Lupus (2015) 0, 1–11

http://lup.sagepub.com

PAPER

Association of BAFF, APRIL serum levels, BAFF-R, TACI and

BCMA expression on peripheral B-cell subsets with clinical

manifestations in systemic lupus erythematosus

DC Salazar-Camarena1, PC Ortiz-Lazareno2, A Cruz1, E Oregon-Romero1, JR Machado-Contreras1,JF Munoz-Valle1, M Orozco-Lopez1, M Marın-Rosales1 and CA Palafox-Sanchez1

1Instituto de Investigacion en Ciencias Biomedicas (IICB), Departamento de Clınicas Medicas, Centro Universitario de Ciencias de la Salud,Universidad de Guadalajara, Guadalajara, Jalisco, Mexico; and 2Division de Inmunologıa, Centro de Investigacion Biomedica de Occidente

(CIBO), Instituto Mexicano del Seguro Social IMSS, Guadalajara, Jalisco, Mexico

Objective: B-cell-activating factor (BAFF) and a proliferation-inducing ligand (APRIL) sig-naling pathways regulate B-cell survival through interactions with their receptors BAFF-R,TACI and BCMA. We evaluated the association of these ligands/receptors on B-cell subsetsaccording to clinical manifestations of systemic lupus erythematosus (SLE). Methods: BAFFand APRIL serum concentrations were measured in 30 SLE patients by enzyme-linkedimmunosorbent assay. The BAFF-R, TACI and BCMA expression was analyzed oneach B cell subset (CD19þCD27-CD38–/þ naıve; CD19þCD27þCD38–/þmemory;CD19þCD27-CD38þþ immature and CD19þCD27þCD38þþ plasma cells) by flowcytometry, and compared among patients with different clinical manifestations as well ashealthy controls (HCs). Results: Serum BAFF and APRIL levels were high in SLE patientsand correlated with the Mex-SLEDAI disease activity index (r¼ 0.584; p¼ 0.001 andr¼ 0.456; p¼ 0.011, respectively). The SLE patients showed an increased proportion ofmemory and plasma B cells (p< 0.05). BAFF-R, TACI and BCMA expression in SLE patientswas decreased in almost all B cell subsets compared to HCs (p< 0.05). A lower BCMAexpression was associated with severe disease activity, glomerulonephritis, serositis and hemo-lytic anemia (p< 0.01). BCMA expression showed a negative correlation with Mex-SLEDAIscore (r¼ –0.494, p¼ 0.006). Conclusions: Decreased BCMA expression on peripheral B cellsaccording to severe disease activity suggests that BCMA plays an important regulatingrole in B-cell hyperactivity and immune tolerance homeostasis in SLE patients. Lupus(2015) 0, 1–11.

Key words: Systemic lupus erythematosus; BAFF; BAFF-R; BCMA; BAFF receptors; B cellsubsets; disease activity

Introduction

Systemic lupus erythematosus (SLE) is the proto-type of systemic autoimmune disease, characterizedby overactive B and T-cell responses and lossof immune tolerance against self-antigens. Theunderlying causes remain unclear; however,

increased autoantibody production, defective elim-ination of apoptotic remnants, tissue deposition ofimmune complexes, complement activation andcytokine overproduction are involved in the patho-genesis of SLE.

B-cell-activating factor (BAFF) and aproliferation-inducing ligand (APRIL) are cyto-kines belonging to the tumor necrosis factor(TNF)-ligand superfamily that have an importantrole in the selection, maturation and survival of Bcells.1,2 Both cytokines bind with different affinitiesto three receptors expressed on B cells: BAFFreceptor (BAFF-R), transmembrane activatorand cyclophilin ligand interactor (TACI) and B-cell maturation antigen (BCMA). BAFF and

Correspondence to: Claudia Azucena Palafox-Sanchez, Instituto de

Investigacion en Ciencias Biomedicas (IICB), Departamento de

Clınicas Medicas, Centro Universitario de Ciencias de la Salud,

Universidad de Guadalajara, Sierra Mojada 950, Edificio Q, 1er.

Piso, Colonia Independencia, C.P. 44340, Guadalajara, Jalisco,

Mexico.

Email: kklaumx@yahoo.com

Received 22 January 2015; accepted 2 September 2015

! The Author(s), 2015. Reprints and permissions: http://www.sagepub.co.uk/journalsPermissions.nav 10.1177/0961203315608254

at UNIV DE GUADALAJARA on September 30, 2015lup.sagepub.comDownloaded from

APRIL bind both with TACI and BCMA,whereas BAFF-R is exclusive for BAFF.3 Thesignal transduced by BAFF and APRIL dependson binding receptors as well as the cellular popu-lation where those are expressed. In general,BAFF-R is involved in germinal center (GC) for-mation, as well as in the selection and survival of Bcells,4 and like TACI can induce immunoglobulin-class switching,5,6 whereas BCMA promotes sur-vival of plasma B cells.4,7

Circulating levels of BAFF, APRIL and mem-brane-bound BAFF, increase in response to Toll-like receptor (TLR)9 activation on human B cells,8

type I interferons (IFNs), IFNg, interleukin (IL)-10and granulocyte colony-stimulating factor (G-CSF).BAFF and APRIL have been found to be elevatedboth in humans with SLE and lupus-pronemice,9–13 as well as other autoimmune diseases,such as Sjogren’s syndrome14–16 and rheumatoidarthritis.17,18 Furthermore, increased BAFF serumlevels have been associated with higher diseaseactivity and anti-double-stranded DNA antibodies(dsDNA) in SLE patients,19,20 and recently inhib-ition of BAFF with monoclonal antibody (beli-mumab) has been approved for treatment of SLEpatients with mild or moderate disease.21–23

In relation to receptors, there have been effortsto analyze their expression in SLE patients andassociate them with clinical characteristics, show-ing controversial results.24–27 Therefore, it isimportant to study the expression of BAFF andAPRIL together with their cognate receptors onB-cell subsets according to clinical spectrums inSLE patients.

