Es

SIMN

a

b

c

R

I

Hh

Human Immunology (2008) 69, 9–15

0d

xpression of human leukocyte antigen-G inystemic lupus erythematosus

ilvia Rosadoa,b,*, Gema Perez-Chacona, Susana Mellor-Pitab,nmaculada Sanchez-Vegazoc, Carmen Bellas-Menendezc,aria Jesus Citoresb, Ignacio Losada-Fernandeza, Trinidad Martin-Donairea,erea Rebolledaa, Paloma Perez-Aciegoa

Fundacion LAIR, Madrid, SpainServicio de Medicina Interna I, Hospital Universitario Puerta de Hierro, Madrid, SpainServicio de Anatomia Patologica, Hospital Universitario Puerta de Hierro, Madrid, Spain

eceived 1 August 2007; received in revised form 27 October 2007; accepted 1 November 2007

Summary The purpose of this study was to examine the expression of human leukocyte antigen-G(HLA-G) in patients with systemic lupus erythematosus (SLE) and its relation with interleukin-10(IL-10) production. The study included 50 female SLE patients and 59 healthy female donors. HLA-Gexpression in peripheral blood and cutaneous biopsies was determined by flow cytometry andimmunohistochemistry, respectively. Soluble HLA-G (sHLA-G) and IL-10 were quantified in serumsamples by enzyme-linked immunosorbent assay. SLE patients presented with serum sHLA-G andIL-10 levels significantly higher than that observed in controls (median [interquartile range (IQR)] �43.6 U/ml [23.2–150.2] vs 26.84 U/ml [6.0–45.2], p � 0.004; and 1.4 pg/ml [0–2.3] vs 0 pg/ml[0–1.5], p � 0.01, respectively). But no correlation was observed between sHLA-G and both IL-10levels and the disease activity index for SLE patients. The expression of membrane HLA-G in periph-eral lymphocytes from SLE patients was low, but higher than in controls (median [IQR] � 1.5%[0.6–1.8] and 0.3% [0.2–0.8], respectively; p � 0.02). Finally, these findings were in accordance withthe weak expression of HLA-G in skin biopsies. Despite the fact that patients present higher levels ofHLA-G than healthy controls, which suggests a possible relevance of this molecule in SLE, it seemsnot to be related to IL-10 production or disease activity.© 2008 American Society for Histocompatibility and Immunogenetics. Published by Elsevier Inc. Allrights reserved.

KEYWORDSHuman leukocyte

antigen-G;

Systemic lupus

erythematosus;

Interleukin-10

flae(G

ntroduction

uman leukocyte antigen-G (HLA-G) is a nonclassic majoristocompatibility complex antigen. HLA-G gene differs

* Corresponding author. Fax: 34 91 373 76 67.

pE-mail address: silvia060500@yahoo.es (S. Rosado).

198-8859/$ -see front matter © 2008 American Society for Histocompatibioi:10.1016/j.humimm.2007.11.001

rom other classic HLA class I molecules because of itsimited polymorphism, the restricted tissue distribution,nd the translation of alternative spliced transcripts thatncode seven distinct isoforms: four as membrane boundG1, G2, G3, and G4) and three as soluble proteins (G5,6, and G7) [1]. The selective HLA-G mRNA expression

attern in tissues suggests a tight transcriptional control

lity and Immunogenetics. Published by Elsevier Inc. All rights reserved.

opHi[

ttcrcdrtikmTcaivitdtd

imIoc[tbtwdttgs

dclilswtd[po

wwoHp

S

P

AdcCwrcpmmsdbaimB

C

Awbg

C

Tmvsme

A

Ti(

10 S. Rosado et al.

f gene expression [2], possibly attributable to the HLA-Gromoter region, which significantly differs from that of otherLA class I genes. Cytokines, such as interferons (IFNs) and

nterleukin-10 (IL-10), induced the expression of this antigen3,4].

