Serum antibodies to human leucocyte antigen (HLA)‐E, HLA‐F and HLA‐G in patients with systemic lupus erythematosus (SLE) during disease flares: Clinical relevance of HLA‐F autoantibodies

V Jucaud; M H Ravindranath; P I Terasaki; L E Morales‐Buenrostro; F Hiepe; T Rose; R Biesen

doi:10.1111/cei.12724

. 2015 Dec 16;183(3):326–340. doi: 10.1111/cei.12724

Serum antibodies to human leucocyte antigen (HLA)‐E, HLA‐F and HLA‐G in patients with systemic lupus erythematosus (SLE) during disease flares: Clinical relevance of HLA‐F autoantibodies

V Jucaud ¹, M H Ravindranath ^1,^✉, P I Terasaki ¹, L E Morales‐Buenrostro ², F Hiepe ³, T Rose ³, R Biesen ³

PMCID: PMC4750595 PMID: 26440212

Summary

T lymphocyte hyperactivity and progressive inflammation in systemic lupus erythematosus (SLE) patients results in over‐expression of human leucocyte antigen (HLA)‐Ib on the surface of lymphocytes. These are shed into the circulation upon inflammation, and may augment production of antibodies promoting pathogenicity of the disease. The objective was to evaluate the association of HLA‐Ib (HLA‐E, HLA‐F and HLA‐G) antibodies to the disease activity of SLE. The immunoglobulin (Ig)G/IgM reactivity to HLA‐Ib and β2m in the sera of 69 German, 29 Mexican female SLE patients and 17 German female controls was measured by multiplex Luminex^®‐based flow cytometry. The values were expressed as mean florescence intensity (MFI). Only the German SLE cohort was analysed in relation to the clinical disease activity. In the controls, anti‐HLA‐G IgG predominated over other HLA‐Ib antibodies, whereas SLE patients had a preponderance of anti‐HLA‐F IgG over the other HLA‐Ib antibodies. The disease activity index, Systemic Lupus Erythematosus Disease Activity Index (SLEDAI)‐2000, was reflected only in the levels of anti‐HLA‐F IgG. Anti‐HLA‐F IgG with MFI level of 500–1999 was associated with active SLE, whereas inactive SLE revealed higher MFI (>2000). When anti‐HLA‐F IgG were cross‐reactive with other HLA‐Ib alleles, their reactivity was reflected in the levels of anti‐HLA‐E and ‐G IgG. The prevalence of HLA‐F‐monospecific antibodies in SLE patients was also associated with the clinical disease activity. Anti‐HLA‐F IgG is possibly involved in the clearance of HLA‐F shed from lymphocytes and inflamed tissues to lessen the disease's severity, and thus emerges as a beneficial immune biomarker. Therefore, anti‐HLA‐Ib IgG should be considered as a biomarker in standard SLE diagnostics.

Keywords: autoantibodies, disease activity, HLA‐F, HLA‐Ib, systemic lupus erythematosus

Introduction

Systemic lupus erythematosus (SLE) is a multi‐organ autoimmune rheumatic disease affecting mainly females 1. Autoantibodies formed against both cellular and extracellular components 2, 3, 4 are implicated in the neurological, cardiovascular, musculoskeletal, dermatological and nephrological inflammation. Immunomodulatory potential of the circulating autoantibodies, shed soluble autoantigens, their immune complexes (ICs) and deposition of the ICs on cell surfaces and endothelial linings (i.e. glomeruli) are attributed to particular inflammations such as nephritis 4, 5.

Terasaki et al. 6 showed that SLE patients generate cytotoxic antibodies against their own lymphocytes and lymphocytes of others, implicating the human leucocyte antigens (HLA) as a target for the cytotoxic antibodies. This finding provides considerable insight into SLE‐associated inflammation in context of the hyperactivity of T and B cells during the clinical progression of SLE 7. Also, inflammation promotes over‐expression of cell surface HLA 8, and antibodies against classical HLA‐Ia are found in normal, non‐alloimmunized males. However, Zhou et al. 9 showed that the existence of HLA‐Ia antibodies was not related significantly to the incidence of SLE disease activity index score or development of SLE, although the involvement of non‐classical HLA‐Ib antibodies remains to be determined.

The genes on chromosome 6 (6p21·1‐q15), coding for HLA proteins, were shown to be the candidate genes for linkage with SLE 10, 11. HLA contains a wide variety of autoantigens, and every individual has a pair each of HLA‐A, HLA‐B and HLA‐Cw (classical HLA‐Ia) alleles, which are polymorphic, and of HLA‐E, HLA‐F and HLA‐G (non‐classical HLA‐Ib) alleles, which are mono‐ or oligomorphic 12. Most importantly, over‐expression of the non‐classical HLA‐Ib alleles is associated with inflammation and may be implicated in multi‐organ inflammation of SLE, as they are capable of immunomodulation. HLA‐E molecules are over‐expressed in T and B cells natural killer (NK) cells, monocytes and macrophages 13 and can bind to immunomodulatory receptors (CD94, NKG2A, NKG2C). Upon inflammation, HLA‐E is shed in circulation as soluble (s) HLA‐E. 13, 14. HLA‐G, expressed commonly on the extravillous cytotrophoblast and the amnion, is also shed as sHLA‐G in biological fluids 15, 16, 17 at a level significantly higher in SLE patients than in the controls 18. The sHLA‐G has the potential of inhibiting NK cytolysis 19 and inducing apoptosis of activated T cells 20. HLA‐F, although found in the cytoplasm of resting T cells 15, are detected in the placenta 21, the tonsils, spleen, bladder, skin, thymus tissue and liver cell lines 22, 23. While surface expression is absent in most tissues, surface expression has been demonstrated on activated T, B and NK cells 24, 25. Such enhanced expression of HLA‐F may lead to their shedding as sHLA‐F 26.

All these findings emphasize the need for a critical examination of the consequences of over‐expression, shedding of HLA‐Ib, particularly from the perspective of HLA‐Ib antibody formation and ICs in SLE. We hypothesize that the antibodies directed against one or more of non‐classical HLA‐Ib over‐expressed on hyperactive lymphocytes of multi‐organ inflammation (as in SLE) patients can serve as an immunomarker for SLE disease activity. The present investigation tested that hypothesis by evaluating the prevalence of anti‐HLA‐Ib antibodies during the course of SLE clinical activity.

Materials and methods

Sera from SLE patients and controls

Sera of German SLE female patients were provided by Dr Robert Biesen from the Department of Rheumatology and Clinical Immunology, Charité Universitätsmedizin Berlin after approval by their Institutional Review Board. The cohort of patients analysed was composed of 69 SLE females from Germany (Supporting information, Table S1). These patients fell into five different age groups: group 1, 20–29 (n = 17); group 2, 30–39 (n = 26); group 3, 40–49 (n = 14); group 4, 50–59 (n = 8); and group 5, 60–69 (n = 4), with a mean disease duration of 10 years [standard deviation (s.d.) = 7 years]. Most of these patients were Caucasian (87%), the remainder were Asian (13%). Sera of patients who met the updated and revised criteria proposed by the American College of Rheumatology (ACR) for the diagnosis of SLE 27, 28 were included into the study. The number of ACR criteria, which reflects the severity of the disease, was determined at the time of first diagnosis. The average number of ACR criteria was 6 ± 1. Although SLE disease activity was assessed at the time of bleed by Dr Biesen and his group, evaluation of the German patients' sera for the results cited in the present report was conducted at the Terasaki Foundation Laboratory (TFL). The clinical data provided by Dr Biesen includes ACR criteria and the following subheadings of the disease activity scores: ‘British Isles Lupus Assessment Group (BILAG) Gesamtscore’ and ‘mBILAG n.YEE’, which may refer to BILAG and BILAG‐2004, respectively, cited by others 29, 30, 31, and ‘SLEDAI letzten 10 tage’ 32 and ‘mSLEDAI’ which, respectively, corresponds to SLEDAI and SLEDAI‐2000 33. Serial bleeds of five female German patients were also analysed.

We also analysed anti‐HLA‐Ib antibodies in the sera of 29 Mexican female SLE patients; the sera were provided by Dr Luis Morales‐Buenrostro from the Department of Nephrology and Mineral metabolism, National Institute of Medical Sciences and Nutrition ‘Salvador Zubirán,’ Mexico City, neither the ACR criteria nor the disease activity scores of Mexican cohort were available, nor was their racial background (Supporting information, Table S2). The Mexican SLE patients fell into six age groups: group 1, 20–20 (n = 5); group 2, 30–39 (n = 6); group 3, 40–49 (n = 7); group 4, 50–59 (n = 6); and group 5, 60–69 (n = 2) and group 6, 70–79 (n = 3).

The sera of control German females were provided by Dr Nils Lachmann, who is at the same Berlin hospital's Institute of Transfusion Medicine. All the control females were Caucasian. Of the 17 controls, 14 belonged in the 20–29 age group (group 1) and three in the 30–39 age group (group 2; Supporting information, Table S3). The controls (recognized as blood donors with no evidence of autoimmune diseases at the time of blood collection by Dr Lachmann) were used as a reference for statistical analysis. All sera were kept at −20˚C before shipment and before analyses of the sera, and none were thawed more than three times. As with the German SLE patients, the sera of the Mexican SLE patients and German controls were analysed for the results of this report at TFL.

Immunoassay with single‐HLA or β2‐microglobulin (β2m) antigen‐coated microbeads

Recombinant HLA‐E, HLA‐F, HLA‐G and β2m folded heavy chains [10 mg/ml in 2‐(N‐morpholino) ethanesulphonic acid (MES) buffer] were from the Immune Monitoring Laboratory, Fred Hutchinson Cancer Research Center (University of Washington, Seattle, WA, USA). Recombinant heavy chains of HLA‐Ib alleles (HLA‐E^R107, HLA‐F and HLA‐G1) were folded and made available for coating microbeads. It is necessary to define the amino acid sequences of HLA‐Ib used in this study; Table 1 shows the amino acid sequences of HLA‐E^R, HLA‐F, HLA‐G and β2m used for coating the beads. All HLA‐Ib alleles have only the extracellular domain without the leader peptide containing 21 amino acids and without transmembrane and intracellular domains. The C‐terminus of all the alleles ended at position 297.

Table 1.