Materials and methods

Patients and healthy controls (HCs)

Thirty SLE patients were recruited from theRheumatology Department of the HospitalGeneral de Occidente in Zapopan, Jalisco,Mexico. The clinical diagnosis in all patients wasmade in accordance with the American College ofRheumatology (ACR) 1997 revised criteria forSLE.28 The Mexican version of the SystemicLupus Erythematosus Disease Activity Index(Mex-SLEDAI)29 and the Systemic LupusInternational Collaborating Clinics index(SLICC)30 were applied to SLE patients at theenrollment of the study. Patients were subdividedinto four groups: inactive (Mex-SLEDAI <2)(n¼ 11), mild and moderate (Mex-SLEDAI 3–7)(n¼ 9) and severe disease activity (Mex-SLEDAI

�8) (n¼ 7) and the treatment-naıve group, whichincluded three newly diagnosed patients with severedisease activity. The first three groups receivedstandard-of-care therapy for their disease, and thelast group (treatment-naıve), had not received priorsteroid or immunosuppressive therapy. FifteenHCs matched by age and gender were recruitedfrom personnel at the Hospital General deOccidente and the Centro Universitario deCiencias de la Salud, Universidad de Guadalajara.The study was performed according the ethics prin-ciples for experiments involving humans stated inthe Declaration of Helsinki, and it was also evalu-ated and approved by the ethics committee (C.I.083-2014).

Laboratory assessments

A clinical assessment was performed in all eligiblepatients at the time of enrollment, including deter-minations of complete blood count (CELL-DYN3500R; Abbott Diagnostics, Abbott Park, IL,USA) and erythrocyte sedimentation rate (ESR)(Wintrobe method).

Antibodies and reagents

Anti-human allophycocyanin/Cy7 (APC/Cy7)-conjugated anti-CD3, peridinin chlorophyll pro-tein-conjugated (PerCP)-anti-CD19, fluoresceinisothiocyanate-conjugated (FITC) anti-CD27,phycoerythrin/Cy7 (PE/Cy7)-conjugated anti-CD38, phycoerythrin (PE)-conjugated anti-268 -(BAFF-R), PE-conjugated anti-267 (TACI) andtheir corresponding isotypes were all purchasedfrom BioLegend Inc (San Diego, CA, USA).Allophycocyanin (APC)-conjugated anti-269(BCMA) and isotype were purchased from R&DSystems (Minneapolis, MN, USA).

Flow cytometry

For identification of B-cell subsets (CD19þCD27-CD38–/þ naıve; CD19þCD27þCD38–/þmemory; CD19þCD27-CD38þþ immature andCD19þCD27þCD38þþ plasma cells) andBAFF-R, TACI and BCMA expression in these,total blood (300 ml) was stained with the appropri-ate combinations of fluorochrome-conjugated anti-bodies to CD3, CD19, CD27, CD38, BAFF-R,TACI and BCMA. At the end of the labeling, redcells were lysed by adding 1X BD FACS Lysingsolution (BD Biosciences, Franklin Lakes, NJ,USA) to each sample. After 15 minutes of incuba-tion at room temperature, cells were pelleted andwashed twice before acquisition on a FACS Aria I

BAFF, APRIL, BAFF-R, TACI and BCMA expression in SLEDC Salazar-Camarena et al.

2

Lupus

at UNIV DE GUADALAJARA on September 30, 2015lup.sagepub.comDownloaded from

cytometer (BD Biosciences, San Jose, CA, USA).For each sample, at least 50,000 lymphocytes wereacquired. Appropriate isotype and FluorescenceMinus One (FMO) controls were used to adjustfor background fluorescence and gates, and resultsare reported as the percentage (%) of expression orgeometric mean fluorescence intensity (MFI). Datawere processed with FACSDiva software (BDBiosciences).

Enzyme-linked immunosorbent assay (ELISA)

All assays were performed on stored (–80�C) serumsamples. Serum BAFF concentration was deter-mined with quantitative sandwich enzyme immuno-assay technique (DBLYS0B; R&D Systems, MN,USA) and according to recommendations made bythe manufacturer. Standard curves were derivedfrom serial dilutions of 40 ng of recombinanthuman BAFF, and the sensitivity of the assayranged from 1.01 to 6.44 pg/ml. Serum concentra-tions were measured in a microplate reader at450 nm (BioTek Instruments Inc, Winooski, VT,USA).

The concentration of human APRIL serum wasmeasured using the human APRIL DuoSet ELISAKit (DY884; R&D Systems, MN, USA). Briefly,polystyrene microplate wells were coated with100 ml per well of the diluted capture antibodyand incubated overnight at room temperature.The next day, serum samples and standards wereadded and incubated for two hours. A seven-pointstandard curve using two-fold serial dilutions of60,000 pg/ml of recombinant human APRIL wascreated. As secondary antibody, biotinylatedmouse anti-human APRIL was used and incubatedfor two hours. The plate was washed three timesand then 100 ml of the working dilution of strepta-vidin-horseradish peroxidase (HRP) was added toeach well. After 20 minutes of incubation the reac-tion was developed with substrate solution. Afteradding the stop solution, the intensity was mea-sured in an automated microplate reader at450 nm (BioTek Instruments Inc, Winooski, VT,USA).

Autoantibodies determination

Serum levels of biologic markers includingdsDNA, anti-Sm (Binazyme, The BindingSite Ltd., Birmingham, UK), anti-Ro, anti-La(ORGENTEC Diagnostika GmbH, Mainz,Germany) autoantibodies were tested in SLEpatients using an ELISA kit. For dsDNA antibo-dies, the range of detection was 12.3–1000 IU/ ml,and sensitivity of the assay was 4.6 IU/ ml, for

anti-Sm sensitivity was 4.0 UI/ ml, while for anti-Ro and anti-La, sensitivity of the assay was 1.0 IU/ml. Determinations were performed according to theinstructions indicated by the manufacturer.

Statistical analyses

Data analysis was performed using PASWStatistics 18 software (IBM Corporation,Armonk, NY, USA) and GraphPad Prism v.6 soft-ware (GraphPad Software Inc, San Diego, CA,USA). Mann-Whitney U test and Kruskal-Wallisanalysis of variance were used for nonparametricdistribution data. Correlations were examined bySpearman’s rank correlation test. P values� 0.05were considered significant.