Physiologic HLA-G expression, in nonpathologic situa-ions, is tissue restricted to extravillous cytotrophoblast [5],hymic epithelial cells [6], and cornea [7]. Under pathologiconditions, HLA-G expression has been documented in non-ejected allograft transplants [8,9], in tumors [10], in theourse of viral infection [11,12], and during inflammatoryiseases (e.g., skin and muscle inflammations, multiple scle-osis) [13–16]. Several studies point toward immunoregula-ory functions of HLA-G, especially at the maternal–fetalnterface during pregnancy. HLA-G is able to inhibit naturaliller-cell-mediated lysis [17,18] and their transendothelialigration properties [19], allogenic T-cell cytotoxicity [20],-cell proliferation [21,22], and maturation of myelomono-ytic cells into functional antigen-presenting cells [23]. Inddition to all these cellular effects, HLA-G may act as anmmunomodulatory molecule by promoting an immune de-iation from Th1 to Th2 response [24–26]. Apart from themmunotolerant role of HLA-G antigens in maternal–fetalolerance, the acceptance of grafts, and probably theissemination of tumors, it has been recently proposedhat they might play a protective role in inflammatoryiseases [27].

The expression of HLA-G is up-regulated by the anti-nflammatory cytokine IL-10 [4], which conversely down-odulates the expression of HLA class I and II molecules.

L-10 is an immunoregulatory cytokine produced by numer-us cell types (monocytes [28], T and B cells [29], Kupfferells [30], keratinocytes [31], and cytotrophoblast cells32]). This pleiotropic cytokine plays an important role inhe regulation of the immune response, mainly by the inhi-ition of proinflammatory cytokine expression [28], the al-eration of antigen presentation on T-cell activation path-ays [29,33], and the stimulation of the proliferation andifferentiation of B cells [34,35]. Both HLA-G and IL-10 seemo be involved in immune modulation and possibly in induc-ion of immune tolerance. Furthermore, it has been sug-ested that IL-10 inhibits cellular responses through expres-

ABBREVIATIONS

B-CLL B-chronic lymphocytic leukemiaELISA enzyme-linked immunosorbent assayHLA-G human leukocyte antigen-GIFN-� interferon-�IL-6 interleukin-6IL-10 interleukin-10mAbs monoclonal antibodiesPBMCs peripheral blood mononuclear cellssHLA-G soluble HLA-GSLE systemic lupus erythematosusSLEDAI disease activity index for systemic lupus ery-

thematosus patientsTNF-� tumor necrosis factor-�

ion of HLA-G [4,24]. m

Systemic lupus erythematosus (SLE) is an autoimmuneisease characterized by B-lymphocyte hyperactivity ac-ompanied by changes in the immunoglobulin repertoire,eading to an increased production of autoantibodies andmpaired cell-mediated immunity, which results from both Tymphocytes and antigen-presenting cells. Patients with SLEpontaneously produce high levels of IL-10 that correlateith parameters of disease activity [36,37]. Moreover, con-

inuous administration of neutralizing anti-IL-10 antibodieselays the onset of autoimmunity in a murine model of SLE38] and improves cutaneous lesions and joint symptoms inatients [39], suggesting a role for IL-10 in the pathophysi-logy of SLE.

Because SLE is characterized by the secretion of IL-10,hich is able to upregulate HLA-G expression, we askedhether HLA-G might be expressed in this disease. The aimf the present study was to investigate the presence ofLA-G molecules in lymphocytes, serum, and skin from SLEatients.

ubjects and methods

atients and controls

total of 50 female SLE patients from Hospital Universitario Puertae Hierro, Madrid (Spain), were included in the study. The inclusionriterion was the fulfillment of four or more of the revised Americanollege of Rheumatology classification criteria [40]. The median ageas 46 (range: 19-64). At the time of the study 14% of patients were

eceiving no treatment; 17% antimalarials (chloroquine or hydroxy-hloroquine); 35% antimalarials and low-dose steroids (�15 mgrednisone/day); 24% immunosuppressive agents (mycophenolateofetil, cyclosporine A, or azathioprine) in combination with anti-alarials or low-dose steroids; and 10% antimalarials and high-dose

teroids (�30 mg prednisone/day). SLE activity was assessed by theisease activity index for lupus patients (SLEDAI), described by Bom-ardier et al. [41]. Eighty percent of patients had active disease (�0t the SLEDAI score, median � 4, range [1–16]), whereas 20% hadnactive lupus. Fifty-nine nonselected and nonrelated healthy fe-ales without any autoimmune history and 28 female patients with-chronic lymphocytic leukemia (B-CLL) were used as controls.

ell separation

fter informed consent was obtained, heparinized whole bloodas collected from patients and healthy subjects. Peripherallood mononuclear cells (PBMCs) were then isolated by densityradient centrifugation.

ell line culture

he choriocarcinoma cell line JEG-3 was cultured in minimum essentialedium (Eagle’s) supplemented with 10% heat-inactivated fetal bo-

ine serum, 2 mM L-glutamine, and 100 U/ml penicillin, and 100 �g/mltreptomycin and was maintained at 37°C in a humidified 5% CO2 at-osphere. This line was used as a positive control of surface HLA-Gxpression.

ntibodies

he following antihuman monoclonal antibodies (mAbs) were usedn flow cytometry studies: anti-HLA-G-fluorescein isothiocyanateMEM-G/9, clone specific for the intact �2m-associated HLA-G, this

onoclonal antibody detects soluble and membrane isoforms of

HfaraCTcSIps

F

FnwpprpuSfc

I

Pfiogntreatbw

S

ScQo

(wStw

S

SCctnTSa

R

Sa

Wtawtdc[1amS

Fi

11Expression of HLA-G in SLE

LA-G without cross-reactivity with HLA-I class) (Serotec Ltd., Ox-ord, UK), anti-CD5-phycoerythrin (BD Biosciences, San Jose, CA),nd anti-CD19-peridinin-chlorophyll-protein complex (Caltag Labo-atories, Burlingame, CA). For immunohistochemistry analysis mAbsgainst the following human molecules were used: CD3, CD4, CD8,D38, and CD68 (Novocastra Laboratories, Ltd., NewCastle uponyne, UK) to determine the infiltration of PBMCs, HLA-G (MEM-G/1lone, which reacts with the denatured HLA-G free heavy chain;erotec Ltd.), and IL-10, IL-6, tumor necrosis factor (TNF)-�, andFN-� (R&D Systems, Inc., Minneapolis, MN) to analyze the cytokinerofile. In all cases an isotype-matched antibody was used underimilar conditions to control nonspecific staining.

low cytometry analysis

or surface staining cells were first incubated for 15 minutes withormal mouse serum to avoid nonspecific binding and then incubatedith mixtures of fluorescein isothiocyanate-, phycoerythrin- anderidinin-chlorophyll-protein complex-conjugated mAbs at room tem-erature for 20 minutes in the dark. The cells were then washed andesuspended in phosphate-buffered saline. A total of 30,000 lym-hocytes were acquired and analyzed by flow cytometry in a FACSortsing the CellQuest research software (BD Immunocytometry Systems,an Jose, CA). For the analysis of HLA-G, lymphocytes were gated inorward scatter/side scatter dot plots and the percentage of HLA-G�

ells in CD19� lymphocytes was analyzed.

mmunohistochemistry

araffin-embedded sections from two skin biopsies were deparaf-nized at 40°C overnight, rehydrated through graded concentrationsf ethanol, and rinsed in distilled water. Cells were cytocentrifuged onlass slides and fixed for 10 minutes in acetone at 4°C. Before immu-ohistochemistry analysis, deparaffinized tissue sections were submit-ed to epitope retrieval by citrate buffer (pH 8). Slides were thenehydrated with phosphate-buffered saline and immunostaining wasvaluated on tissue cells using the Envision kit for leukocyte antigensnd the labeled streptavidin-biotin for HLA-G and cytokines, accordingo the manufacturer’s recommendations. Both kits were suppliedy Dako Cytomation (Glostrup, Denmark). Finally, the sectionsere counterstained with hematoxylin and eosin.

igure 1. Expression of IL-10 and sHLA-G. IL-10 (A) and sH

ndividuals and SLE patients. Line represents the median of the expr

erum IL-10 and soluble HLA-G (sHLA-G) levels

erum IL-10 concentrations were measured in duplicate using theommercially available enzyme-linked immunosorbent assay (ELISA)uantikine HS (R&D Systems), according to the manufacturer’s rec-mmendations. The detection limit of the test was 0.78 pg/ml.

sHLA-G concentrations were measured using a specific ELISAsHLA-G kit; Exbio, Prague, Czech Republic). The optical densitiesere measured at 450 nm (Anthos 2010; Anthos Labtec Instruments,alzburg, Austria). The final concentration was determined by op-ical density according to the standard curves. The detection limitsere 1 U/ml.

tatistics

tatistical analysis was conducted using SPSS software (SPSS, Inc.,hicago, IL). Normality of continuous numeric data outcomes washecked with the Kolmogorov–Smirnoff test, and comparisons be-ween variables were performed by Student’s t test when data wereormally distributed or otherwise with the Mann–Whitney U test.he relationships among the continuous variables were evaluated bypearman rank correlation coefficients. Statistical significance wasssumed for p � 0.05.

