Scaffold Functions of 14-3-3 Adaptors in B Cell Immunoglobulin Class Switch DNA Recombination

Class switch DNA recombination (CSR) of the immunoglobulin heavy chain (IgH) locus crucially diversifies antibody biological effector functions. CSR involves the induction of activation-induced cytidine deaminase (AID) expression and AID targeting to switch (S) regions by 14-3-3 adaptors. 14-3-3 adaptors specifically bind to 5′-AGCT-3′ repeats, which make up for the core of all IgH locus S regions. They selectively target the upstream and downstream S regions that are set to undergo S–S DNA recombination. We hypothesized that 14-3-3 adaptors function as scaffolds to stabilize CSR enzymatic elements on S regions. Here we demonstrate that all seven 14-3-3β, 14-3-3ε, 14-3-3γ, 14-3-3η, 14-3-3σ, 14-3-3τ and 14-3-3ζ adaptors directly interacted with AID, PKA-Cα (catalytic subunit) and PKA-RIα (regulatory inhibitory subunit) and uracil DNA glycosylase (Ung). 14-3-3 adaptors, however, did not interact with AID C-terminal truncation mutant AIDΔ(180–198) or AIDF193A and AIDL196A point-mutants (which have been shown not to bind to S region DNA and fail to mediate CSR). 14-3-3 adaptors colocalized with AID and replication protein A (RPA) in B cells undergoing CSR. 14-3-3 and AID binding to S region DNA was disrupted by viral protein R (Vpr), an accessory protein of human immunodeficiency virus type-1 (HIV-1), which inhibited CSR without altering AID expression or germline IH-CH transcription. Accordingly, we demonstrated that 14-3-3 directly interact with Vpr, which in turn, also interact with AID, PKA-Cα and Ung. Altogether, our findings suggest that 14-3-3 adaptors play important scaffold functions and nucleate the assembly of multiple CSR factors on S regions. They also show that such assembly can be disrupted by a viral protein, thereby allowing us to hypothesize that small molecule compounds that specifically block 14-3-3 interactions with AID, PKA and/or Ung can be used to inhibit unwanted CSR.


