Anti-CD83 Ab capable of Ab-dependent cellular cytotoxicity can deplete activated CD83+ human dendritic cells, thereby inhibiting CD4 T cell–mediated acute graft-versus-host disease. As CD83 is also expressed on the surface of activated B lymphocytes, we hypothesized that anti-CD83 would also inhibit B cell responses to stimulation. We found that anti-CD83 inhibited total IgM and IgG production in vitro by allostimulated human PBMC. Also, Ag-specific Ab responses to immunization of SCID mice xenografted with human PBMC were inhibited by anti-CD83 treatment. This inhibition occurred without depletion of all human B cells because anti-CD83 lysed activated CD83+ B cells by Ab-dependent cellular cytotoxicity and spared resting (CD83) B cells. In cultured human PBMC, anti-CD83 inhibited tetanus toxoid–stimulated B cell proliferation and concomitant dendritic cell–mediated CD4 T cell proliferation and expression of IFN-γ and IL-17A, with minimal losses of B cells (<20%). In contrast, the anti-CD20 mAb rituximab depleted >80% of B cells but had no effect on CD4 T cell proliferation and cytokine expression. By virtue of the ability of anti-CD83 to selectively deplete activated, but not resting, B cells and dendritic cells, with the latter reducing CD4 T cell responses, anti-CD83 may be clinically useful in autoimmunity and transplantation. Advantages might include inhibited expansion of autoantigen- or alloantigen-specific B cells and CD4 T cells, thus preventing further production of pathogenic Abs and inflammatory cytokines while preserving protective memory and regulatory cells.

The type 1 transmembrane glycoprotein CD83 is constitutively expressed on thymic epithelium, where it plays a key role in the development of CD4+ T cells (1). In the periphery, CD83 is expressed on activated, but not resting, dendritic cells (DC), B cells, and CD4 T cells, in association with increased MHC class II (MHC II) and CD86 expression (2). In the mouse it has been reported that the transmembrane domain of CD83 stabilizes MHC II and CD86, but not MHC class I and CD40, on the plasma membrane by binding to the ubiquitin ligase MARCH1 and blocking the ubiquitination and subsequent degradation of MHC II and CD86 (3). However, small interfering RNA knockdown of CD83 on human DC did not affect MHC II and CD86 cell surface expression levels, but T cell allostimulatory capacity was reduced nonetheless (4). The cytoplasmic domain of CD83 has no familiar signaling motifs (5), but interaction with the Golgi protein GRASP55 is reported to be required for its glycosylation and surface expression on human DC (6). A number of potential ligands that bind CD83 have been suggested (7), and its homotypic interaction has been reported to regulate mucosal DC activation (8).

We have previously shown that rabbit polyclonal IgG anti-human CD83 (RA83) inhibits human CD4 T cell proliferation and IFN-γ expression in the allogeneic MLR (9, 10). The mechanism of action of RA83 involves lytic depletion of activated CD83+ DC and possibly, but to a lesser degree, of alloresponding CD83dim CD4 T cell blasts by NK cell–mediated Ab-dependent cellular cytotoxicity (ADCC) (9). Prophylactic treatment of human PBMC-xenografted SCID mice with RA83 prevents acute graft-versus-host disease (GVHD) in this human DC and CD4 T cell–dependent model, suggesting a new approach to the control of acute GVHD in clinical allogeneic hematopoietic cell transplantation (10). To this end, we developed a fully human IgG1 anti-human CD83 mAb, designated 3C12, which, similar to RA83, lyses CD83+ target cells by ADCC (11). To improve efficacy, yeast display technology was used to affinity mature mAb 3C12 by L chain shuffling to produce mAb 3C12C, which we confirmed prevents GVHD in the xenograft model with similar efficacy to polyclonal RA83 (12).

B cells are central to both protective and pathogenic adaptive humoral immunity. Polyclonal reagents such as CD40L (13), anti-CD40, LPS, and anti-IgM (14) stimulate murine B cells to express CD83 on the cell surface, and Ag induces CD83 on Ag-specific murine B cells in vivo (14). Human B cells have been reported to upregulate cell surface CD83 when stimulated with PMA (15), CpG (16), and infectious bacteria (17). We hypothesized that ADCC-capable anti-human CD83 Ab would inhibit human B cell responses to stimulation, either by directly targeting and depleting activated CD83+ B cells and/or indirectly by depleting activated CD83+ DC and thereby reducing CD4 Th cell responses, which strongly influence the magnitude (Ab titer) and quality (affinity for Ag and Ab class/isotype) of B cell responses to stimulation. A potential benefit might be that anti-CD83 mAb finds clinical utility in conditions involving pathogenic B cells, such as rheumatoid arthritis (RA), systemic lupus erythematosus (SLE), Ab-mediated transplant rejection, and chronic GVHD. Compared to current therapies, the simultaneous inhibition of autoantigen- or alloantigen-specific Ab responses and attendant CD4 T cell responses (particularly inflammatory cytokine expression) could have additional, independent benefits in these conditions, for which DC and CD4 T cells as well as B cells have each been shown to contribute to pathogenesis (1821).

Depletion of B cells with the anti-CD20 mAb rituximab can control Ab-mediated autoimmune conditions and Ab-mediated rejection (22, 23). Rituximab can induce long-lasting depletion (6–12 mo) of nearly all B cells in the circulation, including memory B cells, but plasma cells (PC), which are CD20, derived from CD20+ B cells prior to rituximab treatment, continue to express high levels of circulating Igs. However, hypogammaglobulinemia can occur and, in a recent audit, was found in 24% of rituximab-treated patients along with a high infection rate and reduced pathogen-specific Ab levels and impaired humoral immune responses to vaccination (24). Although B cells and Abs play important pathogenic roles in RA, SLE, and renal graft rejection, rituximab therapy does not benefit all patients with these conditions (22, 25, 26). This may be due to ongoing action of pathogenic Abs expressed by long-lived PC (27, 28), to incomplete depletion of pathogenic B cells (29), or to ongoing direct action of pathogenic effector CD4 T cells (19, 20, 30).

In this study, we report the effects of anti-human CD83 Abs on B cell and CD4 T cell responses to Ag stimulation in human PBMC in vitro and in a human PBMC/SCID mouse xenograft vaccination model. Three different ADCC-capable anti-human CD83 Abs were used: the abovementioned rabbit polyclonal RA83 (10), the abovementioned human IgG1 3C12C mAb (12), and a nonfucosylated form (3C12Ckif) with enhanced ADCC capability (31, 32).

RA83 and rabbit negative control IgG (RAneg) were prepared as previously described (9, 10). Recombinant human IgG1 anti-CD83 mAb 3C12C and a human IgG4 isotype version of 3C12C were prepared as described (11, 12). Trastuzumab (Herceptin; Roche) was used as the human IgG1 isotype control mAb. To improve ADCC efficacy, the affinity of the Fc region of 3C12C mAb for the CD16 FcR expressed on human NK cells was enhanced by preventing normal fucosylation of the N-linked oligosaccharide at Asn297 of the IgG1 H chain (31). This was done using the α-mannosidase I inhibitor kifunensine to prepare nonfucosylated 3C12C mAb, denoted 3C12Ckif (33, 34). mAb 3C12Ckif was expressed by culturing Chinese hamster ovary cells transfected with the mAbXpress plasmids (11) encoding 3C12C in medium containing kifunensine (2 μg/ml; GlycoFineChem, Wellington, New Zealand) (34). Conditioned media were harvested after 6–7 d of standard culture and the mAb was purified as described (11).

These Abs included rituximab (MabThera from Roche, Rituxan from Genentech, and chimeric human IgG1 anti-human CD20 from IDEC Pharmaceuticals), alemtuzumab (Campath-1H; human IgG1 anti-human CD52; Schering Health Care), and anti-human CD16 function–blocking mAb 3G8 (murine IgG1; BD Biosciences).

Blood was obtained with informed consent from normal healthy donors either as whole-blood donations or as leukapheresis product (approved by the Mater Human Research Ethics Committee; no. 1407AP). PBMC were prepared by Ficoll-Hypaque centrifugation and used fresh or cryopreserved and stored over liquid nitrogen. NK cells, DC, and B cells were depleted or isolated from PBMC using anti-CD56, CD11c, or CD19 microbeads, respectively (Miltenyi Biotec, Bergisch Gladbach, Germany) according to the supplier’s instructions. Cells were cultured in RPMI 1640 supplemented with 10% heat-inactivated FCS (hiFCS), penicillin, streptomycin, GlutaMAX, and 2-ME (Life Technologies). CD40L-transfected L cells (35) were expanded in 10% hiFCS in DMEM (Life Technologies).

Human PBMC from two donors were cultured together, each at 106/ml, in 10% hiFCS/RPMI 1640 at 37°C in 5% CO2. On day 0, cultures were treated with Nil, anti-CD83, or negative control Ab and/or anti-human CD16 function–blocking mAb 3G8. After 8 d of culture, conditioned media were assayed for total human IgM and IgG by ELISA.

A human PBMC-xenografted SCID mouse model was used to test the in vivo efficacy of anti-CD83 Abs. Briefly, on the day before xenografting, anti-asialo GM1 Ab (ASGM1; Wako Chemicals) was injected i.p. into female SCID mice and total body irradiation (275 cGy) was delivered using a [137Cs] irradiator (Gammacell 40). The next day, human PBMC (30 × 106) were injected i.p. Three days later the mice were treated with RA83 or RAneg Ab (0.125 mg/mouse, i.p.) and, 3 h later, vaccinated i.p. with 200 μl of tetanus toxoid (TT; Equivac T vaccine; Pfizer) or keyhole limpet hemocyanin (KLH; Calbiochem) in CFA/IFA (1:20 v/v) (36). Blood was taken by cardiac puncture on day 13 postxenograft and the mice were then sacrificed and spleens and peritoneal washings taken for flow cytometric analysis. Sera prepared from blood were assayed for total and Ag-specific (TT; KLH) human IgM and IgG by ELISA. Animal experimental procedures were approved by the University of Queensland Animal Ethics Committee (no. MMRI/149/11/CCQ/NHM).

For total IgM and IgG assays, Nunc Immuno plate wells (Thermo Fisher Scientific) were coated with 2 μg/ml mouse anti-human IgM or IgG mAb (BD Biosciences) and blocked with 1% BSA in PBS. For Ag-specific Ig assays, the plate wells were coated with 5 μg/ml TT (CSL, Melbourne, VIC, Australia) or 20 μg/ml KLH. Diluted samples were incubated in the wells for 2 h at room temperature. Between all steps, plates were washed three times with PBS. Bound Ab was detected with HRP-conjugated goat anti-human IgM or IgG (The Jackson Laboratory). Colorimetric reaction was produced by final incubation with SIGMAFAST o-phenylenediamine dihydrochloride (Sigma-Aldrich). OD was measured at 490 nm (iMark microplate reader; Bio-Rad Laboratories).

To assess the effects of anti-CD83 Ab on total IgM and IgG production by allostimulated PBMC, each culture supernatant sample was diluted 50, 25, 12.5, 6.25%, and so forth for ELISA analysis. Total IgM or IgG in each anti-CD83–treated culture was normalized by expression as percentage of that for the appropriate negative control–treated culture, using means of duplicate ELISA OD490 nm values. For each anti-CD83–treated culture, the apparent dilution for midrange OD490 nm means (i.e., approximately linear region of OD490 nm versus dilution curve) was read from the corresponding negative control mean OD490 nm versus dilution curve. The apparent dilution (Dapp) readings were then corrected for actual dilution (Dact): normalized total IgM or IgG = 100%[Dapp/Dact].

