Printer Friendly

New ELISA for B cell-activating factor.

Substantial progress in the understanding of systemic lupus erythematosus (SLE), [3] rheumatoid arthritis (RA), and primary Sjogren syndrome (SS) occurred with the discovery of 2 new tumor necrosis factor family members. The first of these is referred to as BAFF, for B cell-activating factor of the tumor-necrosis factor family (1), and the second as APRIL, for a proliferation-inducing ligand (2).

Soluble variants of BAFF are the major contributors of its activity, through regulating the survival of B cells (3). The importance of BAFF to B cells was confirmed when mice transfected with the gene BAFF manifested excessive lymphocytic proliferation, together with autoantibody production (4). Further to these observations, excessive quantities of BAFF were found in sera and cerebrospinal fluids of SLE patients (5), sera and synovial fluids (SFs) of RA patients (6), and sera and salivary glands of SS patients (7). Additional evidence that BAFF is involved in autoantibody production includes its upregulation by B-cell depletion in patients with RA (8), and the inverse correlation between serum concentrations of BAFF before depletion of B cells and the length of time before their return in the blood of patients with SS (9).

There exists the issue of why the serum concentrations of BAFF remain within, or below, normal range in a proportion of SLE patients (9). Also intriguing is that increased serum concentrations of BAFF have been associated with the production of rheumatoid factor (RF) by some (10) but not other (11) authors, or with that of anti-DNA antibody by some (12) but not other (13) authors.

In addition, estimates of BAFF fluctuate with changes in inflammatory activity, extent of inflammation, and disease classification criteria chosen by the investigators. Pathological features may also influence serum concentrations of BAFF. Thus, receptor occupancy on B cells by BAFF (14), its urinary excretion in case of renal failure (13), and sequestration within immune complexes (15) all hinder assessment of BAFF. More importantly, flaws or variations in an ELISA might explain variations in the assessment of BAFF. For example, the absorbance is affected by serum concentration of Ig, affinity of the monoclonal Ab (mAb) selected for capturing BAFF, specificity of the polyclonal Ab (pAb) used for its detection, and the quality of the calibrators used for constructing the curves. Another potential for variations is that, due to their 50% homology (1, 2), BAFF and APRIL may compete for binding to anti-BAFF Ab, leading to falsely lower concentrations of BAFF. Adding to the complexity is that posttranslational modifications of BAFF may alter its recognition in the ELISA by the Abs. Indeed, evidence exists that BAFF takes on unpredictable glycoforms (16), alternate A3 (17), intergenic splice variants (18), isoforms presently undefined, and combinations of BAFF and APRIL (19).

In practice, the Abs used in an individual immunoassay kit may bind to 1 variant of BAFF, but not to another, while those used in a second kit mayrecognize this latter variant, but not the former. We set out to address deficiencies in the current ELISA protocols and fixed the thresholds for positivity using 2 different kits at 2 SDs above the mean of 100 normal sera. The values above 2 SDs were scored "positive," and those below, "negative." A large group of sera enabled us to compare the 2 kits. There were as many +/- as -/+ mismatches between the 2 series of results, casting doubts on the suitability of the ELISA protocols for measurement of BAFF. This prompted us to develop our own ELISA for BAFF quantification. Given the promise held out for treatment of SLE, RA, and SS with BAFF antagonists, we made every effort to identify differences in the fluid-phase forms of this cytokine and to describe the variants of BAFF characteristic of each autoimmune disorder.

Materials and Methods


Based on their large use in the commercial kits, 2 anti-BAFF mAbs were chosen based on their selection by most of the manufacturers. They were designated M64 (Peprotech) and 137314 (R&D Systems). There were also 2 anti-BAFF pAbs, designated rabbit (Upstate) or biotinylated rabbit (biot-rabbit) (ProSci) and goat or biot-goat (both from Antigenix). Combining any of these 4 Abs as the capture agent (first member of the pair) with any of them as the detection agent (second member) provided 16 different combinations. These mAbs and pAbs were representative of commercially available Abs, since most of them had been generated using the same recombinant BAFF (rBAFF) or similar synthetic peptides.


The full-length molecule of BAFF is a 285-amino acid (285-aa) peptide with 2 potential N-glycosylation sites at residues 124 and 242. Cleavage downstream from aa 124 releases a first form of nonglycosylated BAFF, and possibly a second form glycosylated at aa 242. The so-called [DELTA]BAFF lacks several amino acids, and thereby unmasks a third glycosylation site, since threonine comes to be the second aa downstream from asparagine (20). Glycoforms of BAFF may hence be generated by alternate splicing.