Results

Demographic and clinical characteristics

All SLE patients included in this study werefemales. The average age was 34 years old (range18–67) having 7.3 years of disease duration (0–21years). SLE patients presented an activity diseasescore of 4.5 (range 0–18) for Mex-SLEDAI and0.8 (range 0–3) score for SLICC damage index.Hematological, renal and mucocutaneous involve-ments were the major clinical manifestations. Thetreatment of SLE patients included immunosup-pressive drugs (azathioprine, methotrexate, cyclo-phosphamide and mycophenolate), chloroquineand prednisone (Table 1).

Peripheral B-lymphocyte subsetsand BAFF receptors

Figure 1(a) shows a representative strategy examplefor the analysis of B-cell subsets (CD19þCD27-CD38–/þ naıve; CD19þCD27þCD38–/þmemory; CD19þCD27-CD38þþ immatureand CD19þCD27þCD38þþ plasma cells) inan HC and a treatment-naıve SLE patient.Distribution of peripheral B-cell subsets was differ-ent in SLE compared to HCs. The percentage of Bcells (CD3-CD19þ ) was lower in the SLE group(6.72� 5.55) compared to HCs (9.43� 2.96;p¼ 0.007) except in new-onset treatment-naıvepatients, who had an elevated percentage of Bcells (18.21� 9.08). We analyzed the expression ofBAFF-R, TACI and BCMA in B (CD3-CD19þ )cells in SLE patients and HCs. Both the BAFF-R expression rate in B cells (94.96� 9.56vs 99.45� 1.12%, p¼ 0.003) and mean fluores-cence intensity (MFI) (15,525.5� 13,783.8

BAFF, APRIL, BAFF-R, TACI and BCMA expression in SLEDC Salazar-Camarena et al.

3

Lupus

at UNIV DE GUADALAJARA on September 30, 2015lup.sagepub.comDownloaded from

vs 45,477.5� 2,9321.8, p¼ 0.013) were down-regu-lated significantly in SLE patients in comparisonwith HCs. BCMA expression was also decreasedin the SLE group (6.16� 9.69%) in comparisonwith HCs (17.33� 8.48%), p< 0.001. No differencewas observed in TACI expression (Figure 1(b)).

Subsequently, the different B-cell subsets in SLEpatients and HCs were evaluated. The smallest

subpopulation in both groups corresponded toimmature B lymphocytes and showed no differ-ences. The naıve B subpopulation was the largest;however, it was decreased in the SLE group(52.30� 17.22%) compared to HCs (68.54�9.28%), p¼ 0.003. Subpopulation distribution of per-ipheral memory B cells was expanded in SLE patients(39.54� 18.18%) in comparison with HCs(25.77� 8.61%), p¼ 0.013 (Figure 1(c)). Patientswith active disease (Mex-SLEDAI�3) did not differin their frequency of memory B cells compared tothose with inactive disease (Mex-SLEDAI 0-2),p¼ 0. 478. Finally, as has been reported previously,SLE patients had an increased proportion of plasmacells (3.86� 4.15%) compared to HCs (1.30� 1.19%,p¼ 0.001) (Figure (1c)), and those with severe diseaseactivity showed the highest percentage (6.00� 7.88 vs1.3� 1.1% in HCs), p¼ 0.023.

Afterward, expression of BAFF-R, TACI andBCMA in these B-cell subsets of SLE patientsand HCs was analyzed. In SLE patients, weobserved a lower BAFF-R expression in all B-cellsubsets, except memory cells (p< 0.05). TACIshowed decreased expression in memory andplasma cells of SLE patients, but no differenceswere observed in immature or naıve cells(p< 0.05) in comparison with controls. Finally,we found a lower BCMA expression in naıve,memory and plasma B-cell subsets in SLE patientscompared to HCs (p< 0.05) (Figure 1(d)).

BAFF-R, TACI and BCMA B-cell expression andSLE features

In order to evaluate the possible effects of BAFF/APRIL levels and receptor expression in diseaseactivity, we divided patients into four groups: inac-tive, mild-moderate, and severe disease activity,and new-onset treatment-naıve patients. As shownin Table 2, severe disease activity and treatment-naıve SLE patients had lower BAFF-R andBCMA expression rates than the othergroups (p< 0.05). TACI expression was similarbetween SLE patients and HCs, but in stratifiedgroups, we found that the expression of TACI inHCs (45.41� 25.28%) and inactive patients(45.58� 15.82%) was similar, while the expressionrate in treatment-naıve SLE patients was signifi-cantly decreased (23.97� 13.82%) (p¼ 0.049)(Table 2). Notably, BCMA showed very lowexpression in severe disease activity (Table 2) andtreatment-naıve SLE patients compared to HCsand inactive patients (1.94� 3.87% and0.33� 0.32% vs 17.3� 8.4% and 9.37� 11.89%,respectively) (p< 0.05).

Table 1 Demographic and clinical characteristics in SLEpatients

SLE, n¼ 30

Male/Female 0/100

Age, yearsa 34� 12 (18–67)

Disease duration, yearsa 7.3� 6 (0–21)

Clinical assessment

Mex-SLEDAI, scorea 4.5� 4 (0–18)

SLICC, scorea 0.8� 0.9 (0–3)

Clinical manifestations

Hematologic

Cytopeniasb, n (%) 27 (90)

Hemolytic anemia, n (%) 4 (13.3)

Renal

Renal activity, n (%) 10 (33.3)

Proteinuria, n (%) 7 (23.3)

Mucocutaneousc 11 (36.6)

Serositis, n (%) 4 (13.3)

Arthritis, n (%) 4 (13.3)

Neurologicd, n (%) 2 (6.7)

Autoantibodies

ANA, n (%) 30 (100)

Anti-dsDNA, n (%) 16 (52)

Anti-Sm, n (%) 11 (35.3)

Anti-Ro, n (%) 7 (23.1)

Anti-La, n (%) 7 (23.1)

Treatment

Prednisone, n (%) 20 (66.7)

Azathioprine, n (%) 18 (60)