Except where stated otherwise, flow cytometry Abs were purchased from BD Pharmingen (San Jose, CA), including appropriate isotype controls. The following fluorophore-conjugated mouse anti-human Abs were used: CD4–Pacific Blue (PB) or -allophycocyanin-Cy7 (RPA-T4), CD11b-allophycocyanin-Cy7 (ICRF44), CD14-PE-Cy7 (M5E2), CD19-PB (SJ25-C1; Invitrogen), CD20-V450 (L27), CD27-allophycocyanin (M-T271), CD43-FITC (1G10), CD45-allophycocyanin (2D1), CD83-FITC, -PE, -PE-Cy7, or -allophycocyanin (HB15e), HLA-DR–FITC or –allophycocyanin (G46-6), CD86-PE (2331-FUN1). Cells were washed in MACS buffer containing PBS, 2 mM EDTA, 0.5% BSA, and 0.09% sodium azide. Events were acquired using a BD LSR II (BD Biosciences) or CyAn ADP analyzer (Dako), and FlowJo 8.8.7 was used for analysis. Live lymphocytes in CFSE- and/or fluorophore-conjugated Ab–stained PBMC were identified by low forward scatter and side scatter gating followed by pulse width and forward scatter area for singularity. Live cells were then gated on CD4 to identify CD4 T cells, CD20 or CD19 to identify B cells, and CD11c for DC. Proliferation of CD4 T cells and CD19 B cells was analyzed by CFSE dilution.

In the CD83 kinetics study, human PBMC were cultured with anti-IgG plus anti-IgM (2 μg/ml; BD Pharmingen), CD40L-transfected L cells (CD40L expression confirmed by flow cytometry; 20:1 PBMC/L cell ratio), LPS (2 μg/ml; Sigma-Aldrich) or αδ-dex (10 ng/ml; Fina Biosolutions) for up to 6 d. For experiments investigating B cell and CD4 T cell proliferative responses to Ag stimulation, CFSE-labeled human PBMC (1 μM CFSE, modified from Refs. 37, 38) were cultured in 10% FCS/RPMI 1640 and stimulated with TT (10 μg/ml) in the presence of anti-CD83 (RA83 or mAb 3C12Ckif), rituximab (anti-CD20), or negative control Ab (RANeg or human IgG1 isotype) (1 μg/ml) for 6 d. Proliferation was assessed by CFSE dilution using flow cytometry after staining the cells with CD19-PB and CD4-allophycocyanin-H7.

Primary B cells purified by negative selection were stimulated to express CD83 by 2 d culture with PWM (100 μg/ml) or with CD40L-transfected L cells (10:1 B cell/L cell ratio). As described (9, 11), CD83+ target cells were labeled with 51Cr-Na2CrO4 (Amersham Biosciences) and cultured at 2000 B cell targets per well with either IL-2–activated purified NK effector cells (10,000 per well) or with overnight cultured autologous whole PBMC (100,000 per well) and test or control Ab for 4 h. 51Cr (cpm) released into the culture supernatants due to the action of effector cells was expressed as percentage of that released by Triton X-100.

Cytokines released into media by PBMC (2 × 106/ml) stimulated with TT (10 μg/ml) in 6 d of culture were assayed by the Bio-Plex Pro human cytokine eight-plex immunoassay kit (IL-2, IL-4, IL-6, IL-8, IL-10, IFN-γ, TNF-α and GM-CSF; Bio-Rad Laboratories) and ELISA kits for IL-17A (eBioscience, human IL-17A [homodimer] ELISA Ready-SET-Go!), IL-6 (BD Biosciences), TNF-α (BD Biosciences), and IL-21 (human IL-21 ELISA MAX Deluxe; BioLegend).

CD4 T cell intracellular cytokine expression was assayed by culture of CFSE-labeled human PBMC stimulated with TT (10 μg/ml) for 6 d, with PMA (20 ng/ml; Sigma-Aldrich) and ionomycin (1 μM; Sigma-Aldrich) for the final 6 h and monensin (10 μg/ml, GolgiStop; BD Biosciences) for the final 4 h of culture. The cells were then washed, surface stained with mAbs CD19-PB and CD4-allophycocyanin-Cy7, fixed, permeabilized, and stained with an allophycocyanin conjugate of either anti–IFN-γ, TNF-α, IL-4, IL-10, IL-17a (BioLegend), or with anti–IL-21-Alexa Fluor 647 (BD Biosciences) or isotype controls.

GraphPad Prism version 5 for Windows (GraphPad Software, San Diego, CA) was used for statistical analyses.

Neither polyclonal rabbit IgG anti-CD83 (RA83) nor the human IgG1 anti-CD83 mAb 3C12C inhibited total IgM and IgG production by purified CD83+ B cells stimulated with PWM (data not shown). However, both anti-CD83 Abs at 1 μg/ml inhibited total IgM and IgG production in vitro by allostimulated PBMC (Fig. 1, p < 0.01). Relative to negative control Abs, mAb 3C12C inhibited total IgM production by 96.1% whereas RA83 inhibited total IgM and IgG production by 90.8 and 68.1%, respectively.

FIGURE 1.

Anti-CD83 inhibits total human IgM and IgG production in vitro by allostimulated PBMC. Each dot on graph shows total IgM or IgG in 8-d culture supernatants treated with anti-CD83 (RA83 or mAb 3C12C, 1 μg/ml) or Nil Ab. Each result is expressed as a percentage of that from the corresponding negative control Ab–treated cultures in the same experiment (RAneg or IgG1 isotype control, 1 μg/ml). Diluted culture supernatants were analyzed for total human IgM and IgG by ELISA. The Nil treatment results reveal some experiment/donor-dependent inhibitory or stimulatory effects of negative control Abs on total IgM and IgG production although these did not obscure the inhibitory effects of anti-CD83 Abs. Total human IgG production could not be assessed in the presence of the humanized IgG1 3C12C or isotype control mAbs. Data are from four independent experiments, each using a distinct pair of donor PBMC. Data were analyzed by repeated measures ANOVA (p < 0.0001) and the Bonferroni multiple comparison test (**p < 0.01, ***p < 0.001).

FIGURE 1.

Anti-CD83 inhibits total human IgM and IgG production in vitro by allostimulated PBMC. Each dot on graph shows total IgM or IgG in 8-d culture supernatants treated with anti-CD83 (RA83 or mAb 3C12C, 1 μg/ml) or Nil Ab. Each result is expressed as a percentage of that from the corresponding negative control Ab–treated cultures in the same experiment (RAneg or IgG1 isotype control, 1 μg/ml). Diluted culture supernatants were analyzed for total human IgM and IgG by ELISA. The Nil treatment results reveal some experiment/donor-dependent inhibitory or stimulatory effects of negative control Abs on total IgM and IgG production although these did not obscure the inhibitory effects of anti-CD83 Abs. Total human IgG production could not be assessed in the presence of the humanized IgG1 3C12C or isotype control mAbs. Data are from four independent experiments, each using a distinct pair of donor PBMC. Data were analyzed by repeated measures ANOVA (p < 0.0001) and the Bonferroni multiple comparison test (**p < 0.01, ***p < 0.001).

Close modal

Functional Fc receptor blockade by the anti-human CD16 mAb 3G8 reduced the inhibitory effect of RA83 on total IgM and IgG production by allostimulated PBMC (Fig. 2A). Similarly, 3G8 reduced the inhibitory effect of anti-CD83 mAb 3C12C on total IgM production (Fig. 2B). In the absence of CD16 blockade by 3G8, anti-CD83 mAb 3C12C expressed as IgG4, an isotype with reduced ADCC capability compared with IgG1 (39), failed to inhibit total IgM production (Fig. 2B). These observations are consistent with a major contribution from CD16-mediated ADCC to the mechanism of action of anti-CD83 Abs in inhibiting Ig production, as shown previously for allogeneic T cell proliferation and GVHD (9, 10, 12).

FIGURE 2.

CD16-mediated ADCC contributes to the mechanism by which anti-CD83 inhibits total IgG and IgM production by allostimulated PBMC in vitro. (A) RA83 was less effective in inhibiting total human IgM and IgG production by allostimulated PBMC in the presence of CD16 function–blocking mAb 3G8 at 20 μg/ml (Mann–Whitney U test). (B) Anti-CD83 mAb 3C12C (IgG1) was also less effective in inhibiting IgM production in the presence of CD16-blocking mAb 3G8 in allostimulated PBMC cultures, as was mAb 3C12C expressed as IgG4 in the absence of CD16 blockade by 3G8 (Friedman test with Dunn multiple comparisons). Graphs show data from n ≥ 5 experiments, each using a distinct pair of donor PBMC. Each data point is the mean of duplicate ELISA absorbance values (OD[490 nm]). Data points joined with lines are from the same experiment/pair of donors.

FIGURE 2.

CD16-mediated ADCC contributes to the mechanism by which anti-CD83 inhibits total IgG and IgM production by allostimulated PBMC in vitro. (A) RA83 was less effective in inhibiting total human IgM and IgG production by allostimulated PBMC in the presence of CD16 function–blocking mAb 3G8 at 20 μg/ml (Mann–Whitney U test). (B) Anti-CD83 mAb 3C12C (IgG1) was also less effective in inhibiting IgM production in the presence of CD16-blocking mAb 3G8 in allostimulated PBMC cultures, as was mAb 3C12C expressed as IgG4 in the absence of CD16 blockade by 3G8 (Friedman test with Dunn multiple comparisons). Graphs show data from n ≥ 5 experiments, each using a distinct pair of donor PBMC. Each data point is the mean of duplicate ELISA absorbance values (OD[490 nm]). Data points joined with lines are from the same experiment/pair of donors.

Close modal

We tested the effect of anti-CD83 on Ag-specific Ab responses in vivo. Three days after xenografting with human PBMC, SCID mice were treated with anti-CD83 or negative control Ab and immunized with TT or KLH. The mice were sacrificed on day 13 postxenograft and blood samples were taken for total and Ag-specific human Ig assays. Although anti-CD83 (RA83) had no effect on total circulating human IgG levels, it had significantly reduced day 10 postimmunization TT-specific IgG levels (Fig. 3A). At this time, in preliminary experiments, we found that KLH immunization stimulated Ag-specific human IgM but not IgG responses (data not shown). RA83 treatment significantly reduced KLH-specific IgM levels on day 10 postimmunization, and total circulating human IgM was slightly reduced, although it was not statistically significant (Fig. 3B).

FIGURE 3.

Anti-CD83 treatment inhibits Ag-specific Ab responses in vivo. Human PBMC-xenografted SCID mice were treated on day 3 postxenograft with anti-CD83 (RA83, 0.125 mg/mouse) or with negative control Ab (RAneg) and immunized 3 h later with (A) TT or (B) KLH. The mice were sacrificed on day 13 postxenograft and serum samples were assayed (A) for total and TT-specific IgG or (B) for total and KLH-specific IgM. Each data point is the mean of duplicate ELISA absorbance values (OD[490 nm]). Horizontal lines indicate median values for each treatment (n = 6 per group, combined data shown from two independent donors/experiments, three mice each). The effects of Ab treatments were compared by the Mann–Whitney U test.

FIGURE 3.

Anti-CD83 treatment inhibits Ag-specific Ab responses in vivo. Human PBMC-xenografted SCID mice were treated on day 3 postxenograft with anti-CD83 (RA83, 0.125 mg/mouse) or with negative control Ab (RAneg) and immunized 3 h later with (A) TT or (B) KLH. The mice were sacrificed on day 13 postxenograft and serum samples were assayed (A) for total and TT-specific IgG or (B) for total and KLH-specific IgM. Each data point is the mean of duplicate ELISA absorbance values (OD[490 nm]). Horizontal lines indicate median values for each treatment (n = 6 per group, combined data shown from two independent donors/experiments, three mice each). The effects of Ab treatments were compared by the Mann–Whitney U test.

Close modal

It is not clear in the above experiments whether anti-CD83 targets B cells directly or whether the inhibitory effects on Ig production are mediated via activated CD83+ DC and CD4 Th cells, both of which are known targets of CD83 Ab treatment (9, 12). To address this question, we first characterized CD83 expression by human B cells.