Most of the commercially available rBAFF is not glycosylated, necessitating that we purify glycosylated BAFF. U937 cells were taken as the source of BAFF, on the basis that myeloid cells release BAFF into culture supernatants (21). We isolated purified BAFF (pBAFF) from U937-conditioned medium by affinity-purification through a M64 column, as described (22). We verified the purity of rBAFF and pBAFF by SDS-PAGE, where Sypro[TM] Ruby (Molecular Probes) stained a single 17-kDa band for rBAFF and a single 28-kDa band for pBAFF.

Each ofthe 2 preparations was electroeluted onto 4 strips of polyvinylidene difluoride (PVDF) membrane (Bio-Rad). After saturation with 5% casein in PBS, each of the 4 anti-BAFF Abs was used to probe the 4 rBAFF and 4 pBAFF strips. The Ab binding was revealed with horseradish peroxidase (HRP)-conjugated donkey antimouse Ig, antirabbit Ig, or antigoat Ig (all from Jackson) and visualized by enhanced chemiluminescence (ECL).


Before SDS-PAGE, proteins were cleared from sera with dye-modified affinity Cibacron blue[TM] (CB) resin F3G-A (23), according to Bio-Rad instructions. The sera were then dialyzed overnight against PBS and incubated for 60 min with 5 vol of CB-coated agarose beads. One aliquot of supernatant was subjected to SDS-PAGE and another kept for ELISA to ensure that BAFF had not been adsorbed onto the beads.

In Western blotting (WB), 1 vol of CB-treated samples was mixed with 1 vol of 30 mmol/L Tris-HCl buffer, containing 4% SDS and 1% mercaptoethanol, boiled for 5 min, and loaded onto the SDS-PAGE. After washes with PBS containing 0.1% Tween 20 (PBST), the strips were probed for 16 h at 4[degrees]C with the indicated Ab and visualized byECL using HRP-conjugated Abs toward mouse, rabbit, or goat Ig.


The antigen used to generate anti-BAFF pAb in rabbit was a 13-aa peptide, whereas those used to raise the murine M64 and 137314 mAbs and the goat anti-BAFF pAb were Escherichia coli- derived BAFF recombinants. We first verified that these 4 Abs recognized glycosylated as well as nonglycosylated BAFF. BAFF purified from culture supernatant was the control for glycosylated BAFF, and rBAFF the control for its non-glycosylated form.

To further determine specificityofthe Abs, pBAFF was deglycosylated. We used anhydrous trifluoromethanesulfanic acid (TFMS) to cleave the glycans, with RNase B as a positive control (24). In a second method, we used the enzyme N-glycosidase F (Boehringer) with serum IgA as an affinity-purified positive control, as described (25).


Structural similarities between BAFF and APRIL are greater than would be indicated by sequence homology; this fact prompted us to test the 4 anti-BAFF Abs with rAPRIL (R&D Systems), using an anti-APRIL mAb (Peprotech) as a positive control in the WB and ELISA. We resolved 2 [micro]g rAPRIL by 10% SDS-PAGE and transferred it to a PVDF membrane. Unbound sites were blocked with 3% nonfat dry milk in PBST, and the strips were incubated overnight at 4[degrees]C with the 4 anti-BAFF Ab and anti-APRIL Ab (Peproteck) at 0.1 mg/L. Binding was detected with the corresponding secondary Ab, revealed by HRP-labeled streptavidin, and quantitated byECL. In the ELISA, we coated microplates overnight at 4[degrees]C with 0.5 mg/L of the captive goat anti-BAFF pAb in carbonate bicarbonate buffer. The wells were saturated with BSA for 3 h, and concentrations of 0.05-50 [micro]g/L of rAPRIL were added. After 2-h incubation at 37[degrees]C, we dispensed 0.1 [micro]g/L detection biot-goat into the wells and left the plates at 37[degrees]C for 90 min.


It was not possible to distinguish between the capture mAb and the detection mAb with the HRP-conjugated antimouse Ab, if the 2 mAbs were paired. Consequently, the 137314 + M64 and M64 + 137314 combinations were discounted from the 16 theoretical combinations. The same principle was applied with regard to the M64/M64, 137314 + 137314, goat + goat, and rabbit + rabbit pairings. In addition, the homology between rabbit and goat IgG could induce cross-reactivity of the second-layer antirabbit IgG Ab with goat IgG and of the second-layer antigoat IgG Ab with rabbit IgG. The rabbit + goat and goat + rabbit pairs were thus believed unsuitable for this assay and provisionally excluded.