Antimalarial, n (%) 20 (66.6)

Cyclophosphamide, n (%) 8 (26.7)

Methotrexate, n (%) 2 (6.7)

Mycophenolate, n (%) 2 (6.7)

SLE: systemic lupus erythematosus; Mex-SLEDAI: Mexican version

of the Systemic Lupus Erythematosus Disease Activity Index; SLICC:

Systemic Lupus International Collaborating Clinics; ANA: antinuclear

antibodies; dsDNA: double-stranded DNA.aData provided in mean� standard deviation, minimum and

maximum.bCytopenias included leukopenia, lymphopenia and thrombocytopenia.cMucocutaneous manifestations included: malar erythema, discoid

lupus, oral ulcers and photosensitivity.dNeurologic clinical manifestations included neurologic damage,

psychosis and convulsions.bCytopenias included leucopenia, lymphopenia and thrombocytopenia.cMucocutaneous manifestations included: malar erythema, discoid

lupus, oral ulcers and photosensitivity.dNeurologic clinical manifestations included neurologic damage,

psychosis, and convulsions.

BAFF, APRIL, BAFF-R, TACI and BCMA expression in SLEDC Salazar-Camarena et al.

4

Lupus

at UNIV DE GUADALAJARA on September 30, 2015lup.sagepub.comDownloaded from

Figure 1 Expression of BAFF-R, TACI and BCMA on peripheral subpopulations of B cells in SLE patients and healthy controls.(a) Representative analyses of subpopulations of B cells (CD19þCD27-CD38–/þ naıve; CD19þCD27þCD38–/þmemory;CD19þCD27-CD38þþ immature and CD19þCD27þCD38þþ plasma cells) in an HC and treatment-naıve SLE patient.(b) Percentage of expression of BAFF-R, TACI and BCMA on B cells were compared between SLE and HCs.(c) Subpopulations of B cells (naıve, memory, immature and plasma cells) were compared between SLE and HCs.(d) Expression of BAFF-R, TACI and BCMA in naıve, memory, immature and plasma cells in SLE and HCs were analyzed.BAFF-R: B-cell activating factor receptor; TACI: transmembrane activator and calcium modulator and cyclophilin ligand inter-actor; BCMA: B-cell maturation antigen; SLE: systemic erythematosus lupus; HC: healthy controls.Statistical analysis was performed using Mann-Whitney U test. Values are represented as mean� SD. * p< 0.05 vs HCs. **p< 0.01vs HCs.

BAFF, APRIL, BAFF-R, TACI and BCMA expression in SLEDC Salazar-Camarena et al.

5

Lupus

at UNIV DE GUADALAJARA on September 30, 2015lup.sagepub.comDownloaded from

Moreover, we found that BCMA expressionnegatively correlated with Mex-SLEDAI score(r¼ –0.494, p¼ 0.006) (Figure 2). DecreasedBCMA was associated with different clinical set-tings. When we compared the expression ofBCMA on B cells, those patients with clinical mani-festations showed a lower rate. As shown in Table 3,patients with renal activity (0.28� 0.32% vs9.07� 10.79%, p¼ 0.001), serositis (0.16� 0.05%vs 7.06� 10.12%, p< 0.01) and hemolytic anemia(0.30� 0.21% vs 7.37� 10.46%, p< 0.01) displayeda down-regulated expression of BCMA compared toSLE patients without those clinical manifestations.

BAFF and APRIL and SLE features

As shown in Figure 3(a), elevated serum BAFFlevels were found in SLE compared to HCs(3.19� 4.26 vs 0.97� 0.21 ng/ml, p< 0.05). BAFFwas correlated with the percentage of CD19þBcells (r¼ 0.407, p¼ 0.025) in SLE patients and dis-ease activity score evaluated by Mex-SLEDAI(r¼ 0.584, p¼ 0.001) (Figure 3(b)). Treatment-naıve patients had the highest BAFF levels andthe lowest BAFF-R expression detected on B cells(Table 2), which might indicate over-activation orderegulation of the BAFF axis in this group. SerumAPRIL levels were also increased in SLE comparedto HCs (31.42� 27.42 vs 13.41� 9.84 ng/ml,p< 0.05) (Figure 3(a), right panel). As shown in

Figure 3(c), APRIL correlated both with Mex-SLEDAI (r¼ 0.456, p¼ 0.011) and SLICC scores(r¼ 0.407, p¼ 0.026). Regarding the SLE groups,those with severe disease activity and treatment-

Table 2 Expression of BAFF/APRIL levels and BAFF-R, TACI and BCMA on B cells in systemic lupus erythematosus patientsaccording to disease activity index and HCs

HCsn¼ 15

SLE

Inactive n¼ 11 pMild-moderatedisease activity n¼ 9 p

Severe diseaseactivity n¼ 7 p

Treatment-naıvepatientsa n¼ 3 p

CD19þ B cells (%) 9.43� 2.96 6.63� 4.46 0.03 d 5.27� 1.95 <0.01 b 3.80� 2.01 <0.01 b 18.21� 9.08 <0.05f,g

CD19þ/BAFF-Rþ (%) 99.4� 1.1 92.3� 14.2 <0.01 b 99.1� 0.7 0.17 95.4� 6.2 0.03 b 91.0� 7.4 <0.05 b,f

MFI of BAFF-Rþ 36,426.6� 28,653.6 22,213.5� 16,297.0 0.29 10,796.3� 11,582.5 0.05 b 10,686.2� 7243.4 0.05 9,702.0� 2,904.7 0.16

CD19þ/BCMAþ (%) 17.3� 8.4 9.37� 11.89 <0.05 b,d 7.45� 10.39 0.02 b 1.94� 3.87 <0.05 b,d 0.33� 0.32 <0.01 b

MFI of BCMAþ 8232.7� 5,398.2 9836.8� 10,673.7 0.66 7231.0� 5285.1 0.13 7542.0� 5584.9 0.45 8625.5� 2,865.9 0.61

CD19þ/TACIþ (%) 45.4� 25.2 45.5� 15.8 0.95 35.3� 11.7 0.16 31.6� 11.9 0.70 23.9� 13.8 <0.05 b