When analyzed ex vivo, only 0.91% (0.50% SEM, n = 6) of CD19 B cells in freshly isolated healthy donor PBMC expressed CD83 on the cell surface. After 6 h of unstimulated culture, 19.4% (SEM 2.6%, n = 6) of CD19 B cells expressed CD83. This had decreased to 11.3% (SEM 1.5%) after 24 h (“Nil” stimulation in Fig. 4A). In separate experiments, we found that all B cell subsets expressed CD83 at this time, although the putative human CD11b+ B1 subset (40, 41) had the highest percentage of CD83+ cells (Fig. 4B). A range of polyclonal stimulants upregulated CD83 on B cells in PBMC (Fig. 4A). LPS and CD40L each rapidly upregulated CD83 on B cells and on LinHLA-DR+ DC in the same cultures (Fig. 4A, 4C). The specific recall Ag TT slightly increased B cell CD83 expression above unstimulated control levels (Fig. 4A; at 6 h of TT stimulation, 26.5 ± 3.7% SEM of B cells were CD83+ vs Nil stimulation, 20.2 ± 4.8% SEM; p < 0.05, n = 3, paired t test). Expression was sustained for >18 h by >80% of all B cells in PBMC stimulated via CD40 (by CD40L-transfected L cells or by T cells stimulated with anti-CD3/CD28 in PBMC, thereby inducing CD40L expression by CD4 T cells) (42); otherwise, CD83 expression fell after 6–12 h of stimulation, except when stimulated with dextran-conjugated anti-IgD (αδ-dex). LPS induced CD83 expression by a maximum of 60% of B cells, and the BCR stimulants anti-IgG F(ab′)2, anti-IgM F(ab′)2, and αδ-dex stimulated maxima of 40, 80, and 55% of B cells to express CD83, respectively (Fig. 4A).

FIGURE 4.

Human B cells rapidly upregulate CD83 when stimulated and expression is sustained by CD40 ligation. (A) Graphs show mean percentages of CD19 B cells that expressed CD83 on the cell surface (±SEM, n = 3 experiments/donors) for up to 48 h following stimulation. Freshly isolated human PBMC were cultured with Nil, TT (10 μg/ml), anti-IgG or anti-IgM (2 μg/ml), LPS (2 μg/ml), anti-CD3 plus anti-CD28 (each 1 μg/ml), αδ-dex (10 ng/ml), or CD40L-transfected L cells (L cell/PBMC ratio 1:20). CD19 B cell CD83 expression was assessed by flow cytometry. (B) CD83 expression on B cell subsets in PBMC after 24 h of unstimulated culture. Dots on graph show percentage of cells in each B cell subset that express CD83 for each of four PBMC donors. (Friedman nonparametric comparison of means of matched groups with Dunn posttest). Surface phenotypes include naive B cells (CD20+CD43CD27), memory B cells (CD20+CD43CD27+), and B1 cells (CD20+CD43+CD27+) (40). (C) Mean percentage of LinHLA-DR+ DC-expressing CD83 on the cell surface (±SEM, n = 2 experiments) for the first 48 h following stimulation [as for (A) above]. (D) Summary of n = 3 experiments/donors showing mean percentage of CD19 B cells expressing CD83 in CFSE-labeled PBMC stimulated as described in (A) for up to 6 d (error bars indicate 1 SEM). (E) CD83 expression by proliferated and nonproliferated CD19 B cells after 4 d of stimulation of CFSE-labeled PBMC from (D). Graph shows mean percentage of CFSEdim and CFSEbright CD19 B cells that express CD83 (error bars indicate 1 SEM; n = 3 experiments/donors).

FIGURE 4.

Human B cells rapidly upregulate CD83 when stimulated and expression is sustained by CD40 ligation. (A) Graphs show mean percentages of CD19 B cells that expressed CD83 on the cell surface (±SEM, n = 3 experiments/donors) for up to 48 h following stimulation. Freshly isolated human PBMC were cultured with Nil, TT (10 μg/ml), anti-IgG or anti-IgM (2 μg/ml), LPS (2 μg/ml), anti-CD3 plus anti-CD28 (each 1 μg/ml), αδ-dex (10 ng/ml), or CD40L-transfected L cells (L cell/PBMC ratio 1:20). CD19 B cell CD83 expression was assessed by flow cytometry. (B) CD83 expression on B cell subsets in PBMC after 24 h of unstimulated culture. Dots on graph show percentage of cells in each B cell subset that express CD83 for each of four PBMC donors. (Friedman nonparametric comparison of means of matched groups with Dunn posttest). Surface phenotypes include naive B cells (CD20+CD43CD27), memory B cells (CD20+CD43CD27+), and B1 cells (CD20+CD43+CD27+) (40). (C) Mean percentage of LinHLA-DR+ DC-expressing CD83 on the cell surface (±SEM, n = 2 experiments) for the first 48 h following stimulation [as for (A) above]. (D) Summary of n = 3 experiments/donors showing mean percentage of CD19 B cells expressing CD83 in CFSE-labeled PBMC stimulated as described in (A) for up to 6 d (error bars indicate 1 SEM). (E) CD83 expression by proliferated and nonproliferated CD19 B cells after 4 d of stimulation of CFSE-labeled PBMC from (D). Graph shows mean percentage of CFSEdim and CFSEbright CD19 B cells that express CD83 (error bars indicate 1 SEM; n = 3 experiments/donors).

Close modal

CD83 was maximally upregulated well in advance of any B cell proliferative responses that occurred after 2 d of stimulation (Supplemental Fig. 1). Seventy percent of all B cells in CD40L-treated cultures continued to express CD83 after 4 d (Fig. 4D), and at least half of these had proliferated (Fig. 4E). After 4 d of Nil, LPS, or anti-IgM/G treatment, <20% of B cells had proliferated and a greater percentage of the nonproliferating B cells expressed CD83 compared with the proliferating B cells (e.g., for anti-IgM/G, 11.8 ± 3.2% of CFSEhigh B cells were CD83+ compared with 2.0 ± 0.35% of CFSEdim B cells, Fig. 4E). Less than 2.0% of proliferating and nonproliferating CD4 T cells in the same cultures expressed CD83 (data not shown).

As already known for other Abs (31), we showed that the nonfucosylated IgG1 mAb 3C12Ckif has greater ADCC potency than the normally fucosylated IgG1 mAb 3C12C on CD83+ cell line targets (Supplemental Fig. 2A) and it increased MLR inhibitory capacity (Supplemental Fig. 2B). In further confirmation of the expected enhanced binding of the nonfucosylated Fc region of mAb 3C12Ckif to the FcR CD16 on human NK cells, we observed that the CD16 function–blocking mAb 3G8, which binds to the human IgG1 Fc binding region of the CD16 FcR, was less effective in inhibiting ADCC in the presence of 3C12Ckif compared with 3C12C (Supplemental Fig. 2C).

We assessed the ADCC capability of each mAb on activated (CD83+) purified primary B cells labeled with 51Cr-Na2CrO4. The nonfucosylated variant mAb 3C12Ckif induced greater lysis in the presence of IL-2–activated purified NK cells than normally fucosylated 3C12C (Fig. 5A), consistent with ADCC mediated by human CD16 NK cells (31). 3C12Ckif was more effective than equal concentrations of the anti-CD20 mAb rituximab and the anti-CD52 mAb alemtuzumab when purified B cells stimulated with CD40L transfectants to induce CD83 expression were incubated with overnight-cultured autologous PBMC effectors (Fig. 5B).

FIGURE 5.

Anti-CD83 induces ADCC lysis of activated primary B cells and does not deplete resting B cells. (A) In the presence of IL-2–activated NK cells, nonfucosylated anti-CD83 mAb 3C12Ckif at ≥0.02 μg/ml induced ADCC lysis of PWM-stimulated (CD83+) B cells (IL-2–activated NK cell/stimulated B cell E:T ratio of 5:1) more effectively than did the normally fucosylated mAb 3C12C [p < 0.001, error bars in (A) and (B) indicate 1 SEM; n = 5 replicates]. Data shown are from one of two similar experiments. (B) Anti-CD83 mAb 3C12Ckif at 0.2 μg/ml induced ADCC lysis of CD40L-transfected L cell–stimulated (CD83+) B cells in the presence of autologous PBMC (PBMC/stimulated B cell E:T ratio of 50:1) and was more effective than equal concentrations of rituximab and Campath-1H mAbs (p < 0.001, one-way ANOVA with Dunnett multiple comparison test). Data shown are from one of two similar experiments. The low percentage lysis values in (B) compared with (A) is likely due to unlabeled B cells in the overnight cultured PBMC effectors of (B), ∼12% of which would be expected to upregulate CD83 (Fig. 4A, Nil treatment). These, being unlabeled, would compete with the 51Cr-labeled CD83+-purified B cell targets for Ab and/or for NK effectors. (C) Anti-CD83 did not deplete most CD19 B cells from PBMC, unlike rituximab. PBMC were cultured for 6 d with RA83, RAneg, mAb 3C12Ckif, rituximab, or human IgG1 isotype negative control mAb. Line graph shows fold changes in number of CD19 B cells in live gated PBMC treated with anti-CD83 or rituximab relative to matched negative control Ab-treated cultures (=1.0). Only rituximab treatment reduced CD19 B cell numbers significantly below negative control Ab levels (p < 0.05, Wilcoxin signed rank test, n = 7 experiments/donors). Representative flow cytometry dot plots from one experiment/donor are shown in Supplemental Fig. 3.

FIGURE 5.

Anti-CD83 induces ADCC lysis of activated primary B cells and does not deplete resting B cells. (A) In the presence of IL-2–activated NK cells, nonfucosylated anti-CD83 mAb 3C12Ckif at ≥0.02 μg/ml induced ADCC lysis of PWM-stimulated (CD83+) B cells (IL-2–activated NK cell/stimulated B cell E:T ratio of 5:1) more effectively than did the normally fucosylated mAb 3C12C [p < 0.001, error bars in (A) and (B) indicate 1 SEM; n = 5 replicates]. Data shown are from one of two similar experiments. (B) Anti-CD83 mAb 3C12Ckif at 0.2 μg/ml induced ADCC lysis of CD40L-transfected L cell–stimulated (CD83+) B cells in the presence of autologous PBMC (PBMC/stimulated B cell E:T ratio of 50:1) and was more effective than equal concentrations of rituximab and Campath-1H mAbs (p < 0.001, one-way ANOVA with Dunnett multiple comparison test). Data shown are from one of two similar experiments. The low percentage lysis values in (B) compared with (A) is likely due to unlabeled B cells in the overnight cultured PBMC effectors of (B), ∼12% of which would be expected to upregulate CD83 (Fig. 4A, Nil treatment). These, being unlabeled, would compete with the 51Cr-labeled CD83+-purified B cell targets for Ab and/or for NK effectors. (C) Anti-CD83 did not deplete most CD19 B cells from PBMC, unlike rituximab. PBMC were cultured for 6 d with RA83, RAneg, mAb 3C12Ckif, rituximab, or human IgG1 isotype negative control mAb. Line graph shows fold changes in number of CD19 B cells in live gated PBMC treated with anti-CD83 or rituximab relative to matched negative control Ab-treated cultures (=1.0). Only rituximab treatment reduced CD19 B cell numbers significantly below negative control Ab levels (p < 0.05, Wilcoxin signed rank test, n = 7 experiments/donors). Representative flow cytometry dot plots from one experiment/donor are shown in Supplemental Fig. 3.

Close modal

Anti-CD83 did not deplete all B cells in PBMC. In vitro, in contrast to rituximab, which depleted >80% of CD19 B cells from PBMC, both RA83 and mAb 3C12Ckif depleted <20% of B cells in 6 d of culture (Fig. 5C, Supplemental Fig. 3). In vivo, in the human PBMC-SCID mouse xenograft experiments (Fig. 3), mice treated with anti-CD83 (RA83) 3 d posttransplant had 38% (p < 0.05) and 68% (p < 0.05) fewer peritoneal and splenic human CD19 B cells, respectively, and 53% (p < 0.05) fewer splenic human CD4 T cells at day 13 posttransplant, compared with negative control Ab (RAneg)-treated mice (Supplemental Fig. 4).