Introduction
Immunoglobulin (Ig) class switch DNA recombination (CSR) and somatic hypermutation (SHM) are central to the maturation of the antibody response for the effectiveness of vaccines and the generation of neutralizing antibodies to microbial pathogens (including bacteria and viruses) and tumoral cells as well as the maturation of the autoantibody response in systemic or organspecific autoimmunity. CSR irreversibly substitutes the Ig heavy chain (IgH) constant (C H ) region, for instance, Cm for IgM -which is expressed in all naïve B cells -with a downstream C H region, Cc, Ce or Ca [1,2], thereby giving rise to IgG, IgE or IgA antibodies without altering the structure of the antigen-binding site [3,4]. Class-switched IgG, IgE and IgA antibodies display different tissue distributions and possess diverse biological effector functions. SHM inserts mainly single nucleotide point-mutations into the antibody recombined V(D)J segments at a high rate to provide the structural substrate for positive selection of high affinity mutants by foreign antigen [5,6]. CSR (as well as SHM) occur in B lymphocytes activated in peripheral lymphoid organs. CSR critically requires germline I H -C H transcription of the upstream, e.g., Igm, and downstream Igc, Ige or Iga subloci [4]. Germline I H -C H transcription initiates at the intervening Im, Ic, Ie or Ia promoter and elongates through the I H exon, intronic switch (S) region, and the C H exon cluster, giving rise to germline Im-Cm, Ic-Cc, Ia-Ca or Ie-Ce transcripts. CSR also requires activation-induced cytidine deaminase (AID, encoded by AICDA in humans and Aicda in mice), which is expressed at high levels in activated B lymphocytes, including those in germinal centers [7][8][9]. AID is a member of the AID/ APOBEC cytidine deaminase family; it deaminates deoxycytosines (dCs) in S region DNA, yielding deoxyuracils (dUs) [10,11]. The processing of dUs by uracil DNA glycosylase (Ung) results in abasic sites, nicking of which by apurinic/apyridimic endonucleases (APEs) leads to generation of DNA double-strand breaks (DSBs) in the upstream (donor, e.g., Sm) and downstream (acceptor) S regions that are obligatory intermediates of CSR [12]. CSR then proceeds through DSB resolution [13]; synapsis of the upstream and downstream DSBs occurs through excision of the intervening DNA from the chromosome to form a switch DNA circle and leads to S-S DNA junctions. Switch DNA circles are transiently transcribed, giving rise to circle Ic-Cm, Ie-Cm or Ia-Cm transcripts, which are hallmarks of ongoing CSR [14]. Postrecombined IgH DNA sequences are transcribed, giving rise to post-recombination Im-Cc, Im-Ce or Im-Ca transcripts, and mature V H DJ H -Cc, V H DJ H -Ce or V H DJ H -Ca transcripts, which encode IgG, IgE or IgA, respectively [4].
Triggering of CSR requires both ''primary'' and ''secondary'' CSR-inducing stimuli [4]. Primary CSR-inducing stimuli comprise a T-dependent stimulus, i.e., engagement of CD40 expressed on B cells by trimeric CD154 expressed on CD4 + T cells, or a T-independent stimuli, such as dual engagement of Tolllike receptors (TLRs) and B cell receptor (BCR). Such dual engagement is exemplified by E. coli lipopolysaccharides (LPS), which engages TLR4 and BCR through its monophosphoryl lipid A and polysaccharide moieties, respectively [4,15,16]. Secondary CSR-inducing stimuli consist of cytokines such as interleukin-4 (IL-4), transforming growth factor-b (TGF-b) and interferon-c (IFN-c), which selectively and specifically induce germline Ic-Cc and Ie-Ce (IL-4) or Ic2b-Cc2b and Ia-Ca (TGF-b) or Ic2a-Cc2a (IFN-c, in mouse but not human) transcription. Primary stimuli induce B cells to proliferate and express AID and other CSRrelated genes. They also enable secondary stimuli to direct CSR to specific immunoglobulin isotypes [17,18].
Here, we hypothesized that 14-3-3 adaptors function as structural scaffolds to stabilize multiple enzymatic elements on S regions in CSR. To test our hypothesis, we adapted the bimolecular fluorescence complementation (BiFC) assay to quantify the direct interaction between all seven isoforms of 14-3-3 adaptors with CSR factors, such as AID, PKA, RPA, Ung, or with AID mutants or Ung mutants that cannot mediate CSR. We also used (GST) pull-down assay to show that 14-3-3 interact with AID, PKA and Ung. Finally, we quantified the direct interaction between 14-3-3, AID, PKA or Ung and viral protein R (Vpr), an accessory protein of the human immunodeficiency virus type-1 (HIV-1) and utilized this naturally occurring viral protein to disrupt 14-3-3 and AID binding to S region DNA, which inhibited CSR. Thus, our findings suggest that 14-3-3 adaptors play important scaffold functions and nucleate the assembly of multiple CSR factors on S regions and that small molecule compounds that specifically disrupt their interactions can be used to inhibit unwanted CSR, such as CSR underlying the generation of IgG and IgA autoantibodies in autoimmunity and atopic IgE antibodies in allergies and asthma.

