG+ NK cells in PB blood of pregnant women. We can also presume
that HLA-G+ NK cells could not be detected in PB of healthy donors
by Western blot due to their low content. In contrast, the results of
PCR assay confirmed expression of the HLA-G1 isoform by NK cells
and excluded acquisition of HLA-G from other cells through trogocytosis [15] or binding of soluble HLA-G to its receptor on the surface of NK cells.
Increased HLA-G levels in biological fluids have been associated
with down-regulation of the immune response. There is evidence
that ectopic HLA-G expression in tumor cells contributes to escape
of malignant cells from immune recognition and destruction. T
cells, dendritic cells, and monocytes/macrophages are able to
secrete soluble HLA-G molecules in vitro. Serum HLA-G proteins,
which are derived from membrane-bound HLA-G isoforms as
HLA-G1, and secreted soluble HLA-G isoforms, like HLA-G5, may
affect antitumor immune responses both locally at the tumor site
and systemically by distribution via the circulation. HLA-G plasma
levels are significantly increased in patients with malignant melanoma, glioma, and ovarian and breast carcinoma [20]. HLA-G
expression is significantly more frequent in breast cancer lesions
that are highly infiltrated by host immune cells. Increased HLA-G
expression is attributed mainly to tumor epithelial cells and subsets of infiltrating CD8+ and CD68+ cells [53]. Notably, NK cells
are shown to express CD68. The results suggest that HLA-Gmediated effects are upregulated in human breast cancer. These
observations coincide with our finding that an increased level of
HLA-G+ NK cells in PB of breast cancer patients that may result
in impaired efficiency of antitumor immunity.
In the present study, we also showed that HLA-G+ NK cell subset
was enriched with TGF-b producing cells. The production of potentially regulatory molecules by HLA-G+ NK subset appears to contribute to their immunoregulatory function. Our results also
showed that expression of HLA-G was not associated with the
NK cell maturity marker CD57. HLA-G+ NK cells did not exert cytolytic activity. Similar to in vitro generated NK-ireg cells, PB HLA-G+
NK cells possessed suppressive properties; they suppressed cytol-

Please cite this article in press as: Y.O. Ostapchuk et al., Peripheral blood NK cells expressing HLA-G, IL-10 and TGF-b in healthy donors and breast cancer
patients, Cell. Immunol. (2015), http://dx.doi.org/10.1016/j.cellimm.2015.09.002

Y.O. Ostapchuk et al. / Cellular Immunology xxx (2015) xxxxxx

ysis of K562 cells, as well as expression of perforin, granzyme, and

CD107a by HLA-G
NK cells. Since the effects were shown to
depend on cell-to-cell contact and could be abrogated by HLA-G
blocking mAbs, we presume that membrane-associated HLA-G1
largely contributes to the suppressive effect of HLA-G+ NK cells.
The data are consistent with recent reports that HLA-G1 inhibits
NK cell-mediated cytotoxicity [22], and may also induce fratricidal
killing of NK cells [54]. We can assume that HLA-G1 expressed on
NK cells has similar effect on NK cell cytotoxicity to molecular
events occurring at NK/HLA-G1 expressing target-cell synapse
reported earlier. It has been shown that interaction of ILT2 on
NK-cells with HLA-G1 on target cells strongly inhibits targetcell-induced intracellular mobilization of calcium in NK cells, accumulation of F-actin and LFA-1 at the contact site, and microtubuleorganizing center and perforin granules polarization toward target
cells [22]. These results indicate that HLA-G+ NK cells represent a
regulatory subset which express regulatory molecules and
maintain their functions mostly via HLA-G.
With this study, we addressed whether breast cancer PB NK
cells are able to respond to an unspecific polyclonal stimulation
and increase the expression of HLA-G, IL-10, and TGF-b similar to
NK cells from healthy women. A number of studies have demonstrated that polyclonal stimulation resulted in expansion and activation of NK cells. Our previous study showed that PHA
stimulation in combination with IL-2 induced IL-10 mRNA expression and IL-10 production in NK cells in 2 h [12]. In our hands, PHA
increased expression of HLA-G, i-IL-10, i-TGF-b, and mb-IL-10 in
NK cells both in the control group and breast cancer patients. In
our hands, PHA increased expression of HLA-G, i-IL-10, i-TGF-b,
and mb-IL-10 in NK cells both in the control group and breast cancer patients in nonspecific manner binding glycosylated receptors
[55]. Moreau et al. reported that IL-10 selectively induces HLA-G
expression in human trophoblasts and monocytes [33]. In our
hands, increased expression of both HLA-G and IL-10 was observed
after PHA stimulation of NK cells. We speculate that IL-10 produced in response to mitogen-activated, autoregulated expression
of HLA-G in NK cells.
The increase in the relative content of HLA-G+, i-IL-10+, i-TGF-b+,
and mb-IL-10+ NK cells in breast cancer patients was detected in
both CD57+ and CD57
NK cell subsets and associated with an augmented proportion of circulating NK cells compared to control
donors. Moreover, here we show an increased percentage of
CD3
CD56+ cells in peripheral blood of untreated breast cancer
patients, which is in agreement with a previous study that showed
increased percentage and absolute number of PB CD3
CD56+ cells
in breast cancer patients [56]. However, another group did not find
any significant changes in NK cell numbers in breast cancer based
on expression of CD56 and CD16 [30]. At the same time, the
increased frequency of immature, non-cytotoxic CD56brightCD16