esults

erum sHLA-G, relationship with IL-10 levels,nd SLEDAI

e studied only females because it has been demonstratedhat progesterone upregulates the expression of HLA-G [42],s well as IL-10 [43]. In a preceding study (unpublished data)e had quantified the serum IL-10 expression in 50 SLE pa-

ients and 15 healthy donors and, according to publishedata, we detected higher levels in patients compared withontrols (median [interquartile range (IQR)] � 1.4 pg/ml0-2.3] and 0 pg/ml [0-1.5], respectively; p � 0.01; FigureA). Also, we confirmed that these levels correlated moder-tely with the disease activity index SLEDAI (r and p Spear-an correlation coefficients, r � 0.36 and p � 0.01).

ubsequently, we measured the sHLA-G in 33 of 50 above-

(B) were measured by ELISA in serum samples from healthy

LA-G ession of both molecules. *p � 0.01, **p � 0.05.

da5hhFmlidadmdtm

Io

BSm

pBwwwtr

ES

RHmtiSttctti

FHgH

12 S. Rosado et al.

escribed SLE patients, from which serum samples werevailable, and in a wider group of healthy individuals (n �9). We observed that the levels of sHLA-G were significantlyigher in SLE samples (43.6 U/ml [23.2–150.2]) than inealthy control samples (26.8 U/ml [6.0–45.2]; p � 0.004;igure 1B). Despite the high levels in sera found for botholecules, sHLA-G levels did not correlate either with IL-10

evels (r � 0.29 and p � 0.13) or with the disease activityndex SLEDAI (r � 0.06 and p � 0.76). However, IL-10 pro-uction may be affected by several of the therapeuticgents used in SLE and it is likely the same for HLA-G. Weivided patients into different groups according to the treat-ent received and compared sHLA-G and IL-10 levels. Noifferences existed between nontreated and treated SLE pa-ients or among patients treated with different drugs (anti-alarials, steroids, or immunosuppressive agents; Table 1).

dentification of HLA-G molecules on lymphocytesf SLE patients

ecause soluble HLA-G was easily detected in the serum ofLE patients, we then examined by flow cytometry whetherembrane isoforms were also present in PBMCs from 14 SLE

Table 1 Expression of soluble HLA-G and IL-10 according to

Treatment

sHLA-G (U/ml)

No Yes

Antimalarials 23.7 (19.9–27.8)* 25 (1Steroids 25.1 (13.3–66.7) 20 (1Immunosuppressive agents 20 (11.6–33.7) 26.4 (2

* Median (IQR).

igure 2. Cell surface expression of HLA-G protein was assesLA-G� lymphocytes in normal subjects and B-CLL and SLE patieroup is presented. *p � 0.05, **p � 0.01; n � number of patien

LA-G expression in JEG-3 cells, B-CLL, SLE, and a healthy control.

atients and 20 healthy volunteers. A group of 28 female-CLL patients [44] and the choriocarcinoma cell line JEG-3ere included as positive controls. Although we detected aeak HLA-G expression on lymphocytes from SLE patients, itas significantly higher than that observed in healthy con-

rols (median [IQR] � 1.5% [0.6–1.8] and 0.3% [0.2–0.8],espectively; p � 0.02; Figure 2).

xpression of the HLA-G in skin sections fromLE patients

ecent studies indicate an unexpected expression of theLA-G proteins in skin lesions of chronic cutaneous inflam-atory diseases, such as psoriasis and atopic dermatitis. On

he other hand, the continuous administration of neutraliz-ng anti-IL10 antibodies improved the cutaneous lesions inLE, suggesting a pathophysiologic role of this cytokine inhe skin of patients. Then, if HLA-G and IL-10 are related,he presence of HLA-G is likely in these locations. First, wehecked whether the skin from SLE patients produced Th2-ype cytokines such as IL-6 and IL-10 or, in contrast, Th1-ype cytokines such as IFN-� and TNF-�. Immunohistochem-cal analysis of paraffin-embedded skin sections from two

rent treatments.

IL-10 (pg/ml)

p No Yes p

47.7) 1.0 1.9 (0–2.9) 1.2 (0–2) 0.5933.7) 0.65 2 (0–1.2) 1.1 (0–2.6) 0.3594.5) 0.49 1.7 (0.9–2.5) 0 (0–1.2) 0.05

by flow cytometry using the MEM-G/9 mAb. (A) Distribution ofThe median of the percentage of HLA-G� lymphocytes for eachalyzed. (B) The histograms illustrate one representative case of

diffe

5.5–1.4–0.9–

sednts.ts an

SiTpgtews4ant

D

Tsp

smfsribtaIa

mtHedehwtwemttdadawsuhwow

SseewfasbltptacacssaIsb

FspCtuo

13Expression of HLA-G in SLE

LE patients revealed staining for both cytokines types, butt was more intense for IL-6 and IL-10 than for IFN-� andNF-� (Figure 3). This result suggested a type-2 cytokinerofile in the skin of SLE patients. Because it has been sug-ested that HLA-G is capable of modulating cytokine secre-ion toward the Th2 profile, we next assayed the HLA-Gxpression in skin from SLE patients. The labeling of HLA-Gas detected in inflammatory infiltrate with a cytoplasmic

taining and, to a lesser extent, a membrane pattern (Figure). Although intense staining was detected with MEM-G/1ntibody in the keratinocytes from patients, it seems to beonspecific because it was also detected in healthy skin sec-ions (not shown).

iscussion

o our knowledge this is the first study describing the expres-ion of HLA-G in the surface of PBMCs, serum, and biopsy sam-

igure 3. Immunohistochemical analysis of paraffin-embeddedections of skin from two SLE patients (left and right). A and B:ositivity for Th1-type cytokines IFN-� and TNF-�, respectively.and D: labeling for Th2-type cytokines, IL-10 and IL-6, respec-

ively. The specificity of cytokines staining was confirmed by these of irrelevant isotype-matched mAbs. Original magnificationf all sections are 200�, except for *, which is 400�.

les from SLE patients. HLA-G expression has been studied in a

everal contexts. From the initial studies of HLA-G at theaternal–fetal interface during pregnancy, attention has been

ocused on the expression and possible relevance of HLA-G ineveral malignant and inflammatory diseases [13,14,45] andecently in heart, liver, and kidney transplantation [8,9]. Anmmunomodulatory role has been proposed for this moleculeecause a previous report demonstrated that HLA-G regulateshe Th1/Th2 cytokine balance in a way that may shift toward

relative Th2 dominance [24 –26]. On the other hand,L-10 enhanced HLA-G expression in human trophoblastsnd monocytes [4].

SLE is characterized by elevated IL-10 and impaired cell-ediated immunity. For this reason, in the present study we

ried to analyze whether IL-10 expression could be related toLA-G. In line with other authors [36,37], we demonstratedlevated serum IL-10 levels in SLE patients that correlate withisease activity. If HLA-G and IL-10 were related, it could bexpected to be the same for HLA-G. Although significantlyigher sHLA-G levels were detected in SLE patients comparedith controls, no correlation existed between the expression of

his molecule and both serum IL-10 and SLEDAI. However, aeak association cannot be completely excluded because sev-ral other factors, such as sex hormone levels and treatment,ay affect both serum IL-10 an HLA-G levels. Although the

reatment seemed not to affect sHLA-G expression in patients,he number of samples analyzed was too small to conclude thisefinitively. Also, hormonal levels in our cohort were unknownnd they could differentially affect along the course of theisease, as proposed by some authors [46]. On the other hand,lthough our data are comparable to that of other publishedorks in which serum samples from healthy individuals were

tudied, the real concentration of sHLA-G molecules was likelynderestimated in our patients and controls, because recentlyigher levels of sHLA-G were detected in plasma comparedith serum samples [47]. We do not know whether this meth-dologic difference could also contribute to underestimating aeak association between IL-10 and sHLA-G.

Because IL-10 has a pathophysiologic role in the skin ofLE patients, and based on previous works that demon-trated the HLA-G expression in tissues of inflammatory dis-ase [13,14,45], we analyzed whether HLA-G could be alsoxpressed in skin lesions from SLE patients. We detected aeak HLA-G expression in the infiltrative cells of the skin

rom SLE patients that also seemed to be in accordance withpredominant Th2-type response, although these results

hould be considered with caution, because of the low num-er of skin biopsies analyzed. If they could be confirmed in aarger series, it might be tempting to speculate that secre-ion of IL-10 by the activated keratinocytes in the skin of SLEatients could induce the expression of HLA-G, similar tohat described in monocytes [4], which in turn would induceshift to Th2 cytokines and IL-10 production. These pro-