MFI of TACIþ 5592.5� 9,448.6 2,929.1� 517.7 0.90 3554.2� 2004.4 0.45 3152.7� 2361.4 0.27 2,685.5� 747.4 0.90

BAFF (ng/ml) 0.97� 0.21 1.52� 0.55 <0.01 b,d 1.60� 0.89 <0.05 b,e 3.67� 1.95 <0.05 b,d 13.01� 8.71 <0.05 b,f

APRIL (ng/ml) 13.41� 19.84 21.55� 22.18 0.32 24.69� 26.69 0.31 43.76� 29.35 <0.01 b 58.92� 24.58 0.01 b

BAFF: B-cell activating factor; APRIL: a proliferation-inducing ligand-R: B-cell activating factor receptor; TACI: transmembrane activator and

calcium modulator and cyclophilin ligand interactor; BCMA: B-cell maturation antigen; HCs: healthy controls; SLE: systemic lupus erythemato-

sus; MFI: geometric mean fluorescence intensity; ng/ml: nanograms per milliliter. Bold values represent p< 0.05.aTreatment-naıve patients included newly diagnosed patients who had not received prior steroid or immunosuppressive therapy.bComparison between each SLE group and HCs.cComparison between inactive vs severe disease activity patients.dComparison between inactive vs treatment-naıve patients.eComparison between mild-moderate vs severe disease activity patients.fComparison between mild-moderate disease activity vs treatment-naıve patients.gComparison between severe disease activity vs treatment-naıve patients.

Figure 2 Correlation of the expression rate of BCMA inCD19þB cells with Mex-SLEDAI score in SLE patients.Statistical analysis was performed using Spearman’s correl-ation test.BCMA: B-cell maturation antigen; Mex-SLEDAI: Mexicanversion of the Systemic Lupus Erythematosus DiseaseActivity Index.

BAFF, APRIL, BAFF-R, TACI and BCMA expression in SLEDC Salazar-Camarena et al.

6

Lupus

at UNIV DE GUADALAJARA on September 30, 2015lup.sagepub.comDownloaded from

naıve SLE patients showed higher BAFF- andAPRIL-soluble levels than HCs and other SLEgroups (p< 0.05) (Table 2).

Discussion

SLE is an autoimmune disease characterized by B-cellhyperactivity, and the production of autoantibodiesagainst nuclear components is the hallmark of thedisease. Because these cells are centrally involved inthe pathogenesis of SLE, we analyzed the expressionof B-cell receptors (BCRs) BAFF-R, TACI andBCMA in four different peripheral B cell subsets todetermine their possible correlations with clinical par-ameters and disease activity in SLE patients.

Our study found low B-cell levels in SLEpatients, while for HCs percentages were similarto those reported in other studies.26,31,32 Insideinactive B-cell compartments, patients with SLEhad a significantly lower percentage of naıve Bcells compared to HCs.

Notably, we detected an increased frequency ofcirculating antigen-experienced B cells, like plasmacells and our work, demonstrates that this subset isnotably expanded in SLE patients, even more inthose with active disease. These findings, in add-ition to those of other studies, emphasize the cen-tral role of these cells in the pathogenesis of SLE,mainly due to their ability to produce pathogeni-cally relevant autoantibodies, such as anti-dsDNAantibodies.33,34 Previously, plasma B cells fromSLE patients have been shown to be less susceptibleto immunosuppressive therapy.35 Also, the memoryB-cell subpopulation was significantly elevated inSLE patients, which is consistent with previousanalyses. 36–38 Contributing to these abnormalities,

ectopic GC structures could be involved in T cell-independent activation of memory B cells. In thispathway, TLR engagement and activation of TACIpromote the recruitment of myeloid differentiationprimary-response protein 88 (MYD88), whichinduces the expression of activation-induced cyti-dine deaminase (AID).6

Altogether, an expansion in the subpopulation ofmemory B cells involves a high risk for autoimmun-ity because these cells already exceed negative selec-tion checkpoints of the immune system and havelow levels of activation, allowing them to differen-tiate rapidly in antibody-producing B cells. The dif-ferences found in the B-cell subsets suggestalterations in selection mechanisms and maturationstage control; however, we could not establish if thedefects were intrinsic or related to the inflammatorymilieu.

On the other hand, previous studies havereported BAFF-R as the BAFF receptor with thehighest expression rate on peripheral B cells, whileBCMA had the lowest expression rate.24–26 Wefound both receptors were decreased in SLEpatients, while TACI showed no differences inexpression rates or MFI between SLE and HCs.Previous studies from SLE patients show discord-ant findings, lower BAFF-R expression and ahigher or no differences in TACI and BCMAexpression.26,27,39 Our work demonstrated thatBAFF-R, TACI and BCMA could be co-expressedon the B-cell membrane and vary according to thestage of differentiation of the cell. However,because these three receptors can be occupied byboth cytokines, further work is necessary to eluci-date which one of the ligands binds preferentiallyin vivo according to the differentiation stage. Ourresults are consistent with other studies: BAFF-R is

Table 3 Comparison of serum BAFF/APRIL levels and BAFF-R, TACI, and BCMA expression on B cells of SLE patientsaccording to clinical manifestations

Clinical manifestations

Renal activity Serositis Hemolytic anemia

Present n¼ 10 Absent n¼ 20 p Present n¼ 4 Absent n¼ 26 p Present n¼ 4 Absent n¼ 26 p

BAFF-R (%) 99.02� 0.87 92.96� 11.25 0.027 90.80� 7.05 95.30� 9.97 0.09 92.42� 8.34 95.30� 9.97 0.59

BCMA (%) 0.28� 0.32 9.07� 10.79 <0.01 0.16� 0.05 7.06� 10.12 <0.01 0.30� 0.21 7.37� 10.46 <0.01

TACI (%) 28.74� 35.79 45.70� 26.72 0.07 20.20� 25.38 42.77� 30.29 0.07 30.75� 17.47 42.81� 32.19 0.65

BAFF (ng/ml) 4.13� 4.53 2.73� 4.16 0.21 11.22� 7.96 1.96� 1.27 <0.01 2.87� 1.06 3.33� 4.65 0.19

APRIL (ng/ml) 47.23� 32.30 23.50� 21.36 0.06 47.43� 30.51 28.95� 26.71 0.24 45.05� 17.8 29.89� 28.71 0.20

BAFF: B-cell activating factor; APRIL: a proliferation-inducing ligand; BAFF-R: B-cell activating factor receptor; TACI: transmembrane acti-

vator and calcium modulator and cyclophilin ligand interactor; BCMA: B-cell maturation antigen; SLE: systemic lupus erythematosus; ng/ml:

nanograms per milliliter.