The recall Ag TT stimulates proliferative responses in vitro by TT-specific memory B cells and CD4 T cells in human PBMC (37). Although the percentage of all B cells expressing CD83 was only marginally increased by TT stimulation (Fig. 4A), we found that the ADCC-capable anti-CD83 Abs (RA83 and the human IgG1 isotype mAbs 3C12C and 3C12Ckif) strongly inhibited B cell and CD4 T cell proliferative responses stimulated by TT (Fig. 6). B cell proliferation was not inhibited by the ADCC-deficient human IgG4 isoform of mAb 3C12C and, interestingly, CD4 T cell proliferation was increased compared with donor-matched IgG1 isotype control mAb–treated cultures (Fig. 6C, Supplemental Fig. 5). Rituximab depleted >80% of all B cells and thereby also inhibited the B cell proliferative response to TT but, in contrast to anti-CD83, it had no effect on the strong CD4 T cell proliferation stimulated by TT (Fig. 6A, 6B).

FIGURE 6.

ADCC-capable anti-CD83 strongly inhibits B cell and CD4 T cell proliferative responses to TT stimulation. (A) Representative flow cytometry plots show the effects of anti-CD83 mAb 3C12Ckif and of rituximab on proliferation of CD4 T cells and CD19 B cells in TT (10 μg/ml)-treated CFSE-labeled PBMC cultured for 6 d. Proliferated cells = CFSEdim gated as shown. (B) Histograms show mean proliferated CD4 T cells and CD19 B cells as percentage of total live-gated cells (error bars indicate 1 SEM; n ≥ 7 experiments/donors; Friedman one-way ANOVA with Dunn post hoc multiple comparison tests). (C) MAb 3C12C/IgG1 but not mAb 3C12C/IgG4 inhibited CD19 B cell proliferative responses to TT (relative to donor-matched IgG1 isotype control mAb–treated cultures = 100%; p < 0.0001, Student one-sample t test). mAb 3C12C/IgG4 significantly increased CD4 T (p < 0.05) but not CD19 B cell proliferation relative to IgG1 isotype controls.

FIGURE 6.

ADCC-capable anti-CD83 strongly inhibits B cell and CD4 T cell proliferative responses to TT stimulation. (A) Representative flow cytometry plots show the effects of anti-CD83 mAb 3C12Ckif and of rituximab on proliferation of CD4 T cells and CD19 B cells in TT (10 μg/ml)-treated CFSE-labeled PBMC cultured for 6 d. Proliferated cells = CFSEdim gated as shown. (B) Histograms show mean proliferated CD4 T cells and CD19 B cells as percentage of total live-gated cells (error bars indicate 1 SEM; n ≥ 7 experiments/donors; Friedman one-way ANOVA with Dunn post hoc multiple comparison tests). (C) MAb 3C12C/IgG1 but not mAb 3C12C/IgG4 inhibited CD19 B cell proliferative responses to TT (relative to donor-matched IgG1 isotype control mAb–treated cultures = 100%; p < 0.0001, Student one-sample t test). mAb 3C12C/IgG4 significantly increased CD4 T (p < 0.05) but not CD19 B cell proliferation relative to IgG1 isotype controls.

Close modal

We undertook cell depletion experiments to further investigate the contributions of B cells and DC to the inhibitory action of anti-CD83 in this TT stimulation assay. We found that CD4 T cell proliferation was undiminished in PBMC predepleted of B cells with anti-CD19 mAb–coated immunomagnetic beads (compare Fig. 7Aii with 7Ai). B cell depletion also had no effect on anti-CD83–mediated inhibition of CD4 T cell proliferation (Fig. 7Avi, x). These findings are consistent with the abovementioned non-effect of rituximab in whole PBMC (Fig. 6A, 6B) and suggest that DC rather than B cells are the key cell type that processes and presents TT Ag to CD4 T cells in this in vitro assay. In confirmation, TT failed to stimulate CD4 T cell proliferation significantly above background in PMBC predepleted of myeloid DC using CD11c mAb–coated immunomagnetic beads (Fig. 7Aiii). However, anti-CD83 reduced the background CD4 T cell proliferation in CD11c+ cell–depleted PBMC (compare Fig. 7Avii with 7Av and 7Axi with 7Aix), suggesting either stimulation by CD83+ CD20CD11c APCs or spontaneous CD83 expression and proliferation by a small population of CD4 T cells. TT failed to increase CD4 T cell proliferation above background in PBMC depleted of NK cells (Fig. 7Aiv), and anti-CD83 had no effect in these cell cultures (Fig. 7Aviii, xii). These may be due to the known role of NK cells in activation of DC (43) and consequent failure by DC to upregulate CD83 and costimulatory molecules and to present TT on MHC II.

FIGURE 7.

The roles of CD11c+ DC, B cells, and NK cells in anti-CD83–mediated inhibition of (A) CD4 T cell and (B) CD19 B cell proliferation in TT-stimulated PBMC. B cells, myeloid DC, or NK cells were depleted from whole PBMC with immunomagnetic beads coated with anti-CD19, CD11c, or CD56 mAb, respectively, and cultured with TT as described in Fig. 6. Cell proliferation is shown on the y-axes (CFSEdim CD4+ or CD19+ cells expressed as percentage of total live gated cells). Results are from five or six experiments/donors and lines connect data from the same experiment/donors (*p < 0.05, Wilcoxon matched-pairs signed rank test). First row of graphs in (A) and (B) show effects of TT addition to otherwise untreated cell cultures; subsequent rows show effects of anti-CD83 (RA83, 3C12Ckif) and rituximab (Ritux) on proliferation in the presence of TT compared with appropriate negative control Abs.

FIGURE 7.

The roles of CD11c+ DC, B cells, and NK cells in anti-CD83–mediated inhibition of (A) CD4 T cell and (B) CD19 B cell proliferation in TT-stimulated PBMC. B cells, myeloid DC, or NK cells were depleted from whole PBMC with immunomagnetic beads coated with anti-CD19, CD11c, or CD56 mAb, respectively, and cultured with TT as described in Fig. 6. Cell proliferation is shown on the y-axes (CFSEdim CD4+ or CD19+ cells expressed as percentage of total live gated cells). Results are from five or six experiments/donors and lines connect data from the same experiment/donors (*p < 0.05, Wilcoxon matched-pairs signed rank test). First row of graphs in (A) and (B) show effects of TT addition to otherwise untreated cell cultures; subsequent rows show effects of anti-CD83 (RA83, 3C12Ckif) and rituximab (Ritux) on proliferation in the presence of TT compared with appropriate negative control Abs.

Close modal

TT stimulated B cell proliferation in whole PBMC and, in contrast to CD4 T cells, also in PBMC depleted of CD11c+ cells (Fig. 7Bi, ii). This is consistent with direct targeting of TT Ag–activated CD83+ B cells by anti-CD83. Despite the evidence for an NK cell–mediated ADCC mechanism of action (Figs. 2, 5A) (9, 10, 12), anti-CD83 inhibited TT stimulated B cell proliferation in NK cell–depleted PBMC, although the degree of inhibition was reduced compared with that in whole PBMC (Fig. 7Bvi, ix). Thus, other ADCC effectors in addition to NK cells, such as monocytes/macrophages, may be involved in the inhibition of B cell proliferative responses by anti-CD83.

Given the potent effect of anti-CD83 on Ag-stimulated CD4 T cell proliferation (Fig. 6) and the known roles of cytokines expressed by CD4 T cells in B cell differentiation and Ig expression, and in inflammatory autoimmune disease, we investigated the effects of anti-CD83 on cytokine production in TT-stimulated PBMC cultures. TT stimulated a statistically significant increase above background in the production of only one of the eight cytokines tested (IFN-γ, Fig. 8A). Anti-CD83 mAb 3C12Ckif significantly reduced the production of this cytokine, as well as of IL-2, IL-10, IL-17A, and GM-CSF. IL-4, IL-6, and TNF-α were unaffected (Fig. 8A). RA83 had similar effects except IL-2 and IL-10 inhibition did not reach statistical significance. Rituximab had no effect on the production of any of the cytokines tested (Fig. 8A), consistent with its nil effect on CD4 T cell proliferation (Fig. 6).

FIGURE 8.

Anti-CD83 inhibited IFN-γ and IL-17A but not TNF production in TT-stimulated PBMC cultures. (A) Bar charts show cytokine levels (picograms per milliliter) in media after 6 d of culture of CFSE-labeled PBMC treated with Nil or TT and Abs as shown (mean ± 1 SEM of n ≥ 5 experiments/donors, repeated measures ANOVA and Bonferroni multiple comparison tests). (B) Graph shows fold change in proliferated (CFSEdim), cytokine+ CD4 T cells from (A) above (expressed as percentage of total live cells) due to anti-CD83 mAb 3C12Ckif relative to human IgG1 isotype control mAb (dashed line = 1). Solid horizontal bars are means of n ≥ 5 experiments/donors (**p < 0.002, one-sample t test).

FIGURE 8.

Anti-CD83 inhibited IFN-γ and IL-17A but not TNF production in TT-stimulated PBMC cultures. (A) Bar charts show cytokine levels (picograms per milliliter) in media after 6 d of culture of CFSE-labeled PBMC treated with Nil or TT and Abs as shown (mean ± 1 SEM of n ≥ 5 experiments/donors, repeated measures ANOVA and Bonferroni multiple comparison tests). (B) Graph shows fold change in proliferated (CFSEdim), cytokine+ CD4 T cells from (A) above (expressed as percentage of total live cells) due to anti-CD83 mAb 3C12Ckif relative to human IgG1 isotype control mAb (dashed line = 1). Solid horizontal bars are means of n ≥ 5 experiments/donors (**p < 0.002, one-sample t test).

Close modal

The effect of anti-CD83 on intracellular cytokine expression by CD4 T cells was investigated in TT-stimulated CFSE-labeled PBMC. RA83 significantly reduced IFN-γ+ and IL-17A+ CFSEdim (i.e., proliferated) CD4 T cells by 58 and 73%, respectively (p < 0.005, data not shown). mAb 3C12Ckif also significantly reduced CFSEdim CD4 T cells expressing these cytokines (by 61.4 and 90.4%, respectively, p < 0.002) as well as those expressing IL-4 (by 45.4%), IL-10 (by 64.5%), and IL-21 (by 92.5%) (Fig. 8B). The percentage of TNF-α+ CD4 T cells was not significantly reduced by either anti-CD83 Ab. The percentages of nonproliferated (CFSEbright) cytokine+ CD4 T cells were not significantly affected by anti-CD83, except those expressing IL-10 were reduced by 41.6% by mAb 3C12Ckif (p < 0.02, data not shown).

In this study, we show in preclinical assays using healthy human PBMC that ADCC-capable anti-CD83 Ab can inhibit components of adaptive humoral immunity. In vitro, allostimulated Ab production (Fig. 1) and specific Ag (TT)-stimulated B cell proliferation (Fig. 6) were inhibited without pan–B cell depletion (Fig. 5C, Supplemental Fig. 3). The data presented in the present study show that anti-CD83 can lyse activated (i.e., CD83+) B cells in the presence of NK cells and that the nonfucosylated anti-CD83 mAb 3C12Ckif has greater potency than the normally fucosylated 3C12C mAb (Fig. 5A, Supplemental Fig. 2). Similar improvements in potency have been reported for other nonfucosylated ADCC-capable mAbs, including anti-CD20 mAb rituximab (44, 45), enabling depletion of cells that express lower levels of target Ag (46) in the presence of endogenous inhibitory IgG1 Ab (32).