Amino acid sequences of human leucocyte antigen (HLA)‐E^R, HLA‐F, HLA‐G1 and β2m used for coating the microbeads for monitoring the antibodies in sera

HLA‐E^R107		G S H S L K Y F H ¹⁰	T S V S R P G R G E ²⁰	P R F I S V G Y V D³⁰	D T Q F V R F D N D ⁴⁰	A A S P R M V P R A ⁵⁰	P W M E Q E G S E Y ⁶⁰
α1 * W D R E T R S A R D ⁷⁰	T A Q I F R V N L R ⁸⁰	T L R G Y Y N Q S E ⁹⁰	A G S H T L Q W M H ¹⁰⁰	G C E L G P D R R F¹¹⁰	L R G Y E Q F † A Y D ¹²⁰	G K D Y L T L N E D ¹³⁰	L R S W T A V D T A ¹⁴⁰
α2 * A Q I S E Q K S N D ¹⁵⁰	A S E A E H Q R A Y ¹⁶⁰	L E D T C V E W L H ¹⁷⁰	K Y L E K G K E T L ¹⁸⁰	L H L E P P K T H V¹⁹⁰	T H H P I S D H E A ²⁰⁰	T L R C W A L G F Y ²¹⁰	P A E I T L T W Q Q ²²⁰
D G E G H T Q D T E ²³⁰	L V E T R P A G D G ²⁴⁰	T F Q K W A A V V V ²⁵⁰	P S G E E Q R Y T C ²⁶⁰	H V Q H E G L P E P²⁸⁰	V T L R W K P
HLA‐F		G S H S L R Y F S	T A V S R P G R G E	P R Y I A V E Y V D	D T Q F L R F D S D	A A I P R M E P R E	P W V E Q E G P Q Y
α1 * W E W T T G Y A K A	N A Q T D R V A L R	N L L R R Y N Q S E	A G S H T L Q G M N	G C D M G P D G R L	L R G Y H Q H † A Y D	G K D Y I S L N E D	L R S W T A A D T V
α2 * A Q I T Q R F Y E A	E E Y A E E F R T Y	L E G E C L E L L R	R Y L E N G K E T L	Q R A D P P K A H V	A H H P I S D H E A	T L R C W A L G F Y	P A E I T L T W Q R
D G E E Q T Q D T E	L V E T R P A G D G	T F Q K W A A V V V	P S G E E Q R Y T C	H V Q H E G L P Q P	L I L R W E Q
HLA‐G1		G S H S M R Y F S	A A V S R P G R G E	P R F I A M G Y V D	D T Q F V R F D S D	S A C P R M E P R A	P W V E Q E G P E Y
α1 * W E E E T R N T K A	H A Q T D R M N L Q	T L R G Y Y N Q S E	A S S H T L Q W M I	G C D L G S D G R L	L R G Y E Q Y † A Y D	G K D Y L A L N E D	L R S W T A A D T A
α2 * A Q I S K R K C E A	A N V A E Q R R A Y	L E G T C V E W L H	R Y L E N G K E M L	Q R A D P P K T H V	T H H P V F D Y E A	T L R C W A L G F Y	P A E I I L T W Q R
D G E D Q T Q D V E	L V E T R P A G D G	T F Q K W A A V V V	P S G E E Q R Y T C	H V Q H E G L P E P	L M L R W K Q
β2m
M – I Q R T P K I Q V Y ¹⁰	S R H P A E N G K S ²⁰	N F L N C Y V S G F ³⁰	H P S D I E V D L L ⁴⁰	N G E R I E K V E H⁵⁰	S D L S F S K D W S ⁶⁰	F Y L L Y Y T E F T ⁷⁰	P T E K D E Y A C R ⁸⁰
V N H V T L S Q P K ⁹⁰	I V K W D R D M

Open in a new tab

*IgG antibodies binding to α1 or α 2 or both helices are specific for the antigen due to remarkable differences among the α 1 and α 2 amino acid sequences of HLA‐E, HLA‐F and HLA‐G.

^†IgG antibodies binding to the peptide shared by HLA‐A, HLA‐B, HLA‐Cw (classical HLA‐Ia), HLA‐E, HLA‐F and HLA‐G (non‐classical HLA‐Ib) antigens – and hidden by β2m – recognize all HLA‐Ib antigens.

To detect immunoglobulin (Ig)G/IgM reactivity to HLA‐Ib or β2m in the sera of SLE patients from both German and Mexican cohorts and controls, the multiplex Luminex^®‐based Immuno‐Assay (One Lambda, Inc., Canoga Park, CA, USA) was used 34, 35, 36. All assays were performed at TFL. The recombinant HLA‐E, HLA‐F and HLA‐G heavy chains and β2m were attached individually to differently fluorochromed 5·6‐μm polystyrene microspheres. All serum samples were diluted (1 : 10) in PBS; pH 7·2 and 20‐μl were added to the 2‐μl of antigen‐coated microbeads. Using dual‐laser flow cytometry (Luminex xMap multiplex Technology), single antigen bead assays were performed for analysis of anti‐HLA‐Ib and anti‐β2m antibodies. For HLA‐E, HLA‐F, HLA‐G and β2m, positive (coated with IgG) and negative (coated with human or bovine albumin) were added separately. IgG screening was performed using secondary anti‐human IgG (Fab fragment that recognizes heavy and light chains) (One Lambda, Canoga Park, CA, USA; Cat. no. LS‐AB2; concentration: 0.5 mg/ml); IgM screening was performed using secondary anti‐human IgM (Fc5μ fragment‐specific) (Jackson ImmunoResearch Laboratories, Inc., West Grove, PA, USA; Cat. no. 709‐116‐073; concentration: 0·5 mg/ml); both secondary antibodies were used at a dilution of 1 : 100. Data generated were analysed as reported elsewhere 36, 37, 38, 39. For each analysis, at least 100 beads were counted. Mean and s.d. of mean fluorescence intensity (MFI) were recorded for each allele.

Antigen density on the single antigen beads coated with HLA‐E, ‐F and ‐G molecules was assessed by using HLA‐Ib reactive polyreactive TFL‐006 and TFL‐007 mAbs on the single antigen beads coated with HLA‐E, ‐F and ‐G and checked to determine if the MFI was comparable; the epitope targeted by TFL‐006 and TFL‐007 is found in the heavy chains of HLA‐E,‐F and ‐G coated on the beads. The data were published elsewhere (see Fig. 7d in reference 36). Conventionally, the MFI of HLA antibodies is normalized against the negative control. However, due to the high prevalence of anti‐albumin IgG in SLE patients – and as their levels were significantly higher than those in the sera of normal controls (P ² < 0·0001) (data not shown) – the MFI of HLA autoantibodies were not normalized.

The similarities and dissimilarities in the amino acid sequences of the heavy chains of HLA‐E, HLA‐F, HLA‐G and β2m are critical for this study (Table 1). The binding domains of the antibodies can be categorized as ‘commonly shared epitope region’ (e.g. ¹¹⁷AYDGKDY^{123 126}LNEDLRSWTA¹³⁵) 34, 35, 37, 38, 39 and as ‘specific epitope region’ on the heavy chain (α1 and α2 helices), which differ among the different HLA‐Ib antigens. The peptides specific for HLA‐E are α1: ⁶²DRETRSARDTAQIFRVNLRTLRG⁸⁴; α2: ¹⁴⁴SEQKSNDASEAEHQRAYLEDT¹⁶⁴, those specific for HLA‐F are α1: ⁶²EWTTGYAKANAQTDRVALRNLLR⁸⁴; α2: ¹⁴⁴TQRFYEAEEYAEEFRTYLEGE¹⁶⁴ and peptides specific for HLA‐G are α1: ⁶²EEETRNTKAHAQTDRMNLQTLRG⁸⁴; α2: ¹⁴⁴SKRKCEAANVAEQRRAYLEGT¹⁶⁴ (Table 2). The data may include autoantibodies recognizing both specific and shared peptides. If serum antibodies bind to shared epitopes on the HLA antigens coated on the beads, then reactivity can be observed for all three of the HLA‐Ib alleles. If antibodies bind to specific epitopes on the HLA‐coated on the beads, then the autoantibody reactivity is restricted to a particular HLA. The predominant antibodies are those reacting only to the specific domains of that particular HLA. Both IgG and IgM antibodies to HLA‐E, HLA‐F and HLA‐G were measured as MFI. If the MFI of any one of these HLA antibodies in an individual was higher than that of the antibodies reacting to other two alleles by 10%, that antibody was considered to be dominant.

Table 2.

Amino acid sequences of human leucocyte antigen (HLA)‐Ib: shared and specific immunogenic epitopes.

	Epitopes	HLA alleles (% expressing epitope)
	Amino acid sequence	HLA‐A	HLA‐B	HLA‐Cw	HLA‐E	HLA‐F	HLA‐G
Shared	¹¹⁷AYDGKDY¹²³	259 (100%)	831 (100%)	277 (100%)	9 (100%)	1 (100%)	9 (100%)
Shared	¹²⁶LNEDLRSWTA¹³⁵	259 (51%)	831 (100%)	277 (99.6%)	9 (100%)	1 (100%)	9 (100%)
Specific	α1: ⁶²DRETRSARDTAQIFRVNLRTLRG⁸⁴	0	0	0	9 (100%)	0	0
	α2: ¹⁴⁴SEQKSNDASEAEHQRAYLEDT¹⁶⁴	0	0	0	9 (100%)	0	0
	α1: ⁶²EWTTGYAKANAQTDRVALRNLLR⁸⁴	0	0	0	0	1 (100%)	0
	α2: ¹⁴⁴TQRFYEAEEYAEEFRTYLEGE¹⁶⁴	0	0	0	0	1 (100%)	0
	α1: ⁶²EEETRNTKAHAQTDRMNLQTLRG⁸⁴	0	0	0	0	0	9 (100%)
	α2: ¹⁴⁴SKRKCEAANVAEQRRAYLEGT¹⁶⁴	0	0	0	0	0	9 (100%)

Open in a new tab

Source of alleles and proteins: http://hla.alleles.org/nomenclature/stats.html; Anthony Nolan Research Institute Updated: 14‐01‐2014.

MFI cut‐off points and the grouping of patients based on MFI

MFI is the measure used in Luminex Flow cytometry. The MFI is derived after counting >100 single antigen‐coated microbeads. This technology is used extensively for monitoring HLA antibodies post‐transplantation. Following the manufacturer's (One Lambda Inc.) guidelines, most investigators have used the neat or serum diluted 1 : 3 on 5 µl of beads with a cut‐off of 500 MFI for a weak positive level of antibodies. However, in this investigation we have increased the dilution of sera to 1 : 10 and minimized the bead volume to 2/5th using the same cut‐off (MFI > 500) to signify a positive reaction. Basically, SLE patients were divided into four groups based on the MFI of serum anti‐HLA‐Ib antibody: group 1 comprised those with MFI < 500, indicating no antibodies or immune‐complexed antibodies; group 2, those with low antibody levels, MFI 500‐999; group 3, with medium antibody levels, MFI 1000–1999; and group 4, with high antibody levels, MFI > 2000.

Statistical analysis

All statistical analysis were performed using stata version 11. All data were tested for normality using the skewness/kurtosis test, Shapiro–Wilk W test and Shapiro–Francia W test. Analysis for significance was performed using the Mann–Whitney U, Kruskal–Wallis and Spearman's rank tests. P‐values less than 0·05 were considered significant. All values are expressed as median ± s.d. T‐tests were used to assess the level of significance (with two‐tailed P) among different age groups.

Results

Prevalence of HLA‐Ib autoantibodies

Figure 1 shows the profiles of both IgM and IgG antibodies in the patients directed against the heavy chains of HLA‐E, ‐F and –G and their light chain β2m. IgG antibodies directed against HLA‐E, ‐F and ‐G were significantly higher in SLE patients (P ² < 0·0002) than in the controls. The median MFI in the sera from controls for anti‐HLA‐E, anti‐HLA‐F and anti‐HLA‐G were, respectively, 291, 482 and 528. The median MFI of anti‐HLA‐F IgG (1179 ± 823) in SLE patients was notably higher than that of their HLA‐E (941 ± 1478) or HLA‐G (889 ± 910), indicating that the prevalence of anti‐HLA‐F IgG in SLE patients may be due to the binding of antibodies to HLA‐F‐specific epitopes. Both the profiles of IgM and IgG directed against β2m were significantly different in SLE patients compared with the controls. No such differences were observed for IgM directed against HLA‐E, ‐F or ‐G.

Levels [mean fluorescence intensity (MFI)] of immunoglobulin (Ig)G and IgM autoantibodies directed against human leucocyte antigen (HLA)‐E, HLA‐F, HLA‐G and β2m in German systemic lupus erythematosus (SLE) females (n = 69) compared to control females (n = 17).