Bold values represent p< 0.05.

BAFF, APRIL, BAFF-R, TACI and BCMA expression in SLEDC Salazar-Camarena et al.

7

Lupus

at UNIV DE GUADALAJARA on September 30, 2015lup.sagepub.comDownloaded from

Figure 3 Serum levels of BAFF and APRIL and their association with SLE.(a) Left and right panel show increased BAFF and APRIL serum levels in SLE patients compared to HCs.(b) Left panel displays a positive correlation between disease activity and BAFF serum levels in SLE patients. In the right panel apositive correlation of peripheral percentage of B cells and BAFF serum levels in SLE patients is shown.(c) A positive correlation between disease activity and organ damage with APRIL serum levels in SLE patients is shown in the rightand left panel, respectively.BAFF: B-cell activating factor; APRIL: a proliferation-inducing ligand; SLE: systemic lupus erythematosus; Mex-SLEDAI:Mexican version of the Systemic Lupus Erythematosus Disease Activity Index; SLICC: Systemic Lupus InternationalCollaborating Clinics index; HC: healthy controls.Statistical analysis was performed using Mann-Whitney U test for (a) and Spearman’s correlation test for (b) and (c). Horizontallines and error bars represent median with interquartile range, respectively.

BAFF, APRIL, BAFF-R, TACI and BCMA expression in SLEDC Salazar-Camarena et al.

8

Lupus

at UNIV DE GUADALAJARA on September 30, 2015lup.sagepub.comDownloaded from

the receptor that showed the highest expression rateduring almost all subsets, having elevated levels onimmature, naıve, and memory B subsets.24–26,40

TACI and BCMA raised their expression rates inlater differentiation B-cell stages,24–26 probablybecause their functions are more dependent atthese stages and are also related to immunoglobulinclass switching and plasma B-cell maintenance.7,41

It would be interesting to further analyze theBAFF-R and BCMA expression relationship oreven their possible counter-regulation, as thestudy showed the decreased expression ofBAFF-R coincided with the acquisition of BCMAat later B cell differentiation stages (memory andplasma cells). The analysis of the molecular mech-anisms involved in this phenomenon will help elu-cidate if it is only an event required at the stage ofdifferentiation of B cells, or even more, help deter-mine the biological involvement of BCMA in B-cellhomeostasis.

Because of the systemic characteristic of the dis-ease and the multiple clinical and serological mani-festations of SLE patients, it is important to evaluatethe expression of BAFF-R, TACI and BCMA withthe disease activity score. In general, the level ofexpression of the receptors decreases according tothe increase in disease activity. MFI of BAFF-Rwas decreased in SLE patients with low-mild andsevere disease activity, which had been previouslyreported.26,27,39 Regarding TACI, the expression ofthe receptor is similar between HCs and SLE inac-tive patients, and the expression rate decreased grad-ually as disease activity increased.

BCMA is a receptor for BAFF that underphysiological conditions is important for sustainingenduring antibody protection by mediating survivalof long-lived plasma cells, but is not reported to berequired for B-cell maturation and homeostasis.4,7

Two previous studies of B cells from SLE patientshave shown increased levels of BCMA expressioncompared with those in B cells from healthydonors.27,41 Kim et al. showed increased BCMAexpression and antinuclear antibodies (ANA) secre-tion after TLR9 stimulation on B cells isolatedfrom SLE patients.27 These studies suggest thatincreased BCMA expression levels on peripheralhuman B cells correlates with B-cell activationand responsiveness to BAFF and APRIL, whichcan contribute to autoantibody production fromautoreactive B cells activated through TLR9.

No difference in MFI of BCMA among studygroups was found; nonetheless, expression rates onBCMA receptor in SLE patients with severe diseaseactivity, including treatment-naıve patients, werelow or null. This finding is relevant because so far

it differs from other data published in humans.27

Recently, a study reported the importance ofBCMA in the negative regulation of T follicularhelper (TFH) cell expansion, because BCMA defi-ciency in T cells promotes TFH expansion, GC for-mation, autoantibody production and IFNgproduction by TFH cells through BAFF-R in amurine model of SLE.42 These findings suggest theparticipation of BCMA in addition to BAFF-R forthe maintenance of B- and T-cell homeostasis.

Furthermore, when we compared BCMA expres-sion in SLE patients according to clinical manifest-ations, those SLE patients with renal activity,serositis and hemolytic anemia had the lowest levels.This could indicate the participation of BCMA in SLEpathogenesis is more important than previously attrib-uted. Similar to our results, other work in a murineSLE model demonstrated that lack of BCMA pro-motes lymphoproliferation in mature B cells, anincreased number of short- and long-lived plasmacells, elevated titers of ANA and immune complexdeposition in kidneys.43 Overall, these results suggestthat lack of BCMA expression exacerbates the devel-opment of lymphoproliferation and autoimmunitymanifestations in SLE; additionally it could indicatethat BCMA is a checkpoint of B cell homeostasisand self-tolerance in systemic autoimmunity.