esses would lead to a vicious circle that could explain themplification of the immune activity and maintenance ofutaneous lesions in SLE patients. However, the very lowtaining in the inflammatory infiltrate and the nonspecifictaining of keratinocytes obtained with HLA-G mAb, as wells the absence of correlation of serum levels of HLA-G andL-10, seem not to support this hypothesis. It also remainspeculative whether the release of soluble HLA-G into theloodstream in SLE patients might contribute to systemic

nergy by inducing apoptosis of T cells [48]. Further studies

avH

A

TMfrCt

R

[

[

[

[

[

[

[

[

[

Fpt type-

14 S. Rosado et al.

re needed to establish and understand the possible rele-ance of the presence of membrane and soluble forms ofLA-G in the course of this disease.

cknowledgments

his study was supported by Fundacion LAIR (8108) and by theinisterio de Educacion y Cultura (AP2000-2160). We are grate-

ul to Julia Jorda for providing B-CLL samples and Alberto Du-antez, Maria Josefa Menendez, Juan Antonio Vargas, Beatrizontreras-Martin, Sandra Cuni, and Isabel Martin-Larrauri forheir continuous help.

eferences

[1] Paul P, Cabestre FA, Ibrahim EC, Lefebvre S, Khalil-Daher I,Vazeux G, et al. Identification of HLA-G7 as a new splice variantof the HLA-G mRNA and expression of soluble HLA-G5, -G6, and-G7 transcripts in human transfected cells. Hum Immunol 2000;61:1138.

[2] Gobin SJ, van den Elsen PJ. Transcriptional regulation of theMHC class Ib genes HLA-E, HLA-F, and HLA-G. Hum Immunol2000;61:1102.

[3] Lefebvre S, Berrih-Aknin S, Adrian F, Moreau P, Poea S, Gou-rand L, et al. A specific interferon (IFN)-stimulated responseelement of the distal HLA-G promoter binds IFN-regulatoryfactor 1 and mediates enhancement of this nonclassical class Igene by IFN-beta. J Biol Chem 2001;276:6133.

[4] Moreau P, Adrian-Cabestre F, Menier C, Guiard V, Gourand L,Dausset J, et al. IL-10 selectively induces HLA-G expression inhuman trophoblasts and monocytes. Int Immunol 1999;11:803.

[5] Ellis SA, Sargent IL, Redman CW, McMichael AJ. Evidence for anovel HLA antigen found on human extravillous trophoblast anda choriocarcinoma cell line. Immunology 1986;59:595.

[6] Crisa L, McMaster MT, Ishii JK, Fisher SJ, Salomon DR. Identifi-cation of a thymic epithelial cell subset sharing expression ofthe class Ib HLA-G molecule with fetal trophoblasts. J Exp Med

igure 4. Immunohistochemistry of HLA-G expression in tworedominant cytoplasmic pattern and a strong reactivity in thehese tissue sections was confirmed by the use of irrelevant iso

1997;186:289.

[7] Le DM, Moreau P, Sabatier P, Legeais JM, Carosella ED. Expres-sion of HLA-G in human cornea, an immune-privileged tissue.Hum Immunol 2003;64:1039.

[8] Luque J, Torres MI, Aumente MD, Marin J, Garcia-Jurado G,Gonzalez R, et al. Soluble HLA-G in heart transplantation: theirrelationship to rejection episodes and immunosuppressivetherapy. Hum Immunol 2006;67:257.

[9] Creput C, Durrbach A, Menier C, Guettier C, Samuel D, DaussetJ, et al. Human leukocyte antigen-G (HLA-G) expression inbiliary epithelial cells is associated with allograft acceptance inliver-kidney transplantation. J Hepatol 2003;39:587.

10] Urosevic M, Dummer R. HLA-G and IL-10 expression in humancancer–different stories with the same message. Semin CancerBiol 2003;13:337.

11] Onno M, Pangault C, Le Friec G, Guilloux V, Andre P, Fauchet R.Modulation of HLA-G antigens expression by human cytomega-lovirus: specific induction in activated macrophages harboringhuman cytomegalovirus infection. J Immunol 2000;164:6426.

12] Lozano JM, Gonzalez R, Kindelan JM, Rouas-Freiss N, CaballosR, Dausset J, et al. Monocytes and T lymphocytes in HIV-1-positive patients express HLA-G molecule. AIDS 2002;16:347.

13] Aractingi S, Briand N, Le Danff C, Viguier M, Bachelez H, MichelL, et al. HLA-G and NK receptor are expressed in psoriatic skin:a possible pathway for regulating infiltrating T cells? Am JPathol 2001;159:71.

14] Khosrotehrani K, Le Danff C, Reynaud-Mendel B, Dubertret L,Carosella ED, Aractingi S. HLA-G expression in atopic dermati-tis. J Invest Dermatol 2001;117:750.

15] Wiendl H, Mitsdoerffer M, Weller M. Express and protect your-self: the potential role of HLA-G on muscle cells and in inflam-matory myopathies. Hum Immunol 2003;64:1050.

16] Fainardi E, Rizzo R, Melchiorri L, Vaghi L, Castellazzi M, MarzolaA, et al. Presence of detectable levels of soluble HLA-G mole-cules in CSF of relapsing-remitting multiple sclerosis: relation-ship with CSF soluble HLA-I and IL-10 concentrations and MRIfindings. J Neuroimmunol 2003;142:149.

17] Rouas-Freiss N, Goncalves RM, Menier C, Dausset J, CarosellaED. Direct evidence to support the role of HLA-G in protectingthe fetus from maternal uterine natural killer cytolysis. ProcNatl Acad Sci USA 1997;94:11520.

18] Rouas-Freiss N, Marchal RE, Kirszenbaum M, Dausset J,Carosella ED. The alpha1 domain of HLA-G1 and HLA-G2 inhibits

atients. Positive HLA-G staining in the cellular infiltrate with atinocytes were observed. The specificity of HLA-G staining onmatched mAb. Original magnification, 400�.

SLE pkera

cytotoxicity induced by natural killer cells: is HLA-G the public

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

15Expression of HLA-G in SLE

ligand for natural killer cell inhibitory receptors? Proc NatlAcad Sci USA 1997;94:5249.

19] Dorling A, Monk NJ, Lechler RI. HLA-G inhibits the transendo-thelial migration of human NK cells. Eur J Immunol 2000;30:586.

20] Riteau B, Rouas-Freiss N, Menier C, Paul P, Dausset J, CarosellaED. HLA-G2, -G3, and -G4 isoforms expressed as nonmature cellsurface glycoproteins inhibit NK and antigen-specific CTL cytol-ysis. J Immunol 2001;166:5018.

21] Lila N, Rouas-Freiss N, Dausset J, Carpentier A, Carosella ED.Soluble HLA-G protein secreted by allo-specific CD4� T cellssuppresses the allo-proliferative response: a CD4� T cell reg-ulatory mechanism. Proc Natl Acad Sci USA 2001;98:12150.

22] LeMaoult J, Krawice-Radanne I, Dausset J, Carosella ED. HLA-G1-expressing antigen-presenting cells induce immunosuppres-sive CD4� T cells. Proc Natl Acad Sci USA 2004;101:7064.

23] Horuzsko A, Lenfant F, Munn DH, Mellor AL. Maturation ofantigen-presenting cells is compromised in HLA-G transgenicmice. Int Immunol 2001;13:385.

24] Kanai T, Fujii T, Kozuma S, Yamashita T, Miki A, Kikuchi A, etal. Soluble HLA-G influences the release of cytokines from al-logeneic peripheral blood mononuclear cells in culture. MolHum Reprod 2001;7:195.

25] Kanai T, Fujii T, Unno N, Yamashita T, Hyodo H, Miki A, et al.Human leukocyte antigen-G-expressing cells differently modu-late the release of cytokines from mononuclear cells present inthe decidua versus peripheral blood. Am J Reprod Immunol2001;45:94.

26] Kanai T, Fujii T, Kozuma S, Miki A, Yamashita T, Hyodo H, et al.A subclass of soluble HLA-G1 modulates the release of cyto-kines from mononuclear cells present in the decidua additivelyto membrane-bound HLA-G1. J Reprod Immunol 2003;60:85.

27] Carosella ED, Moreau P, Aractingi S, Rouas-Freiss N. HLA-G: ashield against inflammatory aggression. Trends Immunol 2001;22:553.

28] de Waal MR, Abrams J, Bennett B, Figdor CG, de Vries JE.Interleukin 10(IL-10) inhibits cytokine synthesis by humanmonocytes: an autoregulatory role of IL-10 produced by mono-cytes. J Exp Med 1991;174:1209.