These considerations left 8 pairs for the experiments. The view prevails, however, that mAbs are more efficient than pAbs in capturing proteins. For this reason, we initially chose M64 and 137314 for the assessment. Because the quality of Ab coating the plate is critical, however, we kept the following combinations open for consideration: rabbit + 137314, rabbit + M64, goat + 137314, goat + M64, rabbit + biot-goat, goat + biot-rabbit, rabbit + biot-rabbit, and rabbit + biot-rabbit. As a starting point, we evaluated the 4 Abs for their ability to recognize glycosylated and nonglycosylated BAFF.


Microplates were coated overnight at 4[degrees]C with 3 mg/L per well of anti-BAFF Ab dilutions in carbonate-bicarbonate buffer. We performed 3 washes with PBST in between every 2 assay stages. The wells were saturated with a blocking agent for 3 h and flooded with 100 mL serial dilutions of human rBAFF (Peprotech) or with test sera in 1% BSA-supplemented PBST (PBST/BSA).

After 2-h incubation at 37[degrees]C, we dispensed 3 mg/L of anti-BAFF Ab diluted in PBST/BSA into the wells and left the plates at 37[degrees]C for 90 min. After washes, we filled the wells with 0.1 g/L antigoat Ig biot-donkey, antirabbit Ig biot-donkey, or antimouse Ig biot-goat (all from Jackson), depending on the species of the secondary Ab. After 90 min at 37[degrees]C, we added 0.3 mg/L HRP-con[micro]gated streptavidin (Amersham) diluted in PBST/BSA to each well at 37[degrees]C. The reader was blanked with 0.2% BSA-supplemented PBST. We evaluated Polysorp[TM] and Maxisorp[TM] 96-well plates (Nunc) and the polystyrene Costar[TM] plates (Corning) as microplates in this study. We also determined the effectiveness of 2% BSA, 0.6% gelatin, and 1% casein as blocking agents. We randomly selected another 3 serum samples from the SS patient cohort to calculate the intraassay and interassay CVs with 0.6,2.5, and 10 [micro]g/L rBAFF.


We purchased the Quantikine[TM] (R&D Systems) and APO-54N-020[TM] ELISA kits (Apotech) for comparison purposes. The former kit is an ELISA pairing the 137134 anti-BAFF mAb with a goat HRP-conjugated anti-BAFF pAb, whereas the latter is based on the combination of an unknown capture anti-BAFF mouse mAb with an unknown detection anti-BAFF rat mAb, plus HRP-conjugated goat antirat Ig pAb.


We tested RFs using the latex test (Fumouze). When present, they were removed to prevent interactions with IgG in the ELISA. Each anti-BAFF mouse IgG1-containing well was paralleled with another well coated with an irrelevant mouse IgG1 (26). The absorbance values in control wells were subtracted from the corresponding absorbances for the anti-BAFF-coated wells. In a second method (9), 1 vol of serum was incubated with another of rabbit IgG-coated Sepharose 4B beads (Pharmacia). After 60 min, the suspension was centrifuged and the supernatant collected. This was repeated 3 times, and the absence of RF in the supernatant was validated by a negative latex test.


Sera were collected from 68 normal volunteers and 140 patients referred to our hospital and subsequently shown to fulfill the criteria for RA in 28 cases (27), for SLE in 44 cases (28), or for SS in 58 cases (29).In addition, we assessed BAFF in saliva samples from 15 patients with SS with periodontal disease and 15 patients with xerostomia without SS, as well as in SFs from 10 of the previous 28 patients with RA, 8 patients with ankylosing spondylitis, and 9 patients with osteoarthritis. No exclusions were made on any basis other than inabilityto give informed consent. Informed consent was obtained, and the study was approved by the Institutional Review Board at the Brest University.


All triplicate data were expressed as mean (SD). We calculated percent recovery and all mean absorbance SD and % CVs for intraassay and interassay comparisons using Microsoft Excel.

AP value <0.05 was considered statistically significant. The uncorrected P values were provided along with the effect of the Bonferroni correction whenever that correction would remove statistical significance at the P < 0.05 level. We calculated correlations using Spearman's rank correlation and made comparisons using the Mann-Whitney U-test for unpaired data.



On the basis that the coating Abs must recognize all possible forms of the protein being assessed, selection of a capture Ab that binds most forms of BAFF was a prerequisite step toward the development of a reliable ELISA. For this purpose, WB was used for testing the available anti-BAFF Abs. Sera from SS patients were taken as the source of multiple forms of BAFF. Given that most ofthe anti-BAFF Abs had been raised against rBAFF (Table 1), our prediction was that they would bind to this nonglycosylated immunogen, but not to its glycosylated counterpart.

