Human IgG1 type I CD20 Abs, such as rituximab and ofatumumab (OFA), efficiently induce complement-dependent cytotoxicity (CDC) of CD20+ B cells by binding of C1 to hexamerized Fc domains. Unexpectedly, we found that type I CD20 Ab F(ab′)2 fragments, as well as C1q-binding–deficient IgG mutants, retained an ability to induce CDC, albeit with lower efficiency than for whole or unmodified IgG. Experiments using human serum depleted of specific complement components demonstrated that the observed lytic activity, which we termed “accessory CDC,” remained to be dependent on C1 and the classical pathway. We hypothesized that CD20 Ab-induced clustering of the IgM or IgG BCR was involved in accessory CDC. Indeed, accessory CDC was consistently observed in B cell lines expressing an IgM BCR and in some cell lines expressing an IgG BCR, but it was absent in BCR B cell lines. A direct relationship between BCR expression and accessory CDC was established by transfecting the BCR into CD20+ cells: OFA-F(ab′)2 fragments were able to induce CDC in the CD20+BCR+ cell population, but not in the CD20+BCR population. Importantly, OFA-F(ab′)2 fragments were able to induce CDC ex vivo in malignant B cells isolated from patients with mantle cell lymphoma and Waldenström macroglobulinemia. In summary, accessory CDC represents a novel effector mechanism that is dependent on type I CD20 Ab–induced BCR clustering. Accessory CDC may contribute to the excellent capacity of type I CD20 Abs to induce CDC, and thereby to the antitumor activity of such Abs in the clinic.

This article is featured in In This Issue, p.4507

The type I CD20–specific mAbs such as rituximab (RTX) and ofatumumab (OFA) efficiently recruit executor components of the innate immune system, leading to efficient depletion of CD20+ tumor cells in B cell lymphomas and B cell chronic lymphocytic leukemia (CLL) (1, 2). The mechanisms employed include Ab-dependent cell-mediated cytotoxicity (ADCC), Ab-dependent cell-mediated phagocytosis (ADCP), and, specifically, complement-dependent cytotoxicity (CDC) (36). This is in contrast to type II CD20 Abs (e.g., obinutuzumab, 11B8), which are also potent inducers of ADCC and ADCP but are inefficient inducers of CDC (4, 7). The difference in CDC capacity between these type I and type II CD20 Abs is not due to differences in the intrinsic capacity to activate complement, as all Abs are of the IgG1 isotype. Type II CD20 Abs furthermore have been shown to induce CDC when present at high concentrations in whole blood (8). Therefore, the different capacity to induce CDC upon binding to CD20-expressing cells instead seems related to a different density or orientation of Ab Fc domains upon binding to CD20 (710). Related to this, type I but not type II Abs induce redistribution of CD20 into lipid raft domains.

Ab-mediated CDC is dependent on the classical pathway of complement activation, which starts when complement component C1q binds to the CH2 domain in the Fc region of complexed IgG or IgM (11). C1 binding and activation requires its binding to the C1q binding site on an array of Fc domains. On IgM molecules, the C1q binding site is presented after conformational changes following Ag binding (12, 13). Efficient C1 binding to IgG molecules is dependent on hexamerization of IgG Fc domains of Abs bound to cell surface Ag, which is mediated by extensive intermolecular Fc–Fc contacts between neighboring Abs (11, 1416). For our studies on the mechanism of complement activation by type I CD20 Abs, we generated F(ab′)2 fragments of OFA and RTX. These F(ab′)2 fragments were intended to serve as negative controls in complement activation studies, because with the absence of an IgG Fc domain, they cannot bind C1q and activate the complement pathway. Remarkably, we observed complement-dependent lysis of B cell lymphoma cells upon incubation with these type I CD20 Ab F(ab′)2 fragments. In this study, we investigated the underlying mechanisms behind this seemingly IgG Fc-independent complement activation. We show that type I CD20 Abs, in addition to the recruitment of complement components through the Fc domain, are able to induce CDC by activation of the classical complement pathway in a BCR-dependent manner.

The human lymphoma cell lines Daudi, Raji, WIL2-S, ARH-77, DOHH-2, CA-46, and Ramos were obtained from the European Cell Culture Collection (Porton Down, U.K.), and the human lymphoma cell lines SU-DHL-4, SU-DHL-8, DB, JIYOYE, CL-1, EB-01, HAL-1, MEC-2, WSU-NHL, OCI-Ly7, RI-1, OCI-Ly19, and RC-K8 were obtained from the Deutsche Sammlung von Mikroorganismen und Zellkulturen (Braunschweig, Germany). Alemtuzumab-sensitive Wien 133 cells (Burkitt lymphoma cell line) were provided by Dr. Geoff Hale (BioAnaLab, Oxford, U.K.).

Daudi, Raji, WIL2-S, ARH-77, DOHH-2, Ramos, WSU-NHL, CA-46, and SU-DHL-4 cells were cultured in RPMI 1640 supplemented with 10% heat-inactivated FCS, 1 U/ml penicillin, 1 μg/ml streptomycin, and 4 mM l-glutamine. CL-1 and SU-DHL-8 cells were cultured in RPMI 1640 supplemented with 20% heat-inactivated FCS. RC-K8, EB-01, HAL-1, OCI-Ly18, OCI-Ly19, RI-1, DB, and JIYOYE cells were cultured in RPMI 1640 supplemented with 10% heat-inactivated FCS. MEC-2 and Wien 133 cells were cultured in IMDM supplemented with 10% heat-inactivated FCS, 1 U/ml penicillin, and 1 μg/ml streptomycin. OCI-Ly7 was cultured in IMDM supplemented with 20% heat-inactivated FCS, 1 U/ml penicillin, and 1 μg/ml streptomycin. All media and supplements were obtained from Lonza (Vervier, Belgium).

Mononuclear cells were isolated from peripheral blood of CLL, mantle cell lymphoma (MCL), or Waldenström macroglobulinemia (WM) patients as described previously (17). Blood was drawn after receiving the donors’ written informed consents. Experiments reported in this study were approved by the Ethics Committee of the Christian Albrechts University (Kiel, Germany) in accordance with the Declaration of Helsinki. CLL, MCL, or WM cells were cultured in RPMI 1640 supplemented with 10% heat-inactivated FCS, 1 U/ml penicillin, and 1 μg/ml streptomycin.

Pooled normal human serum (NHS; AB positive) was obtained from Sanquin (Amsterdam, the Netherlands). C1q-, C5-, C8-, or factor B–depleted sera, as well as purified complement components C5 and C8, were obtained from Quidel (San Diego, CA). The procedure to deplete sera of individual complement factors resulted sometimes in a modest reduced activity of these depleted sera. As a control, the activity of depleted serum was checked by assessing the CH50 or AP50 titer after titrating back the depleted component. C1q that had been purified from sera from multiple donors was a gift from Prof. M. Daha (Leiden University Medical Center). Heat-inactivated NHS was prepared by incubating NHS for 30 min at 56°C.

Ca2+-deficient buffer was generated by supplementing PBS with 8 mM EGTA and 2.5 mM MgCl2 (EGTA-Mg2+). Ca2+Mg2+-deficient buffer was prepared by adding 10 mM EDTA to PBS.

Clinical OFA was obtained from GlaxoSmithKline. 7D8 and 11B8 are human, CD20-specific Abs of the IgG1 isotype and were recombinantly produced at Genmab as described (18). Fc mutations to knock out C1q binding were introduced in mAb 7D8. mAb 7D8 only differs at only four amino acid positions from OFA; it binds the same epitope on CD20 and has similar functional characteristics, including similarly efficient induction of CDC. To generate 7D8 variants with reduced or abrogated capacity to activate complement, amino acid mutations in the Fc domain were introduced in the expression plasmids encoding the Ab H chain using the QuikChange XL kit (Stratagene, La Jolla, CA) according to the manufacturer’s guidelines. Mutations were verified by DNA sequencing (LGC Genomics, Berlin, Germany). RTX was obtained from Roche (Grenzach-Whylen, Germany). Alemtuzumab is a humanized CD52-specific IgG1 and was obtained from Genzyme (Cambridge, MA).

F(ab′)2 fragments of OFA, RTX, 7D8, and alemtuzumab were obtained by digestion of IgG with 2 mg/ml pepsin (Sigma-Aldrich) for 30–60 min. Digested material was dialyzed with PBS and loaded on a protein A column (Sigma-Aldrich) to separate the F(ab′)2 fragments from the Fc tails. The purity of the F(ab′)2 fragments was confirmed by SDS-PAGE and ELISA.

Mouse anti-human C1q (mIgG1 clone 85, directed to the globular head region of C1q) was a gift from Prof. Lucien Aarden (Sanquin). A mouse IgG1 isotype control Ab was obtained from BD Biosciences (clone MOPC-21).

Rabbit anti-human C1q (Dako, Glostrup, Denmark), mouse anti-human C4b (Brunschwig, Basel, Switzerland), and peroxidase-conjugated rabbit anti-mouse IgG (1:1000 dilution; Jackson ImmunoResearch Laboratories, Philadelphia, PA) were used for ELISA experiments.

Mouse anti-CD55 (555696; BD Biosciences) and anti-IgM Ab HB-57–allophycocyanin (a gift from Ron Taylor) were used for FACS experiments.

For transfection experiments an anti-human BCR (mouse anti-2F8; BioGenes, Berlin, Germany) followed by rat anti-mouse allophycocyanin (BD Biosciences) and CD20 [OFA-FITC, OFA-F(ab′)2-FITC, or 11B8-FITC; Genmab] were used.

The capacity of immobilized Igs or F(ab′)2 fragments to bind C1q or induce C4b deposition was assessed by ELISA as described previously (18).

CDC was assessed by either a standard chromium release assay utilizing 25% NHS or a propidium iodide (PI) exclusion assay as described previously (18, 19). To test the contribution of C1q in CDC experiments, 10% NHS (v/v) was pretreated with 50 μg/ml mouse anti-human C1q, 50 μg/ml isotype control Ab, or buffer before addition to the cells. To determine the contribution of endogenous Abs to CDC, IgM-deficient cord blood serum (a gift from Lucien Aarden) or IgG-depleted human serum that was obtained by passing NHS over a protein G column (HiTrap; GE Healthcare) was used as complement source. After incubation for 45 min at 37°C, cells were harvested, PI was added, and lysis was detected by flow cytometry (FACSCanto II; BD Biosciences).

Surface expression of CD20, CD55, IgM-BCR, and/or IgG-BCR on B cell lymphoma cell lines and mononuclear cells isolated from patients was assessed by flow cytometry. One hundred thousand target cells were incubated with specific Abs at saturating concentrations for 30 min at 4°C. After washing, cells were analyzed using a FACSCanto II flow cytometer (BD Biosciences).

To separately analyze normal peripheral blood B cell populations with high and low BCR expression, B cells were isolated from buffy coats obtained from healthy donors (Sanquin) using the untouched B cell depletion kit (Dynabeads, 11351D; Thermo Fischer) according to the manufacturer’s instruction. The B cells were subsequently separated into an IgM BCRhigh and an IgM BCRlow population using FITC-labeled Fab fragment targeting Igμ (PL-373; de Beer Medicals, Driessen, the Netherlands) and a FACSAria III cell sorter (BD Biosciences). To assess CDC in the IgM BCRhigh and BCRlow populations, cells were exposed to OFA in the presence of NHS, and CDC was assessed by the PI exclusion assay as described above.

Human CD20+ CEM T cells were obtained by viral transduction as previously described (20). To generate CD20+BCR+ and CD20+BCR cells, CD20+ CEM T cells were cultured to log phase in RPMI 1640 supplemented with 10% FCS and transfected with either IgG1 BCR (pG1f-TM2F8, TM2F8 was cloned into pEE6.4; Lonza, Cologne, Germany) or empty vector (empty plasmid pEE6.4) with the Amaxa Nucleofector kit (Lonza, Cologne, Germany), according to the manufacturer’s instructions.

Two days after transfection, cells were washed and incubated with saturating concentrations of BCR-specific Abs as described above. Cell death was determined by PI exclusion as described above.

Statistical analysis was performed using a one-way ANOVA with a Dunnett multiple comparisons posttest. A p value ≤0.05 was considered significant.

The CD20 Abs, OFA, RTX, and 11B8, as well as OFA-F(ab′)2 fragments, were immobilized on ELISA plates and incubated with NHS to assess their capacity to bind C1q, the first step in the classical pathway of complement activation. All intact Abs showed efficient C1q binding (Fig. 1A). As expected, OFA-F(ab′)2 fragments were unable to bind C1q. Similarly, surface-immobilized OFA, RTX, and 11B8, but not OFA-F(ab′)2 fragments, induced deposition of C4b (Fig. 1B), an activation product of the classical complement pathway that is generated upon C1-dependent cleavage of C4.

FIGURE 1.

Type I CD20 Abs or Ab fragments that lack the capacity to bind C1q are able to elicit lysis of Daudi cells in the presence of NHS. (A and B) OFA, OFA-F(ab′)2 fragments, RTX, and 11B8 were immobilized on ELISA plates in the presence of NHS. C1q binding (A) and C4b deposition (B) were assessed by ELISA. (C) OFA, OFA-F(ab′)2 fragments, RTX, and 11B8 were incubated with Daudi cells in presence of NHS, and tumor cell lysis was assessed by quantifying PI+ cells using flow cytometry. (D) 7D8 and 7D8-IgG4-K322A were incubated with Daudi cells in presence of NHS, and CDC was assessed by quantifying PI+ cells using flow cytometry. (E) OFA and OFA-F(ab′)2 fragments were incubated with Daudi cells in presence of NHS or heat-inactivated NHS (HI-NHS), and CDC was assessed by quantifying PI+ cells using flow cytometry. Data shown are representative of at least three replicate experiments.

FIGURE 1.

Type I CD20 Abs or Ab fragments that lack the capacity to bind C1q are able to elicit lysis of Daudi cells in the presence of NHS. (A and B) OFA, OFA-F(ab′)2 fragments, RTX, and 11B8 were immobilized on ELISA plates in the presence of NHS. C1q binding (A) and C4b deposition (B) were assessed by ELISA. (C) OFA, OFA-F(ab′)2 fragments, RTX, and 11B8 were incubated with Daudi cells in presence of NHS, and tumor cell lysis was assessed by quantifying PI+ cells using flow cytometry. (D) 7D8 and 7D8-IgG4-K322A were incubated with Daudi cells in presence of NHS, and CDC was assessed by quantifying PI+ cells using flow cytometry. (E) OFA and OFA-F(ab′)2 fragments were incubated with Daudi cells in presence of NHS or heat-inactivated NHS (HI-NHS), and CDC was assessed by quantifying PI+ cells using flow cytometry. Data shown are representative of at least three replicate experiments.

Close modal

To assess CDC, we incubated OFA, RTX, 11B8, and OFA-F(ab′)2 fragments with Daudi Burkitt’s lymphoma cells in the presence of NHS. To our surprise, not only the full-length Abs, but also the OFA-F(ab′)2 fragments, induced Daudi cell lysis (Fig. 1C), albeit the OFA-F(ab′)2 fragments induced lysis more slowly and with lower maximal kill. Nevertheless, OFA-F(ab′)2 fragments reproducibly killed at least 50% of the tumor cells. Similarly, RTX-F(ab′)2 fragments induced lysis in Daudi cells in the presence of NHS, although not as efficiently as did RTX (Supplemental Fig. 1A). As previously reported, 11B8 was unable to induce CDC, confirming that type II CD20 Abs are poor inducers of CDC (18). To confirm that CDC induced by the CD20 Ab F(ab′)2 fragments is a general phenomenon that may also be mediated by whole Abs lacking a C1q binding site, we generated a set of IgG1 mutants in the CD20 Ab 7D8. Indeed, an IgG4 variant of 7D8 comprising a lysine to alanine mutation in the Fc fragment (K322A), which completed lacking C1q binding and complement activation when immobilized in ELISA (Supplemental Fig. 1C, 1D), induced CDC of Daudi cells in human serum (Fig. 1D). Similar results were obtained with other 7D8 mutants lacking C1q binding (Supplemental Fig. 1C, 1D). Our observations appeared to be specific for CD20 Abs, as F(ab′)2 fragments or C1q binding-deficient mutants of the potent CDC-inducing CD52 Ab alemtuzumab (21) were unable to induce cytotoxicity (data not shown).

OFA-F(ab′)2 fragment– and 7D8-IgG4-K322A–mediated cytotoxicity was observed in the presence of human serum, but not heat-inactivated human serum (Fig. 1E and data not shown), indicating that active complement was required. In support of this, Daudi cells that survived exposure to OFA-F(ab′)2 fragments showed significantly higher expression of the complement regulatory protein CD55, in comparison with the whole population before OFA-F(ab′)2 fragment treatment (Supplemental Fig. 2A). Similar data were previously obtained in B cell lymphoma cells that survived exposure to RTX-mediated CDC in the presence of active complement, presumably due to CDC resistance in the cell population with the highest CD55 expression (22).

To study the role of the different complement activation pathways in OFA-F(ab′)2 fragment–induced cytotoxicity, a set of experiments was performed in presence of NHS that had been depleted of Ca2+ and Mg2+, resulting in complete abrogation of all complement activity, or of Ca2+ alone, resulting in inactivation of the classical and lectin complement pathways, but not the alternative pathway. Under these conditions, both OFA-F(ab′)2 fragments and intact OFA were unable to induce cytotoxicity in Daudi cells (Fig. 2A). This confirmed that active complement was required and demonstrated that complement activation did not occur through the alternative pathway.

FIGURE 2.

OFA-F(ab′)2 fragment–mediated cytotoxicity is dependent on the classical pathway of complement activation. (AD) OFA and OFA-F(ab′)2 fragments were incubated with Daudi cells in the presence of normal NHS or NHS that had been depleted of Ca2+ and Mg2+ or Ca2+ alone (A), complement components C5 or C8 (B), factor B (C), or C1q (D). Cytotoxicity of OFA and OFA-F(ab′)2 fragments was assessed by flow cytometric quantification of PI+ cells. (E) Cytotoxicity of OFA, 7D8, 7D8-F(ab′)2 fragments, and 7D8-IgG4-K322A in NHS that had been preincubated with a C1q Ab or an isotype control Ab. Cytotoxicity was assessed as described above. All experiments were performed in triplicate; representative experiments are shown.

FIGURE 2.

OFA-F(ab′)2 fragment–mediated cytotoxicity is dependent on the classical pathway of complement activation. (AD) OFA and OFA-F(ab′)2 fragments were incubated with Daudi cells in the presence of normal NHS or NHS that had been depleted of Ca2+ and Mg2+ or Ca2+ alone (A), complement components C5 or C8 (B), factor B (C), or C1q (D). Cytotoxicity of OFA and OFA-F(ab′)2 fragments was assessed by flow cytometric quantification of PI+ cells. (E) Cytotoxicity of OFA, 7D8, 7D8-F(ab′)2 fragments, and 7D8-IgG4-K322A in NHS that had been preincubated with a C1q Ab or an isotype control Ab. Cytotoxicity was assessed as described above. All experiments were performed in triplicate; representative experiments are shown.

Close modal

To study which components of complement were required for the cytotoxicity of OFA-F(ab′)2 fragments, we performed experiments in serum that had been depleted of specific complement components. Neither OFA-F(ab′)2 fragments nor OFA showed CDC activity in sera depleted of C5 or C8 (Fig. 2B), components that are essential for the initiation of terminal complement activation and formation of the membrane attack complex, respectively (23). In contrast, OFA-F(ab′)2 fragments were able to induce lysis of Daudi cells in the presence of NHS that had been depleted of factor B (Fig. 2C). This confirmed that the alternative pathway was not involved in OFA-F(ab′)2 fragment–dependent CDC, as factor B is an essential intermediate in the alternative pathway, but not the classical or lectin pathways.

Finally, OFA-F(ab′)2 fragment–mediated cytotoxicity was shown to be mediated through the classical pathway and not the lectin pathway. OFA-F(ab′)2 fragments were unable to induce cytotoxicity in C1q-depleted serum, whereas reconstitution with C1q restored cytotoxic activity (Fig. 2D). Furthermore, an Ab specific for the globular head of C1q was able to completely inhibit the cytotoxic activity of 7D8, 7D8-F(ab′)2 fragments, and 7D8-IgG4-K322A in NHS (Fig. 2E).

Activation of the classical pathway may also be induced by other serum factors such as naturally occurring autoantibodies (e.g., anti-hinge Abs) or the pentraxins serum amyloid P (24) and C-reactive protein (25, 26). However, a role for these factors in OFA-F(ab′)2 fragment–mediated CDC was excluded by performing experiments in serum devoid of IgM (Supplemental Fig. 3A) or depleted for IgG (Supplemental Fig. 3B), serum amyloid P, or C-reactive protein (data not shown).

Taken together, these results show that type I CD20 F(ab′)2 fragments or Ab variants that are unable to bind C1, retain the capacity to induce complement-mediated lysis via the classical pathway, with, paradoxically, a critical role for C1q.

We hypothesized that an alternative docking site for C1q must be present on the plasma membrane, in or near the CD20-OFA-F(ab′)2 fragment immune complexes. The BCR complex is expressed in close proximity to CD20 on B cells (27) and contains a cell surface–bound Ab (surface Ig) that could provide such a C1q docking site. We addressed the potential role of the BCR in OFA-F(ab′)2 fragment–mediated complement activation by measuring induction of CDC in a panel of human B cell lymphoma cell lines that displayed variable levels of IgM- or IgG-BCRs on the cell surface. OFA-F(ab′)2 fragments were able to induce CDC in seven out of eight cell lines with an IgM BCR (Fig. 3A, Table I), indicating that our observations in Daudi cells seem to represent a common phenomenon for B cell lines expressing a BCR of the IgM subclass. In contrast, OFA-F(ab′)2 fragment–mediated cytotoxicity was observed in only one out of seven IgG-BCR–expressing cell lines, whereas all of these cell lines were susceptible to CDC induced by intact OFA (Fig. 3A, Table I). OFA-F(ab′)2 fragment–mediated CDC was not observed in any of the six B cell lines devoid of BCR expression (Fig. 3A, Table I). The different sensitivity of IgM- and IgG-BCR–expressing cell lines for OFA-F(ab′)2 fragment–mediated CDC could not be attributed to differences in BCR or CD20 expression levels, although analysis of a larger number of cell lines would be required to draw firm conclusions. When examining the lysis of Daudi cells more closely, we observed that the average BCR expression level of cells that survived exposure to OFA-F(ab′)2 fragments in the presence of NHS was lower than the average BCR expression in untreated cells (Supplemental Fig. 2B). This again suggests that susceptibility to OFA-F(ab′)2 fragment–mediated CDC may be linked to its BCR expression.

FIGURE 3.

OFA-F(ab′)2 fragments induces CDC in B cells with different BCR subtypes. (A) CDC activity of OFA and OFA-F(ab′)2 fragments was assessed in the IgM-BCR–expressing OCI-Ly7 cell line (left panel), the IgG-BCR–expressing SU-DHL-4 cell line (middle panel), and the BCR Raji cell line. Error bars indicate the SD of three replicates samples per experiment. Representative examples of at least triplicate experiments are shown. Statistical analysis was done with a one-way ANOVA followed by a Dunnett multiple comparisons posttest to compare the mean of each column to that of the no Ab control. OFA induced lysis of all three cell lines (one-way ANOVA, p < 0.05), whereas OFA-F(ab′)2 fragments only induced significant lysis of OCI-Ly7 cells expressing an IgM BCR and SU-DHL-4 cells expressing an IgG BCR (one-way ANOVA, p < 0.05) but not of BCR Raji cells. *p < 0.05, ***p < 0.0005, ****p < 0.0001. (B) CDC activity of OFA, OFA-F(ab′)2 fragments, and anti-IgM-F(ab′)2 fragments in the OCI-Ly7 DLBCL cell line. CDC was assessed by flow cytometric evaluation of PI+ cells. The figure shows a representative example of at least three experiments.

FIGURE 3.

OFA-F(ab′)2 fragments induces CDC in B cells with different BCR subtypes. (A) CDC activity of OFA and OFA-F(ab′)2 fragments was assessed in the IgM-BCR–expressing OCI-Ly7 cell line (left panel), the IgG-BCR–expressing SU-DHL-4 cell line (middle panel), and the BCR Raji cell line. Error bars indicate the SD of three replicates samples per experiment. Representative examples of at least triplicate experiments are shown. Statistical analysis was done with a one-way ANOVA followed by a Dunnett multiple comparisons posttest to compare the mean of each column to that of the no Ab control. OFA induced lysis of all three cell lines (one-way ANOVA, p < 0.05), whereas OFA-F(ab′)2 fragments only induced significant lysis of OCI-Ly7 cells expressing an IgM BCR and SU-DHL-4 cells expressing an IgG BCR (one-way ANOVA, p < 0.05) but not of BCR Raji cells. *p < 0.05, ***p < 0.0005, ****p < 0.0001. (B) CDC activity of OFA, OFA-F(ab′)2 fragments, and anti-IgM-F(ab′)2 fragments in the OCI-Ly7 DLBCL cell line. CDC was assessed by flow cytometric evaluation of PI+ cells. The figure shows a representative example of at least three experiments.

Close modal
Table I.
Sensitivity of various B cell lymphoma cell lines to OFA F(ab′)2 fragment–mediated CDC
Cell LineOFA-F(ab′)2–Mediated CDCBCR Expression
CD20 Expression (Molecules/Cell) (sABC)
SubtypeMolecules/Cell (sABC)
Wien 133 IgM 52,400 102,000 
MEC-2 − IgM 30,500 168,000 
Daudi ++ IgM 150,000 133,000 
OCI-Ly7 ++ IgM 342,000 392,000 
Ramos ++ IgM 171,000 222,000 
ARH-77 − IgG 20,400 223,000 
WSU-NHL − IgG 117,000 339,000 
DOHH-2 − IgG 23,900 249,000 
SU-DHL-4 IgG 77,500 241,000 
OCI-Ly19 − IgG 22,300 47,900 
SU-DHL-8 − IgG 26,000 83,900 
DB − IgG 233,000 307,000 
HAL-01 − Not expressed <7,000 100,000 
Raji − Not expressed <7,000 174,000 
JIYOYE − Not expressed <7,000 40,000 
RC-K8 − Not expressed <7,000 150,000 
Cell LineOFA-F(ab′)2–Mediated CDCBCR Expression
CD20 Expression (Molecules/Cell) (sABC)
SubtypeMolecules/Cell (sABC)
Wien 133 IgM 52,400 102,000 
MEC-2 − IgM 30,500 168,000 
Daudi ++ IgM 150,000 133,000 
OCI-Ly7 ++ IgM 342,000 392,000 
Ramos ++ IgM 171,000 222,000 
ARH-77 − IgG 20,400 223,000 
WSU-NHL − IgG 117,000 339,000 
DOHH-2 − IgG 23,900 249,000 
SU-DHL-4 IgG 77,500 241,000 
OCI-Ly19 − IgG 22,300 47,900 
SU-DHL-8 − IgG 26,000 83,900 
DB − IgG 233,000 307,000 
HAL-01 − Not expressed <7,000 100,000 
Raji − Not expressed <7,000 174,000 
JIYOYE − Not expressed <7,000 40,000 
RC-K8 − Not expressed <7,000 150,000 

Surface expression of CD20 and the BCR was assessed by flow cytometry. To assess CDC, cells were incubated with OFA-F(ab′)2 fragments in the presence of 20% NHS. Specific lysis was assessed by quantifying PI+ cells using flow cytometry.

−, no lysis; +, percentage specific lysis ≥2-fold the background level; ++, percentage specific lysis ≥3-fold the background level; sABC, specific Ab-binding capacity.

In OCI-Ly7 cells, a cell line with high IgM-BCR surface expression, CDC was induced not only by OFA and OFA-F(ab′)2 fragments, but also by polyclonal IgM-specific F(ab′)2 fragments (Fig. 3B). This indicates that, under certain conditions, crosslinking of the BCR in itself is sufficient to initiate complement activation and complement-mediated lysis.

A direct relationship between BCR expression and OFA-F(ab′)2 fragment–mediated CDC was established by transfection of an IgG1-BCR into CD20+ CEM T cells. Flow cytometry confirmed a population of BCR-expressing (BCR+) cells in the BCR-transfected CD20+ CEM T cells, whereas BCR expression was absent in cells that had been transfected with empty vector (Fig. 4A). CD20+BCR+ CEM T cells were exposed to OFA and OFA-F(ab′)2 fragments in the presence of NHS, and complement-mediated lysis was assessed by quantifying the percentage of PI+ cells within the CD20+BCR+ and CD20+BCR populations. Whereas CD20+BCR+ cells were susceptible to complement-mediated lysis after opsonization with OFA-F(ab′)2 fragments, the CD20+BCR cell population in the same sample was resistant to CDC mediated by OFA-F(ab′)2 fragments but not OFA. The cells in lower panel of Fig. 4A were mock transfected and therefore only contain a CD20-expressing population. As expected, these cells were only lysed in the presence of OFA and not OFA F(ab′)2 fragments (Fig. 4B). For cells treated with OFA, CDC was more efficient in CD20+BCR+ than in CD20+BCR (mock-transfected) cells (p = 0.0397), even though CD20 expression was comparable between the BCR+ and BCR populations. Similarly, OFA-mediated CDC was higher in peripheral blood B cells with high BCR expression than in the population with low BCR expression (Supplemental Fig. 4). This demonstrates that the BCR indeed also contributes to the CDC activity of IgG1 CD20 Abs. As expected, the type II CD20 Ab 11B8 did not induce CDC in either CD20+BCR+ or CD20+BCR cells. The findings demonstrate that the BCR contributes to type I CD20 Ab–mediated complement activation. The BCR-dependent CDC activity in the presence of OFA-F(ab′)2 fragments was named “accessory CDC.”

FIGURE 4.

Transfection of CD20+ CEM T cells with BCR sensitizes cells for OFA-F(ab′)2 fragment–mediated CDC. (A) BCR and CD20 expression in CD20+ CEM T cells that had been transfected with an IgG-BCR (upper panels) or empty vector (lower panels). Expression of BCR and CD20 was assessed by flow cytometry, using allophycocyanin-labeled IgG-specific mouse Fc fragments or FITC-labeled OFA, respectively. P2 indicates the population of BCR cells; P3 indicates the population of BCR+ cells. (B) CDC activity of OFA, OFA-F(ab′)2 fragments, and 11B8 in CD20+BCR+ and CD20+BCR cell populations within the BCR-transfected cultures (orange and blue bars, respectively) and in untransfected cells (gray bars). Bars are means of triplicate experiments; error bars indicate the SD. Statistical analysis was done with a one-way ANOVA followed by a Dunnett multiple comparisons posttest to compare the mean of each column to that of the mean of the control (CD20+Mock). OFA ANOVA, p = 0.0467. Dunnett multiple comparison: CD20+ mock-transfected cells (CD20+Mock) versus CD20+BCR+, p = 0.0397. CD20+Mock versus CD20+IgG+, p = 0.8387. OFA-F(Ab′)2 ANOVA, p = 0.0208. Dunnett multiple comparison: CD20+Mock versus CD20+BCR+, p = 0.0220. CD20+Mock versus CD20+IgG+, p = 0.9790. *p < 0.05.

FIGURE 4.

Transfection of CD20+ CEM T cells with BCR sensitizes cells for OFA-F(ab′)2 fragment–mediated CDC. (A) BCR and CD20 expression in CD20+ CEM T cells that had been transfected with an IgG-BCR (upper panels) or empty vector (lower panels). Expression of BCR and CD20 was assessed by flow cytometry, using allophycocyanin-labeled IgG-specific mouse Fc fragments or FITC-labeled OFA, respectively. P2 indicates the population of BCR cells; P3 indicates the population of BCR+ cells. (B) CDC activity of OFA, OFA-F(ab′)2 fragments, and 11B8 in CD20+BCR+ and CD20+BCR cell populations within the BCR-transfected cultures (orange and blue bars, respectively) and in untransfected cells (gray bars). Bars are means of triplicate experiments; error bars indicate the SD. Statistical analysis was done with a one-way ANOVA followed by a Dunnett multiple comparisons posttest to compare the mean of each column to that of the mean of the control (CD20+Mock). OFA ANOVA, p = 0.0467. Dunnett multiple comparison: CD20+ mock-transfected cells (CD20+Mock) versus CD20+BCR+, p = 0.0397. CD20+Mock versus CD20+IgG+, p = 0.8387. OFA-F(Ab′)2 ANOVA, p = 0.0208. Dunnett multiple comparison: CD20+Mock versus CD20+BCR+, p = 0.0220. CD20+Mock versus CD20+IgG+, p = 0.9790. *p < 0.05.

Close modal

Finally, to assess whether accessory CDC might contribute to the antitumor activity of type I CD20 Abs in the clinic, we performed a set of experiments using PBMCs obtained from three CLL patients, three MCL patients, and one WM patient. Expression of CD20 and the BCR (IgM) was confirmed for all patients (Fig. 5A). Interestingly, OFA-F(ab′)2 fragments were able to induce ex vivo killing of tumor cells isolated from MCL and WM patients, in which OFA-F(ab′)2 fragment–mediated tumor cell killing was particularly efficient in the WM patient, that showed high BCR expression (Fig. 5B). OFA-F(ab′)2 fragments showed statistically significant tumor cell lysis compared with the no Ab control for MCL (p = 0.0214) and WM (p = 0.0062) (Fig. 5C), but not for CLL (p = 0.4129).

FIGURE 5.

Accessory CDC in primary tumor cells. (A) Coexpression of CD20 and the IgM-BCR in tumor material isolated from a patient with CLL (upper panel), MCL (middle panel), and WM (lower panel). Expression of CD20, IgG-BCR, and IgM-BCR was assessed by flow cytometry. (B) Dose-dependent cytotoxicity of OFA and OFA-F(ab′)2 fragments in CLL (left panel), MCL (middle panel), and WM (right panel) patient–derived tumor cells ex vivo. (C) CDC activity of OFA and OFA-F(ab′)2 fragments in tumor material isolated from three CLL patients, three MCL patients, and one WM patient. CDC was assessed using the classical chromium release assay. Error bars indicate the SEM of three replicate samples. Statistical analysis was done with a one-way ANOVA followed by a Dunnett multiple comparisons posttest to compare the means of each column to that of the no Ab control per indication. OFA induced significant lysis in all indications (one-way ANOVA, p < 0.05), whereas OFA-F(ab′)2 fragments only induced significant lysis in patient-derived MCL and WM samples (one-way ANOVA, p < 0.05) but not of patient-derived CLL samples. *p < 0.05, **p < 0.005.

FIGURE 5.

Accessory CDC in primary tumor cells. (A) Coexpression of CD20 and the IgM-BCR in tumor material isolated from a patient with CLL (upper panel), MCL (middle panel), and WM (lower panel). Expression of CD20, IgG-BCR, and IgM-BCR was assessed by flow cytometry. (B) Dose-dependent cytotoxicity of OFA and OFA-F(ab′)2 fragments in CLL (left panel), MCL (middle panel), and WM (right panel) patient–derived tumor cells ex vivo. (C) CDC activity of OFA and OFA-F(ab′)2 fragments in tumor material isolated from three CLL patients, three MCL patients, and one WM patient. CDC was assessed using the classical chromium release assay. Error bars indicate the SEM of three replicate samples. Statistical analysis was done with a one-way ANOVA followed by a Dunnett multiple comparisons posttest to compare the means of each column to that of the no Ab control per indication. OFA induced significant lysis in all indications (one-way ANOVA, p < 0.05), whereas OFA-F(ab′)2 fragments only induced significant lysis in patient-derived MCL and WM samples (one-way ANOVA, p < 0.05) but not of patient-derived CLL samples. *p < 0.05, **p < 0.005.

Close modal

These data indicate that, also for clinically relevant samples, accessory CDC can contribute to the antitumor activity of type I CD20 Abs (Fig. 6).

FIGURE 6.

Proposed model for type I CD20 Ab–mediated CDC in BCR-expressing tumor cells. (A) Upon binding to CD20 on the cell membrane, type I CD20 Abs form hexameric structures that are dependent on intermolecular interactions between Fc domains of bound IgG molecules. These IgG hexamers provide an optimal docking site for C1q. (B) Binding of type I CD20 Abs induces clustering of CD20 and BCR molecules in lipid raft domains, which appears to facilitate complement activation via a process that is independent of the CD20 Ab’s Fc domain. (C) The combination of these two mechanisms of complement activation leads to maximal CDC activity of type I CD20 Abs.

FIGURE 6.

Proposed model for type I CD20 Ab–mediated CDC in BCR-expressing tumor cells. (A) Upon binding to CD20 on the cell membrane, type I CD20 Abs form hexameric structures that are dependent on intermolecular interactions between Fc domains of bound IgG molecules. These IgG hexamers provide an optimal docking site for C1q. (B) Binding of type I CD20 Abs induces clustering of CD20 and BCR molecules in lipid raft domains, which appears to facilitate complement activation via a process that is independent of the CD20 Ab’s Fc domain. (C) The combination of these two mechanisms of complement activation leads to maximal CDC activity of type I CD20 Abs.

Close modal

All IgG1 Abs have an intrinsic capacity to bind C1q and activate complement. However, the ability of specific IgG1 Abs to induce CDC upon target binding is strongly dependent on the characteristics of both the Ab as well as the target and epitope recognized (11). CDC occurs upon formation of ordered IgG hexamers on the plasma membrane of target cells, the agglomerate of IgG Fc parts, thus providing the optimal docking site for C1q (14). In the present study, we demonstrate that, at least for type I CD20 Abs, CDC can also be initiated by F(ab′)2 fragments and other Ab variants that are unable to bind C1q, and that CDC in this case appears dependent on C1q binding by the BCR. This is an intriguing observation, as it is generally assumed that Abs that are able to induce CDC upon binding to a cellular target provide the C1q docking site. Importantly, this phenomenon, which we named accessory CDC, was observed in almost all B cell lines that expressed an IgM BCR, as well as in malignant B cells from patients with MCL and WM. The absence of significant lysis of CLL patient material may be due to low and heterogeneous CD20 expression on these cells.

CD20 and the BCR have been shown to colocalize on the plasma membrane within lipid raft domains (27). Type I but not type II CD20 Abs caused direct association of CD20 with the BCR and enhanced the formation of CD20- and BCR-enriched lipid rafts (2730). Furthermore, type I CD20 mAbs and the BCR were shown to induce similar intracellular signaling pathways, and CD20 mAbs were shown to induce phosphorylation of BCR-specific adaptor proteins, which was dependent on BCR expression (30). Collectively, these data indicate that CD20 and the BCR can reside together in lipid-rich domains, and that cross-talk exists between the two molecules.

Recent data suggest that BCRs of the IgM and IgD subclass are organized in preformed oligomeric structures in distinct protein islands on the plasma membrane of resting B cells (31, 32), and that B cell activation induces remodeling of such BCR protein islands into smaller structures, with a different protein and lipid composition (32). We hypothesize that binding of type I CD20 Abs or F(ab′)2 fragments to the plasma membrane induces remodeling of BCR protein islands, in such a way that the BCR clusters provide sufficient avidity to allow C1q binding and complement activation. Based on this idea, we propose a model for type I CD20 Ab-dependent complement activation in BCR-expressing malignant B cells (Fig. 6). Upon binding to CD20 on the plasma membrane, type I CD20 IgG Abs form hexamers that provide a docking site for C1q and induce CDC (14, 15). Hexamer formation by the CD20 Abs is dependent on the IgG Fc domain and independent of the BCR (Fig. 6A). Ab binding results in clustering of CD20 and the BCR in lipid raft domains, which potentially also facilitates C1q binding by the BCR cluster (Fig. 6B). As illustrated in Fig. 6C, the combination of these two mechanisms leads to maximal CDC activity of type I CD20 Abs. The detailed molecular organization of such CD20–BCR complexes requires further investigation.

C1q binding and complement activation by the BCR was previously observed by Rossbacher and Shlomchik (33) who showed that crosslinking of an IgM-BCR using a highly multivalent Ag induced deposition of C3d on B cells, an effect that was abolished in cells with a mutated BCR that was unable to bind C1q (33). In line with this, we were able to induce CDC of OCI-Ly7 cells by ligation of the IgM-BCR with polyclonal IgM-specific F(ab′)2 fragments.

The concentration of OFA-F(ab′)2 fragments required to observe accessory CDC in cell lines and tumor samples was well within the expected plasma levels of patients that are undergoing treatment with type I CD20 Abs (34). This suggests that accessory CDC may contribute to the antitumor activity of type I CD20 Abs in the clinic.

Whether accessory CDC has a role in normal physiology remains an open question. One could speculate that recognition of a multivalent Ag by the BCR could induce sufficient crosslinking to allow CDC, similar to the CDC we and others observed after crosslinking the BCR with polyclonal anti-IgM. However, we are not aware of studies addressing this subject.

In summary, accessory CDC is a novel effector mechanism of type I CD20 Abs. Accessory CDC occurred through the classical pathway of complement activation, but it was independent of the CD20 Ab Fc domain. Instead, accessory CDC was facilitated by the BCR. Together with known effector mechanisms of type I CDC Abs, such as ADCC, ADCP, and “classical” CDC, accessory CDC may contribute to therapeutic activity of CD20 Abs and potentially other Abs that induce BCR clustering.

The online version of this article contains supplemental material.

Abbreviations used in this article:

ADCC

Ab-dependent cell-mediated cytotoxicity

ADCP

Ab-dependent cell-mediated phagocytosis

CDC

complement-dependent cytotoxicity

CLL

chronic lymphocytic leukemia

MCL

mantle cell lymphoma

NHS

normal human serum

OFA

ofatumumab

PI

propidium iodide

RTX

rituximab

WM

Waldenström macroglobulinemia.

1
Cheson
B. D.
2010
.
Ofatumumab, a novel anti-CD20 monoclonal antibody for the treatment of B-cell malignancies.
J. Clin. Oncol.
28
:
3525
3530
.
2
Maloney
D. G.
,
Grillo-López
A. J.
,
White
C. A.
,
Bodkin
D.
,
Schilder
R. J.
,
Neidhart
J. A.
,
Janakiraman
N.
,
Foon
K. A.
,
Liles
T. M.
,
Dallaire
B. K.
, et al
.
1997
.
IDEC-C2B8 (rituximab) anti-CD20 monoclonal antibody therapy in patients with relapsed low-grade non-Hodgkin’s lymphoma.
Blood
90
:
2188
2195
.
3
Lefebvre
M. L.
,
Krause
S. W.
,
Salcedo
M.
,
Nardin
A.
.
2006
.
Ex vivo-activated human macrophages kill chronic lymphocytic leukemia cells in the presence of rituximab: mechanism of antibody-dependent cellular cytotoxicity and impact of human serum.
J. Immunother.
29
:
388
397
.
4
Glennie
M. J.
,
French
R. R.
,
Cragg
M. S.
,
Taylor
R. P.
.
2007
.
Mechanisms of killing by anti-CD20 monoclonal antibodies.
Mol. Immunol.
44
:
3823
3837
.
5
Taylor
R. P.
,
Lindorfer
M. A.
.
2008
.
Immunotherapeutic mechanisms of anti-CD20 monoclonal antibodies.
Curr. Opin. Immunol.
20
:
444
449
.
6
Zhou
X.
,
Hu
W.
,
Qin
X.
.
2008
.
The role of complement in the mechanism of action of rituximab for B-cell lymphoma: implications for therapy.
Oncologist
13
:
954
966
.
7
Beurskens
F. J.
,
Ruuls
S. R.
,
Engelberts
P. J.
,
Vink
T.
,
Mackus
W. J.
,
van de Winkel
J. G.
,
Parren
P. W.
.
2008
.
Complement activation impacts B-cell depletion by both type I and type II CD20 monoclonal antibodies.
Blood
112
:
4354
4355
.
8
Bologna
L.
,
Gotti
E.
,
Manganini
M.
,
Rambaldi
A.
,
Intermesoli
T.
,
Introna
M.
,
Golay
J.
.
2011
.
Mechanism of action of type II, glycoengineered, anti-CD20 monoclonal antibody GA101 in B-chronic lymphocytic leukemia whole blood assays in comparison with rituximab and alemtuzumab.
J. Immunol.
186
:
3762
3769
.
9
Cragg
M. S.
,
Morgan
S. M.
,
Chan
H. T.
,
Morgan
B. P.
,
Filatov
A. V.
,
Johnson
P. W.
,
French
R. R.
,
Glennie
M. J.
.
2003
.
Complement-mediated lysis by anti-CD20 mAb correlates with segregation into lipid rafts.
Blood
101
:
1045
1052
.
10
Janas
E.
,
Priest
R.
,
Wilde
J. I.
,
White
J. H.
,
Malhotra
R.
.
2005
.
Rituxan (anti-CD20 antibody)-induced translocation of CD20 into lipid rafts is crucial for calcium influx and apoptosis.
Clin. Exp. Immunol.
139
:
439
446
.
11
Melis
J. P.
,
Strumane
K.
,
Ruuls
S. R.
,
Beurskens
F. J.
,
Schuurman
J.
,
Parren
P. W.
.
2015
.
Complement in therapy and disease: regulating the complement system with antibody-based therapeutics.
Mol. Immunol.
67
(
2 Pt A
):
117
130
.
12
Burton
D. R.
1986
.
Is IgM-like dislocation a common feature of antibody function?
Immunol. Today
7
:
165
167
.
13
Collins
C.
,
Tsui
F. W.
,
Shulman
M. J.
.
2002
.
Differential activation of human and guinea pig complement by pentameric and hexameric IgM.
Eur. J. Immunol.
32
:
1802
1810
.
14
Diebolder
C. A.
,
Beurskens
F. J.
,
de Jong
R. N.
,
Koning
R. I.
,
Strumane
K.
,
Lindorfer
M. A.
,
Voorhorst
M.
,
Ugurlar
D.
,
Rosati
S.
,
Heck
A. J.
, et al
.
2014
.
Complement is activated by IgG hexamers assembled at the cell surface.
Science
343
:
1260
1263
.
15
de Jong
R. N.
,
Beurskens
F. J.
,
Verploegen
S.
,
Strumane
K.
,
van Kampen
M. D.
,
Voorhorst
M.
,
Horstman
W.
,
Engelberts
P. J.
,
Oostindie
S. C.
,
Wang
G.
, et al
.
2016
.
A novel platform for the potentiation of therapeutic antibodies based on antigen-dependent formation of IgG hexamers at the cell surface.
PLoS Biol.
14
:
e1002344
.
16
Pawluczkowycz
A. W.
,
Beurskens
F. J.
,
Beum
P. V.
,
Lindorfer
M. A.
,
van de Winkel
J. G.
,
Parren
P. W.
,
Taylor
R. P.
.
2009
.
Binding of submaximal C1q promotes complement-dependent cytotoxicity (CDC) of B cells opsonized with anti-CD20 mAbs ofatumumab (OFA) or rituximab (RTX): considerably higher levels of CDC are induced by OFA than by RTX.
J. Immunol.
183
:
749
758
.
17
Peipp
M.
,
Lammerts van Bueren
J. J.
,
Schneider-Merck
T.
,
Bleeker
W. W.
,
Dechant
M.
,
Beyer
T.
,
Repp
R.
,
van Berkel
P. H.
,
Vink
T.
,
van de Winkel
J. G.
, et al
.
2008
.
Antibody fucosylation differentially impacts cytotoxicity mediated by NK and PMN effector cells.
Blood
112
:
2390
2399
.
18
Teeling
J. L.
,
French
R. R.
,
Cragg
M. S.
,
van den Brakel
J.
,
Pluyter
M.
,
Huang
H.
,
Chan
C.
,
Parren
P. W.
,
Hack
C. E.
,
Dechant
M.
, et al
.
2004
.
Characterization of new human CD20 monoclonal antibodies with potent cytolytic activity against non-Hodgkin lymphomas.
Blood
104
:
1793
1800
.
19
Rösner
T.
,
Lohse
S.
,
Peipp
M.
,
Valerius
T.
,
Derer
S.
.
2014
.
Epidermal growth factor receptor targeting IgG3 triggers complement-mediated lysis of decay-accelerating factor expressing tumor cells through the alternative pathway amplification loop.
J. Immunol.
193
:
1485
1495
.
20
van Meerten
T.
,
van Rijn
R. S.
,
Hol
S.
,
Hagenbeek
A.
,
Ebeling
S. B.
.
2006
.
Complement-induced cell death by rituximab depends on CD20 expression level and acts complementary to antibody-dependent cellular cytotoxicity.
Clin. Cancer Res.
12
:
4027
4035
.
21
de Weers
M.
,
Tai
Y. T.
,
van der Veer
M. S.
,
Bakker
J. M.
,
Vink
T.
,
Jacobs
D. C.
,
Oomen
L. A.
,
Peipp
M.
,
Valerius
T.
,
Slootstra
J. W.
, et al
.
2011
.
Daratumumab, a novel therapeutic human CD38 monoclonal antibody, induces killing of multiple myeloma and other hematological tumors.
J. Immunol.
186
:
1840
1848
.
22
Takei
K.
,
Yamazaki
T.
,
Sawada
U.
,
Ishizuka
H.
,
Aizawa
S.
.
2006
.
Analysis of changes in CD20, CD55, and CD59 expression on established rituximab-resistant B-lymphoma cell lines.
Leuk. Res.
30
:
625
631
.
23
Petersen
B. H.
,
Graham
J. A.
,
Brooks
G. F.
.
1976
.
Human deficiency of the eighth component of complement. The requirement of C8 for serum Neisseria gonorrhoeae bactericidal activity.
J. Clin. Invest.
57
:
283
290
.
24
Hicks
P. S.
,
Saunero-Nava
L.
,
Du Clos
T. W.
,
Mold
C.
.
1992
.
Serum amyloid P component binds to histones and activates the classical complement pathway.
J. Immunol.
149
:
3689
3694
.
25
Kaplan
M. H.
,
Volanakis
J. E.
.
1974
.
Interaction of C-reactive protein complexes with the complement system. I. Consumption of human complement associated with the reaction of C-reactive protein with pneumococcal C-polysaccharide and with the choline phosphatides, lecithin and sphingomyelin.
J. Immunol.
112
:
2135
2147
.
26
Wolbink
G. J.
,
Brouwer
M. C.
,
Buysmann
S.
,
ten Berge
I. J.
,
Hack
C. E.
.
1996
.
CRP-mediated activation of complement in vivo: assessment by measuring circulating complement-C-reactive protein complexes.
J. Immunol.
157
:
473
479
.
27
Petrie
R. J.
,
Deans
J. P.
.
2002
.
Colocalization of the B cell receptor and CD20 followed by activation-dependent dissociation in distinct lipid rafts.
J. Immunol.
169
:
2886
2891
.
28
Polyak
M. J.
,
Li
H.
,
Shariat
N.
,
Deans
J. P.
.
2008
.
CD20 homo-oligomers physically associate with the B cell antigen receptor. Dissociation upon receptor engagement and recruitment of phosphoproteins and calmodulin-binding proteins.
J. Biol. Chem.
283
:
18545
18552
.
29
Holder
M. J.
,
Chamba
A.
,
Hardie
D. L.
,
Deans
J. P.
,
Gordon
J.
.
2004
.
Improved access to CD20 following B cell receptor cross-linking at Burkitt’s lymphoma cell surfaces.
Leuk. Res.
28
:
1197
1202
.
30
Walshe
C. A.
,
Beers
S. A.
,
French
R. R.
,
Chan
C. H.
,
Johnson
P. W.
,
Packham
G. K.
,
Glennie
M. J.
,
Cragg
M. S.
.
2008
.
Induction of cytosolic calcium flux by CD20 is dependent upon B Cell antigen receptor signaling.
J. Biol. Chem.
283
:
16971
16984
.
31
Reth
M.
2001
.
Oligomeric antigen receptors: a new view on signaling for the selection of lymphocytes.
Trends Immunol.
22
:
356
360
.
32
Maity
P. C.
,
Blount
A.
,
Jumaa
H.
,
Ronneberger
O.
,
Lillemeier
B. F.
,
Reth
M.
.
2015
.
B cell antigen receptors of the IgM and IgD classes are clustered in different protein islands that are altered during B cell activation.
Sci. Signal.
8
:
ra93
.
33
Rossbacher
J.
,
Shlomchik
M. J.
.
2003
.
The B cell receptor itself can activate complement to provide the complement receptor 1/2 ligand required to enhance B cell immune responses in vivo.
J. Exp. Med.
198
:
591
602
.
34
Barth
M. J.
,
Czuczman
M. S.
.
2013
.
Ofatumumab: a novel, fully human anti-CD20 monoclonal antibody for the treatment of chronic lymphocytic leukemia.
Future Oncol.
9
:
1829
1839
.

P.J.E., M.V., J.S., T. Vink, E.C.W.B, J.G.J.W., P.W.H.I.P., and F.J.B. are all employees of Genmab. J.M.B and W.J.M.M are former employees of Genmab. P.J.E., M.V., J.S., T. Vink, E.C.W.B, J.G.J.W., P.W.H.I.P., F.J.B., J.M.B., and W.J.M.M all are owners of warrants with Genmab. The other authors have no financial conflicts of interest.

Supplementary data