29] Del Prete G, De Carli M, Almerigogna F, Giudizi MG, Biagiotti R,Romagnani S. Human IL-10 is produced by both type 1 helper(Th1) and type 2 helper (Th2) T cell clones and inhibits theirantigen-specific proliferation and cytokine production. J Im-munol 1993;150:353.

30] Thompson K, Maltby J, Fallowfield J, McAulay M, Millward-Sadler H, Sheron N. Interleukin-10 expression and function inexperimental murine liver inflammation and fibrosis. Hepatol-ogy 1998;28:1597.

31] Enk AH, Katz SI. Identification and induction of keratinocyte-derived IL-10. J Immunol 1992;149:92.

32] Roth I, Corry DB, Locksley RM, Abrams JS, Litton MJ, Fisher SJ.Human placental cytotrophoblasts produce the immunosup-pressive cytokine interleukin 10. J Exp Med 1996;184:539.

33] de Waal M, Haanen J, Spits H, Roncarolo MG, te VA, Figdor C,

et al. Interleukin 10 (IL-10) and viral IL-10 strongly reduce

antigen-specific human T cell proliferation by diminishing theantigen-presenting capacity of monocytes via downregulationof class II major histocompatibility complex expression. J ExpMed 1991;174:915.

34] Mosmann TR. Properties and functions of interleukin-10. AdvImmunol 1994;56:1.

35] Itoh K, Hirohata S. The role of IL-10 in human B cell activation,proliferation, and differentiation. J Immunol 1995;154:4341.

36] Hagiwara E, Gourley MF, Lee S, Klinman DK. Disease severity inpatients with systemic lupus erythematosus correlates with anincreased ratio of interleukin-10:interferon-gamma-secretingcells in the peripheral blood. Arthritis Rheum 1996;39:379.

37] Park YB, Lee SK, Kim DS, Lee J, Lee CH, Song CH. Elevatedinterleukin-10 levels correlated with disease activity in sys-temic lupus erythematosus. Clin Exp Rheumatol 1998;16:283.

38] Ishida H, Muchamuel T, Sakaguchi S, Andrade S, Menon S,Howard M. Continuous administration of anti-interleukin 10antibodies delays onset of autoimmunity in NZB/W F1 mice. JExp Med 1994;179:305.

39] Llorente L, Richaud-Patin Y, Garcia-Padilla C, Claret E, Jakez-Ocampo J, Cardiel MH, et al. Clinical and biologic effects ofanti-interleukin-10 monoclonal antibody administration in sys-temic lupus erythematosus. Arthritis Rheum 2000;43:1790.

40] Tan EM, Cohen AS, Fries JF, Masi AT, McShane DJ, Rothfield NF,et al. The 1982 revised criteria for the classification of systemiclupus erythematosus. Arthritis Rheum 1982;25:1271.

41] Bombardier C, Gladman DD, Urowitz MB, Caron D, Chang CH.Derivation of the SLEDAI. A disease activity index for lupuspatients. The Committee on Prognosis Studies in SLE. ArthritisRheum 1992;35:630.

42] Yie SM, Xiao R, Librach CL. Progesterone regulates HLA-G geneexpression through a novel progesterone response element.Hum Reprod 2006;21:2538.

43] Verthelyi D, Klinman DM. Sex hormone levels correlate with theactivity of cytokine-secreting cells in vivo. Immunology 2000;100:384.

44] Nuckel H, Rebmann V, Durig J, Duhrsen U, Grosse-Wilde H.HLA-G expression is associated with an unfavorable outcomeand immunodeficiency in chronic lymphocytic leukemia. Blood2005;105:1694.

45] Torres MI, Le Discorde M, Lorite P, Rios A, Gassull MA, Gil A, etal. Expression of HLA-G in inflammatory bowel disease providesa potential way to distinguish between ulcerative colitis andCrohn’s disease. Int Immunol 2004;16:579.

46] Verthelyi D, Petri M, Ylamus M, Klinman DM. Disassociation ofsex hormone levels and cytokine production in SLE patients.Lupus 2001;10:352.

47] Rudstein-Svetlicky N, Loewenthal R, Horejsi V, Gazit E. HLA-Glevels in serum and plasma. Tissue Antigens 2006;67:111.

48] Fournel S, Aguerre-Girr M, Huc X, Lenfant F, Alam A, Toubert A,et al. Cutting edge: soluble HLA-G1 triggers CD95/CD95 ligand-mediated apoptosis in activated CD8� cells by interacting with

CD8. J Immunol 2000;164:6100.