Serum BAFF levels were elevated in SLE patientsand correlated with disease activity and the B-cellcount, which led to the explanation of the biologicalimportance of this cytokine in the B-cell function.The overexpression of BAFF could counterbalancethe pro-apoptotic signals induced by BCR in auto-reactive B cells, allowing them to survive and beactivated by autoantigens contributing to auto-immunity. Other authors have reported the associ-ation of higher levels of BAFF with anti-dsDNAautoantibody titers and musculoskeletal manifest-ations in SLE patients,11–13,24 suggesting that dysre-gulation of BAFF could be involved in SLEpathogenesis. On the other hand, serum APRILlevels were also increased in the SLE group andcorrelated with disease activity and SLICC damagescore. Previous reports have shown the associationof disease activity, musculoskeletal manifest-ations9,10,44,45 and neuropsychiatric SLE withAPRIL levels.20 Recently intrarenal APRIL messen-ger RNA (mRNA) levels were associated with pro-teinuria and with resistance to treatment in SLEpatients.46 These results suggest that APRIL maybe an important marker of disease activity and prog-nosis in patients with SLE.

In summary, even though there are few studies ofBAFF and APRIL receptors in humans, TACI andparticularly BAFF-R have been implicated in the

BAFF, APRIL, BAFF-R, TACI and BCMA expression in SLEDC Salazar-Camarena et al.

9

Lupus

at UNIV DE GUADALAJARA on September 30, 2015lup.sagepub.comDownloaded from

differentiation, maturation and homeostasis of Bcells. The third receptor of these ligands, BCMA,has been poorly studied and associated with themaintenance of long-lived plasma cells. Our find-ings demonstrate that BCMA is not only expressedin plasma cells, even if this particularly subset hadthe highest expression rate. Moreover, BCMAdecreased expression rate or absence may have arelevant role in the development or exacerbationof SLE, and may be a useful marker of disease.These findings emphasize the relevance of BCMAin addition to BAFF-R in B-cell homeostasis.

To our knowledge, this is the first study inhumans that demonstrates the importance ofBCMA expression and its relation with diseaseactivity in SLE patients.

Funding

The authors disclosed receipt of the followingfinancial support for the research, authorship,and/or publication of this article: This work wassupported by the National Council of Science andTechnology (grant number 115567).

Declaration of Conflicting Interests

The authors declared no potential conflicts of inter-est with respect to the research, authorship, and/orpublication of this article.

References

1 Mackay F, Schneider P. Cracking the BAFF code. Nat Rev Immunol2009; 9: 491–502.

2 Vincent FB, Saulep-Easton D, Figgett WA, et al. The BAFF/APRIL system: Emerging functions beyond B cell biology and auto-immunity. Cytokine Growth Factor Rev 2013; 24: 203–215.

3 Thompson JS, Bixler SA, Qian F, et al. BAFF-R, a newly identifiedTNF receptor that specifically interacts with BAFF. Science 2001;293: 2108–2111.

4 Schneider P. The role of APRIL and BAFF in lymphocyte activa-tion. Curr Opin Immunol 2005; 17: 282–289.

5 Castigli E, Wilson SA, Scott S, et al. TACI and BAFF-R mediateisotype switching in B cells. J Exp Med 2005; 201: 35–39.

6 He B, Santamaria R, Xu W, et al. The transmembraneactivator TACI triggers immunoglobulin class switching by activat-ing B cells through the adaptor MyD88. Nat Immunol 2010; 11:836–845.

7 O’Connor BP, Raman VS, Erickson LD, et al. BCMA is essentialfor the survival of long-lived bone marrow plasma cells. J Exp Med2004; 199: 91–98.

8 Abu-Rish EY, Amrani Y, Browning MJ. Toll-like receptor 9 acti-vation induces expression of membrane-bound B-cell activatingfactor (BAFF) on human B cells and leads to increased proliferationin response to both soluble and membrane-bound BAFF.Rheumatology (Oxford) 2013; 52: 190–201.

9 Koyama T, Tsukamoto H, Miyagi Y, et al. Raised serum APRILlevels in patients with systemic lupus erythematosus. Ann RheumDis 2005; 64: 1065–1067.

10 Becker-Merok A, Nikolaisen C, Nossent HC. B-lymphocyte acti-vating factor in systemic lupus erythematosus and rheumatoidarthritis in relation to autoantibody levels, disease measures andtime. Lupus 2006; 15: 570–576.

11 Zhang J, Roschke V, Baker KP, et al. Cutting edge: A role for Blymphocyte stimulator in systemic lupus erythematosus. J Immunol2001; 166: 6–10.

12 Pers JO, Daridon C, Devauchelle V, et al. BAFF overexpression isassociated with autoantibody production in autoimmune diseases.Ann N Y Acad Sci 2005; 1050: 34–39.

13 Stohl W, Metyas S, Tan SM, et al. B lymphocyte stimulator over-expression in patients with systemic lupus erythematosus:Longitudinal observations. Arthritis Rheum 2003; 48: 3475–3486.

14 Groom J, Kalled SL, Cutler AH, et al. Association of BAFF/BLySoverexpression and altered B cell differentiation with Sjogren’s syn-drome. J Clin Invest 2002; 109: 59–68.

15 Daridon C, Devauchelle V, Hutin P, et al. Aberrant expression ofBAFF by B lymphocytes infiltrating the salivary glands of patientswith primary Sjogren’s syndrome. Arthritis Rheum 2007; 56:1134–1144.

16 Candon S, Gottenberg JE, Bengoufa D, Chatenoud L, Mariette X.Quantitative assessment of antibodies to ribonucleoproteins in pri-mary Sjogren syndrome: Correlation with B-cell biomarkers anddisease activity. Ann Rheum Dis 2009; 68: 1208–1212.

17 Alsaleh G, Messer L, Semaan N, et al. BAFF synthesis by rheuma-toid synoviocytes is positively controlled by alpha5beta1 integrinstimulation and is negatively regulated by tumor necrosis factoralpha and Toll-like receptor ligands. Arthritis Rheum 2007; 56:3202–3214.

18 Moura RA, Canhao H, Polido-Pereira J, et al. BAFF and TACIgene expression are increased in patients with untreated very earlyrheumatoid arthritis. J Rheumatol 2013; 40: 1293–1302.

19 Petri M, Stohl W, Chatham W, et al. Association of plasma Blymphocyte stimulator levels and disease activity in systemiclupus erythematosus. Arthritis Rheum 2008; 58: 2453–2459.

20 George-Chandy A, Trysberg E, Eriksson K. Raised intrathecallevels of APRIL and BAFF in patients with systemic lupus erythe-matosus: Relationship to neuropsychiatric symptoms. Arthritis ResTher 2008; 10: R97.