NK cells in PB was shown to be associated with characteristics of
invasive and advanced breast cancer [32]. Increased infiltration
by CD56brightCD16
NK cells was observed in thyroid tissue of
patients with papillary thyroid [57] and breast cancer. Malignant
breast tissues had less CD56dimCD16+ cells than did healthy mammary tissues. Tumor-infiltrating NK cells had poor cytotoxic capacity, and their altered activity depended on the tumor
microenvironment, which had the highest correlation with the
presence of PGE2 and TGF-b1 [31]. Furthermore, endogenous TGFb was recently shown to repress development of NK cells from
CD34+ progenitors, inhibits differentiation of CD16+ NK cells, and
trigger conversion of PB CD56brightCD16+ cells into CD56brightCD16

cells, highlighting a possible role of the former as a developmental
intermediator of NK subsets found in blood [58]. IL-10- and TGF-bproduction by NK cells was shown to be induced by HLA-E
expressed by tumor-associated macrophages [19]. Taking together,
we can speculate that tumor immune-editing played the major

role in NK cell subset alterations. The increased pool of the peripheral blood NK cells in breast cancer may be a result of an upregulated subpopulation of immunoregulatory NK cells. Thus, it seems
that in breast cancer, some part of NK cells gain immunoregulatory
properties which may partly explain the low cytotoxic functions of
NK cells in breast cancer reported earlier [2931].
5. Conclusions
In summary, our results demonstrate that HLA-G+ NK cells represent a new immunoregulatory subset of PB NK cells with distinct
characteristics, including production of TGF-b, low expression of
CD57, and suppression of autologous NK cells. The prevalence of
circulating HLA-G+ NK cells, as well as NK cells producing IL-10
and TGF-b cytokines that have been previously shown to have suppressive properties, is considerably increased in breast cancer
patients. Additional experiments need to be performed to study
the mechanisms by which these NK cell subsets are induced and
expanded in breast cancer patients and to investigate their contribution to mechanisms of tumor escape from innate immunity. Our
study highlights that the precise identification of regulatory NK cell
subsets endowed with particular functional capabilities should be
of great help to monitor NK cell-mediated responses in breast cancer and to design future therapies based on anti-tumor immunity.
Acknowledgment
The authors thank Dr. G.K. Zakiryanova for contribution to the
scientific research.
References
[1] J.D. Brandstadter, Y. Yang, Natural killer cell responses to viral infection, J.
Innate Immun. 3 (2011) 274279, http://dx.doi.org/10.1159/000324176.
[2] L. Zamai, C. Ponti, P. Mirandola, G. Gobbi, S. Papa, L. Galeotti, et al., NK cells and
cancer, J. Immunol. 178 (2007) 40114016. 6606.
[3] F. Souza-Fonseca-Guimaraes, M. Adib-Conquy, J.-M. Cavaillon, Natural killer
(NK) cells in antibacterial innate immunity: angels or devils?, Mol Med. 18
(2012) 270285, http://dx.doi.org/10.2119/molmed.2011.00201.
[4] D.S. Korbel, O.C. Finney, E.M. Riley, Natural killer cells and innate immunity to
protozoan pathogens, Int. J. Parasitol. (2004) 15171528, http://dx.doi.org/
10.1016/j.ijpara.2004.10.006.
[5] J. Zimmer, Natural Killer Cells: At the Forefront of Modern Immunology,
Springer-Verlag, Berlin, Heidelberg, 2010.
[6] A. Poli, T. Michel, M. Thrsine, E. Andrs, F. Hentges, J. Zimmer, CD56bright
natural killer (NK) cells: an important NK cell subset, Immunology 126 (2009)
458465, http://dx.doi.org/10.1111/j.1365-2567.2008.03027.x.
[7] J. Punt, J. Owen, M.A. Caligiuri, The biology of human natural killer-cell subsets,
Trends Immunol. 22 (2001) 633640, http://dx.doi.org/10.1016/S1471-4906
(01)02060-9.
[8] L. Moretta, Dissecting CD56dim human NK cells, Blood 116 (2010) 36893691.
[9] M.A. Caligiuri, Human natural killer cells, Blood 112 (2008) 461469, http://dx.
doi.org/10.1182/blood-2007-09-077438.
[10] D. Peritt, S. Robertson, G. Gri, L. Showe, M. Aste-Amezaga, G. Trinchieri, Cutting
edge: differentiation of human NK cells into NK1 and NK2 subsets, J. Immunol.
161 (1998) 58215824.
[11] <http://www.jimmunol.org/content/161/11/5821.abstract?relatedrls=yes&
legid=jimmunol;161/11/5821>.
[12] S. Higuma-Myojo, Y. Sasaki, S. Miyazaki, M. Sakai, A. Siozaki, N. Miwa, et al.,
Cytokine profile of natural killer cells in early human pregnancy, Am. J. Reprod.
Immunol. 54 (2005) 2129, http://dx.doi.org/10.1111/j.1600-0897.2005.
00279.x.
[13] G. Deniz, G. Erten, U.C. Kcksezer, D. Kocacik, C. Karagiannidis, E. Aktas, et al.,
Regulatory NK cells suppress antigen-specific T cell responses, J. Immunol. 180
(2008) 850857, http://dx.doi.org/10.4049/jimmunol.180.2.850.
[14] M. Giuliani, J. Giron-Michel, S. Negrini, P. Vacca, D. Durali, A. Caignard, et al.,
Generation of a novel regulatory NK cell subset from peripheral blood CD34+
progenitors promoted by membrane-bound IL-15, PLoS ONE 3 (2008), http://
dx.doi.org/10.1371/journal.pone.0002241.
[15] J. Caumartin, B. Favier, M. Daouya, C. Guillard, P. Moreau, E.D. Carosella, et al.,
Trogocytosis-based generation of suppressive NK cells, EMBO J. 26 (2007)
14231433, http://dx.doi.org/10.1038/sj.emboj.7601570.
[16] Y.-H. Huang, A.L. Zozulya, C. Weidenfeller, N. Schwab, H. Wiendl, T cell
suppression by naturally occurring HLA-G-expressing regulatory CD4+ T cells
is IL-10-dependent and reversible, J. Leukoc. Biol. 86 (2009) 273281, http://
dx.doi.org/10.1189/jlb.1008649.

Please cite this article in press as: Y.O. Ostapchuk et al., Peripheral blood NK cells expressing HLA-G, IL-10 and TGF-b in healthy donors and breast cancer
patients, Cell. Immunol. (2015), http://dx.doi.org/10.1016/j.cellimm.2015.09.002

10

Y.O. Ostapchuk et al. / Cellular Immunology xxx (2015) xxxxxx

[17] U. Feger, E. Tolosa, Y.H. Huang, A. Waschbisch, T. Biedermann, A. Melms, et al.,

HLA-G expression defines a novel regulatory T-cell subset present in human
peripheral blood and sites of inflammation, Blood 110 (2007) 568577, http://
dx.doi.org/10.1182/blood-2006-11-057125.
[18] V. Rebmann, A. Busemann, M. Lindemann, H. Grosse-Wilde, Detection of HLAG5 secreting cells, Hum. Immunol. (2003) 10171024, http://dx.doi.org/
10.1016/j.humimm.2003.08.354.
[19] M. Marchesi, E. Andersson, L. Villabona, B. Seliger, A. Lundqvist, R. Kiessling, G.
V. Masucci, HLA-dependent tumour development: a role for tumour associate
macrophages?, J Transl. Med. 11 (2013), http://dx.doi.org/10.1186/1479-587611-247.
[20] N. Rouas-Freiss, P. Moreau, S. Ferrone, E.D. Carosella, HLA-G proteins in cancer:
do they provide tumor cells with an escape mechanism?, Cancer Res 65 (2005)
1013910144, http://dx.doi.org/10.1158/0008-5472.CAN-05-0097.
[21] E. Lesport, J. Baudhuin, J. LeMaoult, S. Sousa, C. Doliger, E.D. Carosella, et al.,
Human melanoma cell secreting human leukocyte antigen-G5 inhibit natural
killer cell cytotoxicity by impairing lytic granules polarization toward target
cell, Hum. Immunol. 70 (2009) 10001005, http://dx.doi.org/10.1016/j.
humimm.2009.07.019.
[22] B. Favier, J. Lemaoult, E. Lesport, E.D. Carosella, ILT2/HLA-G interaction impairs
NK-cell functions through the inhibition of the late but not the early events of
the NK-cell activating synapse, FASEB J. 24 (2010) 689699, http://dx.doi.org/
10.1096/fj.09-135194.
[23] C. Li, B.L. Houser, M.L. Nicotra, J.L. Strominger, HLA-G homodimer-induced
cytokine secretion through HLA-G receptors on human decidual macrophages
and natural killer cells, Proc. Natl. Acad. Sci. U.S.A. 106 (2009) 57675772,
http://dx.doi.org/10.1073/pnas.0901173106.
[24] E. Morel, T. Belln, HLA class I molecules regulate IFN-gamma production
induced in NK cells by target cells, viral products, or immature dendritic cells
through the inhibitory receptor ILT2/CD85j, J. Immunol. 181 (2008) 23682381.
[25] A. van der Meer, H.G.M. Lukassen, B. van Cranenbroek, E.H. Weiss, D.D.M. Braat,
M.J. van Lierop, et al., Soluble HLA-G promotes Th1-type cytokine production by
cytokine-activated uterine and peripheral natural killer cells, Mol. Hum. Reprod.
13 (2007) 123133, http://dx.doi.org/10.1093/molehr/gal100.
[26] S. Rajagopalan, Y.T. Bryceson, S.P. Kuppusamy, D.E. Geraghty, A. Van Der Meer,
I. Joosten, et al., Activation of NK cells by an endocytosed receptor for soluble
HLA-G, PLoS Biol. 4 (2006) 00700086, http://dx.doi.org/10.1371/journal.
pbio.0040009.
[27] S.R. White, D.a. Loisel, R. Stern, B. Laxman, T. Floreth, B.a. Marroquin, Human
leukocyte antigen-G expression in differentiated human airway epithelial
cells: lack of modulation by Th2-associated cytokines, Respir. Res. 14 (2013) 4,
http://dx.doi.org/10.1186/1465-9921-14-4.
[28] B.W. Stewart, C.P. Wild, World Cancer Report 2014, 2014. ISBN: 9283204298.
[29] W.L. Garner, J.P. Minton, A.G. James, C.C. Hoffmann, Human breast cancer and
impaired NK cell function, J. Surg. Oncol. 24 (1983) 6466, http://dx.doi.org/
10.1002/jso.2930240115.
[30] I. Caras, A. Grigorescu, C. Stavaru, D.L. Radu, I. Mogos, G. Szegli, et al., Evidence
for immune defects in breast and lung cancer patients, Cancer Immunol.
Immunother. 53 (2004) 11461152, http://dx.doi.org/10.1007/s00262-0040556-2.
[31] E. Mamessier, A. Sylvain, M.L. Thibult, G. Houvenaeghel, J. Jacquemier, R.
Castellano, et al., Human breast cancer cells enhance self tolerance by
promoting evasion from NK cell antitumor immunity, J. Clin. Invest. 121
(2011) 36093622, http://dx.doi.org/10.1172/JCI45816.
[32] E. Mamessier, L.C. Pradel, M.-L.M.-L. Thibult, C. Drevet, A. Zouine, J. Jacquemier,
et al., Peripheral blood NK cells from breast cancer patients are tumor-induced
composite subsets, J. Immunol. 190 (2013) 24242436, http://dx.doi.org/
10.4049/jimmunol.1200140.
[33] P. Paul, N. Rouas-Freiss, I. Khalil-Daher, P. Moreau, B. Riteau, F.A. Le Gal, et al.,
HLA-G expression in melanoma: a way for tumor cells to escape from
immunosurveillance, Proc. Natl. Acad. Sci. U.S.A. 95 (1998) 45104515, http://
dx.doi.org/10.1073/pnas.95.8.4510.
[34] M.R. Rezvany, A. Kazemi, A. Hajifathali, S. Kaviani, H. Mellstedt, Analysis of
HLA-G gene expression in B-lymphocytes from chronic lymphocytic leukemia
patients, Iran. Biomed. J. 11 (2007) 125129.
[35] P. Moreau, F. Adrian-Cabestre, C. Menier, V. Guiard, L. Gourand, J. Dausset,
et al., IL-10 selectively induces HLA-G expression in human trophoblasts and
monocytes, Int. Immunol. 11 (1999) 803811, http://dx.doi.org/10.1093/
intimm/11.5.803.
[36] C.M. Nielsen, M.J. White, M.R. Goodier, E.M. Riley, Functional significance of
CD57 expression on human NK cells and relevance to disease, Front. Immunol.
4 (2013), http://dx.doi.org/10.3389/fimmu.2013.00422.
[37] D.M. Davis, H.T. Reyburn, L. Pazmany, I. Chiu, O. Mandelboim, J.L. Strominger,
Impaired spontaneous endocytosis of HLA-G, Eur. J. Immunol. 27 (1997) 2714
2719, http://dx.doi.org/10.1002/eji.1830271035.
[38] A.L. Veenstra van Nieuwenhoven, A. Bouman, H. Moes, M.J. Heineman, L.F.M.H.
De Leij, J. Santema, et al., Cytokine production in natural killer cells and

[39]

[40]

[41]

[42]

[43]

[44]

[45]

[46]

[47]

[48]

[49]

[50]

[51]

[52]

[53]

[54]

[55]
[56]

[57]

[58]

lymphocytes in pregnant women compared with women in the follicular

phase of the ovarian cycle, Fertil. Steril. 77 (2002) 10321037, http://dx.doi.
org/10.1016/S0015-0282(02)02976-X.
Hamidullah, B. Changkija, R. Konwar, Role of interleukin-10 in breast cancer,
Breast Cancer Res. Treat. 133 (2012) 1121, http://dx.doi.org/10.1007/s10549011-1855-x.
C.A. Crane, S.J. Han, J.J. Barry, B.J. Ahn, L.L. Lanier, A.T. Parsa, TGF-b
downregulates the activating receptor NKG2D on NK cells and CD8+ T cells
in glioma patients, Neuro-Oncology 12 (2010) 713, http://dx.doi.org/
10.1093/neuonc/nop009.
E.B. Wilson, J.J. El-Jawhari, A.L. Neilson, G.D. Hall, A.A. Melcher, J.L. Meade,
et al., Human tumour immune evasion via TGF-b blocks NK cell activation but
not survival allowing therapeutic restoration of anti-tumour activity, PLoS
ONE 6 (2011), http://dx.doi.org/10.1371/journal.pone.0022842.
A. Taylor, J. Verhagen, K. Blaser, M. Akdis, C.A. Akdis, Mechanisms of immune
suppression by interleukin-10 and transforming growth factor-b: the role of T
regulatory cells, Immunology 117 (2006) 433442, http://dx.doi.org/10.1111/
j.1365-2567.2006.02321.x.
S. Mocellin, F.M. Marincola, H.A. Young, Interleukin-10 and the immune
response against cancer: a counterpoint, J. Leukoc. Biol. 78 (2005) 10431051,
http://dx.doi.org/10.1189/jlb.0705358.
S. Mocellin, M.C. Panelli, E. Wang, D. Nagorsen, F.M. Marincola, The dual role of
IL-10, Trends Immunol. 24 (2003) 3643, http://dx.doi.org/10.1016/S14714906(02)00009-1.
K.W. Moore, R. de Waal Malefyt, R.L. Coffman, A. OGarra, Interleukin-10 and
the interleukin-10 receptor, Annu. Rev. Immunol. 19 (2001) 683765, http://
dx.doi.org/10.1146/annurev.immunol.19.1.683.
J.L. Mege, S. Meghari, A. Honstettre, C. Capo, D. Raoult, The two faces of
interleukin 10 in human infectious diseases, Lancet Infect. Dis. 6 (2006) 557
569, http://dx.doi.org/10.1016/S1473-3099(06)70577-1.
M.J. Scott, J.J. Hoth, M. Turina, D.R. Woods, W.G. Cheadle, Interleukin-10
suppresses natural killer cell but not natural killer T cell activation during
bacterial infection, Cytokine 33 (2006) 7986, http://dx.doi.org/10.1016/
j.cyto.2005.12.002.
R. Wang, J.J. Jaw, N.C. Stutzman, Z. Zou, P.D. Sun, Natural killer cell-produced
IFN-c and TNF-a induce target cell cytolysis through up-regulation of ICAM-1,
J. Leukoc. Biol. 91 (2012) 299309, http://dx.doi.org/10.1189/jlb.0611308.
L. Sun, G. Wu, J.K. Willson, E. Zborowska, J. Yang, I. Rajkarunanayake, et al.,
Expression of transforming growth factor beta type II receptor leads to