Chromatin Immunoprecipitation (ChIP)
ChIP assays were performed as we described [23]. B cells were treated with 1% (v/v) formaldehyde for 10 min at 25uC to crosslink chromatin before being washed with cold PBS containing protease inhibitors (Roche) and resuspended in lysis buffer (20 mM Tris-HCl, 200 mM NaCl, 2 mM EDTA, 0.1% w/v SDS and protease inhibitors, pH 8.0). Chromatin was sonicated to yield DNA fragments (about 200 to 600 bps), pre-cleared with protein A agarose beads (Pierce) and then incubated with rabbit anti-14-3-3c Ab (catalog # 18647, IBL, Inc.) or mouse anti-AID mAb (catalog # 39-2500, Invitrogen) overnight at 4uC. Immune complexes were precipitated by Protein A agarose beads, washed and then eluted with elution buffer (50 mM Tris-HCl, 0.5% SDS, 200 mM NaCl, 100 mg/ml proteinase K, pH 8.0), followed by incubation at 65uC for 4 hours to reverse formaldehyde cross-links and digest proteins. DNA in the supernatant was purified using a QIAquick PCR purification kit (Qiagen). Recovered DNA was specified by PCR using the following oligonucleotide primers: Sm, forward 59-GCTAAACTGAGGTGATTACTCTGAGGTAAG-39 and reverse 59-GTTTAGCTTAGCGGCCCAGCTCATTC-CAGT-39; Sc1, forward 59-ATAAGTAGTAGTTGGGGATTC-39 and reverse 59-CTCAGCCTGGTACCTTATACA-39. Data were normalized to input chromatin DNA and depicted as enrichment of each amplicon DNA relative to baseline value obtained using an irrelevant mAb.
Aicda and IgH locus Germline I H -C H , circle Ic1-Cm and Post-recombination Im-C H Transcripts RNA was extracted from 2610 6 cells using the RNeasy Mini Kit (Qiagen) according to the manufacturer's protocol. Levels of Aicda transcripts, germline Im-Cm and Ic1-Cc1 transcripts, circle Ic1-Cm transcripts and post-recombination Im-Cc1 transcripts were quantified by reverse transcription (RT) and real-time quantitative PCR (qPCR). First-strand cDNAs were synthesized from equal amounts of total RNAs (2 mg) using the SuperScript TM III First-Strand Synthesis System (Invitrogen). Real-time qPCR analysis was performed using the DNA Engine Opticon Real-Time PCR Detection System (Bio-Rad Laboratories) to measure SYBR-green (DyNAmo HS SYBR Green) incorporation with the following protocol: 50uC for 2 min, 95uC for 10 min, 40 cycles of 95uC for 10 sec, 60uC for 20 sec, 72uC for 30 sec, 80uC for 1 sec, and data acquisition at 80uC, and 72uC for 10 min. Melting curve analysis was performed from 72u295uC and samples were incubated for another 5 min at 72uC. The DDCt method was used to analyze levels of transcripts and data were normalized to levels of Gapdh transcripts. For qPCR, the following primers were

GST pull-down Assays
Spleen B cells (10 7 ) were stimulated with LPS plus mIL-4 for 48 hours. Cells were collected and resuspended in lysis buffer (20 mM of Tris-Cl, pH 7.5, 150 mM of NaCl, 0.5 mM of EDTA, 0.25% (v/v) of NP-40) supplemented with phosphatase inhibitors Na 4 P 2 O 7 (1 mM), NaF (10 mM) and NaVO 3 (1 mM) and a cocktail of protease inhibitors (Sigma). After sonication and centrifugation, protein lysates were precleared with immobilized Glutathione Magnetic Beads (Pierce) bound to glutathione Stransferase (GST) for 1 hour at 4uC. Precleared whole cell lysates were then probed with immobilized Glutathione Magnetic Beads

Image Analysis
To detect codistributing molecules, human B cells were spun onto cover slips pre-coated with 10 mg/ml poly-D-lysine (Sigma-Aldrich), fixed with 2% paraformaldehyde and permeabilized with 0.25% Triton X-100. After blocking with 1% BSA, cells were stained with a rabbit anti-14-3-3 Ab (catalog # ab6081, Abcam) and a mouse anti-AID mAb (catalog # 39-2500, Invitrogen) or a mouse anti-RPA/p34 mAb (catalog # MS-691-P1, NeoMarkers), followed by staining with FITC-conjugated goat anti-rabbit and TRITC-conjugated goat anti-mouse IgG Ab. Confocal microscopy images of spontaneously switching 4B6 B cells cultured at 5610 4 cell/ml in (FBS-RPMI) and human 2E2 B cells stimulated with nil or anti-hCD40 mAb plus hIL-4 were captured by an LSM510-META NLO multi-photon confocal microscope and analyzed using the laser scanning microscope (LSM) Image Examiner software (Carl Zeiss Microimaging, Inc.). Each pixel in individual cells was analyzed for intensity in green (14-3-3) and red (AID or RPA). The ratio of the number of yellow pixels, to the total number of yellow and green pixels was used to quantify the percentage of 14-3-3 molecules codistributing with AID or RPA.
To detect nuclear foci, spontaneously switching human 4B6 B cells or primary mouse B cells stimulated with LPS plus mIL-4 were washed and then extracted in cytoskeletal (CSK) buffer to deplete soluble cytoplasmic and nuclear components before resuspended in 4% paraformaldehyde in PBS to crosslink chromosomal DNA and bound proteins. Cells were then processed, stained and immunofluorescence images were taken at multiple confocal planes. Such z-stacked images were deconvoluted to generate high-resolution images. All images were pseudo-colored for presentation. B cells were scored as positive for colocalization of 14-3-3 nuclear foci with AID nuclear foci or RPA nuclear foci when showing at least two colocalizing foci, defined as spots within the top 75% pixel intensity of each fluorescence on the pixel intensity map by the AutoquantH X software (Media Cybernetics, Inc.). The percentages of positive cells at 0 hour, 24 hours and 48 hours of stimulation were plotted against time points and the Mander's overlapping coefficiency were calculated to be between 0.91-0.99 (maximum is 1.0), thereby reflecting a timedependent increase of colocalization. Processing and staining of mouse primary B cells or human 4B6 B cells were performed following a protocol for foci formation analysis [31].

Retroviral Transduction
The coding sequence of GFP-Vpr (originally from the pGFP-Vpr plasmid, catalog # 11386, The NIH AIDS Research and Reference Reagents Program) was cloned into the retroviral vector pCSretTAC to generate pTAC-GFP-Vpr (generation of pTAC-GFP was described before [23]). Retroviral constructs were transfected along with the pCL-Eco retrovirus-packaging vector into HEK293T cells using the ProFection Mammalian Transfection SystemH (Promega). Transfected cells were cultured in FBS-RPMI in the presence of chloroquine (25 mM) for 8 hours. After the removal of chloroquine, retrovirus-containing culture supernatants were harvested every 12 hours for 48 hours. For transduction and CSR analysis, mouse B cells were activated with LPS for 24 hours and then centrifuged at 500 g together with viral particles in the presence of 6 mg/ml polybrene (Sigma-Aldrich) for 90 min at 25uC. Transduced B cells were then cultured in virus-free FBS-RPMI in the presence of LPS plus mIL-4 for 48 hours for transcript analysis, or for 72 hours for flow cytometry analysis of surface IgG1 and B220 expression in GFP + B cells.

HIV-1 Vpr (15-mer) Peptides
Twenty-two Vpr peptides, each of 15-amino acids in length and each with a sequential 11-amino acid overlap, covering the entire 96-amino acid viral protein, was provided by The NIH AIDS Research and Reference Reagents Program (catalog # 6447). Vpr peptides were dissolved in DMSO and then added to LPSactivated B cell cultures for 24 hours, after which B cells were stimulated with LPS plus mIL-4 for 72 hours to induce CSR from IgM to IgG1.

Detection of Vpr in Human B Lymphocytes
Tonsils from HIV-1subjects and lymph nodes from HIV-1 + patients (UC Irvine Medical Center) were sectioned (7 mm) onto glass slides and stained with Abs specific to human using rabbit anti-Vpr Ab (catalog #11836, NIH AIDS Research and Reference Reagents Program) alone, or with mouse anti-human CD20 Ab (clone 2H7, eBioscience) or with mouse anti-AID mAb (catalog #  39-2500, Invitrogen). Color was development by DAB+ chromogen (Dako) and slides were dehydrated and counterstained with hematoxylin. Images were captured at 40X magnification using Nikon Eclipse E400.
To detect B cell uptake of Vpr, human 4B6 B cells were cocultured for 7 days with human CEM.NK R -CCR5 T cells infected with nil or HIV-192US657 (a clinical isolate of the primary R5 HIV-1 strain [32][33][34]), fixed with 2% paraformaldehyde and permeabilized with 0.25% Triton X-100 (Sigma-Aldrich). After blocking with 1% BSA, cells were stained with FITC-conjugated anti-m mAb or FITC-conjugated anti-CD4 mAb to identify sIgm + 4B6 cells or CD4 + CEM.NK R -CCR5 T cells, respectively and rabbit anti-Vpr Ab (catalog # 11836, NIH AIDS Research and Reference Reagents Program) and Alexa 594conjugated goat anti-rabbit IgG Ab to identify Vpr uptake for 1 hour. DAPI was used to stain the nucleus. Fluorescence images were captured at 100X magnification. All experiments involving HIV-192US657 were performed in a dedicated Biosafety Level III facility.

Ethics Statement
Human materials. All human subject materials were surgical remnants collected and maintained for research purposes by the School of Medicine Department of Pathology, University of California, Irvine, CA 92697-under patient privacy rule, no information was disclosed on age, sex and identity of the patient sources of the surgical materials. The use of these materials for research associated with this study was specifically approved by the Institutional Review Board of the University of California, Irvine.
Mice. C57BL/6 mice 8-12 weeks old were housed in a pathogen-free facility and provided with autoclaved food and deionized water. The Institutional Animal Care and Use Committee of the University of California, Irvine, CA 92697, specifically approved the use of mice for the isolation of primary murine B cells for research associated with this study.

Statistical Analysis
Statistic analysis was performed using ExcelH software (Microsoft) to determine P values by two-tailed paired student's t-test. P values less than 0.05 were considered significant.
Thus, all seven isoforms of 14-3-3 adaptors interact with AID and a CSR-proficient AID C-terminal point-mutant, but not with CSR-defective AID C-terminal truncation mutant or pointmutants.
Thus, all seven isoforms of 14-3-3 adaptors interact with both catalytic and regulatory subunits of PKA and with Ung in a manner dependent on Ung C-terminal amino acid residues, but not with RPA.

14-3-3c Interacts with AID, PKA-Ca and Ung in B Cells Undergoing CSR
To confirm that 14-3-3c can interact with AID, PKA-Ca and Ung in B cells undergoing CSR, we performed glutathione Stransferase (GST) pull-down experiments using 14-3-3c fused to GST (GST-14-3-3c) and whole cell lysates prepared from mouse B cells stimulated with LPS plus mIL-4 for 48 hours. Using immunoblotting, we readily detected AID, PKA-Ca and Ung among protein molecules after pull-down by GST-14-3-3c, but not GST (Figure 3). GST-14-3-3c was also detected, as expected, and endogenous 14-3-3c was detectable when a higher amount of GST-14-3-3c-precipitated proteins were analyzed (data not shown). Thus, 14-3-3c interacts with AID, PKA-Ca and Ung in switching B cells, consistent with its direct interactions with these CSR factors shown by our BiFC assays.

14-3-3 Adaptors Colocalize with AID and RPA in Nuclei of Switching B Cells
As we have shown, 14-3-3f-AID interaction mainly occurs in the nucleus despite their predominant cytoplasmic localization [23]. Using confocal microscopy we found that, consistent with their direct interaction, 14-3-3 adaptors and AID were expressed and codistributed mainly in the cytoplasm and, to a lower degree, in the nucleus of human 4B6 B cells, which switch spontaneously from IgM to IgG, IgA and IgE [23,45] (Figure 4a). Also, 14-3-3 codistributed with AID and RPA in human 2E2 B cells upon stimulation by an agonistic anti-hCD40 mAb plus hIL-4, which induce human 2E2 B cells to undergo CSR from IgM to IgG and IgE [23] (Figure 4b). To directly visualize the colocalization of 14-3-3 and AID or RPA in the nucleus, where CSR occurs, we adapted a protocol to deplete human and mouse B cells of cytosolic constituents and then identified 14-3-3, AID or RPA by specific Abs. As detected by this approach, 14-3-3, AID and RPA formed nuclear foci in spontaneously switching human 4B6 B cells (Figure 5a, 5b) and in mouse B cells stimulated by LPS plus mIL-4 to undergo CSR from IgM to IgG1 at high levels, after 24 and 48 hours (Figure 5c, 5d). We consistently detected two to four 14-3-3-containing nuclear foci that colocalized with AIDcontaining nuclear foci or RPA-containing nuclear foci ( Figure 5), in spite of no direct interaction between 14-3-3 and RPA (Figure 2b). The percentages of human 2E2 B cells showing 14-3-3 molecules codistributing (%) with AID or RPA that were stimulated with agonistic anti-hCD40 mAb plus hIL-4 were higher than those of 2E2 B cells stimulated with nil ( Figure 6a). The proportion of B cells that displayed colocalization of 14-3-3 nuclear foci with AID nuclear foci or RPA foci increased over time of stimulation from 0% to 4.7% within 24 hours, to 12.7% within 48 hours for AID, or from 0.8% to 6.8% within 24 hours, to 18.1% within 48 hours for RPA, consistent with increased proportions of IgG1 + B cells (typically 5% within 48 hours, 15% within 72 hours and 30% within 96 hours, not shown) under the same stimulation conditions (Figure 6b).
Thus, 14-3-3 adaptors colocalize with AID and RPA in the nucleus of human and mouse B cells undergoing CSR.

14-3-3 and AID Recruitment to S Regions and B Cell CSR are Inhibited by Vpr
Prompted by our findings that 14-3-3 adaptors directly interacted and colocalized with AID and other CSR factors in the nucleus of switching B cells, we hypothesized that 14-3-3mediated recruitment of the AID-centered CSR machinery to S regions would be blocked when interactions of 14-3-3 with CSR factors were disrupted by naturally occurring or synthetic molecules, leading to CSR inhibition. To test this hypothesis, we first analyzed CSR inhibition by utilizing the HIV-1 accessory protein Vpr, which has been suggested to bind to 14-3-3 proteins [46], Ung [43] and possibly PKA [47]. Primary mouse B cells were transduced with pTAC retrovirus to express green fluorescent protein (GFP) alone (pTAC-GFP) or Vpr linked to GFP (pTAC-GFP-Vpr) and then stimulated with LPS and mIL-4 for CSR from IgM to IgG1. CSR was inhibited by Vpr, as shown by reduced proportions of IgG1 + cells (23%) among B cells expressing GFP-Vpr when compared to their counterparts (42%) expressing GFP (Figure 7).
In addition, levels of circle Ic1-Cm and post-recombination Im-Cc1 transcripts, which are accurate parameters of ongoing and completed CSR, respectively, were reduced in the presence of Vpr. However, levels of AID expression and germline Im-Cm and Ic-Cc1 transcription were normal in B cells expressing Vpr (Figure 7b). To further dissect CSR inhibition by Vpr, we used an array of 22 Vpr peptides, each of 15-amino acid in length and each with a sequential 11-amino acid overlap, covering the entire 96-amino acid viral protein. We analyzed CSR inhibition by each peptide in mouse B cells induced by LPS plus mIL-4-uptake of these peptides by B cells was efficient, possibly due to their small size. A differential CSR inhibitory activity was displayed from different Vpr peptides (without affecting B cell proliferation or survival, data not shown), which segregated into three inhibitory clusters (I, II and III) (Figure 8a). The three clusters correlated with the three a-helices of Vpr [48], which are structurally independent (Figure 8b), and perhaps serve as distinct functional domains for the interaction of the retrovirus with host cell proteins.
To address the molecular mechanisms underlying CSR inhibition by Vpr, we analyzed 14-3-3c and AID recruitment to S regions in the presence of Vpr. Binding of 14-3-3c and AID to Sm and Sc1 DNA was inhibited by Vpr, as shown by our ChIP assays using a specific Ab to 14-3-3c or AID and chromatin from B cells expressing GFP-Vpr or GFP that were stimulated with LPS plus mIL-4 ( Figure 7c). Further, BiFC assays were utilized to address the direct interaction between 14-3-3 or AID and Vpr by coexpressing Flag-14-3-3f-EYFP1-154 or Flag-AID-EYFP1-154 and HA-Vpr-EYFP155-238, respectively, in HeLa cells. 14-3-3f and AID indeed interacted with Vpr and they did so in the nucleus (Figure 9a, 9b). Further, in our BiFC assays involving Flag-AIDD(180-198)-EYFP1-154, Flag-PKA-Ca-EYFP1-154, Figure 11. Schematic illustration of 14-3-3 scaffold functions in the assembly of S region DNA-protein complexes and disruption of such complexes by the naturally occurring Vpr molecule. A speculative model of a possible configuration of interacting proteins during CSR, and the disruption of some of these complexes by Vpr, as suggested by our current and previous findings [4,23,42]. The CSR macromolecular complex localizes to chromatin S regions through 14-3-3 adaptors (green) binding to the modified histone tail H3K9acS10ph (PDB ID 2c1j) as well as to 59-AGCT-39 repeats (PDB ID 3afa), which specifically recur in S region DNA. Upon recruitment to S regions by 14-3-3, AID (red, homology modeled after human APOBEC2, PDB ID 2nyt) is phosphorylated at Ser38 by the PKA catalytic subunit (orange, PDB ID 3tnp), leading to the recruitment of RPA (silver, PDB ID 4gop) to S region DNA. Rev1 (brown, PDB ID 3gqc) is important for CSR by virtue of its ability to stabilize Ung (purple, PDB ID 3fci) on S region DNA. Vpr (magenta, PDB ID 1esx) impairs CSR, possibly by disrupting critical DNA-protein interactions, such as S region DNA-14-3-3 binding, and/or protein-protein interactions, such as those between 14-3-3 and AID, PKA or Ung. Graphics were rendered using the UCSF Chimera software (http://www.cgl.ucsf.edu/chimera).  Figure 9c).
Thus, 14-3-3 adaptors function as scaffolds in mediating the recruitment of AID and possibly PKA and Ung to S regions, which can be inhibited by Vpr, likely due to the disruption of interactions between 14-3-3 and with those CSR factors, leading to the inhibition of CSR.

AID Colocalize with Vpr in Germinal Center B Cells from HIV-1 Patients
We next addressed the biological relevance of CSR inhibition by Vpr during HIV infection, in which a decrease in class-switched antibodies has been reported to occur [49]. HIV does not infect B cells, but the HIV component such as Nef have been detected in B cells [49]. This prompted us to hypothesize that Vpr is present in secondary lymphoid structures during HIV infection. Here, we analyzed lymph node sections from HIV-1 + patients and tonsil sections from HIV-1subjects by immunohistochemistry using a specific Ab to Vpr. This revealed Vpr localized within germinal centers of HIV-1 + patients but not HIV-1subjects ( Figure 10a). In addition, AID colocalized with Vpr within germinal center B cells of lymph nodes from HIV-1 + patients but not HIV-1subjects after double staining with an anti-Vpr Ab and anti-AID Ab or anti-CD20 Ab (Figure 10b).
To demonstrate that B cells can acquire Vpr from an HIV-1 + source, we cocultured (HIV-1 -) human 4B6 B cells with (HIV-1 -) human CEM.NK R -CCR5 T cells infected with nil or HIV-192US657 (a clinical isolate of the primary R5 HIV-1 strain [32][33][34]). 4B6 B cells displayed Vpr nuclear and cytoplasmic internalization, as indicated by the intracellular staining with FITC-conjugated anti-m chain mAb staining and with Alexa 594conjugated anti-Vpr Ab, only when cocultured with HIV-1infected T cells, which were specified by staining with FITCconjugated anti-CD4 mAb and positive for the intracellular staining of Vpr (Figure 10c).
Thus, AID colocalizes with Vpr within germinal center B cells of HIV-1 + patients and can be picked up by HIV-1 -B cells upon coculturing with HIV-1 + T cells.

Discussion
The recruitment and scaffold functions of 14-3-3 adaptors for the stabilization of enzymatic activity of AID and other CSR factors on S regions require multiple molecular interactions. Here, we showed that all seven isoforms of 14-3-3 adaptors directly interacted with AID, but failed to interact with AID C-terminal point mutants that have been shown to be defective in mediating CSR [38,39]. In B cells undergoing CSR, 14-3-3 and AID formed nuclear foci of macromolecular complexes, as suggested by the colocalization of 14-3-3 nuclear foci and AID nuclear foci at 24 hours and 48 hours, but not at 0 hour. Non-colocalizing 14-3-3 nuclear foci may be involved in binding to H3K9acS10ph in the genome (outside the IgH locus) for transcription regulation [50]; non-colocalizing AID foci may be responsible for dC deamination outside the Ig locus [51][52][53], particularly in regions where RNA polymerase II stalls [54,55].
All seven isoforms of 14-3-3 adaptors directly interacted with the catalytic and regulatory subunits of the PKA holoenzyme. These would lead to 14-3-3-mediated stabilization of CSR factors on S region DNA and enhancement of AID and PKA activity. 14-3-3 adaptors also directly interacted with Ung, which together with APEs mediates the dominant pathway of DSB generation in CSR [4,12]. The direct interaction of 14-3-3 adaptors with Ung likely serves two important functions. First, 14-3-3 adaptors possibly facilitate AID and Ung reciprocal stabilization on S region DNA [56], as 14-3-3 possibly bridge AID and Ung to form an AID-14-3-3-Ung complex. This notion is further supported by findings that the reciprocal and cooperative stabilization of AID and Ung depends on the AID C-terminal region, which, as we showed [23], mediates 14-3-3-AID interaction. Second, 14-3-3 may enhance the processing of AID-inserted dUs by Ung for the eventual generation of DSBs. This putative 14-3-3 function would complement and/or enhance that of Rev1, which serves as a scaffold for Ung by stabilizing Ung on S region DNA and enhancing its enzymatic activity [42]. Mismatch repair (MMR) proteins Msh2 and Msh6, recognize deoxyuracil:deoxyguanine mismatches and recruit 59R39 exonuclease I (ExoI) to the generate DSBs [57], and they stabilize AID on S region DNA in a manner dependent on AID C-terminal region [56]. 14-3-3 adaptors have been shown to interact with ExoI [58,59], suggesting that 14-3-3 adaptors are also involved in MMRdependent DSB generation. The critical importance of the interaction between 14-3-3 adaptors and essential CSR factors, likely contribute to the assembly of the CSR machinery, which includes AID, Ung, Msh2 and Msh6 [56,57].
14-3-3 nuclear foci colocalized with RPA nuclear foci despite no direct interaction between 14-3-3 and RPA1, the ssDNA-binding subunit, suggesting that RPA1 can only interact with 14-3-3 through another CSR-related factor, such as phosphorylated AID at S38 to form an RPA-AID-14-3-3 complex. RPA has been suggested to mark AID-mediated DNA damage sites [60] and can bind several CSR factors involved in DSB generation, such as Ung [61], MMR elements [62] and the Mre11/Rad50/Nbs1 complex [63]. RPA may also be key to the transition from formation of DSB-generating macromolecular complexes to DSB-resolving complexes on S regions, as RPA can also bind factors that are involved in CSR DSB resolution, such as c-H2AX, 53BP1 and DNA-PKcs [4].
CSR to IgG1 in primary B cells was significantly reduced in the presence of HIV-1 Vpr; Vpr was also reported to inhibit CSR to IgA in CH12F3 B cells [43]. The CSR inhibitory activities of Vpr segregated within three structural clusters (I, II and III) corresponding to the three Vpr helices, therefore, our studies suggest the notion that CSR inhibition by Vpr is through a ''multiprong'' strategy. Interestingly, cluster I peptides encompassed the Ung-binding Vpr region [64] and cluster III peptides straddled the 14-3-3-and PKA-interacting Vpr regions [47,65]. Inhibition of CSR by Vpr did not affect AID expression or germline I H -C H transcription, suggesting that unlike HIV-1 Nef, which inhibits CSR by interfering with CD40-signaling and NF-kB activation [66], signaling events induced by primary CSR-stimuli and/or secondary CSR-stimuli is not affected by Vpr. By contrast, intimate CSR mechanisms was likely disrupted by Vpr, particularly the binding of 14-3-3 to S region 59-AGCT-39 repeats, to H3K9acS10ph, or to important CSR factors, such as AID, PKA and Ung. This notion is supported by our data that recruitment of 14-3-3c to Sm and Sc1 DNA was disrupted by Vpr, and that essential CSR factors such as 14-3-3f, AID, PKA-Ca and Ung directly interacted with Vpr. Reduced binding of 14-3-3c to S regions, perhaps in conjunction with putative blocking of 14-3-3 and AID interaction by Vpr, lead to reduced AID recruitment to Sm and Sc1 DNA. Importantly, proliferation, survival or cell cycle of B cells was not affected in the presence of Vpr, by contrast Vprmediated cycle arrest in T cells has been reported [67].
As we have also shown here, AID colocalized with Vpr in HIV-1 + patients lymph node germinal center B cells. In addition, HIV-1 -B cells internalized Vpr when cocultured with HIV-1infected CD4 + T cells. B cell internalization of HIV-1 Vpr is likely facilitated by CD21 [68,69], CD40 and/or binding of DC-SIGN [70] -Vpr can be released from virions or infected T cells in an extracellular form in an extracellular form [71], which can be internalized by bystander B cells through protein transduction [72]. Irrespective of Vpr entry route, CSR and the generation of class-switched IgG, possibly including HIV neutralizing IgG antibodies [49,66], is efficiently inhibited by Vpr, leading to HIV-1 evading immune response and persistently infecting the host.
Overall, our findings provide evidence that 14-3-3 adaptor proteins nucleate the assembly of the AID-centered CSR machinery on 59-AGCT-39 DNA repeats and H3K9acS10ph marks in S regions that are set to undergo S-S recombination ( Figure 11). They also suggest that CSR DNA-protein and/or protein-protein complexes can be destabilized by naturally occurring molecules, leading to inhibition of CSR, thereby providing a basis for the identification of synthetic small molecule compounds, such as those that disrupt DNA-protein and/or protein-protein interactions involving 14-3-3 [26,73], or biologics that can effectively inhibit unwanted CSR, such as CSR underlying the generation of IgG and IgA autoantibodies in autoimmunity and atopic IgE antibodies in allergies and asthma.