21 Sanz I, Yasothan U, Kirkpatrick P. Belimumab. Nat Rev DrugDiscov 2011; 10: 335–336.

22 Navarra SV, Guzman RM, Gallacher AE, et al. Efficacy and safetyof belimumab in patients with active systemic lupus erythematosus:A randomised, placebo-controlled, phase 3 trial. Lancet 2011; 377:721–731.

23 Furie R, Petri M, Zamani O, et al. A phase III, randomized, pla-cebo-controlled study of belimumab, a monoclonal antibody thatinhibits B lymphocyte stimulator, in patients with systemic lupuserythematosus. Arthritis Rheum 2011; 63: 3918–3930.

24 Carter RH, Zhao H, Liu X, et al. Expression and occupancy ofBAFF-R on B cells in systemic lupus erythematosus. ArthritisRheum 2005; 52: 3943–3954.

25 Darce JR, Arendt BK, Wu X, et al. Regulated expression ofBAFF-binding receptors during human B cell differentiation. JImmunol 2007; 179: 7276–7286.

26 Zhao LD, Li Y, Smith MF, et al. Expressions of BAFF/BAFFreceptors and their correlation with disease activity in ChineseSLE patients. Lupus 2010; 19: 1534–1549.

27 Kim J, Gross JA, Dillon SR, et al. Increased BCMA expression inlupus marks activated B cells, and BCMA receptor engagementenhances the response to TLR9 stimulation. Autoimmunity 2011;44: 69–81.

28 Hochberg MC. Updating the American College of Rheumatologyrevised criteria for the classification of systemic lupus erythemato-sus. Arthritis Rheum 1997; 40: 1725.

29 Guzman J, Cardiel MH, Arce-Salinas A, et al. Measurement ofdisease activity in systemic lupus erythematosus. Prospective val-idation of 3 clinical indices. J Rheumatol 1992; 19: 1551–1558.

30 Gladman D, Ginzler E, Goldsmith C, et al. The development andinitial validation of the Systemic Lupus International

BAFF, APRIL, BAFF-R, TACI and BCMA expression in SLEDC Salazar-Camarena et al.

10

Lupus

at UNIV DE GUADALAJARA on September 30, 2015lup.sagepub.comDownloaded from

Collaborating Clinics/American College of Rheumatology damageindex for systemic lupus erythematosus. Arthritis Rheum 1996; 39:363–369.

31 Odendahl M, Jacobi A, Hansen A, et al. Disturbed peripheral Blymphocyte homeostasis in systemic lupus erythematosus. JImmunol 2000; 165: 5970–5979.

32 Hostmann A, Jacobi AM, Mei H, Hiepe F, et al. Peripheral B cellabnormalities and disease activity in systemic lupus erythematosus.Lupus 2008; 17: 1064–1069.

33 Cheng Q, Mumtaz IM, Khodadadi L, et al. Autoantibodies fromlong-lived ‘‘memory’’ plasma cells of NZB/W mice drive immunecomplex nephritis. Ann Rheum Dis 2013; 72: 2011–2017.

34 Jacobi AM, Odendahl M, Reiter K, et al. Correlation betweencirculating CD27high plasma cells and disease activity in patientswith systemic lupus erythematosus. Arthritis Rheum 2003; 48:1332–1342.

35 Dorner T, Lipsky PE. Correlation of circulating CD27high plasmacells and disease activity in systemic lupus erythematosus. Lupus2004; 13: 283–289.

36 Wei C, Anolik J, Cappione A, et al. A new population of cellslacking expression of CD27 represents a notable component ofthe B cell memory compartment in systemic lupus erythematosus.J Immunol 2007; 178: 6624–6633.

37 Jacobi AM, Reiter K, Mackay M, et al. Activated memory B cellsubsets correlate with disease activity in systemic lupus erythema-tosus: Delineation by expression of CD27, IgD, and CD95.Arthritis Rheum 2008; 58: 1762–1773.

38 Sellam J, Miceli-Richard C, Gottenberg JE, et al. Decreased B cellactivating factor receptor expression on peripheral lymphocytes

associated with increased disease activity in primary Sjogren’s syn-drome and systemic lupus erythematosus. Ann Rheum Dis 2007; 66:790–797.

39 Zhang M, Ko KH, Lam QL, et al. Expression and function ofTNF family member B cell-activating factor in the developmentof autoimmune arthritis. Int Immunol 2005; 17: 1081–1092.

40 Cancro MP. Signalling crosstalk in B cells: Managing worth andneed. Nat Rev Immunol 2009; 9: 657–661.

41 Koarada S, Tada Y, Suematsu R. Phenotyping of P105-negative Bcells subsets in patients with systemic lupus erythematosus. ClinDev Immunol 2012; 2012: 198206.

42 Coquery CM, Loo WM, Wade NS, et al. BAFF regulates T fol-licular helper cells and affects accumulation and IFNg productionin autoimmunity. Arthritis Rheumatol 2015; 67: 773–784.

43 Jiang C, Loo WM, Greenley EJ, et al. B cell maturation antigendeficiency exacerbates lymphoproliferation and autoimmunity inmurine lupus. J Immunol 2011; 186: 136–147.

44 Hegazy M, Darwish H, Darweesh H, et al. Raised serum level ofAPRIL in patients with systemic lupus erythematosus:Correlations with disease activity indices. Clin Immunol 2010;135: 118–124.

45 Boghdadi G, Elewa EA. Increased serum APRIL differentially cor-relates with distinct cytokine profiles and disease activity in sys-temic lupus erythematosus patients. Rheumatol Int 2014; 34:1217–1223.

46 Treamtrakanpon W, Tantivitayakul P, Benjachat T, et al. APRIL,a proliferation-inducing ligand, as a potential marker of lupusnephritis. Arthritis Res Ther 2012; 14: R252.

BAFF, APRIL, BAFF-R, TACI and BCMA expression in SLEDC Salazar-Camarena et al.

11

Lupus

at UNIV DE GUADALAJARA on September 30, 2015lup.sagepub.comDownloaded from