reduced malignancy in human breast cancer MCF-7 cells, J. Biol. Chem. 269
(1994) 2644926455.
J.J. Lebrun, The dual role of TGF-b in human cancer: from tumor suppression to
cancer metastasis, ISRN Mol. Biol. 2012 (2012) 128, http://dx.doi.org/
10.5402/2012/381428.
C. Lidstrm, L. Matthiesen, G. Berg, S. Sharma, J. Ernerudh, C. Ekerfelt, Cytokine
secretion patterns of NK cells and macrophages in early human pregnancy
decidua and blood: implications for suppressor macrophages in decidua, Am. J.
Reprod. Immunol. 50 (2003) 444452, http://dx.doi.org/10.1046/j.87558920.2003.00112.x.
G. Chaouat, S. Zourbas, S. Ostojic, G. Lappree-Delage, S. Dubanchet, N. Ledee, J.
Martal, A brief review of recent data on some cytokine expressions at the
materno-foetal interface which might challenge the classical Th1/Th2
dichotomy, J. Reprod. Immunol. 53 (2002) 241256, http://dx.doi.org/
10.1016/S0165-0378(01)00119-X.
S. Lefebvre, M. Antoine, S. Uzan, M. McMaster, J. Dausset, E.D. Carosella, et al.,
Specific activation of the non-classical class I histocompatibility HLA-G
antigen and expression of the ILT2 inhibitory receptor in human breast
cancer, J. Pathol. 196 (2002) 266274, http://dx.doi.org/10.1002/path.1039.
A. Lindaman, A. Dowden, N. Zavazava, Soluble HLA-G molecules induce
apoptosis in natural killer cells, Am. J. Reprod. Immunol. 56 (2006) 6876,
http://dx.doi.org/10.1111/j.1600-0897.2006.00395.x.
H. Rasmussen, The calcium messenger system (1), N. Engl. J. Med. 314 (1986)
10941101.
F. Mozaffari, C. Lindemalm, A. Choudhury, H. Granstam-Bjrneklett, I.
Helander, M. Lekander, et al., NK-cell and T-cell functions in patients with
breast cancer: effects of surgery and adjuvant chemo- and radiotherapy, Br. J.
Cancer 97 (2007) 105111, http://dx.doi.org/10.1038/sj.bjc.6603840.
F. Gogali, G. Paterakis, G.Z. Rassidakis, C.I. Liakou, C. Liapi, CD3(
)CD16(
)
CD56(bright) immunoregulatory NK cells are increased in the tumor
microenvironment and inversely correlate with advanced stages in patients
with papillary thyroid cancer, Thyroid 23 (2013) 15611568, http://dx.doi.org/
10.1089/thy.2012.0560.
D.S.J. Allan, B. Rybalov, G. Awong, J.C. Ziga-Pflcker, H.D. Kopcow, J.R.
Carlyle, et al., TGF-b affects development and differentiation of human natural
killer cell subsets, Eur. J. Immunol. 40 (2010) 22892295, http://dx.doi.org/
10.1002/eji.200939910.

Please cite this article in press as: Y.O. Ostapchuk et al., Peripheral blood NK cells expressing HLA-G, IL-10 and TGF-b in healthy donors and breast cancer
patients, Cell. Immunol. (2015), http://dx.doi.org/10.1016/j.cellimm.2015.09.002