Figure 2 shows the incidence of dominance of HLA‐E or HLA‐F or HLA‐G antibodies (MFI > 500) among control (German) females and among the two cohorts (German and Mexican) of SLE patients. In the controls (five of the 17 had anti‐HLA‐Ib antibodies with an MFI < 500, hence were not included in the figure), the anti‐HLA‐G IgG autoantibodies predominated (50%) over anti‐HLA‐E (25%) and anti‐HLA‐F (25%) autoantibodies. Such predominance of anti‐HLA‐G IgG in control females may be due to the binding of antibodies to HLA‐G‐specific epitopes. The predominance of anti‐HLA‐F IgG is notably higher – 52% for the German patients and 83% for the Mexicans – than for anti‐HLA‐E or anti‐HLA‐G, indicative of the binding of antibodies to HLA‐F‐specific epitopes.

Predominance of anti‐human leucocyte antigen (HLA)‐Ib autoantibodies in control females (German) and in systemic lupus erythematosus (SLE) female patients in Germany and Mexico. Of the 17 controls studied, five had anti‐HLA‐Ib antibodies with a mean fluorescence intensity (MFI) < 500, hence were not included in the figure. In 50% of the controls, anti‐HLA‐G immunoglobulin (Ig)G predominates over anti‐HLA‐E and anti‐HLA‐F IgG. In the German SLE patients, anti‐HLA‐F IgG predominates over anti‐HLA‐E anti‐HLA‐G IgG in 52%. In the Mexican SLE cohort, anti‐HLA‐F IgG predominates over anti‐HLA‐E and anti‐HLA‐G IgG in 83%. In both SLE cohorts, anti‐HLA‐F IgG autoantibodies may co‐dominate with other anti‐HLA‐Ib antibodies.

Table 3 shows age‐related profiles of HLA‐Ib and β2m IgG and IgM antibodies in the two age groups of controls and five age groups of German SLE patients. Group 1 of SLE patients (age 20–29) showed significantly higher values for all anti‐HLA‐Ib and β2m IgG antibodies than the group 1 controls. Anti‐HLA‐E IgG for SLE patients showed extremely high values and great variability1. Comparing the statistical trend with other SLE age groups, only anti‐HLA‐E IgG – but no other anti‐HLA‐Ib or β2m IgG antibodies – showed an age‐related declining trend. The MFI of anti‐HLA‐E IgG is significantly low in the age groups above the age of 50 (group 4, 50–59) and group 5, > 60), despite the very small sample size. Another interesting finding that emerges from the age‐related comparison of the antibodies is that the IgM formed against all three HLA‐Ib and β2m showed a significant declining trend with ageing. The IgG and IgM antibodies formed against HLA‐Ib and β2m in the Mexican cohort could not be examined for age relationship due to the cohort's smaller sample size and to a s.d. (or 2s.d.) exceeding the mean value of each and every antibody in every age group. Hence, the only purpose that examination of this cohort served was to evaluate the relative prevalence of anti‐HLA‐Ib antibodies in the population, as shown in Fig. 2.

Table 3.

Age group‐dependent changes in the levels [mean fluorescence intensity (MFI)] of immunoglobulin (Ig)G and IgM autoantibodies directed against human leucocyte antigen (HLA)‐E, HLA‐F, HLA‐G and β2m in German Caucasian control females and German female systemic lupus erythematosus (SLE) patients.

Control			Anti‐HLA‐E		Anti‐HLA‐F		Anti‐HLA‐G		Anti‐β2M
Control	Age Categories		IgG	IgM	IgG	IgM	IgG	IgM	IgG	IgM
Group 1	20–29	Mean	337	727	537	430	570	264	278	786
	n = 14	s.d.	109	453	187	274	185	142	226	371

Group 2	30–39	Mean	225	794	393	540	399	354	167	880
	n = 3	s.d.	58	867	260	624	314	327	105	524
[1 vs 2]		P	0.05	n.s.	n.s.	n.s.	n.s.	n.s.	n.s.	n.s.

SLE patients
Group 1	20–29	Mean	1565	1064	1221	627	837	382	427	894
	n = 17	s.d.	2599	1177	490	601	373	334	132	674
(Control 1 versus SLE 1)		P	n.s.	n.s.	0.00003	n.s.	0.01	n.s.	0.02	n.s.

Group 2	30–39	Mean	1409	1189	1285	856	925	523	432	805
	n = 26	s.d.	867	1246	644	1025	414	573	331	773
(Control 2 versus SLE 2)		P	8.50E‐07	n.s.	0.004	n.s.	n.s.	n.s.	0.02	n.s.
(1 versus 2)			n.s.	n.s.	n.s.	n.s.	n.s.	n.s.	n.s.	n.s.

Group 3	40–49	Mean	1625	788	1806	556	1566	307	825	632
	n = 14	s.d.	1167	765	1321	412	1827	232	1337	1075

(1 versus 3)			n.s.	n.s.	n.s.	n.s.	n.s.	n.s.	n.s.	n.s.
(2 versus 3)		P	n.s.	n.s.	n.s.	n.s.	n.s.	n.s.	n.s.	n.s.
Group 4	50–59	Mean	827	439	1381	272	888	170	422	287
	n = 8	s.d.	290	320	899	154	335	165	166	198
(1 versus 4)		P	*0.015*	n.s.	n.s.	*0.035*	n.s.	*0.037*	n.s.	*0.003*
(2 versus 4)			*0.005*	*0.008*		*0.008*		*0.007*		*0.004*
(3 versus 4)			*0.027*	n.s.		*0.033*		n.s.		n.s.

Group 5	60–69	Mean	848	1076	1363	1780	1087	1724	713	1146
	n = 4	s.d.	92	815	299	2072	421	2808	573	1873
(1 versus 5)		P	n.s.	n.s.	n.s.	n.s.	n.s.	n.s.	n.s.	n.s.
(2 versus 5)			*0.003*
(3 versus 5)			*0.028*
(4 versus 5)			n.s.

Open in a new tab

ACR criteria and anti‐HLA‐Ib autoantibodies

American College of Rheumatology (ACR) Criteria for the Classification of Systemic Lupus Erythematosus is based on 11 criteria2 for SLE: malar rash (such as erythema); discoid rash (such as erythematous patches); photosensitivity, oral or nasopharyngeal ulcers; non‐erosive arthritis at peripheral joints; evidence of pleural or pericardial effusion; renal disorders; neurological disorders such as seizures and psychosis; haematological disorders such as haemolytic anaemia, leucopenia, lymphopenia, thrombocytopenia; immunological disorders such as antibodies to DNA, nuclear proteins, phospholipids–cardiolipin and positive anti‐nuclear antibody. For the purpose of identifying patients in clinical studies, a person is defined as having SLE if any four or more of the 11 criteria are present, serially or simultaneously, during any period of observation. As ACR criteria of controls are considered as a reference for comparison purposes, the control level is compared with every level of ACR criteria of SLE. Table 4 presents the antibody levels to HLA‐Ib alleles and β2m in SLE patients in relation to the number of ACR criteria they met (range = 4–9). First, we compared control values with different numbers of ACR criteria (4–9) to those of the controls using the Mann–Whitney U‐test. The IgG antibodies directed against albumin, β2m, HLA‐E, HLA‐F and HLA‐G were significantly higher in every ACR group of SLE patients compared with the controls. Secondly, we compared the levels of anti‐HLA‐Ib antibodies among different numbers of ACR criteria, using Kruskal–Wallis and Spearman's rank tests. The results showed that in SLE patients both IgG and IgM antibodies against all HLA‐Ib – including HLA‐F – are independent of the ACR criteria. In all, therefore, the anti‐HLA‐Ib antibody levels do not correlate with the severity (number of symptoms listed above) of the disease defined by ACR criteria.

Table 4.

Levels (MFI) of IgG and IgM autoantibodies directed against HLA‐E, HLA‐F, HLA‐G and β2m in German systemic lupus erythematosus (SLE) females, control females and German SLE females grouped based on the number of American College of Rheumatology (ACR) criteria [median ± standard deviation (s.d.)] with two‐tailed P‐value) compared with those of control females; n.s.; not significant.

	IgG				IgM
	α‐HLA‐E	α‐HLA‐F	α‐HLA‐G	α‐β2m	α‐HLA‐E	α‐HLA‐F	α‐HLA‐G	α‐β2m
Controls (n = 17)	291 ± 109	482 ± 200	528 ± 212	201 ± 212	585 ± 511	387 ± 334	215 ± 176	751 ± 384
ACR‐4 (n = 6)	1003 ± 830	1285 ± 935	1154 ± 300	359 ± 113	721 ± 1527	746 ± 1926	347 ± 967	301 ± 968
ACR‐4 (n = 6)	<0.0005	<0.0005	<0.0005	<0.02	n.s.	n.s.	n.s.	<0.05
ACR‐5 (n = 18)	901 ± 2495	1096 ± 497	689 ± 327	338 ± 316	360 ± 570	379 ± 1104	198 ± 1350	392 ± 891
ACR‐5 (n = 18)	<0.0001	<0.0001	<0.04	<0.02	n.s.	n.s.	n.s.	<0.02
ACR‐6 (n = 14)	1409 ± 678	1285 ± 901	1049 ± 513	401 ± 220	686 ± 1514	309 ± 548	285 ± 301	578 ± 391
ACR‐6 (n = 14)	<0.0001	<0.0001	<0.001	<0.002	n.s.	n.s.	n.s.	n.s.
ACR‐7 (n = 14)	1011 ± 1252	1205 ± 1213	939 ± 1823	432 ± 1360	1073 ± 741	780 ± 439	423 ± 439	636 ± 695
ACR‐7 (n = 14)	<0.0001	<0.0001	<0.004	<0.0007	n.s.	>0.02	n.s.	n.s.
ACR‐8 (n = 13)	591 ± 503	964 ± 441	676 ± 297	378 ± 148	429 ± 951	372 ± 374	270 ± 227	437 ± 1285
ACR‐8 (n = 13)	<0.0001	<0.0002	<0.03	<0.003	n.s.	n.s.	n.s.	n.s.
ACR‐9 (n = 4)	974 ± 1039	1638 ± 636	1388 ± 821	469 ± 160	1114 ± 1387	563 ± 463	266 ± 202	649 ± 979
ACR‐9 (n = 4)	<0.003	<0.003	<0.02	<0.02	n.s.	n.s.	n.s.	n.s.
Kruskal–Wallis test	n.s.	n.s.	<0.04	n.s.	n.s.	n.s.	n.s.	n.s.
Spearman's rank test	n.s.	n.s.	n.s.	n.s.	n.s.	n.s.	n.s.	n.s.

Open in a new tab

Clinical disease activity and anti‐HLA‐Ib autoantibodies

Table 4 shows the MFI of anti‐HLA‐Ib autoantibodies in relation to disease activity indices. The patients were in groups based on anti‐HLA‐Ib antibody levels (MFI). The group 4 antibody level of anti‐HLA‐Ib (MFI > 2000) was used as reference to compare the median disease activity with other antibody level groups (groups 1, 2 and 3) using the Mann–Whitney U‐test. The median scores of SLEDAI and SLEDAI‐2000 scores were significantly higher only for anti‐HLA‐F IgG in group 3 (respectively, P ² < 0·05, and P ² < 0·03). The SLEDAI and SLEDAI‐2000 median score of anti‐HLA‐F IgG for group 4 were 3 and 0, respectively, and were 6 and 4, respectively, for group 3. No such differences were observed for other anti‐HLA‐Ib IgG and IgM or for β2m.