Anti-CD83 not only inhibited B cell proliferative responses to specific Ag stimulation in PBMC but concomitant DC-mediated CD4 T cell proliferation and cytokine expression were also inhibited (Figs. 6, 8). It is possible that activated CD4 T cells were not targeted directly by anti-CD83 and NK cells in our experiments. Allostimulated human CD4 T cells express CD83 at low levels compared with human DC (9) or when compared with murine lymphocytes (2). Compared to activated human DC, allostimulated (CD83dim) human CD4 T cells are lysed weakly by anti-CD83 (9), but we found that <2% of CD4 T cells expressed CD83 in the experiments reported in the present study (data not shown), and there are no published reports of CD83 expression by specific Ag-stimulated human CD4 T cells. Nevertheless, an indirect reduction in activated CD4 T cells would have occurred due to CD83 Ab-mediated depletion of CD83+ DC that stimulate these cells (9).

TT marginally increased the percentage of CD83+ B cells above background levels (Fig. 4A), consistent with stimulation confined to a small population of TT-specific memory B cells (47). The other stimulants used in the present study were not Ag specific, stimulating via CD40, TLR4, or the BCR, and each one rapidly (<12 h) induced CD83 expression in high percentages of B cells. Thus, any anti-CD83–mediated depletion of recently stimulated B cells could begin well before any subsequent B cell division (≥48 h) and other events that follow B cell division such as Ig expression, affinity maturation, class switch recombination, and differentiation to PC or memory B cells (48).

Notably, CD40L not only rapidly induced CD83 in ∼90% of all B cells but expression was sustained for a longer time (Fig. 4A, 4D) and resulted in significantly more B cell proliferation than that induced by LPS and anti-IgM/G (Supplemental Fig. 1). CD83 upregulation was largely confined to nonproliferated B cells when stimulated with LPS and anti-IgM/G but not when stimulated by CD40L (Fig. 4E). In the latter case, CD83 expression was found to be lost only in those B cells that had undergone four to six divisions in 6 d. In experiments with purified human memory B cells, Henn et al. (16) found that expression of CD83 and of molecules associated with Ag presentation, costimulation, and cytokine production were higher in B cells that did not proliferate when stimulated with CpG and cytokines compared with those that proliferated. In addition to specific Ab development, B cells also, similar to DC, costimulate and present Ag to CD4 T cells in vivo and this is influenced by B cell cytokine expression (49). It has been reported that, in mice, whereas conventional DC are required to prime naive CD4 T cells, T follicular helper cell development is mediated by B cells, not conventional DC, which instead induce subsequent differentiation to Th1 effectors (50). We did not experimentally address the effects of anti-CD83 Ab on these additional B cell functions in human experimental systems. However, as both proliferating and nonproliferating B cells had upregulated CD83 in our experiments, anti-CD83 Ab is likely to target both populations, thereby inhibiting Ag presentation, costimulation, and cytokine production associated with nonproliferating B cells as well as proliferating B cell–associated Ab development (16). The latter can be expected to include CD4 T follicular helper cell and germinal center development and, potentially, BCR remodeling as found for murine memory CD83+ B cells in secondary germinal centers (51).

As our human CD83-specific Abs do not recognize murine CD83 (data not shown), and mouse CD83 specific Abs give distinctly different results in immunocompetent mice (see later discussion), we tested anti-CD83 in vivo in a human PBMC/SCID mouse xenograft model adapted from Sandhu et al. (36). These authors reported human CD4 T cell–dependent human IgM and IgG responses to immunization with neoantigens as well as with recall Ags. Although time limited owing to xenogeneic GVHD, such models have been used to investigate human B cell pathobiology (5254) and to assess the in vivo effects of rituximab (55). We found that anti-CD83 inhibited Ag-specific IgG responses to immunization with the recall Ag TT without affecting total circulating human IgG levels (Fig. 3A). We detected no Ag-specific IgG responses to the neoantigen KLH, but specific IgM responses were detected and these were inhibited by anti-CD83 (Fig. 3B). Anti-CD83 may also reduce total circulating human IgM levels, although our data did not reach statistical significance. Rituximab treatment has been reported to preferentially reduce total IgM compared with IgG in the circulation of patients in some studies (56, 57). Others report that rituximab pretreatment significantly reduced memory B cell Ab responses in vivo but did not affect total circulating Ig or pre-existing Ag-specific Ab levels in human CD20 transgenic mice (58) or in humans (59, 60).

It has been reported that rituximab (2 mg/kg) given 4 d prior to sacrifice depleted human B cells by 75% from the spleens of human PBMC-xenografted SCID mice (61). In comparison, we found anti-CD83 (RA83, 6.25 mg/kg) given 10 d prior to sacrifice had depleted human B cells by 38 and 68% from the mouse peritoneal cavities and spleens, respectively. Any such apparently nonspecific pan–B cell depletion induced by anti-CD83 in the mice may instead result from CD83 upregulation by human B cells stimulated by murine xenoantigens (62), which would not be expected to occur in patients. Splenic human CD4 T cell engraftment was also reduced by RA83 (by 53%, Supplemental Fig. 4). This may similarly be due to anti-CD83–mediated depletion of coengrafted human CD83+ DC and/or B cells that could otherwise stimulate expansion of xenoantigen-responding human CD4 T cells. Alternatively, anti-CD83 Ab may directly deplete blasting CD4 T cells that upregulate CD83 (9). Additional mechanisms, unrelated to xenoantigen stimulation, cannot be ruled out at this time. For example, CD4 T cell migration to, and/or retention and expansion in, the spleen may also be regulated by CD83+ DC and/or B cells by an Ag-independent mechanism.

Our findings regarding anti-CD83–mediated inhibition of TT responses by Ag-specific B cells and CD4 T cells may also apply to pathogenic Ab-expressing B cells and inflammatory cytokine-expressing CD4 T cells undergoing stimulation with autoantigen in patients with autoimmune disease or with alloantigen in transplant patients. If so, anti-CD83 mAb treatment could benefit patients with autoimmune conditions or patients about to undergo allogeneic transplantation for which there is a risk of graft failure.

Along with CD4 T cells and DC, B cells play a central role in a number of autoimmune diseases (18), including SLE (19). Interestingly, the human B cell subset that we found to have the highest spontaneous upregulation of CD83 (CD19+CD27+CD43+CD11b+ B cells, Fig. 4B), controversially described as B1 cells (63), is expanded in SLE patients, has high costimulatory CD86 expression, and is a strong stimulator of CD4 T cells (41). Additionally, circulating apoptotic microparticles from SLE patients upregulate CD83 on myeloid and plasmacytoid DC (64). The latter play a key role in SLE, including type I IFN–induced stimulation of autoreactive memory B cells to differentiate into pathogenic Ab-producing plasmablasts and PC (65). Although ADCC lysis of CD83+ B1 cells and CD83+ plasmacytoid DC by anti-CD83 has yet to be formally demonstrated, anti-CD83 may be highly effective in SLE.

RA pathogenesis also involves DC, B cells (21), and CD4 T cells (20). A significant subset of RA patients responds to treatment with rituximab (25), which directly depletes only B cells. This includes pathogenic autoantigen-specific memory B cells, but most other B cells including pre-existing memory B cells involved in immune protection are also depleted (56, 57). Although derived from CD20+ B cells, PC do not express CD20, and hence pre-existing PC are not depleted by rituximab. Similarly, pre-existing PC would not be depleted by anti-CD83, as they do not express CD83 (6668). Depending on longevity (27, 69), pre-existing autoantigen- or alloantigen-specific PC in bone marrow or in ectopic lymphoid structures (e.g., in inflamed rheumatoid synovial tissue or kidney grafts) (70) continue to produce pathogenic Ab (65). Nevertheless, similar to rituximab, anti-CD83 may be beneficial in the clinic because, as in our TT stimulation experiments, it could deplete Ag-specific IgG+ memory B cells recently activated by autoantigen (or alloantigen) and thus block differentiation to further pathogenic autoantibody (or alloantibody)-producing PC. Note that human memory B cells have a significantly higher proportion of autoantigen-specific IgG+ memory B cells than do IgG-producing PC (71, 72).

In contrast to rituximab, we showed that anti-CD83 does not deplete most B cells, thus preserving resting memory B cells specific for other Ags, including those involved in immune defense induced by prior infections or vaccinations. Also, unlike rituximab, which depletes only B cells, anti-CD83 cannot only deplete activated B cells but also activated DC, thereby reducing CD4 T cell proliferation. Anti-CD83 treatment reduced production of the RA-associated inflammatory cytokines IL-17A, IFN-γ, and GM-CSF produced by CD4 T cells in TT-stimulated healthy donor PBMC (Fig. 8A). TNF and IL-6 production were not significantly inhibited by anti-CD83, possibly because they are expressed by monocytes (73) as well as by CD4 T cells in PBMC. Both TNF and IL-6 can be specifically blocked by existing therapeutics. IL-4 is produced mainly by Th2 CD4 T cells, but it was not inhibited by anti-CD83, which may be beneficial, as IL-4 inhibits Th1 conditions such as RA (74). In the same experiments, both RA83 and 3C12Ckif significantly reduced the proportion of proliferating CD4 T cells that express inflammatory cytokines IFN-γ and IL-17A (by 61 and 90%, respectively, for 3C12Ckif), but the percentage of TNF-α+ CD4 T cells was not significantly affected by either Ab (Fig. 8B). mAb 3C12Ckif also significantly reduced proliferating CD4 T cells that express IL-4 (by 45%), IL-10 (65%), and IL-21 (93%), with this last result suggesting that anti-CD83 may inhibit B cell affinity maturation and class switching mediated by CD4 T follicular helper cells. The increasingly appreciated cytokine coexpression and plasticity of individual CD4 T cells (75, 76) makes it difficult to gauge the extent to which our in vitro findings might apply in vivo to patients with RA or other autoimmune disease. Nevertheless, our data suggest that anti-CD83 may provide not only the clinical benefits of direct (activated) B cell depletion, similar to rituximab, but also additional benefits due to inhibition of CD83+ DC-stimulated CD4 T cell inflammatory cytokine expression, without lasting impairment of pre-existing protective immune memory.

Experimental autoimmune encephalitis (EAE), a model of multiple sclerosis, is induced by immunization of mice with myelin oligodendrocyte glycoprotein. Myelin-specific B cells, but not anti-myelin Abs, are required for the development of EAE (77), although it is not known whether these cells are responsible for inducing the GM-CSF+ CD4 T cells recently shown to be more encephalitogenic than IL-17+ and IFN-γ+ CD4 T cells (78, 79). We found that GM-CSF production was strongly inhibited by anti-CD83 (Fig. 8A). Thus, anti-CD83 may find a role in multiple sclerosis treatment, although it is not clear how our findings fit with the report that synthetic soluble CD83 can prevent EAE (80).

DC (81) and CD4 T cells (30) play key roles in allogeneic solid organ transplant rejection/failure and in GVHD. B cells are clearly important in Ab-mediated rejection (82), and they are also important in chronic and possibly acute GVHD (83). We have previously shown in the allogeneic MLR, an in vitro partial model of allogeneic transplantation, that anti-CD83 depletes activated DC and reduces alloactivated CD4 T cell numbers (9) and cytokine expression (10). In the present study, we show in the allogeneic MLR that anti-CD83 inhibits total IgM and IgG production (Figs. 1, 2). These results together with our TT stimulation data (Fig. 6) suggest that specific B cell responses to alloantigen stimulation would be inhibited by anti-CD83 in transplant recipients. Anti-CD83 mAb administration would best begin immediately prior to transplantation for maximal inhibition of alloantigen-specific B cell responses and to prevent differentiation to pathogenic alloantibody-producing CD83 PC (28). Prophylactic anti-CD83 treatment would simultaneously inhibit DC-mediated stimulation of alloreactive CD4 T cells and expression of IFN-γ and IL-17A, with both cytokines playing key pathogenic roles in acute and chronic graft failure (8487). Despite the report that CD83 is expressed by murine regulatory T cells (88), we observed no reduction in human CD4+CD25+Foxp3+ regulatory T cells in human PBMC-xenografted SCID mice treated with 3C12C mAb (12).

In summary, we found that ADCC-capable anti-human CD83 Ab, as already known for activated CD83+ DC, can directly lyse activated CD83+ human B cells, thereby inhibiting B cell responses in vitro and in vivo in human PBMC-xenografted SCID mice. Potential therapeutic advantages compared with rituximab in autoimmunity and transplantation could arise from simultaneous inhibition of autoantigen- or alloantigen-specific B cell responses and DC-mediated inflammatory CD4 T cell responses while retaining beneficial protective and regulatory properties of nonpathogenic memory B and T cells and resting (CD83) DC.

Our findings contrast strongly with those reported for anti-mouse CD83 mAbs in conventional immunocompetent mice (89, 90). Using the rat IgG1 anti-mouse CD83 mAb Michel-19, a murine isotype that supports ADCC in mice via the CD16 FcR (91), Kretschmer et al. (89, 90) found that anti-CD83 had no effect on thymus-dependent (TD) Ab responses. This is despite mouse follicular B cells readily expressing CD83 on the cell surface when stimulated with TD Ag (14) and recent B cell–specific CD83 conditional knockout mouse experiments showing a number of changes in B cell functionality, including enhanced IgE responses to TD immunization (92), that suggest CD83 has a functional role in germinal center follicular B cells. Although TD responses were unaffected, Kretschmer et al. (89, 90) found that anti-CD83 induced greatly increased IgG1 class-switch responses to thymus-independent type 2 (TI-2) immunization. This required direct engagement of anti-CD83 on marginal zone B cells rather than on follicular B cells (90) or DC (89).

There are important differences between mouse and human marginal zone B cells and TI-2 responses (93). Also, the possible relevance of our observation that the ADCC-incompetent IgG4 isoform of anti-human CD83 mAb 3C12C had increased the CD4 T cell proliferative response to the TD Ag TT (Fig. 6C, Supplemental Fig. 5) is intriguing. CD4 T cell subsets can facilitate class switching by marginal zone B cells responding to TI-2 as well as to TD Ags (93, 94). It will be critical to make appropriate choices of preclinical disease models (e.g., immunocompetent mouse, human cell xenografted immunodeficient mouse, or nonhuman primate) for therapeutic development of anti-CD83 mAbs in autoimmunity and transplantation in humans.

We thank the blood donors, phlebotomy staff, and animal facility staff for contributions to this project.

This work was supported by the Mater Research Institute, the University of Queensland, and the Cooperative Research Centre for Biomarker Translation. Expression of 3C12C and 3C12Ckif utilized infrastructure provided by the National Biologics Facility, University of Queensland, supported by Therapeutic Innovation Australia and the National Collaborative Research Infrastructure Strategy program.

The online version of this article contains supplemental material.

Abbreviations used in this article:

ADCC

Ab-dependent cellular cytotoxicity

DC

dendritic cell

EAE

experimental autoimmune encephalitis

GVHD

graft-versus-host disease

hiFCS

heat-inactivated FCS

KLH

keyhole limpet hemocyanin

MHC II

MHC class II

PB

Pacific Blue

PC

plasma cell

RA

rheumatoid arthritis

RA83

rabbit polyclonal IgG anti-human CD83

RAneg

rabbit negative control IgG

SLE

systemic lupus erythematosus

TD

thymus-dependent

TI-2

thymus-independent type 2

TT

tetanus toxoid.

1
Fujimoto
,
Y.
,
L.
Tu
,
A. S.
Miller
,
C.
Bock
,
M.
Fujimoto
,
C.
Doyle
,
D. A.
Steeber
,
T. F.
Tedder
.
2002
.
CD83 expression influences CD4+ T cell development in the thymus.
Cell
108
:
755
767
.
2
Prazma
,
C. M.
,
T. F.
Tedder
.
2008
.
Dendritic cell CD83: a therapeutic target or innocent bystander?
Immunol. Lett.
115
:
1
8
.
3
Tze
,
L. E.
,
K.
Horikawa
,
H.
Domaschenz
,
D. R.
Howard
,
C. M.
Roots
,
R. J.
Rigby
,
D. A.
Way
,
M.
Ohmura-Hoshino
,
S.
Ishido
,
C. E.
Andoniou
, et al
.
2011
.
CD83 increases MHC II and CD86 on dendritic cells by opposing IL-10–driven MARCH1-mediated ubiquitination and degradation.
J. Exp. Med.
208
:
149
165
.
4
Prechtel
,
A. T.
,
N. M.
Turza
,
A. A.
Theodoridis
,
A.
Steinkasserer
.
2007
.
CD83 knockdown in monocyte-derived dendritic cells by small interfering RNA leads to a diminished T cell stimulation.
J. Immunol.
178
:
5454
5464
.
5
Zhou
,
L. J.
,
R.
Schwarting
,
H. M.
Smith
,
T. F.
Tedder
.
1992
.
A novel cell-surface molecule expressed by human interdigitating reticulum cells, Langerhans cells, and activated lymphocytes is a new member of the Ig superfamily.
J. Immunol.
149
:
735
742
.
6
Stein
,
M. F.
,
K.
Blume
,
C. S.
Heilingloh
,
M.
Kummer
,
B.
Biesinger
,
H.
Sticht
,
A.
Steinkasserer
.
2015
.
CD83 and GRASP55 interact in human dendritic cells.
Biochem. Biophys. Res. Commun.
459
:
42
48
.
7
Hirano
,
N.
,
M. O.
Butler
,
Z.
Xia
,
S.
Ansén
,
M. S.
von Bergwelt-Baildon
,
D.
Neuberg
,
G. J.
Freeman
,
L. M.
Nadler
.
2006
.
Engagement of CD83 ligand induces prolonged expansion of CD8+ T cells and preferential enrichment for antigen specificity.
Blood
107
:
1528
1536
.
8
Bates
,
J. M.
,
K.
Flanagan
,
L.
Mo
,
N.
Ota
,
J.
Ding
,
S.
Ho
,
S.
Liu
,
M.
Roose-Girma
,
S.
Warming
,
L.
Diehl
.
2015
.
Dendritic cell CD83 homotypic interactions regulate inflammation and promote mucosal homeostasis.
Mucosal Immunol.
8
:
414
428
.
9
Munster
,
D. J.
,
K. P.
MacDonald
,
M.
Kato
,
D. J.
Hart
.
2004
.
Human T lymphoblasts and activated dendritic cells in the allogeneic mixed leukocyte reaction are susceptible to NK cell-mediated anti-CD83-dependent cytotoxicity.
Int. Immunol.
16
:
33
42
.
10
Wilson
,
J.
,
H.
Cullup
,
R.
Lourie
,
Y.
Sheng
,
A.
Palkova
,
K. J.
Radford
,
A. M.
Dickinson
,
A. M.
Rice
,
D. N.
Hart
,
D. J.
Munster
.
2009
.
Antibody to the dendritic cell surface activation antigen CD83 prevents acute graft-versus-host disease. [Published erratum appears in 2009 J. Exp. Med. 206: 1203.]
J. Exp. Med.
206
:
387
398
.
11
Jones
,
M. L.
,
T.
Seldon
,
M.
Smede
,
A.
Linville
,
D. Y.
Chin
,
R.
Barnard
,
S. M.
Mahler
,
D.
Munster
,
D.
Hart
,
P. P.
Gray
,
T. P.
Munro
.
2010
.
A method for rapid, ligation-independent reformatting of recombinant monoclonal antibodies.
J. Immunol. Methods
354
:
85
90
.
12
Seldon
,
T. A.
,
R.
Pryor
,
A.
Palkova
,
M. L.
Jones
,
N. D.
Verma
,
M.
Findova
,
K.
Braet
,
Y.
Sheng
,
Y.
Fan
,
E. Y.
Zhou
, et al
.
2016
.
Immunosuppressive human anti-CD83 monoclonal antibody depletion of activated dendritic cells in transplantation.
Leukemia
30
:
692
700
.
13
Kretschmer
,
B.
,
S.
Kühl
,
B.
Fleischer
,
M.
Breloer
.
2011
.
Activated T cells induce rapid CD83 expression on B cells by engagement of CD40.
Immunol. Lett.
136
:
221
227
.
14
Prazma
,
C. M.
,
N.
Yazawa
,
Y.
Fujimoto
,
M.
Fujimoto
,
T. F.
Tedder
.
2007
.
CD83 expression is a sensitive marker of activation required for B cell and CD4+ T cell longevity in vivo.
J. Immunol.
179
:
4550
4562
.
15
Kozlow
,
E. J.
,
G. L.
Wilson
,
C. H.
Fox
,
J. H.
Kehrl
.
1993
.
Subtractive cDNA cloning of a novel member of the Ig gene superfamily expressed at high levels in activated B lymphocytes.
Blood
81
:
454
461
.
16
Henn
,
A. D.
,
M.
Laski
,
H.
Yang
,
S.
Welle
,
X.
Qiu
,
H.
Miao
,
C. T.
Barry
,
H.
Wu
,
M. S.
Zand
.
2012
.
Functionally distinct subpopulations of CpG-activated memory B cells.
Sci. Rep.
2
:
345
.
17
Mahanonda
,
R.
,
N.
Sa-Ard-Iam
,
K.
Yongvanitchit
,
M.
Wisetchang
,
I.
Ishikawa
,
T.
Nagasawa
,
D. S.
Walsh
,
S.
Pichyangkul
.
2002
.
Upregulation of co-stimulatory molecule expression and dendritic cell marker (CD83) on B cells in periodontal disease.
J. Periodontal Res.
37
:
177
183
.
18
Jackson
,
S. W.
,
N. S.
Kolhatkar
,
D. J.
Rawlings
.
2015
.
B cells take the front seat: dysregulated B cell signals orchestrate loss of tolerance and autoantibody production.
Curr. Opin. Immunol.
33
:
70
77
.
19
Crispín
,
J. C.
,
S. N.
Liossis
,
K.
Kis-Toth
,
L. A.
Lieberman
,
V. C.
Kyttaris
,
Y. T.
Juang
,
G. C.
Tsokos
.
2010
.
Pathogenesis of human systemic lupus erythematosus: recent advances.
Trends Mol. Med.
16
:
47
57
.
20
McInnes
,
I. B.
,
G.
Schett
.
2011
.
The pathogenesis of rheumatoid arthritis.
N. Engl. J. Med.
365
:
2205
2219
.
21
Bugatti
,
S.
,
B.
Vitolo
,
R.
Caporali
,
C.
Montecucco
,
A.
Manzo
.
2014
.
B cells in rheumatoid arthritis: from pathogenic players to disease biomarkers.
BioMed Res. Int.
2014
:
681678
.
22
Blüml
,
S.
,
K.
McKeever
,
R.
Ettinger
,
J.
Smolen
,
R.
Herbst
.
2013
.
B-cell targeted therapeutics in clinical development.
Arthritis Res. Ther.
15
(
Suppl. 1
):
S4
.
23
Macklin
,
P. S.
,
P. J.
Morris
,
S. R.
Knight
.
2014
.
A systematic review of the use of rituximab for desensitization in renal transplantation.
Transplantation
98
:
794
805
.
24
Makatsori
,
M.
,
S.
Kiani-Alikhan
,
A. L.
Manson
,
N.
Verma
,
M.
Leandro
,
N. P.
Gurugama
,
H. J.
Longhurst
,
S.
Grigoriadou
,
M.
Buckland
,
E.
Kanfer
, et al
.
2014
.
Hypogammaglobulinaemia after rituximab treatment-incidence and outcomes.
QJM
107
:
821
828
.
25
Edwards
,
J. C.
,
L.
Szczepanski
,
J.
Szechinski
,
A.
Filipowicz-Sosnowska
,
P.
Emery
,
D. R.
Close
,
R. M.
Stevens
,
T.
Shaw
.
2004
.
Efficacy of B-cell-targeted therapy with rituximab in patients with rheumatoid arthritis.
N. Engl. J. Med.
350
:
2572
2581
.
26
Barnett
,
A. N.
,
V. G.
Hadjianastassiou
,
N.
Mamode
.
2013
.
Rituximab in renal transplantation.
Transpl. Int.
26
:
563
575
.
27
Winter
,
O.
,
C.
Dame
,
F.
Jundt
,
F.
Hiepe
.
2012
.
Pathogenic long-lived plasma cells and their survival niches in autoimmunity, malignancy, and allergy.
J. Immunol.
189
:
5105
5111
.
28
Stegall
,
M. D.
,
N.
Moore
,
T.
Taner
,
H.
Li
,
P. G.
Dean
.
2014
.
Down-regulating humoral immune responses: implications for organ transplantation.
Transplantation
97
:
247
257
.
29
Reddy
,
V.
,
G.
Cambridge
,
D. A.
Isenberg
,
M. J.
Glennie
,
M. S.
Cragg
,
M.
Leandro
.
2015
.
Internalization of rituximab and the efficiency of B Cell depletion in rheumatoid arthritis and systemic lupus erythematosus.
Arthritis Rheumatol.
67
:
2046
2055
.
30
Gaughan
,
A.
,
J.
Wang
,
R. P.
Pelletier
,
T.
Nadasdy
,
S.
Brodsky
,
S.
Roy
,
M.
Lodder
,
D.
Bobek
,
S.
Mofatt-Bruce
,
R. L.
Fairchild
, et al
.
2014
.
Key role for CD4 T cells during mixed antibody-mediated rejection of renal allografts.
Am. J. Transplant.
14
:
284
294
.
31
Shields
,
R. L.
,
J.
Lai
,
R.
Keck
,
L. Y.
O’Connell
,
K.
Hong
,
Y. G.
Meng
,
S. H.
Weikert
,
L. G.
Presta
.
2002
.
Lack of fucose on human IgG1 N-linked oligosaccharide improves binding to human FcγRIII and antibody-dependent cellular toxicity.
J. Biol. Chem.
277
:
26733
26740
.
32
Iida
,
S.
,
H.
Misaka
,
M.
Inoue
,
M.
Shibata
,
R.
Nakano
,
N.
Yamane-Ohnuki
,
M.
Wakitani
,
K.
Yano
,
K.
Shitara
,
M.
Satoh
.
2006
.
Nonfucosylated therapeutic IgG1 antibody can evade the inhibitory effect of serum immunoglobulin G on antibody-dependent cellular cytotoxicity through its high binding to FcγRIIIa.
Clin. Cancer Res.
12
:
2879
2887
.
33
van Berkel
,
P. H.
,
J.
Gerritsen
,
E.
van Voskuilen
,
G.
Perdok
,
T.
Vink
,
J. G.
van de Winkel
,
P. W.
Parren
.
2010
.
Rapid production of recombinant human IgG with improved ADCC effector function in a transient expression system.
Biotechnol. Bioeng.
105
:
350
357
.
34
Zhou
,
Q.
,
S.
Shankara
,
A.
Roy
,
H.
Qiu
,
S.
Estes
,
A.
McVie-Wylie
,
K.
Culm-Merdek
,
A.
Park
,
C.
Pan
,
T.
Edmunds
.
2008
.
Development of a simple and rapid method for producing non-fucosylated oligomannose containing antibodies with increased effector function.
Biotechnol. Bioeng.
99
:
652
665
.
35
Garrone
,
P.
,
E. M.
Neidhardt
,
E.
Garcia
,
L.
Galibert
,
C.
van Kooten
,
J.
Banchereau
.
1995
.
Fas ligation induces apoptosis of CD40-activated human B lymphocytes.
J. Exp. Med.
182
:
1265
1273
.
36
Sandhu
,
J.
,
B.
Shpitz
,
S.
Gallinger
,
N.
Hozumi
.
1994
.
Human primary immune response in SCID mice engrafted with human peripheral blood lymphocytes.
J. Immunol.
152
:
3806
3813
.
37
Schneider
,
S.
,
A.
Bruns
,
B.
Moewes
,
B.
Holzknecht
,
G.
Hausdorf
,
G.
Riemekasten
,
A.
Radbruch
,
F.
Hiepe
,
A.
Thiel
.
2002
.
Simultaneous cytometric analysis of (auto)antigen-reactive T and B cell proliferation.
Immunobiology
206
:
484
495
.
38
Mannering
,
S. I.
,
J. S.
Morris
,
K. P.
Jensen
,
A. W.
Purcell
,
M. C.
Honeyman
,
P. M.
van Endert
,
L. C.
Harrison
.
2003
.
A sensitive method for detecting proliferation of rare autoantigen-specific human T cells.
J. Immunol. Methods
283
:
173
183
.
39
Greenwood
,
J.
,
M.
Clark
,
H.
Waldmann
.
1993
.
Structural motifs involved in human IgG antibody effector functions.
Eur. J. Immunol.
23
:
1098
1104
.
40
Griffin
,
D. O.
,
N. E.
Holodick
,
T. L.
Rothstein
.
2011
.
Human B1 cells in umbilical cord and adult peripheral blood express the novel phenotype CD20+CD27+CD43+CD70−. [Published erratum appears in 2011 J. Exp. Med. 208: 67; 409; 871.]
J. Exp. Med.
208
:
67
80
.
41
Griffin
,
D. O.
,
T. L.
Rothstein
.
2011
.
A small CD11b+ human B1 cell subpopulation stimulates T cells and is expanded in lupus.
J. Exp. Med.
208
:
2591
2598
.
42
Skov
,
S.
,
M.
Bonyhadi
,
N.
Odum
,
J. A.
Ledbetter
.
2000
.
IL-2 and IL-15 regulate CD154 expression on activated CD4 T cells.
J. Immunol.
164
:
3500
3505
.
43
Walzer
,
T.
,
M.
Dalod
,
S. H.
Robbins
,
L.
Zitvogel
,
E.
Vivier
.
2005
.
Natural-killer cells and dendritic cells: “l’union fait la force”.
Blood
106
:
2252
2258
.
44
Bowles
,
J. A.
,
S. Y.
Wang
,
B. K.
Link
,
B.
Allan
,
G.
Beuerlein
,
M. A.
Campbell
,
D.
Marquis
,
B.
Ondek
,
J. E.
Wooldridge
,
B. J.
Smith
, et al
.
2006
.
Anti-CD20 monoclonal antibody with enhanced affinity for CD16 activates NK cells at lower concentrations and more effectively than rituximab.
Blood
108
:
2648
2654
.
45
Mössner
,
E.
,
P.
Brünker
,
S.
Moser
,
U.
Püntener
,
C.
Schmidt
,
S.
Herter
,
R.
Grau
,
C.
Gerdes
,
A.
Nopora
,
E.
van Puijenbroek
, et al
.
2010
.
Increasing the efficacy of CD20 antibody therapy through the engineering of a new type II anti-CD20 antibody with enhanced direct and immune effector cell-mediated B-cell cytotoxicity.
Blood
115
:
4393
4402
.
46
Niwa
,
R.
,
M.
Sakurada
,
Y.
Kobayashi
,
A.
Uehara
,
K.
Matsushima
,
R.
Ueda
,
K.
Nakamura
,
K.
Shitara
.
2005
.
Enhanced natural killer cell binding and activation by low-fucose IgG1 antibody results in potent antibody-dependent cellular cytotoxicity induction at lower antigen density.
Clin. Cancer Res.
11
:
2327
2336
.
47
Giesecke
,
C.
,
D.
Frölich
,
K.
Reiter
,
H. E.
Mei
,
I.
Wirries
,
R.
Kuhly
,
M.
Killig
,
T.
Glatzer
,
K.
Stölzel
,
C.
Perka
, et al
.
2014
.
Tissue distribution and dependence of responsiveness of human antigen-specific memory B cells.
J. Immunol.
192
:
3091
3100
.
48
Murphy
,
K. P.
2012
.
Janeway’s Immunobiology.
Garland Science, Taylor & Francis Group
,
New York
.
49
Fillatreau
S.
2015
.
Pathogenic functions of B cells in autoimmune diseases: IFN-γ production joins the criminal gang.
Eur. J. Immunol.
45
:
966
970
.
50
Dahlgren
,
M. W.
,
T.
Gustafsson-Hedberg
,
M.
Livingston
,
H.
Cucak
,
S.
Alsén
,
U.
Yrlid
,
B.
Johansson-Lindbom
.
2015
.
T follicular helper, but not Th1, cell differentiation in the absence of conventional dendritic cells.
J. Immunol.
194
:
5187
5199
.
51
McHeyzer-Williams
,
L. J.
,
P. J.
Milpied
,
S. L.
Okitsu
,
M. G.
McHeyzer-Williams
.
2015
.
Class-switched memory B cells remodel BCRs within secondary germinal centers.
Nat. Immunol.
16
:
296
305
.
52
Woo
,
J.
,
M. P.
Vierboom
,
H.
Kwon
,
D.
Chao
,
S.
Ye
,
J.
Li
,
K.
Lin
,
I.
Tang
,
N. A.
Belmar
,
T.
Hartman
, et al
.
2013
.
PDL241, a novel humanized monoclonal antibody, reveals CD319 as a therapeutic target for rheumatoid arthritis.
Arthritis Res. Ther.
15
:
R207
.
53
Humby
,
F.
,
M.
Bombardieri
,
A.
Manzo
,
S.
Kelly
,
M. C.
Blades
,
B.
Kirkham
,
J.
Spencer
,
C.
Pitzalis
.
2009
.
Ectopic lymphoid structures support ongoing production of class-switched autoantibodies in rheumatoid synovium.
PLoS Med.
6
:
e1
.
54
Moens
,
L.
,
M.
Wuyts
,
I.
Meyts
,
K.
De Boeck
,
X.
Bossuyt
.
2008
.
Human memory B lymphocyte subsets fulfill distinct roles in the anti-polysaccharide and anti-protein immune response.
J. Immunol.
181
:
5306
5312
.
55
Takemura
,
S.
,
P. A.
Klimiuk
,
A.
Braun
,
J. J.
Goronzy
,
C. M.
Weyand
.
2001
.
T cell activation in rheumatoid synovium is B cell dependent.
J. Immunol.
167
:
4710
4718
.
56
Pescovitz
,
M. D.
,
C. J.
Greenbaum
,
H.
Krause-Steinrauf
,
D. J.
Becker
,
S. E.
Gitelman
,
R.
Goland
,
P. A.
Gottlieb
,
J. B.
Marks
,
P. F.
McGee
,
A. M.
Moran
, et al
Type 1 Diabetes TrialNet Anti-CD20 Study Group
.
2009
.
Rituximab, B-lymphocyte depletion, and preservation of β-cell function.
N. Engl. J. Med.
361
:
2143
2152
.
57
Bingham
, III,
C. O.
,
R. J.
Looney
,
A.
Deodhar
,
N.
Halsey
,
M.
Greenwald
,
C.
Codding
,
B.
Trzaskoma
,
F.
Martin
,
S.
Agarwal
,
A.
Kelman
.
2010
.
Immunization responses in rheumatoid arthritis patients treated with rituximab: results from a controlled clinical trial.
Arthritis Rheum.
62
:
64
74
.
58
DiLillo
,
D. J.
,
Y.
Hamaguchi
,
Y.
Ueda
,
K.
Yang
,
J.
Uchida
,
K. M.
Haas
,
G.
Kelsoe
,
T. F.
Tedder
.
2008
.
Maintenance of long-lived plasma cells and serological memory despite mature and memory B cell depletion during CD20 immunotherapy in mice.
J. Immunol.
180
:
361
371
.
59
Pescovitz
,
M. D.
,
T. R.
Torgerson
,
H. D.
Ochs
,
E.
Ocheltree
,
P.
McGee
,
H.
Krause-Steinrauf
,
J. M.
Lachin
,
J.
Canniff
,
C.
Greenbaum
,
K. C.
Herold
, et al
.
2011
.
Effect of rituximab on human in vivo antibody immune responses.
J. Allergy Clin. Immunol.
128
:
1295
1302.e5
.
60
van Assen
,
S.
,
A.
Holvast
,
C. A.
Benne
,
M. D.
Posthumus
,
M. A.
van Leeuwen
,
A. E.
Voskuyl
,
M.
Blom
,
A. P.
Risselada
,
A.
de Haan
,
J.
Westra
, et al
.
2010
.
Humoral responses after influenza vaccination are severely reduced in patients with rheumatoid arthritis treated with rituximab.
Arthritis Rheum.
62
:
75
81
.
61
Breton
,
C. S.
,
A.
Nahimana
,
D.
Aubry
,
J.
Macoin
,
P.
Moretti
,
M.
Bertschinger
,
S.
Hou
,
M. A.
Duchosal
,
J.
Back
.
2014
.
A novel anti-CD19 monoclonal antibody (GBR 401) with high killing activity against B cell malignancies.
J. Hematol. Oncol.
7
:
33
.
62
Williams
,
S. S.
,
T.
Umemoto
,
H.
Kida
,
E. A.
Repasky
,
R. B.
Bankert
.
1992
.
Engraftment of human peripheral blood leukocytes into severe combined immunodeficient mice results in the long term and dynamic production of human xenoreactive antibodies.
J. Immunol.
149
:
2830
2836
.
63
Tangye
,
S. G.
2013
.
To B1 or not to B1: that really is still the question!
Blood
121
:
5109
5110
.
64
Dieker
,
J.
,
J.
Tel
,
E.
Pieterse
,
A.
Thielen
,
N.
Rother
,
M.
Bakker
,
J.
Fransen
,
H.
Dijkman
,
J.
Berden
,
J. M.
de Vries
, et al
.
2015
.
Circulating apoptotic microparticles in SLE patients drive the activation of DC subsets and prime neutrophils for NETosis.
Arthritis Rheumatol.
68
:
462
472
.
65
Hiepe
,
F.
,
T.
Dörner
,
A. E.
Hauser
,
B. F.
Hoyer
,
H.
Mei
,
A.
Radbruch
.
2011
.
Long-lived autoreactive plasma cells drive persistent autoimmune inflammation.
Nat. Rev. Rheumatol.
7
:
170
178
.
66
Shaffer
,
A. L.
,
K. I.
Lin
,
T. C.
Kuo
,
X.
Yu
,
E. M.
Hurt
,
A.
Rosenwald
,
J. M.
Giltnane
,
L.
Yang
,
H.
Zhao
,
K.
Calame
,
L. M.
Staudt
.
2002
.
Blimp-1 orchestrates plasma cell differentiation by extinguishing the mature B cell gene expression program.
Immunity
17
:
51
62
.
67
Tarte
,
K.
,
F.
Zhan
,
J.
De Vos
,
B.
Klein
,
J.
Shaughnessy
Jr
.
2003
.
Gene expression profiling of plasma cells and plasmablasts: toward a better understanding of the late stages of B-cell differentiation.
Blood
102
:
592
600
.
68
Zhan
,
F.
,
E.
Tian
,
K.
Bumm
,
R.
Smith
,
B.
Barlogie
,
J.
Shaughnessy
Jr
.
2003
.
Gene expression profiling of human plasma cell differentiation and classification of multiple myeloma based on similarities to distinct stages of late-stage B-cell development.
Blood
101
:
1128
1140
.
69
Slocombe
,
T.
,
S.
Brown
,
K.
Miles
,
M.
Gray
,
T. A.
Barr
,
D.
Gray
.
2013
.
Plasma cell homeostasis: the effects of chronic antigen stimulation and inflammation.
J. Immunol.
191
:
3128
3138
.
70
Pitzalis
,
C.
,
G. W.
Jones
,
M.
Bombardieri
,
S. A.
Jones
.
2014
.
Ectopic lymphoid-like structures in infection, cancer and autoimmunity.
Nat. Rev. Immunol.
14
:
447
462
.
71
Tiller
,
T.
,
M.
Tsuiji
,
S.
Yurasov
,
K.
Velinzon
,
M. C.
Nussenzweig
,
H.
Wardemann
.
2007
.
Autoreactivity in human IgG+ memory B cells.
Immunity
26
:
205
213
.
72
Scheid
,
J. F.
,
H.
Mouquet
,
J.
Kofer
,
S.
Yurasov
,
M. C.
Nussenzweig
,
H.
Wardemann
.
2011
.
Differential regulation of self-reactivity discriminates between IgG+ human circulating memory B cells and bone marrow plasma cells.
Proc. Natl. Acad. Sci. USA
108
:
18044
18048
.
73
Seshadri
,
S.
,
Y.
Kannan
,
S.
Mitra
,
J.
Parker-Barnes
,
M. D.
Wewers
.
2009
.
MAIL regulates human monocyte IL-6 production.
J. Immunol.
183
:
5358
5368
.
74
Myers
,
L. K.
,
B.
Tang
,
J. M.
Stuart
,
A. H.
Kang
.
2002
.
The role of IL-4 in regulation of murine collagen-induced arthritis.
Clin. Immunol.
102
:
185
191
.
75
Geginat
,
J.
,
M.
Paroni
,
S.
Maglie
,
J. S.
Alfen
,
I.
Kastirr
,
P.
Gruarin
,
M.
De Simone
,
M.
Pagani
,
S.
Abrignani
.
2014
.
Plasticity of human CD4 T cell subsets.
Front. Immunol.
5
:
630
.
76
Nakayamada
,
S.
,
H.
Takahashi
,
Y.
Kanno
,
J. J.
O’Shea
.
2012
.
Helper T cell diversity and plasticity.
Curr. Opin. Immunol.
24
:
297
302
.
77
Molnarfi
,
N.
,
U.
Schulze-Topphoff
,
M. S.
Weber
,
J. C.
Patarroyo
,
T.
Prod’homme
,
M.
Varrin-Doyer
,
A.
Shetty
,
C.
Linington
,
A. J.
Slavin
,
J.
Hidalgo
, et al
.
2013
.
MHC class II-dependent B cell APC function is required for induction of CNS autoimmunity independent of myelin-specific antibodies.
J. Exp. Med.
210
:
2921
2937
.
78
Sheng
,
W.
,
F.
Yang
,
Y.
Zhou
,
H.
Yang
,
P. Y.
Low
,
D. M.
Kemeny
,
P.
Tan
,
A.
Moh
,
M. H.
Kaplan
,
Y.
Zhang
,
X. Y.
Fu
.
2014
.
STAT5 programs a distinct subset of GM-CSF-producing T helper cells that is essential for autoimmune neuroinflammation.
Cell Res.
24
:
1387
1402
.
79
Shiomi
,
A.
,
T.
Usui
.
2015
.
Pivotal roles of GM-CSF in autoimmunity and inflammation.
Mediators Inflamm.
2015
:
568543
.
80
Zinser
,
E.
,
M.
Lechmann
,
A.
Golka
,
M. B.
Lutz
,
A.
Steinkasserer
.
2004
.
Prevention and treatment of experimental autoimmune encephalomyelitis by soluble CD83.
J. Exp. Med.
200
:
345
351
.
81
Solari
,
M. G.
,
A. W.
Thomson
.
2008
.
Human dendritic cells and transplant outcome.
Transplantation
85
:
1513
1522
.
82
Djamali
,
A.
,
D. B.
Kaufman
,
T. M.
Ellis
,
W.
Zhong
,
A.
Matas
,
M.
Samaniego
.
2014
.
Diagnosis and management of antibody-mediated rejection: current status and novel approaches.
Am. J. Transplant.
14
:
255
271
.
83
Shimabukuro-Vornhagen
,
A.
,
M. J.
Hallek
,
R. F.
Storb
,
M. S.
von Bergwelt-Baildon
.
2009
.
The role of B cells in the pathogenesis of graft-versus-host disease.
Blood
114
:
4919
4927
.
84
Heidt
,
S.
,
D. S.
Segundo
,
R.
Chadha
,
K. J.
Wood
.
2010
.
The impact of Th17 cells on transplant rejection and the induction of tolerance. [Published erratum appears in 2010 Curr. Opin. Organ Transplant. 15: 670.]
Curr. Opin. Organ Transplant.
15
:
456
461
.
85
Homs
,
S.
,
H.
Mansour
,
D.
Desvaux
,
C.
Diet
,
M.
Hazan
,
M.
Buchler
,
Y.
Lebranchu
,
D.
Buob
,
C.
Badoual
,
M.
Matignon
, et al
.
2009
.
Predominant Th1 and cytotoxic phenotype in biopsies from renal transplant recipients with transplant glomerulopathy.
Am. J. Transplant.
9
:
1230
1236
.
86
Tellides
,
G.
,
J. S.
Pober
.
2007
.
Interferon-γ axis in graft arteriosclerosis.
Circ. Res.
100
:
622
632
.
87
Serody
,
J. S.
,
G. R.
Hill
.
2012
.
The IL-17 differentiation pathway and its role in transplant outcome.
Biol. Blood Marrow Transplant.
18
(
1
Suppl.
):
S56
S61
.
88
Reinwald
,
S.
,
C.
Wiethe
,
A. M.
Westendorf
,
M.
Breloer
,
M.
Probst-Kepper
,
B.
Fleischer
,
A.
Steinkasserer
,
J.
Buer
,
W.
Hansen
.
2008
.
CD83 expression in CD4+ T cells modulates inflammation and autoimmunity.
J. Immunol.
180
:
5890
5897
.
89
Kretschmer
,
B.
,
K.
Lüthje
,
S.
Schneider
,
B.
Fleischer
,
M.
Breloer
.
2009
.
Engagement of CD83 on B cells modulates B cell function in vivo.
J. Immunol.
182
:
2827
2834
.
90
Kretschmer
,
B.
,
J.
Weber
,
A.
Hutloff
,
B.
Fleischer
,
M.
Breloer
,
A.
Osterloh
.
2015
.
Anti-CD83 promotes IgG1 isotype switch in marginal zone B cells in response to TI-2 antigen.
Immunobiology
220
:
964
975
.
91
Hazenbos
,
W. L.
,
I. A.
Heijnen
,
D.
Meyer
,
F. M.
Hofhuis
,
C. R.
Renardel de Lavalette
,
R. E.
Schmidt
,
P. J.
Capel
,
J. G.
van de Winkel
,
J. E.
Gessner
,
T. K.
van den Berg
,
J. S.
Verbeek
.
1998
.
Murine IgG1 complexes trigger immune effector functions predominantly via FcγRIII (CD16).
J. Immunol.
161
:
3026
3032
.
92
Krzyzak
,
L.
,
C.
Seitz
,
A.
Urbat
,
S.
Hutzler
,
C.
Ostalecki
,
J.
Gläsner
,
A.
Hiergeist
,
A.
Gessner
,
T. H.
Winkler
,
A.
Steinkasserer
,
L.
Nitschke
.
2016
.
CD83 modulates B cell activation and germinal center responses.
J. Immunol.
196
:
3581
3594
.
93
Cerutti
,
A.
,
M.
Cols
,
I.
Puga
.
2013
.
Marginal zone B cells: virtues of innate-like antibody-producing lymphocytes.
Nat. Rev. Immunol.
13
:
118
132
.
94
Vinuesa
,
C. G.
,
P. P.
Chang
.
2013
.
Innate B cell helpers reveal novel types of antibody responses.
Nat. Immunol.
14
:
119
126
.

T.S., M.L.J., and D.J.M. are named as inventors for the use of anti-CD83 mAb in the treatment of GVHD. The other authors have no financial conflicts of interest.

Supplementary data