Ig-Independent Ig Expression on the Surface of B Lymphocytes after B Cell Receptor Aggregation
http://www.100md.com
免疫学杂志 2005年第3期
Abstract
In order for humoral immune responses to develop, B cells must be able to recognize, bind, and internalize Ags. These functions are performed by the BCR, which is also responsible for initiating and transducing activation signals necessary for B cell proliferation and differentiation. We have examined surface expression patterns of individual components of the BCR following anti-Ig- and Ag-induced aggregation. Specifically, the localization and expression levels of the Ag-binding component, surface Ig (sIg), and the Ig component of the Ig/Ig signaling unit were investigated to determine their individual participation in the internalization and signal transduction. Using primary murine B cells, we found that while >95% of the sIg is internalized following anti-Ig-induced aggregation, 20–30% of Ig remains on the surface. These results suggest that sIg and Ig may function independently following the initial stages of signal transduction.
Introduction
Development of a humoral immune response requires the B cell to receive two separate signals. The first is the result of the Ag binding the clonally expressed BCR, and the second is provided as a consequence of cognate interaction with an activated CD4 Th cell or, in some cases, signaling through TLRs. In order for productive interaction with MHC-restricted Ag-specific T cells to occur, the BCR itself has to perform two essential functions: communicate the activating signal to the inside of the cell, and internalize the Ag for processing and presentation.
The BCR is a multisubunit complex composed of surface Ig (sIg)3 noncovalently associated with the Ig/Ig heterodimer. sIg functions as the Ag recognition and binding component, while Ig/Ig transduces the Ag-induced signal (1). The association of sIg with the Ig/Ig heterodimer is believed to be necessary for trafficking the BCR to the plasma membrane (2). Although different forms of BCR are expressed throughout B cell development, the Ig/Ig heterodimer is an essential component of all of them (3). Presence of a specific form of BCR is required to proceed to the next stage of development, and in mature cells surface expression of BCR is necessary for survival (4).
BCR aggregation as a consequence of Ag-sIg binding and oligomerization results in tyrosine phosphorylation of specific protein-signaling motifs called ITAMs present in the cytoplasmic regions of both Ig and Ig. ITAM phosphorylation initiates recruitment of cytoplasmic signaling proteins and formation of a stable signaling complex that functions to link Ag binding to B cell activation. Activation refers to a complex process associated with generation of cytoplasmic second messengers, elevations in RNA and protein synthesis, changes in gene expression, and initiation of programs leading to the ability to elicit and receive secondary signals from T cells, culminating with the growth and proliferation of the B cell (5).
Shortly after the ligand-induced aggregation of the BCR, the stimulated B cell concentrates the majority of its BCR complexes to a single pole on the cell surface in a process commonly referred to as capping (6). Following the formation of these tight unipolar aggregated complexes, the BCR together with the bound Ag is internalized (7). Recent studies have indicated that in mature stage B cells, internalization occurs in compartments associated with clathrin and organized lipid rafts (8). Other studies also indicate that under some conditions of Ag binding, there is a functional and physical uncoupling of sIg and the Ig/Ig signaling complex (9, 10). It is unclear how this uncoupling affects the internalization of the BCR complex. In this study, we monitored the dynamics of surface expression and subsequent internalization of the signaling and Ag recognition components of the BCR. We report in this work that following BCR cross-linking, a significant amount of the Ig pool (presumably together with Ig) remains on the surface independent of sIg internalization. This retention occurs in response to both anti-Ig stimulation of polyclonal splenic B cells as well as Ag stimulation of anti-arsonate-specific Ig transgenic B cells.
Materials and Methods
Cell preparation and cell lines
Female BALB/c mice were used for all experiments, unless specified. Transgenic mice expressing human Ig were obtained from S. Liebhaber (University of Pennsylvania). A-deficient mice were provided by T. Laufer (University of Pennsylvania) and also were purchased from Taconic Farms. Transgenic mice expressing BCRs specific for p-azophenylarsonate (Ars) were a gift from T. Manser (Jefferson University, Philadelphia, PA). Mature B cells were purified from spleens, as described previously (11). J558L is a murine plasmacytoma cell line. J558L-μm3 is a variant that has been engineered to express the BCR (12).
Abs and other reagents
B cells were stimulated using goat anti-mouse IgG F(ab')2 (anti-Ig) (25 μg/ml) purchased from Jackson ImmunoResearch Laboratories. Ars-specific B cells were treated with Ars-keyhole limpet hemocyanin (KLH) (20 μg/ml) for specified amounts of time and stained for flow cytometry, as described below. Brefeldin A (BFA; Epicentre Technologies) treatment (10 μg/ml) was done simultaneously with anti-BCR stimulation at 37°C. The following Abs were used for immunofluorescence microscopy: hamster anti-mouse Ig FITC (Serotec), hamster anti-mouse Ig FITC (Southern Biotechnology Associates), anti-mouse IgM biotin (Southern Biotechnology Associates), and streptavidin Alexa Fluor 633 (Molecular Probes). Cy5- and Cy3-conjugated F(ab')2 of donkey anti-goat IgG and FITC-conjugated F(ab')2 of goat anti-mouse IgM, μ-chain specific, were purchased from Jackson ImmunoResearch Laboratories. For flow cytometric analysis, the following Abs were used: B220 allophycocyanin (BD Pharmingen), R-phycoerythrin (RPE)-conjugated rat anti-mouse IgD (Southern Biotechnology Associates), RPE-conjugated rat anti-mouse (Southern Biotechnology Associates), RPE-conjugated mouse anti-human Ig (BD Pharmingen), and RPE-conjugated anti-mouse I-Ab (BD Pharmingen).
B cell stimulation and staining for fluorescence microscopy and flow cytometry
After washing twice in cold PBS/2% FCS, cells (1–2 x 106) were warmed to 37°C for 20 min. For BCR aggregation, prewarmed anti-mouse IgG (H and L chain) F(ab')2 anti-Ig was added to purified B cells, and the reaction was stopped at indicated times by adding ice-cold PBS/0.2% BSA/0.02% sodium azide (staining buffer). Samples stimulated for 0 min were incubated at 37°C for 20 min and washed with staining buffer, as for the other samples, and then incubated with anti-Ig on ice for 20 min. All samples were washed twice with staining buffer and then incubated with primary Ab on ice for 30 min and washed twice, followed by incubation with secondary Ab for 20 min on ice, if needed. Following secondary Ab incubation, samples were washed twice and fixed in 4% formaldehyde. The cells were then loaded onto CytoSpin (Thermo Shandon) for adherence onto glass slides and mounted with ProLong Anti-fade solution (Molecular Probes). Samples were examined using a Zeiss Axiovert 200M inverted epifluorescence microscope equipped with a Sensicam QE high performance camera. Images were captured and analyzed using Slidebook image analysis software (Intelligent Imaging Innovations). The "No Neighbors" deconvolution module of the software was used to remove the out-of-focus light. Specified samples were examined using a Zeiss LSM510 META laser-scanning confocal microscope on a Zeiss Axiovert 200M inverted microscope (Biomedical Imaging Core Laboratory, University of Pennsylvania School of Medicine). FITC fluorescence was examined by exciting with an argon laser and collected with a BP 505- to 530-nm filter. Cy5 fluorescence was examined by exciting with a HeNe laser and collected with an LP 650-nm filter.
Results
Ig is retained on the surface, while sIg is internalized following BCR aggregation
The recent finding that the sIg and the Ig/Ig signaling complex can be functionally and physically uncoupled following Ag binding (9, 10) prompted us to assess how this uncoupling could affect the internalization of the BCR complex. Primary B cells were stimulated using an Ab directed against their L chains (anti-Ig) (Fig. 1A). At the indicated times, the remaining surface BCR bound by anti-Ig was detected on live cells by binding of fluorochrome-conjugated secondary Ab recognizing the stimulating Ab. As has been previously reported, one observes a marked clustering of the sIg at 5 min after receptor cross-linking. Receptor clustering is followed by internalization and loss of detectable sIg by 60 min. The level of sIg expression was quantified during this time frame by flow cytometric analysis. Because the stimulating Ab used in these experiments is specific for IgG (H and L chains), it stimulates primary IgM- and IgD-expressing B cells through their L chain, which in these splenic populations are >90% Ig. In Fig. 1B are FACS histograms depicting the relative levels of sIg expression, as measured by anti- binding at 0, 5, and 60 min after anti-Ig cross-linking at 37°C. By 60 min, 99% of the sIg has been modulated and is therefore undetectable by surface staining as compared with cells treated with anti-Ig for 0 min at 37°C (kept on ice). To also monitor individually the IgM- and IgD-containing BCR complexes, we again modulated BCR using anti-IgG F(ab')2, but this time we used Abs directed to the μ or H chain to determine the relative levels of sIg expression. As shown in Fig. 1C, consistent with the data, there is a detectable degree of modulation of both μ and H chain at 5 min of stimulation, but after 60 min of stimulation, surface and μ expression is reduced to 0.1 and 6%, respectively. It is also important to note that primary murine B cells are believed to express 5–10 times the amount of sIgD compared with IgM (13), indicating that only 1–2% of the sIg expressed on resting B cells remains on the surface after 60 min of stimulation.
FIGURE 1. sIg is internalized following BCR cross-linking. A, Primary mature B cells were stimulated at 37°C with goat anti-mouse IgG F(ab')2 for 0, 5, or 60 min. sIg was detected at indicated time points by staining with a secondary anti-goat IgG-Cy5 Ab to the stimulating Ab. Representative immunofluorescent images of sIg staining are shown. B, B cells were stained with anti-mouse PE to determine surface expression levels of sIg of all isotypes using flow cytometry. Samples stimulated for 0 min (shown as shaded gray) were cross-linked with anti-mouse IgG on ice for 20 min. C, B cells were stimulated as in B, and then stained with either anti-IgD RPE (-specific) or anti-IgM (μ-specific) Abs to determine surface expression levels of each isotype.
BCR expression was monitored until this point by detection of the sIg component (μ, , and chains). To determine the effect of anti-Ig-induced BCR aggregation on the signaling component of the complex, we also monitored levels of sIg. Purified splenic B cells were again stimulated with anti-Ig for 0, 5, and 60 min and then surface expression levels of Ig and L chain, and thus total sIg (IgM + IgD) were determined. By 5 min of anti-Ig-induced aggregation, surface expression levels of both Ig and Ig are slightly decreased compared with the 0-min time point (Fig. 2A). By 60 min at 37°C, expression level of L chain was decreased to 0.5% and remained so through 120 and 150 min (longest time point examined). Surprisingly, surface levels of Ig decreased to a much lesser degree. Although surface expression was modulated by the anti-Ig treatment to 0.5% of resting levels and approached background staining by flow cytometry, 29% of Ig was remaining on the surface. This difference in relative modulation cannot be attributed to differential FcR binding of primary or secondary Abs, because addition of the FcIII/IIR-blocking Ab 2.4G2 did not change the detected expression levels of Ig or Ig at any time point (data not shown).
FIGURE 2. Ig is retained on the surface, while sIg is internalized following BCR cross-linking. A, As in Fig. 1, primary mature B cells were treated with anti-Ig for the indicated amount of time at 37°C and then stained with anti-Ig FITC or anti-Ig PE Abs and analyzed by flow cytometry. B, B cells were treated with anti-Ig and stained with a secondary Ab (Cy5-conjugated anti-goat IgG) to visualize sIg, and FITC-conjugated Ig Ab to visualize sIg. Localization of BCR components was examined using confocal microscopy. C, B cells were treated on ice with goat anti-mouse IgG F(ab')2 for 20 min, followed by staining with anti-goat IgG conjugated to Cy3 on ice for 20 min, and then warmed up to 37°C for indicated amount of time to visualize internalized sIg. Cells were then stained for sIg and sIgM using Ig FITC and μ-bio, followed by streptavidin-Alexa Flour 633, and visualized using epifluorescence microscopy.
sIg and Ig distribution and levels were also compared using immunofluorescence microscopy at different time points after cross-linking with anti-Ig. sIg and Ig were monitored on individual cells using directly conjugated anti-Ig and a secondary Ab to the stimulating Ab (anti-goat Cy5) to detect sIg. At 0 min of anti-Ig treatment (cells kept on ice), sIg and Ig are coexpressed, and expression for both is diffuse (Fig. 2B, top panels). Cells stimulated for 5 min show tight aggregates, when stained for both sIg and Ig, and they appear to be colocalized (middle panels). By 60 min of stimulation, there is little if any detectable sIg, while Ig is clearly expressed and has returned to a diffuse surface distribution (bottom panels).
In the experiments described above, addition of a fluorochrome-labeled secondary reagent to B cells (anti-goat Cy5) that had been stimulated with goat anti-mouse IgG F(ab')2 at 37°C only allowed for the detection of Ig remaining on the cell surface. To directly observe Ig internalization, BCR was cross-linked with a goat anti-mouse Ig Ab and stained with a secondary fluorescently labeled anti-goat Ab on ice, and then warmed up to 37°C (prelabeled Ig) (Fig. 2C). Following the stimulation at 37°C, cells were stained with directly conjugated anti-Ig and anti-μ Abs to detect sIg and sIg, respectively. At 0 min (maintained on ice), each shows diffuse distribution, and prelabeled Ig and sIg detected at the surface post-anti-Ig treatment are colocalized. By 5 min of stimulation, the tight aggregations of BCR components can be observed, as shown by colocalization of the sIg pools and Ig at the same pole of the cell. By 60 min of stimulation, prelabeled Ig appears internalized, and there is only minimal staining for the surface form of Ig. Again, Ig surface expression is clearly maintained despite modulation of most, if not all, sIg.
All of the above studies used the same mAb directed to the murine Ig ectodomain and, therefore, their interpretation depends upon the specificity of that reagent for Ig. To corroborate these studies, we isolated splenic B cells from a transgenic mouse that coexpresses murine Ig (mIg) and human Ig (hIg). This strategy allowed us to use an additional reagent to detect Ig (14, 15). As before, purified splenic B cells were stimulated with anti-Ig for 0, 5, and 60 min, and levels of surface L chain, hIg, and mIg were then determined by flow cytometry (Fig. 3A). Both mIg and hIg were detected on the surface at 60-min stimulation, despite nearly complete modulation of sIg. Fig. 3B shows quantification of surface expression levels of these BCR components based on median fluorescence intensity (with median fluorescence intensity of the isotype control subtracted from each number), as determined by flow cytometry. In this experiment, hIg and mIg are maintained at 18 and 23% of their resting levels, respectively, whereas sIg levels are decreased to <1%.
FIGURE 3. sIg-independent retention of the signaling component is not limited to cross-linking Abs and can also be detected using a transgenic mouse model. A, As in Fig. 1, B cells were treated with anti-Ig for indicated amount of time and then stained with anti- PE, anti-murine Ig FITC, or anti-human Ig PE. B, Quantification of surface expression levels of BCR components as indicated by median fluorescence intensity at each time point. C, J558L and its μM3 variant were stained with anti-mouse Ig Ab to show specificity of Ig binding. J558L is shown in black, μm3 in shaded gray, and isotype control in black dotted line. D, B cells were isolated from mice expressing BCR specific for Ars, treated with Ars-KLH for indicated amounts of time at 37°C, and then stained with anti-Ig and anti-Ig Abs at each time point. Relative expression of IgB (left) and IgK (right) was determined based on the median fluorescence intensity.
To rule out nonspecific binding of the anti-murine Ig Ab, we used J558L B cell line, which does not express Ig or any other components of the BCR on its surface. Indeed, studies using this cell line did not show any surface staining when stained with the anti-Ig Ab (Fig. 3C). This finding was not the case, however, with the variant of this cell line, J558L μm3, which has been engineered to express all the components of the BCR, including Ig, on its surface.
The experimental system used in the above studies uses B cell stimulation with a high affinity cross-linking Ab to the Ig. Such a strong interaction is unlikely to be very common in the physiological setting of the immune response, and therefore we also examined how the BCR components respond to the aggregation resulting from a specific Ag binding Ag-specific B cells. For this purpose, we used B cells from Ig transgenic mice that express BCRs specific for p-azophenylarsonate (Ars) (16, 17). B cells from these mice were stimulated in vitro with Ars-KLH conjugates at 37°C for 0–90 min, and the levels of sIg and sIg at each time point were quantified as in previous experiments. Although the overall internalization of sIg is somewhat lower in this system (most likely due to the fact that Ars-KLH does not result in the same degree of cross-linking as the anti-BCR Abs), the levels of Ig remaining on the surface are still significantly higher (66%) than the levels of Ig (13%) (Fig. 3D).
Expression of MHC class II is not required for the retention of Ig on the cell surface
The proposed secondary structure of the Ig transmembrane region argues for a thermodynamically unstable association of the Ig/Ig complex with the hydrophobic milieu of the plasma membrane, unless it is associated with the sIg H chain (18). The ability of Ig to remain on the surface after complete or nearly complete internalization of sIg would thus be predicted to be thermodynamically unstable, and therefore raised the possibility of a complex with another cell surface protein. Lang et al. (10) recently reported the association of Ig/Ig heterodimer with MHC class II on the surface, subsequent to BCR internalization.
To investigate a requirement for MHC class II molecules for the retention of Ig in the absence of sIg, we assessed the ability to retain Ig on the B cells isolated from C57BL/6J A-deficient mice. These mice lack surface expression of all forms of MHC class II (both I-A and I-E) (19) (Fig. 4A). B cells purified from these and wild-type C57BL/6J (B6) mice were stimulated with anti-Ig, as described before, for 0, 5, and 60 min. The cells were then analyzed for IgD, IgM, and Ig expression by flow cytometry (Fig. 4B). The Ig, μ, and expression patterns in B6 mice were indistinguishable from wild-type BALB/c mice (data not shown). By 5 min of stimulation with anti-Ig, there is a slight decrease in the expression levels of each protein, possibly due to the formation of tight caps and a decrease in the exposed epitopes that could bind the detecting Ab. However, by 60 min of stimulation, the modulation of IgM and IgD from the surface is nearly complete, resulting in expression levels approaching the level of the isotype control. In contrast, Ig levels are only slightly decreased, indicating that the majority of Ig is retained on the surface and that this retention does not depend upon its association with MHC class II.
FIGURE 4. Surface expression of MHC class II molecules is not required for the retention of Ig on the surface following sIg internalization. A, Purified B cells from wild-type and A-deficient mice were stained with anti-mouse I-Ab to verify lack of MHC class II expression. B, Splenic B cells were isolated from A-deficient mice and treated with anti-Ig for indicated times, as in previous figures. Cells were then stained with anti-IgM FITC, anti-IgD PE, and anti-Ig biotin, followed by streptavidin-Red 670.
To follow the dynamics of the BCR components trafficking more closely, we examined the expression levels of both Ig and Ig at additional time points following anti-Ig stimulation. As shown in Fig. 5A, there is an initial decrease of both Ig and Ig surface expression levels, resulting partly from internalization of both components and partly from the formation of a tight BCR cap. At the later stages of the time course, Ig expression levels continue to decrease to practically undetectable levels, while the expression levels of sIg start to increase. This increase can be attributed partially to the relief of the steric inhibition, due to the removal of the BCR cap from the cell surface. In support of this argument, in the Bal 17 cell line model that shows the same general behavior of the BCR components, but does not exhibit the tight BCR cap formation, this initial dip in sIg expression and the gradual increase that follows at later time points are much less apparent (data not shown).
FIGURE 5. Internalization dynamics and rates of both BCR components are similar in wild-type and MHC class II-deficient B cells. Primary B cells isolated from wild-type (A) and A-deficient (B) mice were treated with anti-Ig, as described in previous figures, for indicated amounts of time, and then stained with anti-Ig and Ig Abs. The percentage remaining on the surface was calculated based on median fluorescence intensity at each time point relative to that at 0-min time point.
Some of the sIg may be trafficking to the cell surface from the intracellular compartments. Although we were able to rule out MHC class II as a requirement for Ig retention, it is still possible that MHC class II plays a role in this process when present and that its absence results in Ig retention using an alternative pathway. We thus looked at the patterns of Ig trafficking in B cells from the MHC class II-deficient mice to see whether they are similar to their wild-type counterparts. Indeed, the internalization dynamics of both Ig and Ig in the MHC class II-deficient B cells are very similar to the wild-type B cells (Fig. 5B), indicating that MHC class II molecules are not responsible for either transporting Ig to the surface or keeping it there.
To assess how much of the surface-expressed Ig was coming from the intracellular compartments, we used a pharmacologic agent, BFA, to inhibit traffic to the cell surface. BFA is believed to inhibit both the endoplasmic reticulum to Golgi traffic as well as recycling of the endocytosed molecules back to the cell surface, while not affecting internalization of the surface receptors (20, 21). The concentration of BFA used in these experiments was determined based on its ability to block sIg recycling to the surface following washing out of anti-Ig (Fig. 6A). B cells were incubated with anti-Ig at 37°C for 30 min, which normally results in internalization of sIg. At that point, anti-Ig was washed out and cells were reincubated at 37°C to allow for re-expression of sIg. Addition of BFA during the reincubation period resulted in nearly complete inhibition of sIg re-expression, thus demonstrating the ability of BFA to inhibit trafficking of the BCR components to the cell surface. Next, B cells were treated with BFA during the anti-Ig stimulations, and levels of Ig were quantified as before. As shown in Fig. 6B, BFA treatment resulted in a 30% decrease in the amount of Ig retained on the surface. This result indicates that some of the surface-expressed Ig is the result of new expression. However, the majority of Ig seen at the surface after sIg internalization results from the original surface-expressed pool present before BCR aggregation.
FIGURE 6. Majority of Ig retained on the cell surface is not being directed there from the intracellular compartments. A, Primary B cells were treated with anti-Ig for 30 min, washed, and then reincubated at 37°C for indicated time. Another set of samples was treated with BFA during the reincubation period to inhibit trafficking to the cell surface. Relative expression levels of sIg at each time point were determined by staining with anti-Ig Ab. B, Primary B cells were treated with anti-Ig and anti-Ig plus BFA at concentration shown in A (10 μg/ml), and the relative surface expression of Ig was determined at each time point by staining with anti-Ig Ab.
Discussion
Structurally, the two components of the BCR are ideally designed for their individual functions: sIg for interacting with the Ag, and the Ig/Ig heterodimer for transmitting the signals. Assembly into a receptor complex allows the BCR to effectively perform both functions. Our results show that by 60 min of anti-Ig cross-linking, approximately one-third of the initially expressed Ig is still detectable on the surface, while at the same time <1% of the sIg portion of the receptor is surface expressed. These data led us to conclude that at some point before receptor internalization, the complex disassociates, allowing sIg with the bound ligand to be internalized and leaving the signaling complex (as shown by Ig expression) on the surface. Alternatively, the complete BCR complex is internalized, and subsequently, new Ig is released to the cell surface. Based on our BFA studies and the fact that we did not observe a complete loss of Ig that might indicate prior internalization before re-expression, we conclude that new sIg expression is only making a small contribution to the overall levels of sIg at this point of B cell activation. Therefore, the most likely scenario is that the Ig that remains on the cell surface represents molecules that together with Ig are released following BCR aggregation.
A study by Vilen et al. (9) showed that upon binding of low to moderate affinity Ag to the BCR, the Ig/Ig complex becomes destabilized, and this destabilization is coincident with an inability of the BCR to signal in response to Ag. Our data support and extend this study by confirming the ability of the receptor to dissociate following ligand binding, and demonstrating that this destabilization is associated with the separation of the Ag-binding and signaling components of the BCR. In support of the ability of the BCR complex to dissociate in response to antigenic stimulation, Lang et al. (10) have demonstrated that under some conditions, the Ig/Ig heterodimer is transferred from the BCR to MHC class II. Our studies do not support a requirement for MHC class II in maintaining sIg expression. Furthermore, our results argue that the dissociation and exchange of Ig complexes from sIg occur at the plasma membrane rather than in intracellular compartments, as was indicated in these prior studies (10).
The conclusion that as much as 30% of Ig (and thus presumably the Ig/Ig heterodimer) remains on the surface despite the nearly complete absence of sIg implies that the two components of the BCR can be expressed independently following anti-Ig stimulation. The current model of BCR structure, however, predicts that due to the polar residues in their transmembrane domains, individual BCR components are not thermodynamically stable in the hydrophobic environment of plasma membrane, unless they are expressed as part of the BCR complex. This model was based in part on theoretical modeling of the BCR structure as well as on the fact that past attempts to express individual BCR components on the cell surface have failed, unless specific regions in the transmembrane domains of these molecules were mutated (22). Our study raises the possibility that the Ig/Ig heterodimer needs to be in the complex with sIg only for the initial transport to the plasma membrane. Once at the membrane, the lipid environment or other membrane constituents may lessen the thermodynamic requirements for Ig association. Another possibility is that BCR aggregation results in a conformational change in one or both of the BCR components, thus allowing for sIg-independent expression of Ig/Ig heterodimer. At this time, we can only speculate on the role of the Ig complexes that remain on the surface after Ag-sIg internalization. Retention of the signaling component of the BCR on the surface may be necessary for maintaining the activation state of the cell. In addition, surface expression of the BCR is required for survival of B cells in vivo, and thus, the retained Ig may play a role in providing basal signals for survival of B cells that have modulated their sIg for Ag processing (4, 23). It will be important in further studies to establish the role of non-Ig-associated Ig in B cell physiology.
Acknowledgments
We thank Dr. Isabela Cajiao for providing hIg-expressing transgenic mice, Xiaohe Liu for providing Ars-specific Ig-expressing mice, Alison Man for breeding A-deficient mice, Dr. Xinyu Zhao at the core imaging facility for the technical assistance with the confocal microscopy, and Dr. Leslie King for critical reading of the manuscript. We also thank Justina Standalick for editorial assistance.
Footnotes
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 This work was supported by National Institutes of Health Grants AI 322592, CA 093615, and AI 43620 to J.G.M. M.K. was supported by a training grant from the National Cancer Institute.
2 Address correspondence and reprint requests to Dr. John G. Monroe, 311 BRB II/III, 421 Curie Boulevard, Philadelphia, PA 19104. E-mail address: monroej@mail.med.upenn.edu
3 Abbreviations used in this paper: sIg, surface Ig; Ars, p-azophenylarsonate; BFA, brefeldin A; hIg, human Ig; KLH, keyhole limpet hemocyanin; mIg, murine Ig; RPE, R-phycoerythrin.
Received for publication March 12, 2004. Accepted for publication November 24, 2004.
References
Williams, G. T., C. J. Peaker, K. J. Patel, M. S. Neuberger. 1994. The / sheath and its cytoplasmic tyrosines are required for signaling by the B-cell antigen receptor but not for capping or for serine/threonine-kinase recruitment. Proc. Natl. Acad. Sci. USA 91:474.
Matsuuchi, L., M. R. Gold, A. Travis, R. Grosschedl, A. L. DeFranco, R. B. Kelly. 1992. The membrane IgM-associated proteins MB-1 and Ig- are sufficient to promote surface expression of a partially functional B-cell antigen receptor in a nonlymphoid cell line. Proc. Natl. Acad. Sci. USA 89:3404.
Fuentes-Panana, E. M., G. Bannish, J. G. Monroe. 2004. Basal B-cell receptor signaling in B lymphocytes: mechanisms of regulation and role in positive selection, differentiation, and peripheral survival. Immunol. Rev. 197:26.
Lam, K. P., R. Kuhn, K. Rajewsky. 1997. In vivo ablation of surface immunoglobulin on mature B cells by inducible gene targeting results in rapid cell death. Cell 90:1073
Kurosaki, T.. 1997. Molecular mechanisms in B cell antigen receptor signaling. Curr. Opin. Immunol. 9:309.
Young-Karlan, B. R., R. F. Ashman. 1981. Order of events leading to surface immunoglobulin capping: analysis of a transmembrane signal. J. Immunol. 127:1177
Clark, M. R., D. Massenburg, M. Zhang, K. Siemasko. 2003. Molecular mechanisms of B cell antigen receptor trafficking. Ann. NY Acad. Sci. 987:26.
Stoddart, A., M. L. Dykstra, B. K. Brown, W. Song, S. K. Pierce, F. M. Brodsky. 2002. Lipid rafts unite signaling cascades with clathrin to regulate BCR internalization. Immunity 17:451
Vilen, B. J., T. Nakamura, J. C. Cambier. 1999. Antigen-stimulated dissociation of BCR mIg from Ig-/Ig-: implications for receptor desensitization. Immunity 10:239.
Lang, P., J. C. Stolpa, B. A. Freiberg, F. Crawford, J. Kappler, A. Kupfer, J. C. Cambier. 2001. TCR-induced transmembrane signaling by peptide/MHC class II via associated Ig-/ dimers. Science 291:1537.
Norvell, A., J. G. Monroe. 1996. Acquisition of surface IgD fails to protect from tolerance-induction: both surface IgM- and surface IgD-mediated signals induce apoptosis of immature murine B lymphocytes. J. Immunol. 156:1328.
Justement, L. B., K. S. Campbell, N. C. Chien, J. C. Cambier. 1991. Regulation of B cell antigen receptor signal transduction and phosphorylation by CD45. Science 252:1839.
Havran, W. L., D. L. DiGiusto, J. C. Cambier. 1984. mIgM:mIgD ratios on B cells: mean mIgD expression exceeds mIgM by 10-fold on most splenic B cells. J. Immunol. 132:1712.
Su, Y., S. A. Liebhaber, N. E. Cooke. 2000. The human growth hormone gene cluster locus control region supports position-independent pituitary- and placenta-specific expression in the transgenic mouse. J. Biol. Chem. 275:7902.
Cajiao, I., A. Zhang, E. J. Yoo, N. E. Cooke, S. A. Liebhaber. 2004. Bystander gene activation by a locus control region. EMBO J. 23:3854.
Notidis, E., L. Heltemes, T. Manser. 2002. Dominant, hierarchical induction of peripheral tolerance during foreign antigen-driven B cell development. Immunity 17:317.
Snyder, C. M., X. Zhang, L. J. Wysocki. 2002. Negligible class II MHC presentation of B cell receptor-derived peptides by high density resting B cells. J. Immunol. 168:3865.
Michnoff, C. H., V. S. Parikh, D. L. Lelsz, P. W. Tucker. 1994. Mutations within the NH2-terminal transmembrane domain of membrane immunoglobulin (Ig) M alters Ig and Ig association and signal transduction. J. Biol. Chem. 269:24237.(Marina Kremyanskaya and J)
In order for humoral immune responses to develop, B cells must be able to recognize, bind, and internalize Ags. These functions are performed by the BCR, which is also responsible for initiating and transducing activation signals necessary for B cell proliferation and differentiation. We have examined surface expression patterns of individual components of the BCR following anti-Ig- and Ag-induced aggregation. Specifically, the localization and expression levels of the Ag-binding component, surface Ig (sIg), and the Ig component of the Ig/Ig signaling unit were investigated to determine their individual participation in the internalization and signal transduction. Using primary murine B cells, we found that while >95% of the sIg is internalized following anti-Ig-induced aggregation, 20–30% of Ig remains on the surface. These results suggest that sIg and Ig may function independently following the initial stages of signal transduction.
Introduction
Development of a humoral immune response requires the B cell to receive two separate signals. The first is the result of the Ag binding the clonally expressed BCR, and the second is provided as a consequence of cognate interaction with an activated CD4 Th cell or, in some cases, signaling through TLRs. In order for productive interaction with MHC-restricted Ag-specific T cells to occur, the BCR itself has to perform two essential functions: communicate the activating signal to the inside of the cell, and internalize the Ag for processing and presentation.
The BCR is a multisubunit complex composed of surface Ig (sIg)3 noncovalently associated with the Ig/Ig heterodimer. sIg functions as the Ag recognition and binding component, while Ig/Ig transduces the Ag-induced signal (1). The association of sIg with the Ig/Ig heterodimer is believed to be necessary for trafficking the BCR to the plasma membrane (2). Although different forms of BCR are expressed throughout B cell development, the Ig/Ig heterodimer is an essential component of all of them (3). Presence of a specific form of BCR is required to proceed to the next stage of development, and in mature cells surface expression of BCR is necessary for survival (4).
BCR aggregation as a consequence of Ag-sIg binding and oligomerization results in tyrosine phosphorylation of specific protein-signaling motifs called ITAMs present in the cytoplasmic regions of both Ig and Ig. ITAM phosphorylation initiates recruitment of cytoplasmic signaling proteins and formation of a stable signaling complex that functions to link Ag binding to B cell activation. Activation refers to a complex process associated with generation of cytoplasmic second messengers, elevations in RNA and protein synthesis, changes in gene expression, and initiation of programs leading to the ability to elicit and receive secondary signals from T cells, culminating with the growth and proliferation of the B cell (5).
Shortly after the ligand-induced aggregation of the BCR, the stimulated B cell concentrates the majority of its BCR complexes to a single pole on the cell surface in a process commonly referred to as capping (6). Following the formation of these tight unipolar aggregated complexes, the BCR together with the bound Ag is internalized (7). Recent studies have indicated that in mature stage B cells, internalization occurs in compartments associated with clathrin and organized lipid rafts (8). Other studies also indicate that under some conditions of Ag binding, there is a functional and physical uncoupling of sIg and the Ig/Ig signaling complex (9, 10). It is unclear how this uncoupling affects the internalization of the BCR complex. In this study, we monitored the dynamics of surface expression and subsequent internalization of the signaling and Ag recognition components of the BCR. We report in this work that following BCR cross-linking, a significant amount of the Ig pool (presumably together with Ig) remains on the surface independent of sIg internalization. This retention occurs in response to both anti-Ig stimulation of polyclonal splenic B cells as well as Ag stimulation of anti-arsonate-specific Ig transgenic B cells.
Materials and Methods
Cell preparation and cell lines
Female BALB/c mice were used for all experiments, unless specified. Transgenic mice expressing human Ig were obtained from S. Liebhaber (University of Pennsylvania). A-deficient mice were provided by T. Laufer (University of Pennsylvania) and also were purchased from Taconic Farms. Transgenic mice expressing BCRs specific for p-azophenylarsonate (Ars) were a gift from T. Manser (Jefferson University, Philadelphia, PA). Mature B cells were purified from spleens, as described previously (11). J558L is a murine plasmacytoma cell line. J558L-μm3 is a variant that has been engineered to express the BCR (12).
Abs and other reagents
B cells were stimulated using goat anti-mouse IgG F(ab')2 (anti-Ig) (25 μg/ml) purchased from Jackson ImmunoResearch Laboratories. Ars-specific B cells were treated with Ars-keyhole limpet hemocyanin (KLH) (20 μg/ml) for specified amounts of time and stained for flow cytometry, as described below. Brefeldin A (BFA; Epicentre Technologies) treatment (10 μg/ml) was done simultaneously with anti-BCR stimulation at 37°C. The following Abs were used for immunofluorescence microscopy: hamster anti-mouse Ig FITC (Serotec), hamster anti-mouse Ig FITC (Southern Biotechnology Associates), anti-mouse IgM biotin (Southern Biotechnology Associates), and streptavidin Alexa Fluor 633 (Molecular Probes). Cy5- and Cy3-conjugated F(ab')2 of donkey anti-goat IgG and FITC-conjugated F(ab')2 of goat anti-mouse IgM, μ-chain specific, were purchased from Jackson ImmunoResearch Laboratories. For flow cytometric analysis, the following Abs were used: B220 allophycocyanin (BD Pharmingen), R-phycoerythrin (RPE)-conjugated rat anti-mouse IgD (Southern Biotechnology Associates), RPE-conjugated rat anti-mouse (Southern Biotechnology Associates), RPE-conjugated mouse anti-human Ig (BD Pharmingen), and RPE-conjugated anti-mouse I-Ab (BD Pharmingen).
B cell stimulation and staining for fluorescence microscopy and flow cytometry
After washing twice in cold PBS/2% FCS, cells (1–2 x 106) were warmed to 37°C for 20 min. For BCR aggregation, prewarmed anti-mouse IgG (H and L chain) F(ab')2 anti-Ig was added to purified B cells, and the reaction was stopped at indicated times by adding ice-cold PBS/0.2% BSA/0.02% sodium azide (staining buffer). Samples stimulated for 0 min were incubated at 37°C for 20 min and washed with staining buffer, as for the other samples, and then incubated with anti-Ig on ice for 20 min. All samples were washed twice with staining buffer and then incubated with primary Ab on ice for 30 min and washed twice, followed by incubation with secondary Ab for 20 min on ice, if needed. Following secondary Ab incubation, samples were washed twice and fixed in 4% formaldehyde. The cells were then loaded onto CytoSpin (Thermo Shandon) for adherence onto glass slides and mounted with ProLong Anti-fade solution (Molecular Probes). Samples were examined using a Zeiss Axiovert 200M inverted epifluorescence microscope equipped with a Sensicam QE high performance camera. Images were captured and analyzed using Slidebook image analysis software (Intelligent Imaging Innovations). The "No Neighbors" deconvolution module of the software was used to remove the out-of-focus light. Specified samples were examined using a Zeiss LSM510 META laser-scanning confocal microscope on a Zeiss Axiovert 200M inverted microscope (Biomedical Imaging Core Laboratory, University of Pennsylvania School of Medicine). FITC fluorescence was examined by exciting with an argon laser and collected with a BP 505- to 530-nm filter. Cy5 fluorescence was examined by exciting with a HeNe laser and collected with an LP 650-nm filter.
Results
Ig is retained on the surface, while sIg is internalized following BCR aggregation
The recent finding that the sIg and the Ig/Ig signaling complex can be functionally and physically uncoupled following Ag binding (9, 10) prompted us to assess how this uncoupling could affect the internalization of the BCR complex. Primary B cells were stimulated using an Ab directed against their L chains (anti-Ig) (Fig. 1A). At the indicated times, the remaining surface BCR bound by anti-Ig was detected on live cells by binding of fluorochrome-conjugated secondary Ab recognizing the stimulating Ab. As has been previously reported, one observes a marked clustering of the sIg at 5 min after receptor cross-linking. Receptor clustering is followed by internalization and loss of detectable sIg by 60 min. The level of sIg expression was quantified during this time frame by flow cytometric analysis. Because the stimulating Ab used in these experiments is specific for IgG (H and L chains), it stimulates primary IgM- and IgD-expressing B cells through their L chain, which in these splenic populations are >90% Ig. In Fig. 1B are FACS histograms depicting the relative levels of sIg expression, as measured by anti- binding at 0, 5, and 60 min after anti-Ig cross-linking at 37°C. By 60 min, 99% of the sIg has been modulated and is therefore undetectable by surface staining as compared with cells treated with anti-Ig for 0 min at 37°C (kept on ice). To also monitor individually the IgM- and IgD-containing BCR complexes, we again modulated BCR using anti-IgG F(ab')2, but this time we used Abs directed to the μ or H chain to determine the relative levels of sIg expression. As shown in Fig. 1C, consistent with the data, there is a detectable degree of modulation of both μ and H chain at 5 min of stimulation, but after 60 min of stimulation, surface and μ expression is reduced to 0.1 and 6%, respectively. It is also important to note that primary murine B cells are believed to express 5–10 times the amount of sIgD compared with IgM (13), indicating that only 1–2% of the sIg expressed on resting B cells remains on the surface after 60 min of stimulation.
FIGURE 1. sIg is internalized following BCR cross-linking. A, Primary mature B cells were stimulated at 37°C with goat anti-mouse IgG F(ab')2 for 0, 5, or 60 min. sIg was detected at indicated time points by staining with a secondary anti-goat IgG-Cy5 Ab to the stimulating Ab. Representative immunofluorescent images of sIg staining are shown. B, B cells were stained with anti-mouse PE to determine surface expression levels of sIg of all isotypes using flow cytometry. Samples stimulated for 0 min (shown as shaded gray) were cross-linked with anti-mouse IgG on ice for 20 min. C, B cells were stimulated as in B, and then stained with either anti-IgD RPE (-specific) or anti-IgM (μ-specific) Abs to determine surface expression levels of each isotype.
BCR expression was monitored until this point by detection of the sIg component (μ, , and chains). To determine the effect of anti-Ig-induced BCR aggregation on the signaling component of the complex, we also monitored levels of sIg. Purified splenic B cells were again stimulated with anti-Ig for 0, 5, and 60 min and then surface expression levels of Ig and L chain, and thus total sIg (IgM + IgD) were determined. By 5 min of anti-Ig-induced aggregation, surface expression levels of both Ig and Ig are slightly decreased compared with the 0-min time point (Fig. 2A). By 60 min at 37°C, expression level of L chain was decreased to 0.5% and remained so through 120 and 150 min (longest time point examined). Surprisingly, surface levels of Ig decreased to a much lesser degree. Although surface expression was modulated by the anti-Ig treatment to 0.5% of resting levels and approached background staining by flow cytometry, 29% of Ig was remaining on the surface. This difference in relative modulation cannot be attributed to differential FcR binding of primary or secondary Abs, because addition of the FcIII/IIR-blocking Ab 2.4G2 did not change the detected expression levels of Ig or Ig at any time point (data not shown).
FIGURE 2. Ig is retained on the surface, while sIg is internalized following BCR cross-linking. A, As in Fig. 1, primary mature B cells were treated with anti-Ig for the indicated amount of time at 37°C and then stained with anti-Ig FITC or anti-Ig PE Abs and analyzed by flow cytometry. B, B cells were treated with anti-Ig and stained with a secondary Ab (Cy5-conjugated anti-goat IgG) to visualize sIg, and FITC-conjugated Ig Ab to visualize sIg. Localization of BCR components was examined using confocal microscopy. C, B cells were treated on ice with goat anti-mouse IgG F(ab')2 for 20 min, followed by staining with anti-goat IgG conjugated to Cy3 on ice for 20 min, and then warmed up to 37°C for indicated amount of time to visualize internalized sIg. Cells were then stained for sIg and sIgM using Ig FITC and μ-bio, followed by streptavidin-Alexa Flour 633, and visualized using epifluorescence microscopy.
sIg and Ig distribution and levels were also compared using immunofluorescence microscopy at different time points after cross-linking with anti-Ig. sIg and Ig were monitored on individual cells using directly conjugated anti-Ig and a secondary Ab to the stimulating Ab (anti-goat Cy5) to detect sIg. At 0 min of anti-Ig treatment (cells kept on ice), sIg and Ig are coexpressed, and expression for both is diffuse (Fig. 2B, top panels). Cells stimulated for 5 min show tight aggregates, when stained for both sIg and Ig, and they appear to be colocalized (middle panels). By 60 min of stimulation, there is little if any detectable sIg, while Ig is clearly expressed and has returned to a diffuse surface distribution (bottom panels).
In the experiments described above, addition of a fluorochrome-labeled secondary reagent to B cells (anti-goat Cy5) that had been stimulated with goat anti-mouse IgG F(ab')2 at 37°C only allowed for the detection of Ig remaining on the cell surface. To directly observe Ig internalization, BCR was cross-linked with a goat anti-mouse Ig Ab and stained with a secondary fluorescently labeled anti-goat Ab on ice, and then warmed up to 37°C (prelabeled Ig) (Fig. 2C). Following the stimulation at 37°C, cells were stained with directly conjugated anti-Ig and anti-μ Abs to detect sIg and sIg, respectively. At 0 min (maintained on ice), each shows diffuse distribution, and prelabeled Ig and sIg detected at the surface post-anti-Ig treatment are colocalized. By 5 min of stimulation, the tight aggregations of BCR components can be observed, as shown by colocalization of the sIg pools and Ig at the same pole of the cell. By 60 min of stimulation, prelabeled Ig appears internalized, and there is only minimal staining for the surface form of Ig. Again, Ig surface expression is clearly maintained despite modulation of most, if not all, sIg.
All of the above studies used the same mAb directed to the murine Ig ectodomain and, therefore, their interpretation depends upon the specificity of that reagent for Ig. To corroborate these studies, we isolated splenic B cells from a transgenic mouse that coexpresses murine Ig (mIg) and human Ig (hIg). This strategy allowed us to use an additional reagent to detect Ig (14, 15). As before, purified splenic B cells were stimulated with anti-Ig for 0, 5, and 60 min, and levels of surface L chain, hIg, and mIg were then determined by flow cytometry (Fig. 3A). Both mIg and hIg were detected on the surface at 60-min stimulation, despite nearly complete modulation of sIg. Fig. 3B shows quantification of surface expression levels of these BCR components based on median fluorescence intensity (with median fluorescence intensity of the isotype control subtracted from each number), as determined by flow cytometry. In this experiment, hIg and mIg are maintained at 18 and 23% of their resting levels, respectively, whereas sIg levels are decreased to <1%.
FIGURE 3. sIg-independent retention of the signaling component is not limited to cross-linking Abs and can also be detected using a transgenic mouse model. A, As in Fig. 1, B cells were treated with anti-Ig for indicated amount of time and then stained with anti- PE, anti-murine Ig FITC, or anti-human Ig PE. B, Quantification of surface expression levels of BCR components as indicated by median fluorescence intensity at each time point. C, J558L and its μM3 variant were stained with anti-mouse Ig Ab to show specificity of Ig binding. J558L is shown in black, μm3 in shaded gray, and isotype control in black dotted line. D, B cells were isolated from mice expressing BCR specific for Ars, treated with Ars-KLH for indicated amounts of time at 37°C, and then stained with anti-Ig and anti-Ig Abs at each time point. Relative expression of IgB (left) and IgK (right) was determined based on the median fluorescence intensity.
To rule out nonspecific binding of the anti-murine Ig Ab, we used J558L B cell line, which does not express Ig or any other components of the BCR on its surface. Indeed, studies using this cell line did not show any surface staining when stained with the anti-Ig Ab (Fig. 3C). This finding was not the case, however, with the variant of this cell line, J558L μm3, which has been engineered to express all the components of the BCR, including Ig, on its surface.
The experimental system used in the above studies uses B cell stimulation with a high affinity cross-linking Ab to the Ig. Such a strong interaction is unlikely to be very common in the physiological setting of the immune response, and therefore we also examined how the BCR components respond to the aggregation resulting from a specific Ag binding Ag-specific B cells. For this purpose, we used B cells from Ig transgenic mice that express BCRs specific for p-azophenylarsonate (Ars) (16, 17). B cells from these mice were stimulated in vitro with Ars-KLH conjugates at 37°C for 0–90 min, and the levels of sIg and sIg at each time point were quantified as in previous experiments. Although the overall internalization of sIg is somewhat lower in this system (most likely due to the fact that Ars-KLH does not result in the same degree of cross-linking as the anti-BCR Abs), the levels of Ig remaining on the surface are still significantly higher (66%) than the levels of Ig (13%) (Fig. 3D).
Expression of MHC class II is not required for the retention of Ig on the cell surface
The proposed secondary structure of the Ig transmembrane region argues for a thermodynamically unstable association of the Ig/Ig complex with the hydrophobic milieu of the plasma membrane, unless it is associated with the sIg H chain (18). The ability of Ig to remain on the surface after complete or nearly complete internalization of sIg would thus be predicted to be thermodynamically unstable, and therefore raised the possibility of a complex with another cell surface protein. Lang et al. (10) recently reported the association of Ig/Ig heterodimer with MHC class II on the surface, subsequent to BCR internalization.
To investigate a requirement for MHC class II molecules for the retention of Ig in the absence of sIg, we assessed the ability to retain Ig on the B cells isolated from C57BL/6J A-deficient mice. These mice lack surface expression of all forms of MHC class II (both I-A and I-E) (19) (Fig. 4A). B cells purified from these and wild-type C57BL/6J (B6) mice were stimulated with anti-Ig, as described before, for 0, 5, and 60 min. The cells were then analyzed for IgD, IgM, and Ig expression by flow cytometry (Fig. 4B). The Ig, μ, and expression patterns in B6 mice were indistinguishable from wild-type BALB/c mice (data not shown). By 5 min of stimulation with anti-Ig, there is a slight decrease in the expression levels of each protein, possibly due to the formation of tight caps and a decrease in the exposed epitopes that could bind the detecting Ab. However, by 60 min of stimulation, the modulation of IgM and IgD from the surface is nearly complete, resulting in expression levels approaching the level of the isotype control. In contrast, Ig levels are only slightly decreased, indicating that the majority of Ig is retained on the surface and that this retention does not depend upon its association with MHC class II.
FIGURE 4. Surface expression of MHC class II molecules is not required for the retention of Ig on the surface following sIg internalization. A, Purified B cells from wild-type and A-deficient mice were stained with anti-mouse I-Ab to verify lack of MHC class II expression. B, Splenic B cells were isolated from A-deficient mice and treated with anti-Ig for indicated times, as in previous figures. Cells were then stained with anti-IgM FITC, anti-IgD PE, and anti-Ig biotin, followed by streptavidin-Red 670.
To follow the dynamics of the BCR components trafficking more closely, we examined the expression levels of both Ig and Ig at additional time points following anti-Ig stimulation. As shown in Fig. 5A, there is an initial decrease of both Ig and Ig surface expression levels, resulting partly from internalization of both components and partly from the formation of a tight BCR cap. At the later stages of the time course, Ig expression levels continue to decrease to practically undetectable levels, while the expression levels of sIg start to increase. This increase can be attributed partially to the relief of the steric inhibition, due to the removal of the BCR cap from the cell surface. In support of this argument, in the Bal 17 cell line model that shows the same general behavior of the BCR components, but does not exhibit the tight BCR cap formation, this initial dip in sIg expression and the gradual increase that follows at later time points are much less apparent (data not shown).
FIGURE 5. Internalization dynamics and rates of both BCR components are similar in wild-type and MHC class II-deficient B cells. Primary B cells isolated from wild-type (A) and A-deficient (B) mice were treated with anti-Ig, as described in previous figures, for indicated amounts of time, and then stained with anti-Ig and Ig Abs. The percentage remaining on the surface was calculated based on median fluorescence intensity at each time point relative to that at 0-min time point.
Some of the sIg may be trafficking to the cell surface from the intracellular compartments. Although we were able to rule out MHC class II as a requirement for Ig retention, it is still possible that MHC class II plays a role in this process when present and that its absence results in Ig retention using an alternative pathway. We thus looked at the patterns of Ig trafficking in B cells from the MHC class II-deficient mice to see whether they are similar to their wild-type counterparts. Indeed, the internalization dynamics of both Ig and Ig in the MHC class II-deficient B cells are very similar to the wild-type B cells (Fig. 5B), indicating that MHC class II molecules are not responsible for either transporting Ig to the surface or keeping it there.
To assess how much of the surface-expressed Ig was coming from the intracellular compartments, we used a pharmacologic agent, BFA, to inhibit traffic to the cell surface. BFA is believed to inhibit both the endoplasmic reticulum to Golgi traffic as well as recycling of the endocytosed molecules back to the cell surface, while not affecting internalization of the surface receptors (20, 21). The concentration of BFA used in these experiments was determined based on its ability to block sIg recycling to the surface following washing out of anti-Ig (Fig. 6A). B cells were incubated with anti-Ig at 37°C for 30 min, which normally results in internalization of sIg. At that point, anti-Ig was washed out and cells were reincubated at 37°C to allow for re-expression of sIg. Addition of BFA during the reincubation period resulted in nearly complete inhibition of sIg re-expression, thus demonstrating the ability of BFA to inhibit trafficking of the BCR components to the cell surface. Next, B cells were treated with BFA during the anti-Ig stimulations, and levels of Ig were quantified as before. As shown in Fig. 6B, BFA treatment resulted in a 30% decrease in the amount of Ig retained on the surface. This result indicates that some of the surface-expressed Ig is the result of new expression. However, the majority of Ig seen at the surface after sIg internalization results from the original surface-expressed pool present before BCR aggregation.
FIGURE 6. Majority of Ig retained on the cell surface is not being directed there from the intracellular compartments. A, Primary B cells were treated with anti-Ig for 30 min, washed, and then reincubated at 37°C for indicated time. Another set of samples was treated with BFA during the reincubation period to inhibit trafficking to the cell surface. Relative expression levels of sIg at each time point were determined by staining with anti-Ig Ab. B, Primary B cells were treated with anti-Ig and anti-Ig plus BFA at concentration shown in A (10 μg/ml), and the relative surface expression of Ig was determined at each time point by staining with anti-Ig Ab.
Discussion
Structurally, the two components of the BCR are ideally designed for their individual functions: sIg for interacting with the Ag, and the Ig/Ig heterodimer for transmitting the signals. Assembly into a receptor complex allows the BCR to effectively perform both functions. Our results show that by 60 min of anti-Ig cross-linking, approximately one-third of the initially expressed Ig is still detectable on the surface, while at the same time <1% of the sIg portion of the receptor is surface expressed. These data led us to conclude that at some point before receptor internalization, the complex disassociates, allowing sIg with the bound ligand to be internalized and leaving the signaling complex (as shown by Ig expression) on the surface. Alternatively, the complete BCR complex is internalized, and subsequently, new Ig is released to the cell surface. Based on our BFA studies and the fact that we did not observe a complete loss of Ig that might indicate prior internalization before re-expression, we conclude that new sIg expression is only making a small contribution to the overall levels of sIg at this point of B cell activation. Therefore, the most likely scenario is that the Ig that remains on the cell surface represents molecules that together with Ig are released following BCR aggregation.
A study by Vilen et al. (9) showed that upon binding of low to moderate affinity Ag to the BCR, the Ig/Ig complex becomes destabilized, and this destabilization is coincident with an inability of the BCR to signal in response to Ag. Our data support and extend this study by confirming the ability of the receptor to dissociate following ligand binding, and demonstrating that this destabilization is associated with the separation of the Ag-binding and signaling components of the BCR. In support of the ability of the BCR complex to dissociate in response to antigenic stimulation, Lang et al. (10) have demonstrated that under some conditions, the Ig/Ig heterodimer is transferred from the BCR to MHC class II. Our studies do not support a requirement for MHC class II in maintaining sIg expression. Furthermore, our results argue that the dissociation and exchange of Ig complexes from sIg occur at the plasma membrane rather than in intracellular compartments, as was indicated in these prior studies (10).
The conclusion that as much as 30% of Ig (and thus presumably the Ig/Ig heterodimer) remains on the surface despite the nearly complete absence of sIg implies that the two components of the BCR can be expressed independently following anti-Ig stimulation. The current model of BCR structure, however, predicts that due to the polar residues in their transmembrane domains, individual BCR components are not thermodynamically stable in the hydrophobic environment of plasma membrane, unless they are expressed as part of the BCR complex. This model was based in part on theoretical modeling of the BCR structure as well as on the fact that past attempts to express individual BCR components on the cell surface have failed, unless specific regions in the transmembrane domains of these molecules were mutated (22). Our study raises the possibility that the Ig/Ig heterodimer needs to be in the complex with sIg only for the initial transport to the plasma membrane. Once at the membrane, the lipid environment or other membrane constituents may lessen the thermodynamic requirements for Ig association. Another possibility is that BCR aggregation results in a conformational change in one or both of the BCR components, thus allowing for sIg-independent expression of Ig/Ig heterodimer. At this time, we can only speculate on the role of the Ig complexes that remain on the surface after Ag-sIg internalization. Retention of the signaling component of the BCR on the surface may be necessary for maintaining the activation state of the cell. In addition, surface expression of the BCR is required for survival of B cells in vivo, and thus, the retained Ig may play a role in providing basal signals for survival of B cells that have modulated their sIg for Ag processing (4, 23). It will be important in further studies to establish the role of non-Ig-associated Ig in B cell physiology.
Acknowledgments
We thank Dr. Isabela Cajiao for providing hIg-expressing transgenic mice, Xiaohe Liu for providing Ars-specific Ig-expressing mice, Alison Man for breeding A-deficient mice, Dr. Xinyu Zhao at the core imaging facility for the technical assistance with the confocal microscopy, and Dr. Leslie King for critical reading of the manuscript. We also thank Justina Standalick for editorial assistance.
Footnotes
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 This work was supported by National Institutes of Health Grants AI 322592, CA 093615, and AI 43620 to J.G.M. M.K. was supported by a training grant from the National Cancer Institute.
2 Address correspondence and reprint requests to Dr. John G. Monroe, 311 BRB II/III, 421 Curie Boulevard, Philadelphia, PA 19104. E-mail address: monroej@mail.med.upenn.edu
3 Abbreviations used in this paper: sIg, surface Ig; Ars, p-azophenylarsonate; BFA, brefeldin A; hIg, human Ig; KLH, keyhole limpet hemocyanin; mIg, murine Ig; RPE, R-phycoerythrin.
Received for publication March 12, 2004. Accepted for publication November 24, 2004.
References
Williams, G. T., C. J. Peaker, K. J. Patel, M. S. Neuberger. 1994. The / sheath and its cytoplasmic tyrosines are required for signaling by the B-cell antigen receptor but not for capping or for serine/threonine-kinase recruitment. Proc. Natl. Acad. Sci. USA 91:474.
Matsuuchi, L., M. R. Gold, A. Travis, R. Grosschedl, A. L. DeFranco, R. B. Kelly. 1992. The membrane IgM-associated proteins MB-1 and Ig- are sufficient to promote surface expression of a partially functional B-cell antigen receptor in a nonlymphoid cell line. Proc. Natl. Acad. Sci. USA 89:3404.
Fuentes-Panana, E. M., G. Bannish, J. G. Monroe. 2004. Basal B-cell receptor signaling in B lymphocytes: mechanisms of regulation and role in positive selection, differentiation, and peripheral survival. Immunol. Rev. 197:26.
Lam, K. P., R. Kuhn, K. Rajewsky. 1997. In vivo ablation of surface immunoglobulin on mature B cells by inducible gene targeting results in rapid cell death. Cell 90:1073
Kurosaki, T.. 1997. Molecular mechanisms in B cell antigen receptor signaling. Curr. Opin. Immunol. 9:309.
Young-Karlan, B. R., R. F. Ashman. 1981. Order of events leading to surface immunoglobulin capping: analysis of a transmembrane signal. J. Immunol. 127:1177
Clark, M. R., D. Massenburg, M. Zhang, K. Siemasko. 2003. Molecular mechanisms of B cell antigen receptor trafficking. Ann. NY Acad. Sci. 987:26.
Stoddart, A., M. L. Dykstra, B. K. Brown, W. Song, S. K. Pierce, F. M. Brodsky. 2002. Lipid rafts unite signaling cascades with clathrin to regulate BCR internalization. Immunity 17:451
Vilen, B. J., T. Nakamura, J. C. Cambier. 1999. Antigen-stimulated dissociation of BCR mIg from Ig-/Ig-: implications for receptor desensitization. Immunity 10:239.
Lang, P., J. C. Stolpa, B. A. Freiberg, F. Crawford, J. Kappler, A. Kupfer, J. C. Cambier. 2001. TCR-induced transmembrane signaling by peptide/MHC class II via associated Ig-/ dimers. Science 291:1537.
Norvell, A., J. G. Monroe. 1996. Acquisition of surface IgD fails to protect from tolerance-induction: both surface IgM- and surface IgD-mediated signals induce apoptosis of immature murine B lymphocytes. J. Immunol. 156:1328.
Justement, L. B., K. S. Campbell, N. C. Chien, J. C. Cambier. 1991. Regulation of B cell antigen receptor signal transduction and phosphorylation by CD45. Science 252:1839.
Havran, W. L., D. L. DiGiusto, J. C. Cambier. 1984. mIgM:mIgD ratios on B cells: mean mIgD expression exceeds mIgM by 10-fold on most splenic B cells. J. Immunol. 132:1712.
Su, Y., S. A. Liebhaber, N. E. Cooke. 2000. The human growth hormone gene cluster locus control region supports position-independent pituitary- and placenta-specific expression in the transgenic mouse. J. Biol. Chem. 275:7902.
Cajiao, I., A. Zhang, E. J. Yoo, N. E. Cooke, S. A. Liebhaber. 2004. Bystander gene activation by a locus control region. EMBO J. 23:3854.
Notidis, E., L. Heltemes, T. Manser. 2002. Dominant, hierarchical induction of peripheral tolerance during foreign antigen-driven B cell development. Immunity 17:317.
Snyder, C. M., X. Zhang, L. J. Wysocki. 2002. Negligible class II MHC presentation of B cell receptor-derived peptides by high density resting B cells. J. Immunol. 168:3865.
Michnoff, C. H., V. S. Parikh, D. L. Lelsz, P. W. Tucker. 1994. Mutations within the NH2-terminal transmembrane domain of membrane immunoglobulin (Ig) M alters Ig and Ig association and signal transduction. J. Biol. Chem. 269:24237.(Marina Kremyanskaya and J)