Contrary to expectations (Fig. 1A), M64 bound not only to nonglycosylated or deglycosylated BAFF, but also to glycosylated BAFF. Even worse, 137314 recognized exclusively the glycosylated form of rBAFF. This method was previously validated in our laboratory for its ability to N-deglycosylate IgA (25). Here, there was a shift from 28 to 21 kDa. Presumably, the interval from 21-kDa to 17-kDa rBAFF reveals the presence of O-glycosylation sites.

In contrast, the 2 pAbs bound to both forms of soluble BAFF. Interestingly, rabbit was raised against a synthetic peptide, and hence was capable of recognizing linear sequences of BAFF, irrespective of its glycosylation status. In addition, there was a weak 23-kDa band, possibly due to recognition of residual IgG light chain, as suggested by others (30). Ultimately, goat appeared to be the most efficient in binding the 2 forms of BAFF.

These paradoxical findings were verified (Fig. 1B) by WB analysis ofglycosylated pBAFF and its form deglycosylated by TFMS. The mobility shift of RNase from 15.0 to 13.5 kDa proved that this deglycosylation was effective. On the other hand (Fig. 1C), neither the mAbs nor the pAbs bound to APRIL. Based on these observations, the pAb format was preferred to the mAb format to capture BAFF.


The microtiter plates were Polysorp, and BSA was used as a blocking agent. Checkerboard combinations helped identify the best compromise between adequate sensitivity and specificity for the capture and the detection Abs at 3 mg/L. Because of the low quantity of pBAFF, it could not be used in the ensuing experiments. Although desirable, it would be more difficult to use pBAFF on a routine BAFF, since the molecule is fragile. Both pAbs were good as capture reagents and functioned successively with each of the 2 mAbs added as detection reagents. Generally, poor sensitivity was found for M64 and a high background for 137314, using either the rabbit (Fig. 2A) or the goat pAb (Fig. 2B) as the coating reagent.

Each of the 2 capture pAbs worked well with the other pAb as a detector, yet there was a loss of sensitivity. The capture rabbit pAbs with detection biot-goat were poor in retaining BAFF on the plates, possibly because of the need to be absorbed before being made commercially acceptable (32).

Of the remaining homologous 2 pairs, the combination of 2 rabbit pAbs was relatively insensitive. In contrast, the combination of 2 goat pAbs not only proved sensitive, but also yielded sigmoid-shaped curves with a linear segment (dotted area in Fig. 2B). Consequently, this latest pair of capture goat and detection biot-goat was used thereafter in developing the ELISA. This ELISA was also used to confirm (Fig. 2C) that the anti-BAFF pAbs did not recognize APRIL.



Calibration curves (Fig. 3) showed that the optimal concentration of the capture Ab was 0.5 mg/L with an arbitrarily fixed dose of 0.5 mg/L for the detection biot-Ab (Fig. 3A), whereas the optimal concentration of detection biot-Ab was 0.1 mg/L with an arbitrarily fixed dose of 0.5 mg/L for the capture Ab (Fig. 3B).

We next evaluated the efficiency of different ELISA microplates to retain the goat pAb used for coating. This is reflected by the amount of added BAFF captured by the coating Ab and the consequent amount of bound biot-goat pAb (Fig. 3C). There were fewer molecules of BAFF bound by the coating Ab when the Costar plates were used than when using the high-affinity plates. The 2 high-affinity plates (Polysorp and Maxisorp) produced equivalent dose-response curves; the former was selected because of its slightly higher sensitivity.



Finally, different proteins were examined for their ability to quench free sites. Better dose-response curves were obtained with BSA compared with gelatin (increased background) and casein (decreased sensitivity); BSA was therefore considered optimal for the ELISA.


Intraassay and interassay CVs were calculated by constructing a calibration curve using 0.6,2.5, and 10 [micro]g/L BAFF, and data were expressed as the ratio ofthe SD to the mean of 5 measurements. The intraassay CV ranged from 3.5% to 7.5%, and the interassay CV from 4.8% to 9.7% (see Supplemental Table 1, which accompanies the online version of this article at http://www. These results also showed the lowest detection limit at 3 SD above the mean of 0, 0.1, 0.2, and 0.4 [micro]g/L rBAFF. The background absorbance range of 0.041-0.071 did not cross over the absorbance range of 0.221-0.311 for 0.1 [micro]g/L rBAFF. This low background reflects the setting of the plate reader. The detection limit was thus confirmed to be 100 ng/L in serum. Detection limits were 111 ng/L in saliva and 30 ng/L in SF.

Albeit less precise, a regression analysis was carried out from the linear parts of the curve by plotting the concentrations of rBAFF in [micro] g/L against the logs of [(absorbance with BAFF)--(absorbance without BAFF)] in the sera. As above, the linear part of the curve extended from 0.04 to 4.0 [micro]g/L.


To evaluate the relationship of BAFF and disease, we tested the ELISA by screening sera (32) from 68 normal controls, 44 patients with SLE, 28 patients with RA, and 58 patients with SS. The 3 patient groups displayed higher concentrations than the controls (Table 2). We detected salivary BAFF in 15 cases of SS, but not in 15 cases ofxerostomia without SS, nor in 10 patients with periodontal disease (33). BAFF was also found in 9 of 10 SFs from RA, 6 of 8 SFs from ankylosing spondylitis, and 0 of 9 SFs from osteoarthritis. Interestingly, as described (6), the concentrations of BAFF in the sera of 9 RA patients correlated with those in their SFs (P = 0.041).

We compared the calibration curves for the in-house ELISA with those of 2 commercial kits. The results indicated (Fig. 3D) that our ELISA was suitable for detecting low concentrations of rBAFF. In contrast, the slope of the curve of kit 1 was insufficient to discriminate intermediate values, and the curve generated by kit 2 began at too high concentrations of BAFF to detect low concentrations.

Titers of RF correlated with the absorbances of anti-BAFF Ab-coated wells (Fig. 4A), but not with absorbances ofirrelevant IgG-coated wells. This discrepancy suggests that RF does not compete with BAFF for the capture Ab. Had RF influenced the assay, an inverse correlation would have been observed between RF titers and absorbances for irrelevant IgG.

Two experiments confirmed this interpretation. The first (Fig. 4B) consisted in comparing sera without and with preabsorption with rabbit IgG, the second in testing serum samples twice, once in an anti-BAFF--coated well and once in a well coated with irrelevant IgG, and subtracting the background absorbance from the test absorbance. Clearly, RF interfered less in the in-house than in the commercial ELISAs: absorption of RF reduced the test absorbance by only 14.2% (24.0%), demonstrating that the values had not been nonspecifically augmented by serum RF.



The development of our ELISA for the measurement of BAFF helped provide clues as to why the concentrations of BAFF vary between patients with the same disease using a given assay, and for the same patient using different assays. The success of such a test depends largely on the quality of the Abs selected. In this respect, our most noticeable result was the exclusive binding of 137314 to the glycosylated form of BAFF despite that it had been raised against nonglycosylated BAFF. One explanation is that dissociating conditions of SDS-PAGE destroy the epitopes against which the Abs were produced in response to immunization with 3-dimensional rBAFF. Furthermore, assembly of these molecules into inclusion bodies within E. coli might draw together 2 amino acids originally separated, create artificial N-or O-glycosylation sites in the immunized animal (34), and induce the production of Abs to glycosylated BAFF.

Treatment with N-glycanase ofthe 134-aa 285 domain does not reduce its molecular weight, indicating that the threonine at site 242 is not glycosylated in the main form of BAFF (1). Alternatively, this site might be accessible in a restricted number of proteins. Assuming that some forms are glycosylated, neoconformational epitopes may be generated by glycosylation at this site. Some pAbs present in the assay might detect such epitopes. This is crucial, as carbohydrates prevent unfolding of globular proteins. The 28-kDa version of BAFF could represent this glycosylated form of the cytokine. Parenthetically, it is not surprising that M64 recognized glycosylated, nonglycosylated BAFF, since we selected it to purify BAFF, and since the manufacturer possibly used the glycosylated form of BAFF to raise the Ab response in the mouse.

Heterogeneity in the concentration of circulating BAFF protein found in SLE patients could arise through different routes (34). The correlation between concentrations of BAFF and disease activity remains modest at best (35). The most tenable explanation is that none of the assays provides sensitive and specific means of measurements, as suggested by the observation that disease activity correlates with the number of transcripts for BAFF in blood cells from SLE patients, but not with the protein concentration (15). It is also possible that the lack of increase in BAFF concentrations in the serum of some SLE patients, notwithstanding the activity of their disease, reflects the diminished release of the cytokine by an increased proportion of [DELTA]3BAFF (36).

Another bias might be an excess of nonglycosylated BAFF at the expense of glycosylated BAFF. The absorbance would thus underestimate the concentration of BAFF, because the capture Ab may mainly recognizes glycosylated BAFF. Previous studies have suggested that changes in N-glycan synthesis cause aberrant glycosylation in SLE-prone mice (37). High proportions ofheterotrimers of BAFF and APRIL in SLE patients may also cause underestimation of BAFF (20).

To conclude, the measurement of BAFF is useful for following progress of diseases. Differences in the distribution ofthe forms of BAFF denote the potential of SLE patients to respond to, or to resist, BAFF antagonist therapy (38). Thus reliable indicators for predicting this behavior (39) are becoming increasingly important.

Author Contributions: All authors confirmed they have contributed to the intellectual content of this paper and have met the following 3 requirements: (a) significant contributions to the conception and design, acquisition of data, or analysis and interpretation of data; (b) drafting or revising the article for intellectual content; and (c) final approval of the published article.

Authors' Disclosures of Potential Conflicts of Interest: No authors declared any potential conflicts of interest.

Role of Sponsor: The funding organizations played no role in the design of study, choice of enrolled patients, review and interpretation of data, or preparation or approval of manuscript.

Acknowledgments: The help of Simone Forest and Cindy Sene with the writing of the manuscript is appreciated. Thanks are also due to Prof. Peter M. Lydyard, Westminster University, and Prof. Rizgar A. Mageed, Queen Mary School of Medicine, London, UK, for editorial assistance.


(1.) Schneider P, MacKay F, Steiner V, Hofmann K, Bodmer JL, Holler N, et al. BAFF, a novel ligand of the TNF family, stimulates B cell growth. J Exp Med 1999;189:1747-56.

(2.) Hahne M, Kataoka T, Schroter M, Hofmann K, Irmler M, Bodmer JL, et al. APRIL, a new ligand of the TNF family, stimulates tumor cell growth. J Exp Med 1998;188:1185-90.

(3.) Stadanlick JE, Cancro MP. BAFF and the plasticity of peripheral B cell tolerance. Curr Opin Immunol 2008;20:158-61.

(4.) Mackay F, Woodcock SA, Lawton P, Ambrose C, Baetscher M, Schneider P, et al. Mice transgenic for BAFF develop lymphocytic disorders, along with autoimmune manifestations. J Exp Med 1999;190:1697-710.

(5.) George-Chandy A, Trysberg E, Eriksson K. Raised intrathecal levels of APRIL and BAFF in patients with systemic lupus erythematosus: relationship to neuropsychiatric symptoms. Arthritis Res Ther 2008;10:R97.

(6.) Tan SM, Xu D, Roschke V, Perry JW, Arkfeld DG, Ehresmann GR, et al. Local production of B lymphocyte stimulator protein and APRIL in arthritic joints of patients with inflammatory arthritis. Arthritis Rheum 2003;48:982-92.

(7.) Groom J, Kalled SL, Cutler AH, Olson C, Woodcock SA, Schneider P, et al. Association of BAFF/ BLyS overexpression and altered B cell differentiation with Sjogren's syndrome. J Clin Invest 2002; 109:59-68.

(8.) Cambridge G, Stohl W, Leandro MJ, Migone TS, Hilbert DM, Edwards JC. Circulating levels of B lymphocyte stimulator in patients with rheumatoid arthritis following rituximab treatment: relationships with B cell depletion, circulating antibodies, and clinical relapse. Arthritis Rheum 2006;54:723-32.

(9.) Pers JO, Devauchelle V, Daridon C, Bendaoud B, Le Berre R, Bordron A, et al. BAFF-modulated repopulation of B lymphocytes in the blood and salivary glands of rituximab-treated patients with Sjogren's syndrome. Arthritis Rheum 2007;56: 1464-77.

(10.) Bosello S, Youinou P, Daridon C, Tolusso B, Bendaoud B, Pietrapertosa D, et al. Concentrations of BAFF correlate with autoantibody levels, clinical disease activity, and response to treatment in early RA. J Rheumatol 2008;35:1256-64.

(11.) Becker-Merok A, Nikolaisen C, Nossent HC. B-lymphocyte activating factor in systemic lupus erythematosus and rheumatoid arthritis in relation to autoantibody levels, disease measures and time. Lupus 2006;15:570-6.

(12.) Toubi E, Kessel A, Rosner I, Rozenbaum M, Paran D, Shoenfeld Y. The reduction of serum B-lymphocyte activating levels following quinacrine add-on therapy in systemic lupus erythe matosus. Scand J Immunol 2006;63:299-303.

(13.) Collins CE, Gavin AL, Migone TS, Hilbert DM, Nemazee D, Stohl W. B lymphocyte stimulator (BLyS) isoforms in systemic lupus erythematosus: disease activity correlates better with blood leukocyte BLyS mRNA levels than with plasma BLyS protein levels. Arthritis Res Ther 2006;8:R6.

(14.) Carter RH, Zhao H, Liu X, Pelletier M, Chatham W, Kimberly R, Zhou T. Expression and occupancy of BAFF-R on B cells in systemic lupus erythematosus. Arthritis Rheum 2005;52:3943-54.

(15.) Gao H, Bian A, Zheng Y, Li R, Ji Q, Huang G, et al. sBAFF mutants induce neutralizing antibodies against BAFF. FEBS Lett 2007;581:581-6.

(16.) Bossen C, Schneider P. BAFF, APRIL and their receptors: structure, function and signaling. Semin Immunol 2006;18:263-75.

(17.) Gavin AL, Ait-Azzouzene D, Ware CF, Nemazee D. DeltaBAFF, an alternate splice isoform that regulates receptor binding and biopresentation of BAFF. J Biol Chem 2003;278:38220-8.

(18.) Pradet-Balade B, Medema JP, Lopez-Fraga M, Lozano JC, Kolfschoten GM, Picard A, et al. An endogenous hybrid mRNA encodes TWE-PRIL, a functional cell surface TWEAK-APRIL fusion protein. EMBO J 2002;21:5711-20.

(19.) Roschke V, Sosnovtseva S, Ward CD, Hong JS, Smith R, Albert V, et al. BLyS and APRIL form biologically active heterotrimers that are expressed in patients with systemic immune-based rheumatic diseases. J Immunol 2002;169:4314-21.

(20.) Diao Z, Ye T, Cao P, Zhang J, Mei J, Lin Z, Zhang S. Expression, purification, and characterization of recombinant human soluble BAFF secreted from the yeast Pichia pastoris. Protein Expr Purif 2007;54: 11-7.

(21.) Nardelli B, Belvedere O, Roschke V, Moore PA, Olsen HS, Migone TS, et al. Synthesis and release of B-lymphocyte stimulator from myeloid cells. Blood 2001;97:198-204.

(22.) Le Pottier L, Bendaoud B, Dueymes M, Daridon C, Youinou P, Shoenfeld Y, et al. BAFF, a new target for IVIg in autoimmunity and cancer. J Clin Immunol 2007;27:257-65.

(23.) Zolotarjova N, Martosella J, Nicol G, Bailey J, Boyes BE, Barrett WC. Differences among techniques for high-abundant protein depletion. Pro teomics 2005;5:3304-13.

(24.) Patel T, Bruce J, Merry A, Bigge C, Wormald M, Jacques A, et al. Use of hydrazine to release in intact and unreduced form both N- and O-linked oligosaccharides from glycoproteins. Biochemistry 1993;32:679-93.

(25.) Basset C, Durand V, Mimassi N, Pennec YL, Youi nou P, Dueymes M. Enhanced sialyltransferase activity in B lymphocytes from patients with primary Sjogren's syndrome. Scand J Immunol 2000; 51:307-11.

(26.) Mariette X, Roux S, Zhang J, Bengoufa D, Lavie F, Zhou T, et al. The level of BLyS (BAFF) correlates with the titre of autoantibodies in human Sjogren's syndrome. Ann Rheum Dis 2003;62: 168-71.

(27.) A committee of the American Rheumatism Association. 1958 revision of diagnostic criteria for rheumatoid arthritis: by a committee of the American Rheumatism Association. Arthritis Rheum 2008;58(2 Suppl):S15-9.

(28.) Hochberg MC Updating the American College of Rheumatology revised criteria for the classification of systemic lupus erythematosus. Arthritis Rheum 1997;40:1725.

(29.) Vitali C, Bombardieri S, Jonsson R, Moutsopoulos HM, Alexander EL, Carsons SE, et al. Classification criteria for Sjogren's syndrome: a revised version of the European criteria proposed by the American-European Consensus Group. Ann Rheum Dis 2002;61:554-8.

(30.) Zhang J, Roschke V, Baker KP, Wang Z, Alarcon GS, Fessler BJ, et al. Cutting edge: a role for B lymphocyte stimulator in systemic lupus erythematosus. J Immunol 2001;166:6-10.

(31.) Joshi KS, Hoffmann LG, Butler JE. The immunochemistry of sandwich ELISAs--V. the capture antibody performance of polyclonal antibody-enriched fractions prepared by various methods. Mol Immunol 1992;29:971-81.

(32.) Pers JO, Daridon C, Devauchelle V, Jousse S, Saraux A, Jamin C, Youinou P. BAFF overexpression is associated with autoantibody production in autoimmune diseases. Ann N Y Acad Sci 2005; 1050:34-9.

(33.) Pers JO, d'Arbonneau F, Devauchelle-Pensec V, Saraux A, Pennec YL, Youinou P. Is periodontal disease mediated by salivary BAFF in Sjogren's syndrome? Arthritis Rheum 2005;52:2411-4.

(34.) Cao P, Mei JJ, Diao ZY, Zhang S Expression, refolding, and characterization of human soluble BAFF synthesized in Escherichia coli. Protein Expr Purif 2005;41:199-206.

(35.) Stohl W. Systemic lupus erythematosus: a bliss-less disease of too much BLyS (B lymphocyte stimulator) protein. Curr Opin Rheumatol 2002; 14:522-8.

(36.) Gavin AL, Duong B, Skog P, Ait-Azzouzene D, Greaves DR, Scott ML, Nemazee D. DeltaBAFF, a splice isoform of BAFF, opposes full-length BAFF activity in vivo in transgenic mouse models. J Immunol 2005;175:319-28.

(37.) Hashii N, Kawasaki N, Itoh S, Nakajima Y, Kawanishi T, Yamaguchi T. Alteration of N-glycosylation in the kidney in a mouse model of systemic lupus erythematosus: relative quantification of N-glycans using an isotope-tagging method. Immunology 2008;126:336-45.

(38.) Furie R, Stohl W, Ginzler EM, Becker M, Mishra N, Chatham W, et al. Biologic activity and safety of belimumab, a neutralizing anti-B-lymphocyte stimulator (BLyS) monoclonal antibody: a phase I trial in patients with systemic lupus erythematosus. Arthritis Res Ther 2008;10:R109.

(39.) Youinou P. Is BAFF the murderer in lupus? Lupus 2008;17:613-4.

Laetitia Le Pottier, [1] Boutahar Bendaoud, [2] Yves Renaudineau, [1,2] Pierre Youinou, [1,2] * Jacques-Olivier Pers, [1,2] and Capucine Daridon [1]

[1] EA 2216 "Immunologie et Pathologie" and IFR 418 ScInBioS, Universitede Brest, Brest, France, and University Europeenne de Bretagne, Rennes, France; [2] Brest University Medical School Hospital (CHU Augustin Morvan), Brest, France.

[3] Nonstandard abbreviations: SLE, systemic lupus erythematosus; RA, rheumatoid arthritis; SS, Sjogren syndrome; BAFF, B cell-activating factor; APRIL, proliferation-inducing ligand; SF, synovial fluid; RF, rheumatoid factor; mAb, monoclonal antibody; pAb, polyclonal antibody; rBAFF, recombinant BAFF; biot, biotinylated; aa, amino acid; pBAFF, purified BAFF; PVDF, polyvinylidene difluoride; HRP, horseradish peroxidase; ECL, enhanced chemiluminescence; CB, Cibacron blue; WB, Western blotting; PBST, PBS containing 0.1% Tween 20; TFMS, trifluoromethanesulfanic acid.

* Address correspondence to this author at: Laboratory of Immunology, Brest University Medical School Hospital, BP824, F29609, Brest, France. Fax +33298-22-38-47; e-mail

Received April 30, 2009; accepted July 23, 2009.

Previously published online at DOI: 10.1373/clinchem.2009.129940
Table 1. Unconjugated monoclonal and polyclonal
anti-BAFF Abs used in the study.

Ab Specificity Immunogen Origin Isotype

 M64 Human rBAFF Mouse IgG1
 137314 Human rBAFF Mouse IgG1
 Rabbit Human and aa 254-269 Rabbit IgG
 Goat Human rBAFF Goat IgG


Table 2. Concentrations of BAFF in serum, saliva,
and SF of patients with SLE, RA, SS, periodontal
disease, xerostomia, ankylosing spondylitis,
and osteoarthritis.

 Mean BAFF
Sample and Samples, concentration Positive,
population n (SD), [micro]g/L n (%)

 SLE 44 10.6 (8.5) 23 (52) (a)
 RA 28 9.7 (1.5) 5(18) (b)
 SS 58 15.8(12.9) 35 (51)
 Normal 68 4.8 (3.8)
 SS 15 7.4 (2.1) 10 (67) (c)
 Periodontal disease 10 2.2 (0.6) 5(50)
 Xerostomia 15 1.0 (0.4) 4(27)
 RA 10 13.2 (3.5) 9(90)
 Ankylosing spondylitis 8 11.7 (3.9) 6(75)
 Osteoarthritis 9 5.9(1.1) 0(0)

(a) P < 0.001 and (b) P = 0.017 compared with normal controls.

(c) P = 0.0016 compared with both groups of disease controls.
COPYRIGHT 2009 American Association for Clinical Chemistry, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2009 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:Clinical Immunology
Author:Pottier, Laetitia Le; Bendaoud, Boutahar; Renaudineau, Yves; Youinou, Pierre; Pers, Jacques-Olivier;
Publication:Clinical Chemistry
Date:Oct 1, 2009
Previous Article:Increased tissue factor expression is associated with reduced survival in non-small cell lung cancer and with mutations of TP53 and PTEN.
Next Article:Methylation status and neurodegenerative markers in parkinson disease.

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters