B Cell Rab7 Mediates Induction of Activation-Induced Cytidine Deaminase Expression and Class-Switching in T-Dependent and T-Independent Antibody Responses

The maturation of the Ab response is critical to effective host defense against microbial infections and tumors. It depends on two B lymphocyte differentiation processes: Ig class switch DNA recombination (CSR) and somatic hypermutation (SHM) (1). CSR replaces an IgH constant (C_H) region (e.g., Cμ, with a downstream C_H region [Cγ, Cα, or Cε]), thereby diversifying the biological effector functions of an Ab without changing its specificity for Ag (2). SHM inserts mainly point mutations in the Ig V(D)J DNA, thereby providing the structural substrate for the positive selection by Ag for higher affinity Ab mutants (1). CSR and SHM require deamination of deoxycytosines in IgH switch (S) region and V(D)J region DNA, respectively, by activation-induced cytidine deaminase (AID, encoded by AICDA/Aicda) (3). In CSR, AID is targeted to an upstream (donor) S and a downstream (acceptor) S regions by 14-3-3 adaptors, which display high affinities for 5′-AGCT-3′ repeats, as recurring in all S regions, as well as for H3K9acS10ph histone posttranslational modifications, as induced specifically in the donor and acceptor S regions—these S regions also undergo germline I_H-S-C_H transcription initiated upon activation of their respective upstream intervening I_H promoter (4, 5). High densities of deoxyuracils, as generated through AID-mediated DNA deamination, are processed by uracil DNA glycosylase (Ung) recruited by Rev1 (6-8), leading to the generation of S region dsDNA breaks, which are obligatory CSR intermediates. Repair of such dsDNA breaks results in looping out of the intervening DNA as an S circle, formation of an S-S junction, and juxtaposition of the expressed V_HDJ_H DNA to the downstream C_H DNA. Transient transcription of the S circle gives rise to circle transcripts, such as Iγ-Cμ, Iα-Cμ or Iε-Cμ, which are hallmarks of ongoing CSR from IgM to IgG, IgA, or IgE; transcription of the IgH locus gives rise to postrecombination Iμ-Cγ, Iμ-Cα, or Iμ-Cε transcripts. After undergoing CSR and SHM, B cells differentiate into plasma cells and memory B cells for mature Ab production and generation of long-term immunity, respectively (9–12).

In T-independent Ab responses, B cells are induced to express AID and undergo CSR upon engagement of their BCRs by repetitive antigenic ligands together with engagement of their TLRs and other innate immune receptors by microbe-associated molecular patterns (13). TLR signaling also plays an important role in T-dependent responses (e.g., by activating B cells before the emergence of specific Th cells) (14). Once specific Th cells emerge, they will bear surface trimeric CD154 for engagement of CD40, which is constitutively expressed on B cells. CD154:CD40 engagement leads to further B cell differentiation, including increased AID expression and CSR, which can be enhanced by BAFF/BlyS and APRIL (15–18). The primary CSR-inducing stimuli (e.g., dual TLR/BCR engagement and CD40 engagement) enable secondary stimuli (i.e., cytokines IL-4 and TGF-β [as well as IFN-γ in the mouse]) to induce IgH germline I_H-S-C_H transcription and histone posttranslational modifications in the donor and acceptor S regions (19–23), thereby directing CSR to specific Ig isotypes. IL-4 and TGF-β also can enhance AID expression, as induced by primary CSR-inducing stimuli, through activation of Stat6 and Smad transcription factors, respectively, which are recruited to the Aicda promoter and enhancers (24, 25). T-independent and T-dependent primary CSR-inducing stimuli activate NF-κB through both the canonical and noncanonical pathways, leading to recruitment of NF-κB to the Aicda promoter for induction of AID expression, which is restricted to activated B cells (2, 24–27).

In B cells, signals from TLRs, BCR, or CD40 are transduced by multiple pathways, including those involving TRAF6 or PI(3)K (13, 28, 29). These pathways mediate NF-κB activation, thereby linking receptor signals with AID induction. Genetic, biochemical, and structural studies have furthered our understanding of the recruitment of signal adaptors through “signalosomes” along plasma membrane lipid rafts (30–32). Nevertheless, the preservation of selected signals in B cells that are virtually ablated in plasma membrane signalosomes (e.g., the intact ERK activation in phospholipase Cγ2–deficient B cells (33)) indicates that a B cell can use signaling pathways mediated by intracellular membranes. These would include the endoplasmic reticulum membrane, which could mediate NF-κB activation by different surface receptors, such as CD40 (BL41 B cells), TNFR (HEK 293T cells), and TCR (Jurkat T cells) (34). In addition, autophagy-related double-membrane structures, which originate from endoplasmic reticulum or mitochondria membranes (35), play a role in MAPK p38 activation triggered by BCR and TLR9 (36). Finally, a role of intracellular membranes in B cell signal transduction is suggested by the regulation of CD40 and BCR signaling as well as immunity and inflammation by autophagy-related (Atg) factors (37–40), including Atg5 (41, 42).

The Rab7 small GTPase mediates the maturation of endosomes by replacing Rab5 through a “GTPase switch” process. It also promotes the conversion of endosomes to lysosomes as well as fusion of endosomes with autophagosomes to form amphisomes in different cell types (43). In stressed cells, such as those having phagocytosed large extracellular particles or engulfed a portion of the cytoplasm in response to unfavorable metabolic conditions (e.g., serum starvation), Rab7 mediates the fusion of autophagosomes or amphisomes with lysosomes to form autolysosomes, in which the cargo is degraded. Rab7 also promotes cell death induced by growth factor withdrawal and clearance of apoptotic bodies (44–46). In this study, we reasoned that in proliferating or differentiating immune cells, which are not deprived of nutrients or growth factors, Rab7 would play additional and specific roles. This was prompted by the putative role of intracellular membranes in NF-κB activation and the association of Rab7 with those membranes (43).

Rab7 has been shown to regulate T cell functions (47), but its role in B cells is unknown. To address the B cell–intrinsic role of Rab7 in the Ab response, we constructed conditional Igh^+/Cγ^1-creRab7^fl/fl mice, in which Rab7 expression is abrogated only in B cells undergoing Igh^Cγ^1-cre Iγ1-Sγ1-Cγ1-cre transcription, as induced by IL-4 in conjunction with a primary stimulus. We stimulated Igh^+/Cγ^1-creRab7^fl/fl B cells with CD154 to activate CD40 signaling and T-independent stimuli to activate both TLR and BCR signaling, including LPS (engaging TLR4 and BCR through its lipid A and polysaccharidic moieties, respectively) or CpG ODN plus anti-δ mAb/dex (engaging TLR9 and BCR, respectively) to address the role of Rab7 in NF-κB activation, AID expression, and CSR. We also used Igh^+/Cγ^1-creRab7^fl/fl mice and B cells to analyze the contribution of Rab7 to other B cell functions and differentiation processes, such as plasma cell differentiation, in vivo and in vitro. Our findings suggest an important role of Rab7 in induction of AID expression and CSR for the maturation of T-dependent and T-independent Ab responses.

Igh^+/Cγ^1-creRab7^fl/fl mice and their Igh^+/Cγ^1-creRab7^+/fl littermate controls (Figs. 1, 2A) were generated by the breeding of Igh^+/Cγ^1-cre mice (Ref. 48; JAX stock number 010611) with Rab7^fl/fl mice (Ref. 47; JAX stock number 021589), both of which had been backcrossed onto the C57BL/6 background for at least nine generations. Mice were maintained in pathogen-free vivaria and were used for experiments at 8–12 wk of age and without any apparent infection or disease. For immunization, mice were injected i.p. with 100 μg 4-hydroxy-3-nitrophenyl acetyl (NP)–conjugated chicken gammaglobulin (CGG) (in average 16 molecules of NP coupled to 1 molecule of CGG; BioSearch Technologies) in 100 μl alum (Imject Alum; Pierce), 100 μg OVA in 100 μl alum, or 25 μg NP-conjugated LPS (in average 0.2 NP molecule conjugated with one LPS molecule; BioSearch Technologies) in 100 μl PBS. All protocols were in accordance with the rules and regulations of the Institutional Animal Care and Use Committee of the University of Texas Health Science Center at San Antonio and those of the Institutional Animal Care and Use Committee of the University of California (Irvine, CA).

Single-cell suspensions were prepared from spleens or pooled axillary, inguinal, and cervical lymph nodes using a 70-μm cell strainer. Lymph node cells were directly resuspended in RPMI 1640 medium (Invitrogen) supplemented with FBS (10% v/v, Hyclone), penicillin-streptomycin (1% v/v, Invitrogen) and amphotericin B (1% v/v, Invitrogen) (RPMI-FBS) before purification of B cells. Spleen cells were resuspended in ACK Lysis Buffer (Lonza) to lyse RBCs and, after quenching with RPMI-FBS, were resuspended in RPMI-FBS before B cell purification. B cells were purified by depletion of cells expressing CD43, CD4, CD8, CD11b, CD 49b, CD90.2, Gr-1, or Ter-119 using the EasySep Mouse B cell Isolation kit (StemCell Technologies) following the manufacturer’s protocol. For CFSE (Invitrogen) labeling, B cells were resuspended in PBS at 10⁷ cell/ml and incubated with 4 μM CFSE at 37°C for 5 min and immediately quenched with RPMI-FBS. After pelleting, B cells were resuspended and cultured at 3 × 10⁵ cell/ml in RPMI-FBS supplemented with 50 μM 2-ME.

For CSR induction, B cells were stimulated with the following primary stimuli: CD154 (3 U/ml, mouse CD154-containing membrane fragments of baculovirus-infected Sf21 insect cells (5, 49)), LPS (3 μg/ml, from Escherichia coli, serotype 055:B5, deproteinized by chloroform extraction; Sigma-Aldrich), TLR1/2 ligand Pam₃CSK₄ (100 ng/ml; InvivoGen), TLR4 ligand monophophoryl lipid A (lipid A, 1 μg/ml; Sigma-Aldrich), TLR7 ligand R-848 (30 ng/ml; InvivoGen), or TLR9 ligand CpG ODN1826 with a phosphorothioate backbone (CpG, 1 μM, sequence 5′-TCCATGACGTTCCTGACGTT-3′; Operon). Anti-Igδ mAb (clone 11-26c) conjugated to dextran (anti-δ mAb/dex; Fina Biosolutions) was added, as indicated, to cross-link BCRs. IL-4 (3 ng/ml) was added for CSR to IgG1 and IgE, IFN-γ (50 ng/ml) for CSR to IgG2a, and TGF-β (2 ng/ml) for CSR to IgG2b (all from R&D Systems). In selected experiments, Igh^+/Cγ^1-creRab7^fl/fl B cells and their Igh^+/Cγ^1-creRab7^+/fl B cell counterparts were stimulated with CD154 or LPS plus IL-4 for 48 h. After washing for three times with RPMI-FBS to remove residual stimuli, the B cells were recultured in LPS plus TGF-β and anti-δ mAb/dex for induction of CSR to IgA.

After stimulation, B cells were harvested, stained with fluorochrome-conjugated mAbs (Supplemental Table I) in HBSS for 15 min, and analyzed by flow cytometry for Igγ3, Igγ1, Igγ2b, and Igγ2a expression (CSR to IgG3, IgG1, IgG2b, and IgG2a) and other B cell surface molecules. For intracellular Igε analysis, B cells (2 × 10⁶) stimulated by CD154 or LPS plus IL-4 were harvested and treated with 200 μl trypsin for 2 min to cleave off FcεRI, followed by quenching with 1 ml RPMI-FBS, as described previously (50). Cells were then fixed in 200 μl formaldehyde (3.7% v/v) at 25°C for 10 min and, after quenching with 22 μl 1 M glycine (pH 7), were permeabilized in 500 μl 90% cold methanol on ice for 30 min before staining with fluorochrome-conjugated Abs and flow cytometry analysis. FACS data were analyzed by the FlowJo software (Tree Star).

For analysis of B cell proliferation in vivo in mice immunized with NP-CGG, mice were injected i.p. with BrdU twice, 2 mg (in 200 μl PBS) each time, within a 20-h interval, and sacrificed 4 h after the second injection. B cells were analyzed by flow cytometry for BrdU incorporation and DNA contents in spleen B cells using a BrdU Flow Kit (BD Biosciences), following the manufacturer’s instructions. Briefly, spleen cells were stained with fluorochrome-conjugated anti-B220 mAb and, after washing, resuspended in the BD Cytofix/Cytoperm buffer for 15 min on ice. After washing, cells were incubated with 7-aminoactinomycin D (7-AAD) and allophycocyanin-conjugated anti-BrdU mAb for staining of DNA and incorporated BrdU, respectively, and then analyzed by flow cytometry. For analysis of B cell proliferation in vitro, purified CFSE-labeled naive B cells were cultured for 96 h in the presence of appropriate stimuli and then analyzed by flow cytometry for CFSE intensity (CFSE distributes equally into the two daughter cells when a cell divides). The number of B cell divisions was determined by the “Proliferation Platform” of the FlowJo software. CSR as a function of the cell division number was the ratio of class-switched B cells in each division over the total number of B cells in that division. For analysis of B cell viability, cells were stained with 7-AAD and/or fluorochrome-conjugated Annexin V in HBSS and analyzed by flow cytometry.

To visualize Rab7 expression, B cells were spun onto coverslips precoated with poly-d-lysine (10 μg/ml; Sigma-Aldrich) and fixed with 2% paraformaldehyde for 10 min. After blocking with 1% BSA, cells were stained with FITC-conjugated anti-B220 mAb. After washing with HBSS and permeabilization with 0.5% Triton X-100 for 10 min, cells were stained with a rabbit anti-Rab7 mAb (Supplemental Table I) at 25°C for 1 h, followed by staining with Alexa Fluor 647–conjugated goat anti-rabbit Ab. After washing, cells were mounted using ProLong Gold with DAPI (Invitrogen) for confocal microscopic analysis.

To analyze Rab7 and CD40 distributions at high resolution, B cells were activated by CD154 and IL-4 for 48 h and then fixed with paraformaldehyde on the poly-d-lysine–coated cover slips. After permeabilization with 1% Triton X-100 and 10% normal donkey serum, B cells were stained with the rabbit anti-Rab7 mAb or a rat anti-CD40 mAb, followed by staining with an Alexa Fluor 532–conjugated goat anti-rabbit Ab or an Alexa Fluor 488–conjugated goat anti-rat Ab. A Leica TCS SP8 confocal laser scanning microscope equipped with a time-gated stimulated emission depletion module was used to achieve superresolution of Rab7 and CD40 staining by a 592-nm depletion laser.

To visualize germinal centers, spleens from immunized mice were embedded in OCT (Tissue-Tek) and snap-frozen on dry ice. Cryostat sections (7 μm) were fixed in prechilled acetone at −80°C for 30 min and then air-dried at 25°C for 30 min. Sections were stained with PE-conjugated anti-B220 mAb (Supplemental Table I) and FITC-conjugated peanut agglutinin (PNA; EY Laboratories) at 25°C for 1 h in a moist chamber. After washing with HBSS, sections were mounted using ProLong Gold Antifade Reagent (Invitrogen) and examined under an Olympus FluoView 1000 confocal microscope.

The pTAC and pTAC-AID retroviral vectors were as we described previously (13). For the generation of retrovirus, retroviral vectors were used to transfect along with the pCL-Eco retrovirus–packaging vector (Imgenex) HEK293T cells using the ProFection Mammalian Transfection System (Promega). Transfected cells were cultured in RPMI-FBS in the presence of chloroquine (25 μM) for 8 h. After the removal of chloroquine, retrovirus-containing culture supernatants were harvested every 12 h. For transduction and CSR analysis, mouse B cells were activated with CD154 plus IL-4 for 48 h and then incubated with viral particles that were premixed with 6 μg/ml DEAE–dextran at 25°C for 30 min (Sigma-Aldrich). After incubation at 37°C for 5 h with gentle mixing every hour, cells were centrifuged at 500 × g for 1 h and then 1000 × g for 4 min. Transduced B cells were cultured in virus-free RPMI-FBS in the presence of CD154 plus IL-4 for 72 h and then harvested for flow cytometry analysis of viability (7-AAD⁻) and surface IgG1 and CD19 expression. Dead (7-AAD⁺) cells were excluded from IgG1 and CD19 analysis.

To determine titers of total IgM, IgG1, IgG2a, IgG2b, IgG3, IgA, or IgE, sera and culture supernatants were first diluted 4–100-fold and 4–10-fold, respectively, with PBS (pH 7.4) plus 0.05% (v/v) Tween 20 (PBST). Two-fold serially diluted samples and standards for each Ig isotypes were incubated in 96-well plates coated with preadsorbed goat anti-IgM, -IgG (to capture IgG1, IgG2a, IgG2b, and IgG3), -IgA, or -IgE Abs (all 1 mg/ml; Supplemental Table I). After washing with PBST, captured Igs were detected with biotinylated anti-IgM, -IgG1, -IgG2a, -IgG2b, -IgG3, -IgA, or -IgE Abs (Supplemental Table I), followed by reaction with HRP-labeled streptavidin (Sigma-Aldrich), development with o-phenylenediamine, and measurement of absorbance at 492 nm. Ig concentrations were determined using Prism (GraphPad Software) or Excel (Microsoft) software.

To analyze titers of high-affinity NP-specific Abs, sera were diluted 1,000-fold in PBST. Two-fold serially diluted samples were incubated in a 96-well plate preblocked with BSA and coated with NP₃-BSA (in average three NP molecules on one BSA molecule). Captured Igs were detected with biotinylated Ab to IgM, IgG1, IgG2a, IgG2b, IgG3, or IgA. To analyze titers of OVA-specific IgG1 and IgE, sera were diluted 100- (for IgG1) or 10-fold (for IgE) and incubated in plates coated with anti-IgG1 or anti-IgE. Captured Igs were detected with biotinylated OVA. Data are relative values based on end-point dilution factors.

MultiScreen filter plates (Millipore) were activated with 35% ethanol, washed with PBS, and coated with NP₃-BSA. Single-cell suspensions were cultured at 50,000 cells/ml in RPMI-FBS supplemented with 50 μM 2-ME at 37°C for 16 h. After supernatants were removed, plates were incubated with biotinylated goat anti-mouse IgM or anti-IgG1 Ab for 2 h and, after washing, incubated with HRP-conjugated streptavidin. After washing, plates were developed using the Vectastain AEC peroxidase substrate kit (Vector Laboratories) to stain Ab-forming cells (AFCs). The stained area in each well was quantified using the CTL Immunospot software (Cellular Technology) and depicted as the percentage of total area of each well. This was used to quantify the production of AFCs.

Total RNA was extracted from 5 × 10⁶ B cells using the RNeasy Mini Kit (Qiagen). First-strand cDNA was synthesized from 2 μg RNAs using the SuperScript III System (Invitrogen) and analyzed by qPCR using SYBR Green (Dynamo HS kit; New England Biolabs) and appropriate primers (Supplemental Table II). PCR was performed in a MyiQ Real-Time PCR System (Bio-Rad Laboratories) according to the following protocol: 95°C for 30 s, 40 cycles of 95°C for 10 s, 60°C for 30 s, and 72°C for 30 s. Melting curve analysis was performed at 72–95°C. The ΔΔ cycle threshold method was used to analyze levels of transcripts, and data were normalized to the level of Cd79b, which encodes the BCR Igβ-chain constitutively expressed in B cells.

B cells (1 × 10⁷) were harvested by centrifugation at 500 × g for 5 min, resuspended in 0.5 ml lysis buffer (20 mM Tris-Cl [pH 7.5], 150 mM NaCl, 0.5 mM EDTA, and 1% [v/v] Nonidet P-40) supplemented with phosphatase inhibitors, including sodium pyrophosphate (1 mM), NaF (10 mM), and NaVO₃ (1 mM) and a mixture of protease inhibitors (Sigma-Aldrich). Cell lysates were separated by SDS-PAGE and transferred onto nitrocellulose membranes for immunoblotting involving specific Abs (Supplemental Table I). Membranes were then stripped with 200 mM glycine (pH 2.5) for reimmunoblotting with an anti–β-actin mAb.

Unlike mice with Rab7 deletion in germ cells (47, 51), Igh^+/Cγ^1-creRab7^fl/fl mice were born at Mendelian ratio because of their intact Rab7 locus throughout embryogenesis. They were indistinguishable from their Igh^+/+Rab7^+/fl, Igh^+/+Rab7^fl/fl and Igh^+/Cγ^1-creRab7^+/fl littermates in the size, fertility, and organ morphology during development and maturation (Figs. 1, 2A; data not shown). This was expected, because in Igh^+/Cγ^1-creRab7^fl/fl mice, efficient Cre recombinase expression and Cre-mediated Rab7 knockout (KO) occurred exclusively in B cells activated to undergo germline Iγ1-Sγ1-Cγ1-cre transcription in the Igh^Cγ^1-cre allele (Fig. 1). Such transcription process, like IgH germline Iγ1-Sγ1-Cγ1 transcription, was induced by IL-4 in the presence of a primary stimulus, such as CD154, LPS, or CpG ODN plus anti-δ mAb/dex, leading to DNA deletion in the Rab7^fl allele in 89 and 97% of cells after 36 and 48 h, respectively (Fig. 2B), as well as a virtual abrogation of Rab7 protein expression in Igh^+/Cγ^1-creRab7^fl/fl B cells after 48 h of stimulation (Fig. 2C, 2D). Notably, Rab7, which localized in the cytoplasm (Fig. 2D), formed distinct “foci” structures, as revealed by superresolution microscopy analysis, in contrast to the more diffused pattern of CD40 (Fig. 2E).

Consistent with the lack of Igh^Cγ^1-cre germline Iγ1-Sγ1-Cγ1-cre transcription in B cells and T cells during their development and maturation, Igh^+/Cγ^1-creRab7^fl/fl mice displayed normal frequencies, cellularity, and viability of B and T cells at different developmental stages in central and peripheral lymphoid organs, including B1 cells in the peritoneal cavity (Fig. 3; data not shown). Thus, Igh^+/Cγ^1-creRab7^fl/fl mice show normal B and T cell development and maturation; Igh^+/Cγ^1-creRab7^fl/fl B cells show abrogated Rab7 expression upon stimulation by a primary CSR-inducing stimulus plus IL-4.

In Igh^+/Cγ^1-creRab7^fl/fl mice and their Igh^+/Cγ^1-creRab7^+/fl littermate controls (which showed phenotypes comparable to those of Igh^+/Cγ^1-creRab7^+/+; data not shown), IgG1 expression was driven solely by the Igh⁺ allele that had undergone a productive V_HDJ_H recombination and then CSR from Igμ to Igγ1, because the Cre sequence would interfere with the expression of V_HDJ_H-Cγ1 in the Igh^Cγ^1-cre allele (48). As compared with their Igh^+/Cγ^1-creRab7^+/fl littermates, Igh^+/Cγ^1-creRab7^fl/fl mice showed normal titers of serum IgM but much reduced (by 85%) titers of IgG1 (Fig. 4A). They had normal titers of other switched IgG isotypes (IgG2a, IgG2b, and IgG3) and IgA, consistent with our prediction that B cells that were set to switch to these Ig isotypes maintained an intact Rab7 locus because of their germline Iγ2a-Sγ2a-Cγ2a, Iγ2b-Sγ2b-Cγ2b, or Iγ3-Sγ3-Cγ3 transcription and lack of Iγ1-Sγ1-Cγ1-cre transcription.

We next addressed the role of Rab7 in a specific T-dependent response by injecting mice with NP-CGG, which elicits predominantly NP-binding IgG1. Igh^+/Cγ^1-creRab7^fl/fl mice showed a severe defect (>96% reduction) in generation of (high-affinity) NP-specific IgG1 but normal levels of NP-specific IgM and other IgG isotypes (IgG2a, IgG2b, IgG3) and IgA (Fig. 4B). Likewise, Igh^+/Cγ^1-creRab7^fl/fl mice could not produce T-independent NP-specific IgG1 response to NP-LPS but were competent in generating NP-specific IgM, IgG2a, IgG2b, and IgG3 (Fig. 4C)—CSR to IgG2b and IgG3 are characteristically induced by LPS.

Consistent with an important role of B cell Rab7 in class-switched Ab responses, Igh^+/Cγ^1-creRab7^fl/fl mice injected with NP-CGG were defective in generation of AFCs that secreted NP-specific IgG1 but could generate AFCs that secreted NP-binding IgM (Fig. 5A). These mice showed normal germinal center development, normal B cell proliferation and survival, as well as normal differentiation of B cells into germinal center (PNA^hi) B cells and plasma cells (Fig. 5B–D). They, however, failed to generate IgG1⁺ B cells (Fig. 5E). By contrast, they displayed normal frequencies of IgG3⁺ B cells in the spleen and IgA⁺ B cells in the peritoneal cavity and Peyer’s patches (Figs. 3A, 5E; data not shown), consistent with specific ablation of Rab7 in B cells that were to undergo CSR to IgG1. Thus, Rab7 mediates B cell class-switching and generation of AFCs that secrete class-switched Abs in T-independent and T-dependent responses. It does not play a major role in B cell proliferation, survival, or differentiation into plasma cells.

The notion that B cell Rab7 plays an important role in CSR was emphasized by the induction of Rab7 in B cells activated to undergo CSR (Fig. 2C). To address the intrinsic role of Rab7 in CSR, we isolated naive B cells from Igh^+/Cγ^1-creRab7^fl/fl mice (which expressed Rab7) and cultured them in the presence of a T-dependent or T-independent primary stimulus plus IL-4 to ablate Rab7. Upon stimulation by CD154, LPS alone, or a TLR ligand combined with anti-δ mAb/dex (for dual engagement of TLR1/2, TLR4, TLR7, or TLR9 and BCR), Igh^+/Cγ^1-creRab7^fl/fl B cells gave rise to much fewer IgG1⁺ cells (up to 80% reduction, overall and across all B cell divisions) and secreted less IgG1, as compared with their Igh^+/Cγ^1-creRab7^+/fl and Igh^+/+Rab7^fl/fl B cell counterparts (Figs. 6, 7A). Consistent with what we reported in C57BL/6 B cells (13), CSR to IgG1 occurred at marginal levels in Igh^+/Cγ^1-creRab7^+/fl B cells stimulated by a TLR ligand alone plus IL-4. It, however, was virtually abolished in Igh^+/Cγ^1-creRab7^fl/fl B cells exposed to the same stimuli (Fig. 6A). Upon stimulation by a primary CSR-inducing stimulus plus a cytokine that direct CSR to IgG3, IgG2a, and IgG2b, Igh^+/Cγ^1-creRab7^+/fl B cells showed normal CSR, as expected because of their intact Rab7 locus (Fig. 7B; data not shown). They, however, became Rab7 deficient after stimulation by CD154 or LPS plus IL-4 for 48 h (Fig. 2B, 2C) and defective in CSR to IgA, as subsequently induced by LPS plus TGF-β and anti-δ mAb/dex (Fig. 7C). Despite their deficiency in CSR, Igh^+/Cγ^1-creRab7^fl/fl B cells showed normal IgM expression and plasma cell differentiation (Fig. 7D, 7E).

In addition to inducing Igh⁺ germline Iγ1-Sγ1-Cγ1 transcription and direct CSR to IgG1, IL-4 in conjunction with a primary stimulus (particularly CD154) induces Igh⁺ Iε-Sε-Cε transcription and CSR to IgE. The abrogation of Rab7 expression in Igh^+/Cγ^1-creRab7^fl/fl B cells stimulated with CD154 or LPS plus IL-4 (Fig. 2B) prompted us to analyze the impact of Rab7 deficiency on CSR to IgE. As expected (5, 50), Igh^+/Cγ^1-creRab7^+/fl B cells underwent CSR to IgE upon stimulation by CD154 plus IL-4 and less efficiently upon stimulation by LPS plus IL-4 (Fig. 8A). By contrast, Igh^+/Cγ^1-creRab7^fl/fl B cells failed to undergo CSR to IgE, despite largely normal B cell dividing rates (Fig. 8A). Their defect in CSR to IgE was comparable in magnitude to their defect in CSR to IgG1 (Fig. 8B), reflecting an Ig isotype-independent role of Rab7 in CSR. Accordingly, in Igh^+/Cγ^1-creRab7^fl/fl mice injected with OVA, the levels of OVA-specific IgE and IgG1 as well as total IgE and IgG1 were significantly reduced, as compared with those in Igh^+/Cγ^1-creRab7^+/fl littermates, whereas IgM levels were normal (Fig. 8C). These in vivo and in vitro experiments further show that Rab7 plays an important role in mediating CSR independently of Ig isotypes. Such a role is central to the CSR machinery, as induced by all the T-dependent and T-independent primary stimuli we tested.

Our studies strongly suggested that Rab7-mediated processes that were induced mainly by primary CSR-inducing stimuli but not secondary stimuli (which specify the isotype that a B cell will switch to). Indeed, Igh^+/Cγ^1-creRab7^fl/fl B cells underwent normal germline Iγ1-Sγ1-Cγ1 and Iε-Sε-Cε transcription when induced in vivo and in vitro by IL-4 plus a primary stimulus (Fig. 9A–C). They, however, showed significantly reduced expression of the Aicda transcripts and protein (Figs. 9A–C, 10A) as well as reduced levels of circle Iγ1-Cμ and Iε-Cμ transcripts and postrecombination Iμ-Cγ1 and Iμ-Cε transcripts. Rab7 deficiency also decreased the expression of 14-3-3γ adaptor molecules, which stabilize AID and other CSR factors on the donor and receptor S region DNA for CSR to unfold (4, 52). By contrast, it did not decrease (or even increased) expression of Blimp-1 (encoded by Prdm1) and Xbp-1 (encoded by Xbp1), which drive plasma cell differentiation and/or functions (Figs. 9A–C, 10B), consistent with the normal differentiation of Igh^+/Cγ^1-creRab7^fl/fl B cells into plasma cells—expression of autophagy-related genes was also normal with the exception of Bcln1 (Fig. 9D). Enforced expression of AID in Igh^+/Cγ^1-creRab7^fl/fl B cells that were prestimulated with CD154 plus IL-4 for 48 h (to abrogate Rab7) rescued CSR, resulting in >10% of cells being IgG1⁺ after 72 h of stimulation with CD154 plus IL-4 (Fig. 9E)—the proportion (10%) of IgG1⁺ B cells was within the range of CSR levels in normal B cells stimulated for 72 h. Thus, Rab7 plays an important role in mediating AID induction and, therefore, CSR.

As we and others have shown (13, 24, 25), NF-κB activation is required for AID induction. This, together with the role of Rab7 in the induction of AID expression [as well as 14-3-3γ expression, which also depends on NF-κB (49)], prompted us to analyze NF-κB activation in Igh^+/Cγ^1-creRab7^fl/fl B cells, as stimulated by CD40, LPS, or CpG plus anti-δ mAb/dex in the presence of IL-4. These B cells were defective in activation of p65 of the canonical NF-κB pathway (Fig. 10A), which, as we have shown, is induced by CD40 or TLR signaling but not BCR signaling (13). By contrast, they were normal in the activation of the noncanonical NF-κB pathway (i.e., upregulation of p100 expression and processing of p100 to p52), which is induced by CD40 and BCR signaling, but not TLR signaling, and also plays an important role in Aicda induction (13). Stimulated Igh^+/Cγ^1-creRab7^fl/fl B cells were normal in activation of p38 and ERK1/2 MAPK pathways, which are associated with cell metabolism and induction of Prdm1 expression (53, 54), consistent with their normal proliferation/survival and plasma cell differentiation (Fig. 10B). Finally, they were normal in induction of LC3 (LC3-I) and conversion of LC3-I to LC3-II (Fig. 10A). Thus, Rab7 plays a specific role in the activation of the canonical NF-κB pathway, which is indispensable for AID induction.

In this study, we have unveiled an important (B cell–intrinsic) role of Rab7 in the Ab response by constructing conditional KO mice that lack Rab7 only in B cells induced to undergo Iγ1-Sγ1-Cγ1 transcription. This design has circumvented any complications potentially arising from possible roles of Rab7 in B cell and T cell development. It also has allowed us to outline an important role of Rab7 in CSR, through activation of the canonical NF-κB pathway and induction of AID expression. The role of Rab7 in the generation of class-switched B cells and AFCs, but not plasma cells, contrasts the role of Atg5 in mediating the development of peritoneal B1 cells and generation/function of plasma cells but not CSR (55–57). Rab7 may play a role in the bone marrow homing of plasma cells originating in secondary lymphoid organs, as suggested by our (unpublished) observations in Igh^+/Cγ^1-creRab7^fl/fl mice; it might also mediate the differentiation of IgM⁺ and residual class-switched B cell into memory B cells and the maintenance/function of memory B cells—the Atg7 autophagic protein has been shown to play such a role (58). Addressing these possibilities would entail the generation of new conditional Rab7 KO mice, including those with tamoxifen-inducible Cre expression, long after primary immunization (59) to ablate Rab7 in plasma cells and memory B cells—use of Tg(Prdm1-cre) mice would not be advisable because of the critical role of Rab7 in embryogenesis (47, 51) and possibly high Cre expression (together with Blimp-1 expression) during this process (60, 61). Finally, the normal IgM expression, proliferation, and survival of Igh^+/Cγ^1-creRab7^fl/fl B cells suggests that the role of Rab7 is redundant with that of other small GTPases in mediating basal membrane functions and maintaining B cell homeostasis.

The role of Rab7 in CSR induction was intrinsic to B cells, as shown by the failure of Igh^+/Cγ^1-creRab7^fl/fl B cells to undergo CSR in vivo in a T-independent Ab response and in vitro. In T-dependent Ab responses, Rab7 might mediate, in addition to CSR, autophagy in B cells for Ag presentation (62–65), thereby activating Th cells, which, in turn, induce further B cell activation and differentiation. In Igh^+/Cγ^1-creRab7^fl/fl mice injected with (T-dependent) NP-CGG, however, B cells displayed normal proliferation, survival, germinal center reaction, IgM production, and plasma cell differentiation, suggesting that Rab7-deficient B cells were competent in activating Th cells, possibly by using a different Rab GTPase (e.g., Rab9) for autophagy-mediated Ag presentation, thereby receiving Th cell help. Igh^+/Cγ^1-creRab7^fl/fl B cells showed impaired CSR to not only IgG1 but also IgE in vivo and in vitro despite normal Igh⁺ germline Iε-Sε-Cε transcription, further suggesting that mediating induction of AID expression is the mechanism by which Rab7 plays a role in CSR. Decreased AID expression in these B cells would impair direct Sμ→Sε (IgM→IgE) switching and sequential Sμ→Sγ1→Sε (IgM→IgG1 and then →IgE) switching (Fig. 11), although the latter would be a minor contributor to the overall CSR to IgE in vitro, as suggested by the intact or even increased IgE production in B cells lacking the Iγ1 promoter or Sγ1 DNA (66, 67). Also, the normal IgG3, IgG2a, IgG2b, or IgA levels in Igh^+/Cγ^1-creRab7^fl/fl mice indicated that CSR to IgG2a, IgG2b, or IgA is not dependent on sequential switching Sμ→Sγ1→Sγ2a, Sγ2b, or Sα. As, to best of our knowledge, no other Igγ or Igα sublocus–specific cre knockin mice exist, verification of Rab7 role in CSR to non-IgG1 or IgE Ig isotypes in vitro would entail Rab7 gene ablation through retroviral transduction to express Cre in Igh^+/+Rab7^fl/fl B cells or, as shown in this study, through stimulation of Igh^+/Cγ^1-creRab7^fl/fl B cells with IL-4 in the presence of a primary CSR-inducing stimulus for 36–48 h, followed by culturing of B cells with stimuli that induce CSR to those Ig isotypes. The in vivo verification would entail generation of conditional Rab7 KO mice using Aicda^+/cre (68) or BacTg(Aicda-cre)1Rcas allele (69). In these mice, Cre is induced in B cells activated to transcribe Aicda in a fashion likely dependent on the initially available Rab7 (once Rab7 is ablated, Cre expression would be extinguished).

The role of Rab7 in CSR induction is tightly associated with its role in AID induction because enforced expression of AID in Igh^+/Cγ^1-creRab7^fl/fl B cells, which showed impaired AID expression, rescued CSR. The CSR rescue was perhaps incomplete because expression of selected other important CSR factors, such as 14-3-3 adaptor proteins, whose induction also dependent on the canonical NF-κB activation (49), likely remained suboptimal. As we and others have shown, AID expression is tightly regulated, at several levels, including transcription, posttranscription, targeting, and enzymatic activity (27, 70). Primary inducing stimuli induced expression or activation of upstream factors, including Rab7 and NF-κB, resulting in high levels of Aicda expression. The function of the canonical NF-κB pathway in AID induction can be potentiated, but not replaced, by the noncanonical NF-κB pathway, a nonredundancy emphasized by the inability of BCR cross-linking alone to induce AID or CSR despite its efficient activation of the noncanonical NF-κB pathway (13). Accordingly, the defective AID induction in Igh^+/Cγ^1-creRab7^fl/fl B cells would at least partially result from impaired canonical NF-κB activation. By contrast, the noncanonical NF-κB pathway in these B cells was normal and would have mediated—perhaps in a compensatory manner—germline Iγ1-Sγ1-Cγ1 transcription. Consistent with this notion, this transcription process is critically regulated by the κB sites in the Iγ1 promoter and the 3′ locus control region (71, 72) and can be induced by BCR cross-linking (13, 73). The normal noncanonical NF-κB pathway would also support homeostasis of Igh^+/Cγ^1-creRab7^fl/fl B cells. These B cells showed lower levels of CSR than Igh^+/+Rab7^fl/fl B cells, suggesting that their defective AID induction does not result from modulation of Aicda transcription by the loxP sites inserted in the Rab7 locus, which lies 30 Mbp 5′ of Aicda locus, including a distal regulatory region (24, 25).

As suggested by our findings in this study, NF-κB activation in B cells would be mediated by Rab7-dependent transduction of signals triggered by surface or intracellular receptors. The predominant localization of this small GTPase in mature endosomes and endosome-derived membranes suggests an important role of intracellular membranes in signal transduction in B cells. Such a role has been underappreciated for CD40 or surface TLRs, which are also likely endocytosed into endosomes upon ligand engagement. CD40 can be internalized by B cells stimulated by an anti-CD40 Ab, and the TLR4-LPS complex can be endocytosed by non-B cells (perhaps together with the TLR4 coreceptor CD14) to regulate IRF3 activation through Rab11a small GTPase and Ca²⁺ signaling (74–78). Rab7-containing endosomes, which may be the source of Rab7⁺ foci-like structures we detected by superresolution microscopy, would house internalized receptors and recruit signaling adaptors necessary for specific activation of the canonical NF-κB pathway, such as the E3 ubiquitin ligase TRAF6. Thus, Rab7 would nucleate the formation of macromolecular complexes, perhaps within the foci structures, to further stabilize interactions of immune receptors and adaptors, thereby enhancing signaling strength and specificity. In B cells stimulated through CD40 or surface TLRs (e.g., TLR1/2 and TLR4), canonical NF-κB activation mediated by intracellular membranes would complement and strengthen that mediated by signals triggered at the plasma membrane. The normal induction of autophagy-related genes and normal conversion of LC3-I to LC3-II in Igh^+/Cγ^1-creRab7^fl/fl B cells suggest that induction of autophagosome formation is independent of Rab7—lack of significant increase of LC3-II in these B cells suggests that the overall degree of Rab7-mediated generation of autolysosomes (by fusion of autophagosomes with lysosomes) in B cells is low or another Rab protein plays a compensatory role in the fusion process to complete the autophagic flux. Our observation that ERK1/2 activation was normal Rab7-deficient B cells suggests that autophagosomes may recruit ERK1/2 and localize them onto their cytosolic/extraluminal side to mediate ERK phosphorylation (79), likely in an LC3-dependent but Rab7-independent manner. Rab7 dependency of CSR and Atg5 dependency of plasma cell differentiation reflect putative roles of different intracellular membranes in transducing different signals from some of the same receptors. Although internalized BCR might not trigger signals and merely be processed in lysosomes for Ag presentation (80), it could collaborate with TLR9–CpG complexes in double-membrane structures to activate NF-κB and MAPK (36). This raises the possibility that intracellular membranes provide a platform for relay of signals from individual receptors as well as integration of signals from different receptors.

Our findings outline an important role of Rab7 in Ab responses (i.e., by regulating expression of genes critical to B cell differentiation, namely Aicda) possibly through biogenesis of endosomes, trafficking of immune receptors (CD40 or TLRs) along these membranes, and stabilization of receptor colocalization with specific signaling molecules (e.g., TRAF6). This would reflect a complex role of Rab7, a highly conserved molecule in phylogeny (81). In plants and early animals, engulfment and subsequent degradation of microbes by Rab7-dependent xenophagy would be an important component of immunity. In later animals that use the TLR-NF-κB and CD40-NF-κB signaling modules (82), Rab7 gained a new function (i.e., in signal transduction and gene regulation). Such functions would endow some nonmammals and all mammals with a sophisticated adaptive immunity, as centered on induction of AID for CSR and/or SHM in T-dependent and T-dependent and Ab responses.

We thank Christie-Lynn Mortales for help with ELISPOT analysis, Dr. Peilin Zheng for expert help with stimulated emission depletion imaging analysis, and Dr. Andrew Lees (Fina Biosolutions) for anti-δ mAb/dextran.

The authors have no financial conflicts of interest.