As the median score of SLEDAI‐2000 was significant, Fig. 3 is generated using two different disease activity thresholds (≤ 4 or ≤ 6) of SLEDAI‐2000 in relation to different levels of anti‐HLA‐F IgG; the percentage of 69 patients with a score > 4 and > 6 is important, because these two scores have been used to mark the threshold of active SLE 40. A striking difference in disease activity related to the MFI levels of anti‐HLA‐F IgG was observed. SLE was inactive when the MFI of anti‐HLA‐F IgG was > 2000 (inactive in 92% of the patients if SLEDAI‐2000 was ≤ 4 and in 100% if SLEDAI‐2000 was ≤ 6). Conversely, the disease was active in 38% of patients with an anti‐HLA‐F IgG MFI 500–999. Moreover, when the SLEDAI‐2000 cut‐off was 6, the incidence of patients with active disease was higher (33%) for MFI 500‐999 and non‐existent (0%) for MFI > 2000. None of the patients had an anti‐HLA‐F MFI < 500 (Table 4), while 100% of the patients with an anti‐HLA‐F IgG MFI > 2000 (Fig. 3) had an SLEDAI‐2000 score ≤ 6. As no correlation between anti‐HLA‐E or anti‐HLA‐G or of anti‐β2m IgG or IgM was observed, the data are not presented.

A different threshold of disease activity [Systemic Lupus Erythematosus Disease Activity Index (SLEDAI)−2000 ≤ 4 or ≤ 6] in relation to the levels of anti‐HLA‐F IgG [mean fluorescence intensity (MFI): 500–1999 *versus* > 2000] in German systemic lupus erythematosus (SLE) females. The percentage of patients with active SLE (SLEDAI‐2000 > 4 or > 6) increases as the MFI level of anti‐human leucocyte antigen (HLA)‐F IgG decreases. Eight per cent of the patients with high levels of anti‐HLA‐F IgG (MFI > 2000) are active (SLEDAI‐2000 > 4) compared with 38% of the patients with low levels of anti‐HLA‐F IgG (range = 500–999). Similarly, 0% of the patients with high levels of anti‐HLA‐F IgG (MFI > 2000) are active (SLEDAI‐2000 > 6) compared with 33% of the patients with low levels of anti‐HLA‐F IgG (range = 500–999). The patients with inactive disease (SLEDAI‐2000 ≤ 4 or ≤ 6) who have low levels of anti‐HLA‐F IgG (range = 500–999) may be progressing towards a flare episode, and those inactive patients (SLEDAI‐2000 ≤ 4 or ≤ 6) with medium levels of anti‐HLA‐F IgG (range = 1000–1999) may be progressing away from a flare episode.

Anti‐HLA‐F IgG autoantibodies during the longitudinal course of clinical disease activity

Figure 4a compares the MFI of anti‐HLA‐F IgG profiles of five patients during the course of approximately 8 months with the corresponding trend of fluctuations in the scores of clinical disease activity (SLEDAI, which essentially followed the pattern of SLEDAI‐2000). Although there were individual variations, it is evident from the record of patient (no. 1) that when the MFI of anti‐HLA‐F IgG was > 2000 the SLEDAI score was < 4. With another patient (no. 20), when the SLEDAI score fluctuated between 14 and 20, the MFI of anti‐HLA‐F IgG remained < 1000. In other patients (nos 14, 15, 20 and 28), whenever the SLEDAI was low the level of anti‐HLA‐F IgG tended to increase. These observations indicate that when the anti‐HLA‐F IgG level declines the incidence of SLE disease activity increases. Figure 4b shows that the predominance of anti‐HLA‐F‐specific IgG antibodies is greater when patients' SLE is inactive than when it is active (67 versus 50%; marked dark blue in the vertical bars below the active and inactive SLE).

(a) The mean fluorescence intensity (MFI) of anti‐human leucocyte antigen (HLA)‐F immunoglobulin (Ig)G with the corresponding trend of fluctuations of the scores of clinical disease activity [Systemic Lupus Erythematosus Disease Activity Index (SLEDAI)], which essentially followed the pattern of SLEDAI‐2000) in five patients over the course of approximately 8 months. Patient 1 has high levels of anti‐HLA‐F IgG (MFI > 2000) and a low disease activity score. In contrast, patient 20 has low levels of anti‐HLA‐F IgG (range = 500–999) and high disease activity score. Patients 14, 15 and 28 are in transition and could be moving towards or away from a flare episode. (b). The events that may occur in systemic lupus erythematosus (SLE) patients during a flare episode based on anti‐HLA‐F IgG levels (MFI) and disease activity score during the course of the disease. Beginning of a flare episode is characterized by: anti‐HLA‐F IgG MFI levels ranging between 500 and 999; sharp increase of disease activity score; and high prevalence of anti‐HLA‐F IgG compared with controls. The flare episode is characterized by: anti‐HLA‐F IgG MFI levels ranging between 1000 and 1999; stabilization of disease activity score above active threshold; and a prevalence of anti‐HLA‐F IgG comparable to the beginning of the flare. Inactive SLE is characterized by: anti‐HLA‐F IgG MFI levels above 2000; steady decrease of disease activity score below the active threshold; and increased prevalence of anti‐HLA‐F IgG compared with its prevalence during the flare episode.

Discussion

Association of anti‐HLA‐F IgG autoantibodies with SLE disease activity, but not disease severity

This is the first report documenting the prevalence of anti‐HLA‐Ib autoantibodies in SLE females. The IgG antibodies against HLA‐E and the IgM formed against all three HLA‐Ib (HLA‐E/‐F/‐G) and β2m showed a statistically significant declining trend with ageing. It is known that fewer antibodies are produced in response to antigens during ageing, possibly contributing to the increased incidence of infection, inflammation and autoimmune diseases. This age‐related declining trend is also interesting in view of the role of sex hormones on immune response and age‐related onset of SLE. A detailed comparison of the hormonal status of patients and controls (menopause and hormone replacement therapy) with the levels of anti‐HLA‐E IgG and IgM formed against HLA‐Ib and β2m deserves attention. Due to the paucity of this information about the German SLE patients, this examination was not possible in the present study. We should note, however, that no such age‐related trend was observed with reference of anti‐HLA‐F IgG, which this study documents as a major factor in SLE. Moreover, our study emphasizes the need to investigate all antibodies often implicated in the disease status of SLE (such as anti‐dsDNA and anti‐phospholipid antibodies) in relation to age and hormonal status of the SLE patients.

Most strikingly, the MFI level of anti‐HLA‐F IgG was detectable (MFI > 500) in all SLE patients (Tables 4 and 5) in contrast to the levels of anti‐HLA‐E and anti‐HLA‐G IgG in the same SLE patients, and also in contrast to the levels of anti‐HLA‐Ib IgG in controls. This prevalence of anti‐HLA‐F IgG compared to anti‐HLA‐F IgM and other anti‐HLA‐Ib IgG and IgM antibodies, is a unique characteristic of female SLE patients.

Table 5.

The clinical systemic lupus erythematosus (SLE) activity scores in relation to different levels [mean fluorescence intensity (MFI)] of immunoglobulin (Ig)G and IgM autoantibodies directed against human leucocyte antigen (HLA)‐E, HLA‐F, HLA‐G and β2m.

	Group 1 (MFI<499)			Group 2 (500<MFI<999)			Group 3 (1000<MFI<1999)			Group 4 (2000<MFI)			Kruskal‐Wallis test
α‐HLA‐E IgG			M–W test			M–W test			M–W test			M‐W Test	Kruskal‐Wallis test
SLEDAI	(n = 8)	12 ± 7	n.s.	(n = 29)	4 ± 5	n.s.	(n = 19)	4 ± 5	n.s.	(n = 13)	4 ± 3	Ref.	n.s.
SLEDAI‐2000		9 ± 6	n.s.		2 ± 4	n.s.		0 ± 4	n.s.		2 ± 3		n.s.
BILAG		6 ± 5	n.s.		3 ± 5	n.s.		4 ± 5	n.s.		5 ± 3		n.s.
BILAG‐2004		11 ± 9	n.s.		3 ± 10	n.s.		4 ± 8	n.s.		10 ± 8		n.s.
α‐HLA‐F IgG	Group 1 does not occur					M–W test			M–W test			M‐W Test	Kruskal‐Wallis test
SLEDAI				(n = 24)	4 ± 7	n.s.	(n = 33)	6 ± 4	<0.05	(n = 12)	3 ± 2	Ref.	n.s.
SLEDAI‐2000					3 ± 6	n.s.		4 ± 4	<0.03		0 ± 2		n.s.
BILAG					4 ± 6	n.s.		5 ± 5	n.s.		2 ± 4		n.s.
BILAG‐2004					9 ± 10	n.s.		10 ± 10	n.s.		2 ± 6		n.s.
α‐HLA‐G IgG			M–W test			M–W test			M–W test			M‐W Test	Kruskal‐Wallis test
SLEDAI	(n = 6)	4 ± 4	n.s.	(n = 36)	4 ± 6	n.s.	(n = 24)	4 ± 5	n.s.	(n = 3)	4 ± 2	Ref.	n.s.
SLEDAI‐2000		2 ± 3	n.s.		2 ± 5	n.s.		1 ± 4	n.s.		2 ± 2		n.s.
BILAG		3 ± 3	n.s.		5 ± 5	n.s.		4 ± 4	n.s.		5 ± 4		n.s.
BILAG‐2004		6 ± 7	n.s.		10 ± 10	n.s.		7 ± 9	n.s.		10 ± 6		n.s.
α‐β2m IgG			M–W test			M–W test			M–W test			M‐W Test	Kruskal‐Wallis test
SLEDAI	(n = 47)	4 ± 5	n.s.	(n = 19)	4 ± 5	n.s.	(n = 2)	6 ± 8	Ref.	(n = 1)	2 NA	NA	n.s.
SLEDAI‐2000		2 ± 5	n.s.		2 ± 4	n.s.		5 ± 7			0 NA		n.s.
BILAG		4 ± 4	n.s.		5 ± 5	n.s.		9 ± 11			2 NA		n.s.
BILAG‐2004		8 ± 9	n.s.		10 ± 9	n.s.		13 ± 17			2 NA		n.s.

Open in a new tab

SLEDAI = ; BILAG = British Isles Lupus Assessment Group.

Most importantly, anti‐HLA‐F IgG MFI levels had a significant association with disease activity (SLEDAI and SLEDAI‐2000) in the SLE patients (Table 4), in contradistinction to anti‐HLA‐F IgM as well as anti‐HLA‐E and ‐G IgG and IgM, which showed no association with disease activity. A high level (MFI > 2000) of anti‐HLA‐F IgG is associated with low disease activity, thus pointing out the functional relevance of anti‐HLA‐F IgG during flare episodes in SLE. Therefore, anti‐HLA‐F IgG may reflect the state of inflammation, which is comparable to the disease activity of SLE. Conversely, levels of anti‐HLA‐F IgG were low with disease severity (number of ACR criteria), as were those of other anti‐HLA‐Ib antibodies, thus emphasizing the association of anti‐HLA‐F IgG with the immune status of SLE patients, rather than the symptomatic manifestation or damage linked with SLE.

It is known that the hyperactive immune responses to autoantigens are often associated with flare episodes, increasing the disease activity score and the inflammation of target organs in patients 11. Earlier investigations of SLE have focused on organ‐specific markers of disease activity, pathogenic antibodies or genetic linkage to SLE (HLA genes in particular) 11. In contrast, this investigation focuses on the overall inflammation associated with hyperactive immune cells. HLA‐I genes offer a wide variety of autoantigens expressed on immune cells, which are shed into circulation post‐inflammation 13, 14, 26. Shedding of antigens is known to expose cryptic epitopes upon dissociation from β2m 41 and eliciting an immune response leading to antibody production.

Significance of the predominance of anti‐HLA‐F IgG in SLE

Autoantibodies to HLA are produced against shed HLA heavy chain polypeptides and β2m. The shed HLA heavy chains can produce antibodies directed against either unique amino acid sequences (called ‘monospecific HLA antibodies’) or amino acid sequences shared by several HLA alleles (called ‘polyspecific HLA antibodies’). The shared amino acid sequences (¹¹⁷AYDGKDY^{123 126}LNEDLRSWTA¹³⁵, Table 2) are cryptic in an intact HLA. Shedding of the HLA exposes the cryptic shared epitopes and can elicit the production of antibodies. Poly‐HLA‐reactive antibodies can bind to both HLA‐Ib and HLA‐Ia alleles, whereas monospecific HLA antibodies can bind to unique amino acid sequences of the α1 and α2 helices restricted to, or specific for, HLA‐F. If anti‐HLA‐F IgG were formed against shared epitopes, then a higher or equal MFI would be observed for all HLA‐Ib molecules.

In the controls, anti‐HLA‐G IgG antibodies predominated (50%) over anti‐HLA‐E (25%) and HLA‐F (25%) antibodies, while in both the German and Mexican SLE cohorts, anti‐HLA‐F autoantibodies predominated over other HLA‐Ib autoantibodies, anti‐HLA‐F being, respectively, 52 and 83% (Fig. 2). The striking predominance of anti‐HLA‐F IgG antibodies over other anti‐HLA‐Ib antibodies in SLE suggests that the anti‐HLA‐F IgG produced are monospecific. In addition, the proportion of monospecific anti‐HLA‐F IgG antibodies increases with decrease in the disease activity (Fig. 4b), due possibly to clearance of shed sHLA‐F, which may increase their level during the disease progresses and during flares.

All these observations demonstrate the immunogenicity of HLA‐F in patients with SLE. As HLA‐F is a marker of activated lymphocytes and monocytes 24, 42 and shed into circulation, the anti‐HLA‐F autoantibodies could serve as a potential and novel immunomarker for the hyperactivity of immune cells, thus reflecting the disease activity of SLE.

Are anti‐HLA‐F IgG autoantibodies protective?

The association between anti‐HLA‐F IgG autoantibodies and the course of SLE clinical activity is extrapolated from the longitudinal profiles (approximately 8 months) of five different SLE patients, numbers 1, 14, 15, 20 and 28. This model represents the longitudinal profiles of two parameters: anti‐HLA‐F IgG MFI levels and disease activity score during the course of the disease (Fig. 4a,b). When SLE patients had a MFI of anti‐HLA‐F IgG ranging between 500 and 1999, they experienced a flare, i.e. having a SLEDAI score > 4 40. In contrast, when the MFI of anti‐HLA‐F IgG was > 2000, SLE was inactive, with SLEDAI score ≤ 4. These observations clearly document an inverse relationship between SLE disease activity and anti‐HLA‐F IgG, and it is the first line of evidence suggesting that anti‐HLA‐F antibodies are protective towards flares of SLE.

Strikingly, only 8% of the patients in this study's cohorts who had high MFI of anti‐HLA‐F IgG (>2000) experienced active SLE (SLEDAI‐2000 > 4), whereas 38% of the SLE patients who had low MFI levels for anti‐HLA‐F IgG (500‐999) had active SLE, as did 30% of those with medium anti‐HLA‐F IgG MFI levels (1000–1999) (Fig. 3). There is clearly a predictive potential of high levels (MFI > 2000) of anti‐HLA‐F IgG in characterizing inactive SLE. Conversely, the seeming exception of SLE patients who had low (62%) or medium (70%) anti‐HLA‐F IgG MFI levels but did not have active SLE, can be categorized into two groups: those moving towards the active disease and those moving away from it. First, they may have been at the very beginning of a flare episode, and would experience active SLE in the near future (e.g. patient no. 14 in Fig. 4a) or, secondly, they may have been undergoing remission of the flare episode, with anti‐HLA‐F IgG MFI levels returning to the level of controls (e.g. patient no. 15). Withal, it seems that anti‐HLA‐F IgG MFI could be a beneficial tool, together with other validated clinical or serological markers of active SLE, in predicting flare episodes as well as managing severe flares.

Figure 4b also illustrates the importance of anti‐HLA‐F IgG in SLE patients. In this model of SLE flare, the predominance of anti‐HLA‐F‐specific IgG antibodies is greater when patients' SLE is inactive than when it is active (67 versus 50%; marked dark blue in the vertical bars below the active and inactive SLE in Fig. 4b), suggesting that abatement of SLE flares is associated with an increasing proportion of free antibodies against HLA‐F in contrast to antigen‐bound antibodies or immune complex. The preponderance of anti‐HLA‐F IgG antibodies recognizing specific and shared epitopes of HLA‐F in the pool of circulating anti‐HLA‐Ib IgG antibodies in SLE patients leads to the hypothesis that SLE patients may produce specific anti‐HLA‐F IgG antibodies as an innate response to shed sHLA‐F during the immune dysregulation of T/B cells specific to SLE. The overall function of both the mono‐ and polyreactive pool of anti‐HLA‐Ib IgG antibodies could be to clear the HLA‐F shed in circulation of SLE patients. The functional capabilities of monospecific and polyreactive HLA‐F IgG may differ and remain to be elucidated.

SLE can become inactive (with no clinical or serological evidence of flare episodes), yet patients may experience flares of the disease even after an inactive state 43. However, as inactive SLE correlates with a high level of anti‐HLA‐F IgG antibodies, maintenance of an inactive state may perhaps be prolonged in order to eliminate the infrequent appearance of sHLA‐F shed as a consequence of inflammatory responses by T/B cells 24, 25, 26. This prolongation of SLE inactivity, achieved by maintaining a high level of anti‐HLA‐F IgG antibodies, is surely beneficial for patients because an inactive SLE may lead eventually to complete remission and permanent reduction of anti‐HLA‐F IgG to normal levels. Of course, SLE patients are susceptible to innate failure of immune regulation upon exposure to specific stimuli (environmental, genetic, viral or bacterial), and consequently may experience recrudescence of the flares. Nevertheless, any protocol that may promote progression to complete (and possibly permanent) remission of SLE is clearly a boon to patients. Although much preclinical research must still be performed for a full understanding of the implications of HLA‐F in SLE (e.g. the monitoring of sHLA‐F levels, mentioned below), clinical trials will be needed.

The importance of monitoring circulating soluble HLA‐F

The low MFI of anti‐HLA‐F could be due either to cessation of the production of anti‐HLA‐F antibodies by treatment protocols or due to the formation of an immune complex of anti‐HLA‐F IgG with sHLA‐F. Accordingly, circulating sHLA‐F must be monitored in SLE patients in addition to monitoring anti‐HLA‐F antibodies. Possible release of sHLA‐F by immune cells and tissues during inflammation or during flares (severity) of disease is well supported by other investigators. Bresciani and others 43 were the first to observe the levels of serum sHLA‐I antigens in patients with SLE were significantly higher than in controls and, strikingly, that these levels correlated with disease activity. It is therefore logical to infer a low level of free antibodies in patients with disease severity due to the association of antibodies with sHLA shed as result of flares and associated inflammation. Therefore, the low levels of anti‐HLA‐F IgG antibodies are due to the formation of an immune complex between the antibodies and the sHLA‐F. Furthermore, sHLA‐Ia and Ib molecules, found in the serum of healthy individuals at low levels, increase in the level during inflammation and pathological conditions that include SLE, pregnancy, acute rejection episodes following organ allografts, acute graft‐versus‐host disease following bone marrow transplantation, autoimmune disease, malignant melanoma and viral infections 44, 45, 46, 47, 48, 49.

The observations of Murdaca et al. 50, 51, 52 are very much to this point. This group reported that the serum level of sHLA‐I and sHLA‐G1 and ‐G5 molecules are higher in HIV‐positive subjects than in healthy controls. The serum level of the sHLA‐A, ‐B and ‐C molecules in HIV‐positive patients is correlated with disease stage, increased with disease progression and was predictive of AIDS development 53. Moreover, the peripheral blood mononuclear cells (PBMC) from HIV‐infected individuals spontaneously shed a significantly higher amount of sHLA‐I molecules than peripheral blood mononuclear cells (PBMC) from HIV‐negative subjects. Most importantly, the sHLA‐I level in HIV patients decreased after 36 months of highly active anti‐retroviral therapy, which correlated with the increase in T cell population 50, 51, 52.

Importantly, during the course of an immune response, the sHLA‐I antigens are known to inhibit cytotoxic lysis mediated by both T cells and NK cells and induce apoptosis of activated T cells 20, 54, 55, 56, 57, 58, 59, 60, thus supporting an immune regulatory function of sHLA‐I present in the serum of SLE patients 44, 61, 62, 63. The sHLA‐F in circulation may emanate from activated lymphocytes. Antibodies can be produced when sHLA‐F increase in circulation. Monitoring the levels of the sHLA‐F in circulation and its correlation with patients' SLEDAI and SLEDAI‐2000 scores may validate the hypothesis that clearance of sHLA‐F from the circulation may be the primary function of anti‐HLA‐F IgG antibodies. The development of HLA‐F monospecific mAbs to monitor the circulating sHLA‐F is required.

In conclusion, the immunogenicity of HLA‐F in SLE leads to the dynamic changes in anti‐HLA‐F IgG autoantibodies in relation to clinical SLE activity establishing the autoimmune nature of the disease, and possible beneficial autoimmunity. Monitoring both sHLA‐F and anti‐HLA‐F IgG autoantibodies in SLE patients may provide a better understanding of the clinical disease activity and enable the development of strategies to maintain SLE in remission.

Disclosure

There are no disclosures for any of the authors.

Author contributions

V. J. performed the assays, critically analyzed the data and edited the manuscript. M. H. R. formulated the hypothesis with P. I. T. and V. J., designed the protocol for testing the hypothesis with V. J., co‐analysed the data and prepared the manuscript. P. I. T. evaluated the hypothesis, co‐analysed the data and edited the manuscript. L. E. M.‐B. provided the treatment information for the Mexican female SLE patients. F. K., T. R. and R. B. performed the analysis of the clinical and some of the serological data for the German cohort and reviewed the manuscript and offered comments for improvement.

Supporting information

Additional Supporting information may be found in the online version of this article at the publisher's web‐site:

Table S1. Characteristics and anti‐HLA‐lb profile of German SLE females.

Table S2. Characteristics and anti‐HLA‐lb profile of Mexican SLE females.

Table S3. Characteristics and anti‐HLA‐lb profile of German control females.

Click here for additional data file.^{(7.3MB, pdf)}

Acknowledgements

We thank Dr Nils Lachmann, Institute of Transfusion Medicine, Charité Universitätsmedizin Berlin, Germany, for providing control German females sera and Mr Curtis Maehara for screening the antibodies of the Mexican cohort. The project is supported by grants from the Terasaki Family Foundation and facilities at the Terasaki Foundation Laboratory. Part of this study was presented as an E‐poster (Abstract 771) at the 9th International Congress of Autoimmunity in Nice, France, on 27 March 2014.

Footnotes

As the s.d. exceeded the mean of anti‐HLA‐E IgG in the 20–29 age group, the t‐test performed to compare with same age control did not show any significant difference.

See Hochberg [28].

References

1. Lu LJ, Wallace DJ, Ishimori ML, Scofield RH, Weisman MH. Male systemic lupus erythematosus: a review of sex disparities in this disease. Lupus 2010; 19:119–29. [DOI] [PMC free article] [PubMed] [Google Scholar]
2. Tan EM. Autoantibodies to nuclear antigens (ANA): their immunobiology and medicine. Adv Immunol 1982; 33:167–240. [DOI] [PubMed] [Google Scholar]
3. Ng KP, Manson JJ, Rahman A, Isenberg DA. Association of anti‐nucleosome antibodies with disease flare in serologically active clinically quiescent patients with systemic lupus erythematosus. Arthritis Rheum 2006; 55:900–4. [DOI] [PubMed] [Google Scholar]
4. Mannik M, Merrill CE, Stamps LD, Wener MH. Multiple autoantibodies form the glomerular immune deposits in patients with systemic lupus erythematosus. J Rheumatol 2003; 30:1495–504. [PubMed] [Google Scholar]
5. Amital H, Shoenfeld Y. Autoimmunity and autoimmune diseases such as systemic lupus erythematosus In: Lahita RG, ed. Systemic lupus erythematosus, 3rd edn San Diego, CA: Academic Press, 1999: 1–16. [Google Scholar]
6. Terasaki PI, Mottironi VD, Barnett EV. Cytotoxins in disease autocytotoxins in lupus. N Engl J Med 1970; 283:724–8. [DOI] [PubMed] [Google Scholar]
7. Mok CC, Lau CS. Pathogenesis of systemic lupus erythematosus. J Clin Pathol 2003; 56:481–90. [DOI] [PMC free article] [PubMed] [Google Scholar]
8. Müller‐Suur C, Larsson PH, Larsson K. T cell activation by organic dust in vitro . Respir Med 2000; 94:821–7.] [DOI] [PubMed] [Google Scholar]
9. Zhou B, Saito S, Nakazawa N et al Existence of an immunoglobulin G component of naturally occurring HLA class I antibodies that are not directed against self‐antigens in human serum. Tissue Antigens 2008; 72:98–104. [DOI] [PubMed] [Google Scholar]
10. De Azevedo Silva J, Addobbati C, Sandrin‐Garcia P, Crovella S. Systemic lupus erythematosus: old and new susceptibility genes versus clinical manifestations. Curr Genomics 2014; 15:52–65. [DOI] [PMC free article] [PubMed] [Google Scholar]
11. Rahman A, Isenberg DA. Systemic lupus erythematosus. N Engl J Med 2008; 358:929–39. [DOI] [PubMed] [Google Scholar]
12. European Bioinformatics Institute. HLA‐I Alleles . Available at: http://www.ebi.ac.uk/ipd/imgt/hla (accessed 2013).
13. Coupel S, Moreau A, Hamidou M, Horejsi V, Soulillou JP, Charreau B. Expression and release of soluble HLA‐E is an immunoregulatory feature of endothelial cell activation. Blood 2007; 109:2806–14. [DOI] [PubMed] [Google Scholar]
14. Allard M, Oger R, Vignard V et al Serum soluble HLA‐E in melanoma: a new potential immune‐related marker in cancer. PLOS ONE 2011; 6:e21118. [DOI] [PMC free article] [PubMed] [Google Scholar]
15. Wei XH, Orr HT. Differential expression of HLA‐E, HLA‐F, and HLA‐G transcripts in human tissue. Hum Immunol 1990; 29:131–42. [DOI] [PubMed] [Google Scholar]
16. Kovats S, Main EK, Librach C, Stubblebine M, Fisher SJ, DeMars R. A class I antigen, HLA‐G, expressed in human trophoblasts. Science 1990; 248:220–3. [DOI] [PubMed] [Google Scholar]
17. Kovats S, Librach C, Fisch P et al Expression and possible function of HLA‐G chain in human cytotrophoblasts In: Chaouat G, Mowbray J, eds. Cellular and molecular biology of the materno–fetal relationship. London: INSERM/John Libbey Eurotxt Ltd, 1991:21–9. [Google Scholar]
18. Rosado S, Perez‐Chacon G, Mellor‐Pita S et al Expression of human leukocyte antigen‐G in systemic lupus erythematosus. Hum Immunol 2008; 69:9–15. [DOI] [PubMed] [Google Scholar]
19. Park GM, Lee S, Park B et al Soluble HLA‐G generated by proteolytic shedding inhibits NK‐mediated cell lysis. Biochem Biophys Res Commun 2004; 313:606–11. [DOI] [PubMed] [Google Scholar]
20. Bahri R, Hirsch F, Josse A et al Soluble HLA‐G inhibits cell cycle progression in human alloreactive T lymphocytes. J Immunol 2006; 176:1331–9. [DOI] [PubMed] [Google Scholar]
21. Ishitani A, Sageshima N, Lee N et al Protein expression and peptide binding suggest unique and interacting functional roles for HLA‐E, F, and G in maternal‐placental immune recognition. J Immunol 2003; 171:1376–84. [DOI] [PubMed] [Google Scholar]
22. Lee N, Geraghty DE. HLA‐F surface expression on B cell and monocyte cell lines is partially independent from tapasin and completely independent from TAP. J Immunol 2003; 171:5264–71. [DOI] [PubMed] [Google Scholar]
23. Lepin EJ, Bastin JM, Allan DS et al Functional characterization of HLA‐F and binding of HLA‐F tetramers to ILT2 and ILT4 receptors. Eur J Immunol 2000; 30:3552–61. [DOI] [PubMed] [Google Scholar]
24. Lee N, Ishitani A, Geraghty DE. HLA‐F is a surface marker on activated lymphocytes. Eur J Immunol 2010; 40:2308–18. [DOI] [PMC free article] [PubMed] [Google Scholar]
25. Goodridge JP, Burian A, Lee N, Geraghty DE. HLA‐F complex without peptide binds to MHC class I protein in the open conformer form. J Immunol 2010; 184:6199–208. [DOI] [PMC free article] [PubMed] [Google Scholar]
26. Morandi F, Cangemi G, Barco S et al Plasma levels of soluble HLA‐E and HLA‐F at diagnosis may predict overall survival of neuroblastoma patients. Biomed Res Int 2013; 2013:956878. [DOI] [PMC free article] [PubMed] [Google Scholar]
27. Tan EM, Cohen AS, Fries JF et al The 1982 revised criteria for the classification of systemic lupus erythematosus. Arthritis Rheum 1982; 25:1271–7. [DOI] [PubMed] [Google Scholar]
28. Hochberg MC. Updating the American College of Rheumatology revised criteria for the classification of systemic lupus erythematosus. Arthritis Rheum 1997; 40:1725. [DOI] [PubMed] [Google Scholar]
29. Isenberg DA, Rahman A, Allen E et al BILAG 2004: development and initial validation of an updated version of the British Isles Lupus Assessment Group's disease activity index for patients with systemic lupus erythematosus. Rheumatology (Oxf) 2005; 44:902–6. [DOI] [PubMed] [Google Scholar]
30. Yee CS, Farewell V, Isenberg DA et al Revised British Isles Lupus Assessment Group 2004 Index. Arthritis Rheum 2006; 54:3300–5. [DOI] [PubMed] [Google Scholar]
31. Yee CS, Cresswell L, Farewell V et al Numerical scoring for the BILAG‐2004 index. Rheumatology 2010; 49:1665–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
32. Gladmann DD, Ibañez D, Urowitz MB. Systemic lupus erythematosus disease activity index 2000. J Rheumatol 2002; 29:288–91. [PubMed] [Google Scholar]
33. Romero‐Diaz J, Isenberg DA, Ramsey‐Goldman R. Measures of adult systemic lupus erythematosus. Arthritis Care Res 2011; 63:S37–46. [DOI] [PMC free article] [PubMed] [Google Scholar]
34. Ravindranath MH, Selvan SR, Terasaki PI. Augmentation of anti‐HLA‐E antibodies with concomitant HLA‐Ia reactivity in IFNγ‐treated autologous melanoma cell vaccine recipients. J Immunotoxicol 2012; 9:282–91. [DOI] [PubMed] [Google Scholar]
35. Ravindranath MH, Pham T, Ozawa M, Terasaki PI. Antibodies to HLA‐E may account for the non‐donor‐specific anti‐HLA class‐Ia antibodies in renal and liver transplant recipients. Int Immunol 2012; 24:43–57. [DOI] [PubMed] [Google Scholar]
36. Ravindranath MH, Terasaki PI, Pham T, Jucaud V, Kawakita S. Therapeutic preparations of IVIg contains naturally occurring anti‐HLA‐E Abs that react with HLA‐Ia (HLA‐A/‐B/‐Cw) alleles. Blood 2013; 121:2013–28. [DOI] [PubMed] [Google Scholar]
37. Ravindranath MH, Pham T, El‐Awar N, Kaneku H, Terasaki PI. Anti‐HLA‐E mAb 3D12 mimics MEM‐E/02 in binding to HLA‐B and HLA‐C alleles: Web‐tools validate the immunogenic epitopes of HLA‐E recognized by the antibodies. Mol Immunol 2011; 48:423–30. [DOI] [PubMed] [Google Scholar]
38. Ravindranath MH, Taniguchi M, Chen CW et al HLA‐E monoclonal antibodies recognize shared peptide sequences on classical HLA class Ia: relevance to human natural HLA antibodies. Mol Immunol 2010; 47:1121–31. [DOI] [PubMed] [Google Scholar]
39. Ravindranath MH, Kaneku H, El‐Awar N, Morales‐Buenrostro LE, Terasaki PI. Antibodies to HLA‐E in nonalloimmunized males: pattern of HLA‐Ia reactivity of anti‐HLA‐E‐positive sera. J Immunol 2010; 185:1935–48. [DOI] [PubMed] [Google Scholar]
40. Yee CS, Farewell VT, Isenberg DA et al The use of Systemic Lupus Erythematosus Disease Activity Index‐2000 to define active disease and minimal clinically meaningful change based on data from a large cohort of systemic lupus erythematosus patients. Rheumatology (Oxf) 2011; 50:982–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
41. Ravindranath MH, Terasaki PI, Kawakita S. Naturally occurring anti‐HLA‐E autoantibodies: evidences of HLA‐Ia reactivity of anti‐HLA‐E antibodies, Chapter 35. In: Shoenfeld Y, Gershwin ME, Meroni PI, eds. Autoantibodies. Amsterdam, Netherlands: Elsevier, 2014:1–9. [Google Scholar]
42. Wainwright SD, Biro PA, Holmes CH. HLA‐F is a predominantly empty, intracellular, TAP‐associated MHC class Ib protein with a restricted expression pattern. J Immunol 2000; 164:319–28. [DOI] [PubMed] [Google Scholar]
43. Bresciani A, Pirozzi G, Spera M et al Increased level of serum HLA class I antigens in patients with systemic lupus in patients with systemic lupus erythematosus. Correlation with disease activity. Tissue Antigens 1998; 52:44–50. [DOI] [PubMed] [Google Scholar]
44. Adamashvili I, Wolf R, Aultman D et al Soluble HLA‐I (s‐HLA‐I) synthesis in systemic lupus erythematosus. Rheumatol Int 2003; 23:294–300. [DOI] [PubMed] [Google Scholar]
45. Puppo F, Scudeletti M, Indiveri F, Ferrone S. Serum HLA class I antigens: marker and modulators of an immune response? Immunol Today 1995; 16: 124–7. [DOI] [PubMed] [Google Scholar]
46. Tabayoyong WB, Zavazava N. Soluble HLA revisited. Leuk Res 2007; 31: 121–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
47. Fainardi E, Rizzo R, Melchiorri L et al CSF levels of soluble HLA‐G and Fas molecules are inversely associated to MRI evidence of disease activity in patients with relapsing‐remitting multiple sclerosis. Mult Scler 2008; 14: 446–54. [DOI] [PubMed] [Google Scholar]
48. Campoli M, Ferrone S. Tumour escape mechanisms: potential role of soluble HLA antigens and NK cells activating ligands. Tissue Antigens 2008; 72: 321–34. [DOI] [PMC free article] [PubMed] [Google Scholar]
49. Migliaresi S, Bresciani A, Ambrosone L et al Increased serum concentrations of soluble HLA‐class I antigens in hepatitis C virus related mixed cryoglobulinaemia. Ann Rheum Dis 2000; 59:20–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
50. Murdaca G, Contini P, Setti M et al Soluble human leukocyte antigen‐G serum levels in patients with acquired immune deficiency syndrome affected by different disease‐defining conditions before and after antiretroviral treatment. Hum Immunol 2011; 72:712–6. [DOI] [PubMed] [Google Scholar]
51. Murdaca G, Contini P, Setti M et al Behavior of non‐classical soluble HLA class G antigens in human immunodeficiency virus 1‐infected patients before and after HAART: comparison with classical soluble HLA‐A, ‐B, ‐C antigens and potential role in immune‐reconstitution. Clin Immunol 2009; 133:238–44. [DOI] [PubMed] [Google Scholar]
52. Murdaca G, Contini P, Setti M et al Behavior of serum human major histocompatibility complex class I antigen levels in human immunodeficiency virus‐infected patients during antiretroviral therapy: correlation with clinical outcome. Hum Immunol 2007; 68:894–900. [DOI] [PubMed] [Google Scholar]
53. Puppo F, Brenci S, Lanza L et al Increased level of serum HLA class I antigens in HIV infection. Correlation with disease progression. Hum Immunol 1994; 40: 259–66. [DOI] [PubMed] [Google Scholar]
54. Gros F, Cabillic F, Toutirais O, Maux AL, Sebti Y, Amiot L. Soluble HLA‐G molecules impair natural killer/dendritic cell crosstalk via inhibition of dendritic cells. Eur J Immunol 2008; 38:742–9.] [DOI] [PubMed] [Google Scholar]
55. Zavazava N, Krönke M. Soluble HLA class I molecules induce apoptosis in alloreactive cytotoxic T lymphocytes. Nat Med 1996; 2:1005–10. [DOI] [PubMed] [Google Scholar]
56. Lindaman A, Dowden A, Zavazava N. Soluble HLA‐G molecules induce apoptosis in natural killer cells. Am J Reprod Immunol 2006; 56:68–76. [DOI] [PubMed] [Google Scholar]
57. Spaggiari GM, Contini P, Dondero A et al Soluble HLA class I induces NK cell apoptosis upon the engagement of killer‐activating HLA class I receptors through FasL‐Fas interaction. Blood 2002; 100:4098–107. [DOI] [PubMed] [Google Scholar]
58. Spaggiari GM, Contini P, Carosio R et al Soluble HLA class I molecules induce natural killer cell apoptosis through the engagement of CD8: evidence for a negative regulation exerted by members of the inhibitory receptor superfamily. Blood 2002; 99:1706–14. [DOI] [PubMed] [Google Scholar]
59. Fournel S, Aguerre‐Girr M, Huc X et al Cutting edge: soluble HLA‐G1 triggers CD95/CD95 ligand‐mediated apoptosis in activated CD8+ cells by interacting with CD8. J Immunol 2000; 164:6100–4. [DOI] [PubMed] [Google Scholar]
60. Puppo F, Contini P, Ghio M et al Soluble human MHC class I molecules induce soluble Fas ligand secretion and trigger apoptosis in activated CD8(+) Fas (CD95)(+) T lymphocytes. Int Immunol 2000; 12:195–203. [DOI] [PubMed] [Google Scholar]
61. Hagihara M, Shimura T, Yamamoto K et al Clinical significance of serum soluble HLA class I antigens in systemic lupus erythematosus. Tokai J Exp Clin Med 1993; 18:61–4. [PubMed] [Google Scholar]
62. Tsuchiya N, Shiota M, Yamaguchi A, Ito K. Elevated serum level of soluble HLA class I antigens in patients with systemic lupus erythematosus. Arthritis Rheum 1996; 39:792–6. [DOI] [PubMed] [Google Scholar]
63. Kippner L, Klint C, Sturfelt G, Bengtsson AA, Eriksson H, Truedsson L. Increased level of soluble HLA class I antigens in systemic lupus erythematosus: correlation with anti‐DNA antibodies and leukopenia. J Autoimmun 2001; 16:471–8. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Additional Supporting information may be found in the online version of this article at the publisher's web‐site:

Table S1. Characteristics and anti‐HLA‐lb profile of German SLE females.

Table S2. Characteristics and anti‐HLA‐lb profile of Mexican SLE females.

Table S3. Characteristics and anti‐HLA‐lb profile of German control females.

Click here for additional data file.^{(7.3MB, pdf)}

[cei12724-bib-0001] 1. Lu LJ, Wallace DJ, Ishimori ML, Scofield RH, Weisman MH. Male systemic lupus erythematosus: a review of sex disparities in this disease. Lupus 2010; 19:119–29. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cei12724-bib-0002] 2. Tan EM. Autoantibodies to nuclear antigens (ANA): their immunobiology and medicine. Adv Immunol 1982; 33:167–240. [DOI] [PubMed] [Google Scholar]

[cei12724-bib-0003] 3. Ng KP, Manson JJ, Rahman A, Isenberg DA. Association of anti‐nucleosome antibodies with disease flare in serologically active clinically quiescent patients with systemic lupus erythematosus. Arthritis Rheum 2006; 55:900–4. [DOI] [PubMed] [Google Scholar]

[cei12724-bib-0004] 4. Mannik M, Merrill CE, Stamps LD, Wener MH. Multiple autoantibodies form the glomerular immune deposits in patients with systemic lupus erythematosus. J Rheumatol 2003; 30:1495–504. [PubMed] [Google Scholar]

[cei12724-bib-0005] 5. Amital H, Shoenfeld Y. Autoimmunity and autoimmune diseases such as systemic lupus erythematosus In: Lahita RG, ed. Systemic lupus erythematosus, 3rd edn San Diego, CA: Academic Press, 1999: 1–16. [Google Scholar]

[cei12724-bib-0006] 6. Terasaki PI, Mottironi VD, Barnett EV. Cytotoxins in disease autocytotoxins in lupus. N Engl J Med 1970; 283:724–8. [DOI] [PubMed] [Google Scholar]

[cei12724-bib-0007] 7. Mok CC, Lau CS. Pathogenesis of systemic lupus erythematosus. J Clin Pathol 2003; 56:481–90. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cei12724-bib-0008] 8. Müller‐Suur C, Larsson PH, Larsson K. T cell activation by organic dust in vitro . Respir Med 2000; 94:821–7.] [DOI] [PubMed] [Google Scholar]

[cei12724-bib-0009] 9. Zhou B, Saito S, Nakazawa N et al Existence of an immunoglobulin G component of naturally occurring HLA class I antibodies that are not directed against self‐antigens in human serum. Tissue Antigens 2008; 72:98–104. [DOI] [PubMed] [Google Scholar]

[cei12724-bib-0010] 10. De Azevedo Silva J, Addobbati C, Sandrin‐Garcia P, Crovella S. Systemic lupus erythematosus: old and new susceptibility genes versus clinical manifestations. Curr Genomics 2014; 15:52–65. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cei12724-bib-0011] 11. Rahman A, Isenberg DA. Systemic lupus erythematosus. N Engl J Med 2008; 358:929–39. [DOI] [PubMed] [Google Scholar]

[cei12724-bib-0012] 12. European Bioinformatics Institute. HLA‐I Alleles . Available at: http://www.ebi.ac.uk/ipd/imgt/hla (accessed 2013).

[cei12724-bib-0013] 13. Coupel S, Moreau A, Hamidou M, Horejsi V, Soulillou JP, Charreau B. Expression and release of soluble HLA‐E is an immunoregulatory feature of endothelial cell activation. Blood 2007; 109:2806–14. [DOI] [PubMed] [Google Scholar]

[cei12724-bib-0014] 14. Allard M, Oger R, Vignard V et al Serum soluble HLA‐E in melanoma: a new potential immune‐related marker in cancer. PLOS ONE 2011; 6:e21118. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cei12724-bib-0015] 15. Wei XH, Orr HT. Differential expression of HLA‐E, HLA‐F, and HLA‐G transcripts in human tissue. Hum Immunol 1990; 29:131–42. [DOI] [PubMed] [Google Scholar]

[cei12724-bib-0016] 16. Kovats S, Main EK, Librach C, Stubblebine M, Fisher SJ, DeMars R. A class I antigen, HLA‐G, expressed in human trophoblasts. Science 1990; 248:220–3. [DOI] [PubMed] [Google Scholar]

[cei12724-bib-0017] 17. Kovats S, Librach C, Fisch P et al Expression and possible function of HLA‐G chain in human cytotrophoblasts In: Chaouat G, Mowbray J, eds. Cellular and molecular biology of the materno–fetal relationship. London: INSERM/John Libbey Eurotxt Ltd, 1991:21–9. [Google Scholar]

[cei12724-bib-0018] 18. Rosado S, Perez‐Chacon G, Mellor‐Pita S et al Expression of human leukocyte antigen‐G in systemic lupus erythematosus. Hum Immunol 2008; 69:9–15. [DOI] [PubMed] [Google Scholar]

[cei12724-bib-0019] 19. Park GM, Lee S, Park B et al Soluble HLA‐G generated by proteolytic shedding inhibits NK‐mediated cell lysis. Biochem Biophys Res Commun 2004; 313:606–11. [DOI] [PubMed] [Google Scholar]

[cei12724-bib-0020] 20. Bahri R, Hirsch F, Josse A et al Soluble HLA‐G inhibits cell cycle progression in human alloreactive T lymphocytes. J Immunol 2006; 176:1331–9. [DOI] [PubMed] [Google Scholar]

[cei12724-bib-0021] 21. Ishitani A, Sageshima N, Lee N et al Protein expression and peptide binding suggest unique and interacting functional roles for HLA‐E, F, and G in maternal‐placental immune recognition. J Immunol 2003; 171:1376–84. [DOI] [PubMed] [Google Scholar]

[cei12724-bib-0022] 22. Lee N, Geraghty DE. HLA‐F surface expression on B cell and monocyte cell lines is partially independent from tapasin and completely independent from TAP. J Immunol 2003; 171:5264–71. [DOI] [PubMed] [Google Scholar]

[cei12724-bib-0023] 23. Lepin EJ, Bastin JM, Allan DS et al Functional characterization of HLA‐F and binding of HLA‐F tetramers to ILT2 and ILT4 receptors. Eur J Immunol 2000; 30:3552–61. [DOI] [PubMed] [Google Scholar]

[cei12724-bib-0024] 24. Lee N, Ishitani A, Geraghty DE. HLA‐F is a surface marker on activated lymphocytes. Eur J Immunol 2010; 40:2308–18. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cei12724-bib-0025] 25. Goodridge JP, Burian A, Lee N, Geraghty DE. HLA‐F complex without peptide binds to MHC class I protein in the open conformer form. J Immunol 2010; 184:6199–208. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cei12724-bib-0026] 26. Morandi F, Cangemi G, Barco S et al Plasma levels of soluble HLA‐E and HLA‐F at diagnosis may predict overall survival of neuroblastoma patients. Biomed Res Int 2013; 2013:956878. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cei12724-bib-0027] 27. Tan EM, Cohen AS, Fries JF et al The 1982 revised criteria for the classification of systemic lupus erythematosus. Arthritis Rheum 1982; 25:1271–7. [DOI] [PubMed] [Google Scholar]

[cei12724-bib-0028] 28. Hochberg MC. Updating the American College of Rheumatology revised criteria for the classification of systemic lupus erythematosus. Arthritis Rheum 1997; 40:1725. [DOI] [PubMed] [Google Scholar]

[cei12724-bib-0029] 29. Isenberg DA, Rahman A, Allen E et al BILAG 2004: development and initial validation of an updated version of the British Isles Lupus Assessment Group's disease activity index for patients with systemic lupus erythematosus. Rheumatology (Oxf) 2005; 44:902–6. [DOI] [PubMed] [Google Scholar]

[cei12724-bib-0030] 30. Yee CS, Farewell V, Isenberg DA et al Revised British Isles Lupus Assessment Group 2004 Index. Arthritis Rheum 2006; 54:3300–5. [DOI] [PubMed] [Google Scholar]

[cei12724-bib-0031] 31. Yee CS, Cresswell L, Farewell V et al Numerical scoring for the BILAG‐2004 index. Rheumatology 2010; 49:1665–9. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cei12724-bib-0032] 32. Gladmann DD, Ibañez D, Urowitz MB. Systemic lupus erythematosus disease activity index 2000. J Rheumatol 2002; 29:288–91. [PubMed] [Google Scholar]

[cei12724-bib-0033] 33. Romero‐Diaz J, Isenberg DA, Ramsey‐Goldman R. Measures of adult systemic lupus erythematosus. Arthritis Care Res 2011; 63:S37–46. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cei12724-bib-0034] 34. Ravindranath MH, Selvan SR, Terasaki PI. Augmentation of anti‐HLA‐E antibodies with concomitant HLA‐Ia reactivity in IFNγ‐treated autologous melanoma cell vaccine recipients. J Immunotoxicol 2012; 9:282–91. [DOI] [PubMed] [Google Scholar]

[cei12724-bib-0035] 35. Ravindranath MH, Pham T, Ozawa M, Terasaki PI. Antibodies to HLA‐E may account for the non‐donor‐specific anti‐HLA class‐Ia antibodies in renal and liver transplant recipients. Int Immunol 2012; 24:43–57. [DOI] [PubMed] [Google Scholar]

[cei12724-bib-0036] 36. Ravindranath MH, Terasaki PI, Pham T, Jucaud V, Kawakita S. Therapeutic preparations of IVIg contains naturally occurring anti‐HLA‐E Abs that react with HLA‐Ia (HLA‐A/‐B/‐Cw) alleles. Blood 2013; 121:2013–28. [DOI] [PubMed] [Google Scholar]

[cei12724-bib-0037] 37. Ravindranath MH, Pham T, El‐Awar N, Kaneku H, Terasaki PI. Anti‐HLA‐E mAb 3D12 mimics MEM‐E/02 in binding to HLA‐B and HLA‐C alleles: Web‐tools validate the immunogenic epitopes of HLA‐E recognized by the antibodies. Mol Immunol 2011; 48:423–30. [DOI] [PubMed] [Google Scholar]

[cei12724-bib-0038] 38. Ravindranath MH, Taniguchi M, Chen CW et al HLA‐E monoclonal antibodies recognize shared peptide sequences on classical HLA class Ia: relevance to human natural HLA antibodies. Mol Immunol 2010; 47:1121–31. [DOI] [PubMed] [Google Scholar]

[cei12724-bib-0039] 39. Ravindranath MH, Kaneku H, El‐Awar N, Morales‐Buenrostro LE, Terasaki PI. Antibodies to HLA‐E in nonalloimmunized males: pattern of HLA‐Ia reactivity of anti‐HLA‐E‐positive sera. J Immunol 2010; 185:1935–48. [DOI] [PubMed] [Google Scholar]

[cei12724-bib-0040] 40. Yee CS, Farewell VT, Isenberg DA et al The use of Systemic Lupus Erythematosus Disease Activity Index‐2000 to define active disease and minimal clinically meaningful change based on data from a large cohort of systemic lupus erythematosus patients. Rheumatology (Oxf) 2011; 50:982–8. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cei12724-bib-0041] 41. Ravindranath MH, Terasaki PI, Kawakita S. Naturally occurring anti‐HLA‐E autoantibodies: evidences of HLA‐Ia reactivity of anti‐HLA‐E antibodies, Chapter 35. In: Shoenfeld Y, Gershwin ME, Meroni PI, eds. Autoantibodies. Amsterdam, Netherlands: Elsevier, 2014:1–9. [Google Scholar]

[cei12724-bib-0042] 42. Wainwright SD, Biro PA, Holmes CH. HLA‐F is a predominantly empty, intracellular, TAP‐associated MHC class Ib protein with a restricted expression pattern. J Immunol 2000; 164:319–28. [DOI] [PubMed] [Google Scholar]

[cei12724-bib-0043] 43. Bresciani A, Pirozzi G, Spera M et al Increased level of serum HLA class I antigens in patients with systemic lupus in patients with systemic lupus erythematosus. Correlation with disease activity. Tissue Antigens 1998; 52:44–50. [DOI] [PubMed] [Google Scholar]

[cei12724-bib-0044] 44. Adamashvili I, Wolf R, Aultman D et al Soluble HLA‐I (s‐HLA‐I) synthesis in systemic lupus erythematosus. Rheumatol Int 2003; 23:294–300. [DOI] [PubMed] [Google Scholar]

[cei12724-bib-0045] 45. Puppo F, Scudeletti M, Indiveri F, Ferrone S. Serum HLA class I antigens: marker and modulators of an immune response? Immunol Today 1995; 16: 124–7. [DOI] [PubMed] [Google Scholar]

[cei12724-bib-0046] 46. Tabayoyong WB, Zavazava N. Soluble HLA revisited. Leuk Res 2007; 31: 121–5. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cei12724-bib-0047] 47. Fainardi E, Rizzo R, Melchiorri L et al CSF levels of soluble HLA‐G and Fas molecules are inversely associated to MRI evidence of disease activity in patients with relapsing‐remitting multiple sclerosis. Mult Scler 2008; 14: 446–54. [DOI] [PubMed] [Google Scholar]

[cei12724-bib-0048] 48. Campoli M, Ferrone S. Tumour escape mechanisms: potential role of soluble HLA antigens and NK cells activating ligands. Tissue Antigens 2008; 72: 321–34. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cei12724-bib-0049] 49. Migliaresi S, Bresciani A, Ambrosone L et al Increased serum concentrations of soluble HLA‐class I antigens in hepatitis C virus related mixed cryoglobulinaemia. Ann Rheum Dis 2000; 59:20–5. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cei12724-bib-0050] 50. Murdaca G, Contini P, Setti M et al Soluble human leukocyte antigen‐G serum levels in patients with acquired immune deficiency syndrome affected by different disease‐defining conditions before and after antiretroviral treatment. Hum Immunol 2011; 72:712–6. [DOI] [PubMed] [Google Scholar]

[cei12724-bib-0051] 51. Murdaca G, Contini P, Setti M et al Behavior of non‐classical soluble HLA class G antigens in human immunodeficiency virus 1‐infected patients before and after HAART: comparison with classical soluble HLA‐A, ‐B, ‐C antigens and potential role in immune‐reconstitution. Clin Immunol 2009; 133:238–44. [DOI] [PubMed] [Google Scholar]

[cei12724-bib-0052] 52. Murdaca G, Contini P, Setti M et al Behavior of serum human major histocompatibility complex class I antigen levels in human immunodeficiency virus‐infected patients during antiretroviral therapy: correlation with clinical outcome. Hum Immunol 2007; 68:894–900. [DOI] [PubMed] [Google Scholar]

[cei12724-bib-0053] 53. Puppo F, Brenci S, Lanza L et al Increased level of serum HLA class I antigens in HIV infection. Correlation with disease progression. Hum Immunol 1994; 40: 259–66. [DOI] [PubMed] [Google Scholar]

[cei12724-bib-0054] 54. Gros F, Cabillic F, Toutirais O, Maux AL, Sebti Y, Amiot L. Soluble HLA‐G molecules impair natural killer/dendritic cell crosstalk via inhibition of dendritic cells. Eur J Immunol 2008; 38:742–9.] [DOI] [PubMed] [Google Scholar]

[cei12724-bib-0055] 55. Zavazava N, Krönke M. Soluble HLA class I molecules induce apoptosis in alloreactive cytotoxic T lymphocytes. Nat Med 1996; 2:1005–10. [DOI] [PubMed] [Google Scholar]

[cei12724-bib-0056] 56. Lindaman A, Dowden A, Zavazava N. Soluble HLA‐G molecules induce apoptosis in natural killer cells. Am J Reprod Immunol 2006; 56:68–76. [DOI] [PubMed] [Google Scholar]

[cei12724-bib-0057] 57. Spaggiari GM, Contini P, Dondero A et al Soluble HLA class I induces NK cell apoptosis upon the engagement of killer‐activating HLA class I receptors through FasL‐Fas interaction. Blood 2002; 100:4098–107. [DOI] [PubMed] [Google Scholar]

[cei12724-bib-0058] 58. Spaggiari GM, Contini P, Carosio R et al Soluble HLA class I molecules induce natural killer cell apoptosis through the engagement of CD8: evidence for a negative regulation exerted by members of the inhibitory receptor superfamily. Blood 2002; 99:1706–14. [DOI] [PubMed] [Google Scholar]

[cei12724-bib-0059] 59. Fournel S, Aguerre‐Girr M, Huc X et al Cutting edge: soluble HLA‐G1 triggers CD95/CD95 ligand‐mediated apoptosis in activated CD8+ cells by interacting with CD8. J Immunol 2000; 164:6100–4. [DOI] [PubMed] [Google Scholar]

[cei12724-bib-0060] 60. Puppo F, Contini P, Ghio M et al Soluble human MHC class I molecules induce soluble Fas ligand secretion and trigger apoptosis in activated CD8(+) Fas (CD95)(+) T lymphocytes. Int Immunol 2000; 12:195–203. [DOI] [PubMed] [Google Scholar]

[cei12724-bib-0061] 61. Hagihara M, Shimura T, Yamamoto K et al Clinical significance of serum soluble HLA class I antigens in systemic lupus erythematosus. Tokai J Exp Clin Med 1993; 18:61–4. [PubMed] [Google Scholar]

[cei12724-bib-0062] 62. Tsuchiya N, Shiota M, Yamaguchi A, Ito K. Elevated serum level of soluble HLA class I antigens in patients with systemic lupus erythematosus. Arthritis Rheum 1996; 39:792–6. [DOI] [PubMed] [Google Scholar]

[cei12724-bib-0063] 63. Kippner L, Klint C, Sturfelt G, Bengtsson AA, Eriksson H, Truedsson L. Increased level of soluble HLA class I antigens in systemic lupus erythematosus: correlation with anti‐DNA antibodies and leukopenia. J Autoimmun 2001; 16:471–8. [DOI] [PubMed] [Google Scholar]

PERMALINK

Serum antibodies to human leucocyte antigen (HLA)‐E, HLA‐F and HLA‐G in patients with systemic lupus erythematosus (SLE) during disease flares: Clinical relevance of HLA‐F autoantibodies

V Jucaud

M H Ravindranath

P I Terasaki

L E Morales‐Buenrostro

F Hiepe

T Rose

R Biesen

Summary

Introduction

Materials and methods

Sera from SLE patients and controls

Immunoassay with single‐HLA or β2‐microglobulin (β2m) antigen‐coated microbeads

Table 1.

Table 2.

MFI cut‐off points and the grouping of patients based on MFI

Statistical analysis

Results

Prevalence of HLA‐Ib autoantibodies

Figure 1.

Figure 2.

Table 3.

ACR criteria and anti‐HLA‐Ib autoantibodies

Table 4.

Clinical disease activity and anti‐HLA‐Ib autoantibodies

Figure 3.

Anti‐HLA‐F IgG autoantibodies during the longitudinal course of clinical disease activity

Figure 4.

Discussion

Association of anti‐HLA‐F IgG autoantibodies with SLE disease activity, but not disease severity

Table 5.

Significance of the predominance of anti‐HLA‐F IgG in SLE

Are anti‐HLA‐F IgG autoantibodies protective?

The importance of monitoring circulating soluble HLA‐F

Disclosure

Author contributions

Supporting information

Acknowledgements

Footnotes

References

Associated Data

Supplementary Materials

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases