Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Immunity. Author manuscript; available in PMC 2010 April 1.
Published in final edited form as:
PMCID: PMC2716660

Initial clonal expansion of germinal center B cells takes place at the perimeter of follicles


Current models of the germinal center (GC) response propose that following stimulation at the edges of T cell zones, pre-GC B cells directly migrate to the center of follicles and proliferate to form GCs. We followed the interrelationship of proliferation, differentiation and micro environmental locale in populations of pre-GC B cells responding to antigen. In contrast to the predictions of current models, after accumulation at the T-B interface, these cells appeared at the perimeter of follicles adjacent to the marginal zone. There, they rapidly proliferated for several days, but underwent no V gene hypermutation and little heavy chain class switching. Their chemokine receptor expression pattern indicated that these cells were sessile, yet they had begun to acquire many phenotypic characteristics of GC B cells. The expanded clones were subsequently observed in the center of follicles, suggesting that GCs are created by coalescence of B cells from this follicular perimeter response.


During T cell dependent (TD) immune responses, B cells undergo proliferation, differentiation and selection in the microenvironment of the GC leading to the formation of the memory B cell compartment. According to many models, the GC response takes place in a series of discrete stages (Kelsoe and Zheng, 1993; MacLennan, 1994; Liu and Arpin, 1997; Camacho et al., 1998). After their antigen receptors (BCRs) engage antigen, follicular B cells migrate to near the border of T zones and undergo cognate interaction with helper T cells. Subsequently, pre-GC B cells travel to the center of the follicle and rapidly proliferate in association with follicular dendritic cells (FDCs), resulting in formation of the GC proper. It has been suggested that during this early phase of clonal expansion, changes in the structure and function of BCRs do not occur, and this results in the generation of a sufficient pool of precursors to sustain the somatic hypermutation-selection and differentiation processes that ensue (Jacob et al., 1991). Histologic and flow cytometric GC “time course” studies have provided support for this idea (Berek et al., 1991; Jacob et al., 1991; Jacob and Kelsoe, 1992; McHeyzer-Williams et al., 1993; Camacho et al., 1998; Vora et al., 1999), but their limited resolution has not allowed a distinction between initial B cell clonal expansion in and recruitment to the GC. Moreover, recent multi-photon imaging studies of ongoing GC reactions in lymph nodes have revealed a far greater degree of plasticity of GC B cell behavior than predicted by previous models of this response (Allen et al., 2007; Schwickert et al., 2007; Hauser et al., 2007). In particular, these analyses demonstrated extensive migration of individual GC B cells within, around and into and out of the GC proper.

Given the above considerations, we performed a detailed analysis of the interrelationship of proliferation, differentiation and micro environmental locale in homogenous populations of pre-GC B cells during the early stages of TD immune responses. These studies revealed a previously unappreciated stage of the GC B cell developmental pathway, in which pre-GC B cells undergo rapid clonal expansion at the perimeter of follicles without significant induction of somatic hypermutation, heavy chain class switching, or expression of genes necessary for homing to the center of follicles. Our findings necessitate significant revision of previous models for the GC response.


A previously described line of VH “knockin” transgenic mice termed HKI65 (Heltemes-Harris et al., 2004), whose B cells express a high frequency of BCRs that bind the hapten arsonate (Ars) and are termed “canonical”, was used for our studies. VH knockin loci efficiently undergo somatic hypermutation (SHM) and class switch recombination (CSR) (Taki et al., 1995; Shih et al., 2002). For most experiments, HKI65 mice were mated to a previously described line of Ig L chain transgenic mice expressing the canonical Vk10A-Jk1 L chain gene (Liu et al., 2007), creating double transgenic mice we term HKI65/Vk10. Canonical B cells are readily detected in histological and flow cytometric analyses using the anti-clonotypic mAbs E4 (Notidis et al., 2002; Heltemes-Harris et al., 2004) or 5Ci (Wysocki and Sato, 1991; B.A., unpublished observations).

During the early stages of the anti-Ars response HKI65/Vk10 B cells are located predominantly at the perimeter of follicles

Splenic B cells enriched from HKI65/Vk10 mice were injected into immunosufficient, unirradiated, nontransgenic, syngeneic recipients that had or had not been immunized i.p. with Ars-KLH in alum one week earlier. All recipient mice used in these studies were incapable of expressing canonical BCRs due to absence of the requisite VH gene segment. This pre-immunization approach was initially taken as we anticipated that it would result in prior development of a KLH-specific helper T cell compartment sufficient to induce synchronous activation of many Ars-specific donor B cells, allowing effective evaluation of various stages of their subsequent response to antigen in vivo. In some experiments, donor cells were labeled with CFSE to allow monitoring of cell division (Lyons and Parish, 1994). Also, in most studies, recipient mice were strain C57BL/6.CD45.1 (B6.CD45.1), permitting distinction of donor and host B cells using the CD45.1/CD45.2 allotype system, irrespective of the structure and expression of their BCRs.

One day after transfer, donor B cells were mainly found by histological analyses to be scattered throughout follicles in unimmunized chimeric mice, a distribution that persisted for ten days. Flow cytometric analysis at all time points after transfer did not reveal proliferation or expression of the GC activation marker GL7 by donor B cells in these mice (data not shown). B cells transferred to immunized mice were also observed in follicles on day one. However, a major subset of these cells was concentrated at the interface of follicles and the periarteriolar lymphoid sheath (PALS) T cell zones (Figure 1A, panels a and b), where initial cognate interaction of antigen specific B cells and T cells most probably takes place (Jacob et al., 1991; Liu et al., 1991; Jacob and Kelsoe, 1992; Martin and Goodnow, 2002; Pape et al., 2003).

Figure 1
Micro environmental locale of HKI65/Vk10 donor B cells during the first four days of the immune response in Ars-KLH pre-immunized mice

By day two, staining with anti-CD45.2 (Figure 1A, panels c and d) and 5Ci (data not shown) revealed small numbers of donor B cells located at the perimeter of follicles, near the border with the marginal zone (MZ) (defined by staining of metallophillic MZ cells with the mAb MOMA1). Most donor cells in this locale were distal to the PALS (Figure 1A, panel d). By day three, donor B cells were numerous in this region (Figure 1A, panels e and f). At day four, donor B cells were abundant in the center of follicles (Figure 1A, panels g and h). By day five, most of these cells were located in GCs, as defined by staining with GL7 for GC B cells, and FDC-M2 for follicular dendritic cells (FDCs) (example shown in Figure 1B).

HKI65/Vκ10 transgenic B cells undergo rapid proliferation at the perimeter of follicles prior to entering the GC

The above results suggested that canonical HKI65/Vk10 B cells were undergoing clonal expansion at the perimeter of follicles prior to entering the GC reaction. To test this idea, histological studies using a mAb specific for the nuclear proliferation antigen Ki67, and flow cytometric analyses for dilution of CFSE in donor B cells were performed. Figure 2A (panels a and b and magnified panel) illustrates that only a small percentage of donor B cells present in and around the T cell zone at day one were Ki67+. In marked contrast, by days two and three, the majority of these B cells, now located at the perimeter of follicles, were Ki67+ (Figure 2A, panels c-f and magnified panels). The results of quantification of such data are presented in Figure 2B. As expected from previous studies (Camacho et al., 1998; Rahman et al., 2003), by day four nearly all donor B cells, now located in GCs in the center of follicles were Ki67+ (Figure 2A, panels g and h and magnified panel). In addition, BrdU pulse labeling studies showed that anti-clonotype+ B cells in S phase were numerous in the locale of the perimeter of follicles in the day two to four time frame (data not shown).

Figure 2
Proliferation of HKI65/Vk10 donor B cells at the perimeter of follicles in pre-immunized mice

Pre-immunization is not required for localization of proliferating pre-GC B cells to the perimeter of follicles

Introduction of naïve donor B cells into mice with ongoing immune responses and GC reactions clearly does not directly mimic conditions at the onset of a conventional immune response. As such, we conducted an extensive series of studies in which naïve HKI65/Vk10 splenic B cells were injected into nonirradiated, syngeneic recipients, followed by i.p. immunization of the resulting chimeric mice with Ars-KLH in alum 12 hours later. Control studies revealed few, if any GCs in the spleens of naïve recipient mice prior to cell transfer and immunization (data not shown). Figure 3 illustrates that results entirely analogous to those obtained using the pre-immunization protocol were obtained using this post-immunization approach, although donor B cell localization to and proliferation in the follicular perimeter and subsequent entry into GCs were delayed by one to two days as compared to the pre-immunization situation, and the response of these donor cells appeared less synchronous.

Figure 3
HKI65/Vk10 B cell localization to and proliferation at the follicular perimeter does not require pre-immunization

We next combined CFSE labeling of donor B cells with histological and flow cytometry analyses to extend the resolution of these post-immunization protocol studies. At day two after immunization donor B cells were largely located at the interface of the follicle and the PALS and had not undergone detectable levels of proliferation (data not shown). Figure 4A illustrates that at day three after immunization donor B cells were located both at the interface of the follicles and PALS and in the follicular perimeter. Flow cytometric measurement of CFSE dilution and other markers revealed that the majority of donor B cell had not undergone proliferation at this time point (Figure 4C, upper left panel). By day four, however, Figure 4B shows that most donor B cells were located at the follicular perimeter and many were Ki67+ (Figure 4B, panel c and magnified area of panel c in panel d). Figure 4B panel b illustrates that all of the donor B cells located in this region lacked CFSE fluorescence, most probably due to the fact that active proliferation had diluted their levels of CFSE below the detection limit of fluorescence microscopy. Flow cytometry demonstrated that a major fraction of these cells had divided two to four times (Figure 4C, lower panels). Figure 4C also illustrates that while proliferating donor B cells at days three and four had begun to acquire the GL7 activation marker and expressed progressively lower levels of surface IgD, they had not yet become PNA+ and did not express detectable levels of surface IgG.

Figure 4
HKI65/Vk10 B cell proliferation and changes in cell surface phenotype during the response at the follicular perimeter

Analogous results were obtained using the pre-immunization protocol (Figure S1). However, in these studies nearly all donor B cells were recruited into the proliferative response, and this response was accelerated by several days. Also, proliferating donor B cells expressed higher, and very uniform levels of the GL7 marker, intermediate amounts of PNA binding, and lower levels of sIgD than those observed using the post-immunization protocol. A minor fraction of such cells also appeared to acquire expression of sIgG. We speculate that these differences are due to the high levels of antigen-specific T cell help available to donor B cells as they are incorporated into an already ongoing immune response.

Some B cells participating in the early stages of certain TD immune responses are destined for development into plasmablasts (Liu et al., 1991; Jacob and Kelsoe, 1992; McHeyzer-Williams et al., 1993; Smith et al., 1996; Shih et al., 2002; Pape et al., 2003). Since such cells would not express GL7, and might only express low levels of B220 but high levels of the AFC marker syndecan (CD138) we also monitored the dilution of CFSE in B220+, GL7-, and B220- and CD138+ donor cells in the chimeric mice. Significant levels of division were only observed in the B220high and CD138- subpopulations (Figure S1C panels a and b and data not shown). Moreover, using both protocols we did not observe many proliferating donor B cells in the blood during the time of rapid proliferation in the spleen (Figure S2). Taken together with the Ki67 staining studies described above, this argues that most, if not all of the pre-GC B cell proliferative response takes place at the follicular perimeter. Finally, we also investigated the influence of canonical B cell precursor frequency on the results obtained in both protocols. For these studies, HKI65/Vk10 B cells were injected in various doses. As few as 2×104 donor B cells could be injected and subsequently detected via histological analyses (see legend to Figure 1 for examples). The results obtained with this small number of cells were analogous to those obtained with larger number of cells.

Canonical B cells in Ars-KLH immunized HKI65 heterozygous mice proliferate at the follicular perimeter

HKI65 single transgenic, H chain knockin locus hemizygous mice (HKI65+/-), have a “semi-diverse” BCR repertoire due to the expression of endogenous L chain genes. Nonetheless, canonical B cells can be identified in these mice using the anti-clonotypic E4 mAb. To investigate the possibility that localization of proliferating canonical B cells to the follicular perimeter in the studies described above was influenced by the adoptive transfer protocols, HKI65+/- mice were directly immunized with Ars-KLH in alum and histological analysis of the canonical B cell response in the spleen was performed at several time points.

As expected, the canonical B cell response in this case was far less uniform and synchronous as compared to that observed using adoptive transfer protocols. Nonetheless, at day four after immunization many E4+ B cells were observed at the follicle-PALS interface and at day five and into day 6 this distribution had changed dramatically in that a major subpopulation canonical B cells were now observed at the follicular perimeter (Figure S3, panel A). Ki67 staining analysis showed that many of the donor B cells in this locale were proliferating (Figure S3, panels B and C).

HKI65/Vκ10 B cells undergoing rapid proliferation at the perimeter of follicles have a unique expression profile of genes involved in regulation of antigen driven B cell development and migration

The above studies suggested that canonical B cells at the perimeter of follicles in immunized mice were intermediates between those initially activated at the T-B interface, and GC B cells. We further examined this issue by analyzing the expression of various genes required for GC and AFC development and B cell migration via Q-RT-PCR in four subpopulations of B cells purified by FACS: naive B cells from C57BL/6 mice (used as a standard); and three fractions of B cells from chimeric mice at day three after donor cell injection of immunized recipient mice: GL7-, CFSE+, (donor B cells that had not divided), GL7+/-, CFSE+/- (donor cells that had divided 4-6 times), and GL7+, CFSE- cells (donor and host GC B cells that had divided extensively, see lower right inset, Figure 5).

Figure 5
Gene expression profile of HKI65/Vk10 donor B cells after different numbers of cell divisions during the follicular perimeter response

Figure 5A and 5B illustrate that, as predicted from previous studies, GL7+, CFSE-, GC B cells expressed high levels of RNA encoding activation induced cytidine deaminase (AID) (Muramatsu et al., 1999), IgG H chain and Bcl-6 (Onizuka et al., 1995; Cattoretti et al., 1995; Allman et al., 1996; Flenghi et al, 1996), CXCR4 (Allen et al., 2004) and CXCR5, as compared to naive B cells. These cells expressed moderate levels of BLIMP-1 and nearly undetectable levels of CCR7 RNA. In contrast, GL7-, CFSE+ donor B cells had undetectable levels of RNA encoding AID, low levels for IgG H chain, modest levels for Bcl-6, BLIMP-1 and CXCR5; and very low levels for CXCR4 and high levels for CCR7. This expression profile is consistent with B cells that had recently left the interface of B and T cell zones, since CCR7 is required for B cell migration to this locale (Reif et al., 2002; Cyster, 2005). Cells with a GL7+/-, CFSE+/- phenotype expressed low levels of RNA for AID and Bcl-6, and very low to undetectable levels of RNA for IgG, BLIMP-1, CCR7, CXCR4 and CXCR5. The low level expression of RNA for the chemokine receptors indicates that these cells, predominantly located at the perimeter of follicles, are sessile. Histological analyses confirmed that few donor B cells in this locale expressed IgG at the day three time point (data not shown).

B cell proliferation at the perimeter of follicles in not accompanied by extensive somatic hypermutation

To determine if donor HKI65/Vk10 B cells that had undergone various numbers of divisions in the chimeric mice had activated SHM to an extent characteristic of GC B cells, B cells present in different CFSE intensity peaks were purified by FACS. The HKI65 VH knockin locus was PCR amplified from these fractions and the products cloned and sequenced. No mutations were found in VH clones obtained from division peaks zero and five isolated three days after donor cell injection (Supplemental Table 1). In contrast, anti-clonotype+ GCs micro dissected from sections of spleens isolated at day five after donor cell transfer yielded HKI65 PCR VH clones with an average mutation frequency of 0.4%, and 30% of these mutations were found in somatic hypermutation “hotspots”. These mutations were rather randomly distributed, and those known to increase affinity for Ars were rare (data not shown), as would be expected at this early stage of the GC reaction (Berek et al., 1991; Jacob et al; 1993; McHeyzer-Williams, 1993). These data are consistent with those in Figure 5 showing that low, and very high levels of AID expression are apparent in donor GL7+/-, CFSE+/-, and GL7+, CFSE- (GC) B cells, respectively.

Anti-HEL Ig transgenic B cells rapidly proliferate at the perimeter of follicles in response to HEL immunization

We next considered that the rapid proliferative response of pre-GC B cells at the perimeter of follicles might only take place during immune responses to certain antigens. To investigate this possibility, B cells expressing a transgene-encoded BCR specific for hen egg lysozyme (HEL) were injected into immunosufficient, nonirradiated, nontransgenic, syngeneic recipients either primed three days earlier with HEL in alum i.p., or 12 hours after cell transfer. Results analogous to those obtained using the HKI65/Ars system were found in this study (Figure 6). Specifically, at days two and three of the response in the pre-immunization protocol (Figure 6A, panels a-d) and day six of the post-immunization protocol (Figure 6B, panels a and c), the majority of donor (IgMa+) B cells and were located at the perimeter of follicles. By day four of the pre-immunization protocol and day seven of the post-immunization protocol many of these donor B cells were becoming GL7+ and had entered or formed GCs (Figure 6A, panel f and Figure 6B, panels b and d-j). Detailed staining analysis of the GCs obtained using the postimmunization protocols showed that they contained a low frequency of CD4 T cells (Figure 6B, panel b) and extensive FDC networks (Figure 6B, panels k and l). Flow cytometric studies using CFSE labeled anti-HEL B cells revealed that at the time points where such cells were mainly located at the follicular perimeter they had divided several times and were GL7+ (data not shown).

Figure 6
Proliferation and locale of anti-HEL transgenic B cells during the immune response to HEL

Accessory cells for B cell activation and differentiation are not abundant in the follicular perimeter during the pre-GC B cell proliferative response

CD4 T cells, FDCs and certain subclasses of other dendritic cells (DC) have all been shown to provide growth, survival and differentiation signals to antigen activated B cells (Clarke and Lane, 1991; Berney et al., 1999; Tew et al., 2001; MacLennan et al., 2003; Haynes, 2008). To determine if any of these cells type were co-localized at the follicular perimeter during the height of the pre-GC B cell proliferative response there, more extensive histological analyses were performed using the post-immunization protocol. CD11c+ interdigitating DCs were clearly absent from the follicular perimeter (Figure 7A, left panels). Sparsely scattered T cells (Figure 7A, right panels) and FDCs (Figure 7B) were found in this locale, but did not appear to be directly associated with most donor B cells. Also, staining for the marginal zone B cell marker CD1d (Figure 7C) suggested that donor B cells had not acquired this phenotype. Analogous results were obtained using the pre-immunization protocol (data not shown).

Figure 7
Location of accessory cells during the HKI65/Vk10 B cell proliferative response at the follicular perimeter


The data presented here demonstrate that during the initial phases of GC B cell development, rapid proliferation takes place at the perimeter of follicles. Antigen activated B cells in this locale display a gene and cell surface antigen expression profile consistent with recent departure from the T cell zone (i.e. moderate levels of CCR7 prior to proliferation), and ongoing development via the GC pathway (i.e. low to moderate levels of AID, BCL-6, and GL7 expression and sIgD down regulation). Moreover, highly proliferative B cells in this locale express very low levels of RNA encoding CXCR4, CXCR5 and CCR7, indicating that they are not simply “in transit” from the T cell zone to the GC. The subsequent migration of large numbers these B cells to the center of the follicle in which they are resident suggests that GCs are initially formed by B cell coalescence, rather than proliferation in situ.

Our results support previous speculations (Jacob et al., 1993; McHeyzer-Williams et al., 1993) that SHM is not induced during the initial stages of GC B cell proliferation. Past studies on this issue were hampered by lack of knowledge of the time in which GC precursors entered the GC reaction, and the degree to which these precursor clones had expanding prior to entry. More importantly, our data indicate that the majority of the “mutation-free” period of pre-GC B cell proliferation takes place at the perimeter of follicles. Very shortly after entering GCs, however, donor B cells in our experiments must have substantially up regulated SHM, since the majority of the transgenic VH clones we isolated from micro dissected GCs five days after cell transfer contained substantial levels of somatic mutation. These data are incompatible with the idea that high rate SHM occurs as a function of GC B cell division alone. In contrast, they support the notion that a B cell extrinsic factor(s) present in the GC proper is required to promote this process.

Our results demonstrating very low or undetectable levels of heavy chain class switching in B cells rapidly proliferating at the perimeter of follicles contrast with past findings also obtained using a transgenic B cell adoptive transfer approach. In these studies, a major fraction of donor B cells were found to have switched to IgG expression at the T-B interface and in the follicles of the spleen four days after immunization, and to have undergone more extensive expansion than their IgM expressing counterparts (Pape et al., 2003). Another previous study suggested that the cytoplasmic tail of membrane IgG confers a proliferative advantage to responding B cells in the AFC pathway (Martin and Goodnow, 2002). Nonetheless, our data demonstrate that expression of an IgG form of the BCR is not a prerequisite for the dramatic burst of proliferation characteristic of the early stages of pre-GC B cell development.

These considerations emphasize the importance of future studies to determine how changes in the many conditions known to affect the quantity and quality of B cell responses impinge upon proliferation and differentiation at the perimeter of follicles. For example, the response of most anti-HEL B cells in this locale was characterized by both rapid proliferation and induction of GL7 and PNA binding, both classical GC B cell markers. In contrast, the proliferative response of HKI65/Vκ10 B cells in this locale was slower and accompanied by a more gradual induction of GL7, as well as reduction of IgD expression. It is tempting to speculate that variation in the affinity of the BCR(s) expressed by donor B cells for the driving antigen resulted in the differences in cellular behavior described here (e.g. the Ka of the HKI65/Vk10A BCR for Ars is approximately 2×105 (Rothstein et al., 1983), whereas the Ka of the anti-HEL BCR for HEL is > 108 (Goodnow et al., 1988)) and in previous studies.

In total, our results demonstrate that a previously unrecognized microenvironment exists at the perimeter of follicles opposite the T cell zone that can promote the initial proliferative stages of the GC B cell response: namely, after primary interaction with T cells at the interface of follicles and the T cell zone, and prior to formation of or entry into the GC. While our studies were restricted to the spleen, others have observed antigen specific B cell proliferation in the sub capsular perimeter of lymph node follicles after immunization with a TD antigen (Garside et al., 1998), and have described a novel class of antigen-transporting DCs that localizes to this region of LNs shortly after immunization (Berney et al., 1999). Future studies on the factors that regulate the migration of recently activated B cells to, and retention and activity of pre-GC B cells in this microenvironment are clearly required.

What might be the function of this intermediate stage of rapid clonal expansion during TD B cell responses in which the structure and function of BCRs remain largely unaltered? In the context of natural infections, Cohn noted many years ago (Cohn, 1972) that unless a large pool of pathogen-specific lymphocytes was generated in a short period of time, the infection would overwhelm the host due to the high replicative capacity of most pathogens. Since that time, numerous studies have shown that the GC reaction is an extremely dynamic process in which BCR structures are modified by SHM and CSR, and stringent selective forces are brought to bear on GC B cells based on the affinity and specificity of their BCRs (Kelsoe and Zheng, 1993; MacLennan, 1994; Liu and Arpin, 1997; Camacho et al., 1998). SHM alters the structure of antibody V regions rather randomly (Winter and Gearhart, 1998), and for this reason the majority of somatic mutations are expected to disrupt expression of BCRs, or alter their affinity and specificity such that the GC B cells expressing them are counter selected (Shlomchik et al., 1989, Casson and Manser, 1995, Wiens et al., 1998). Given such considerations, the average rate of SHM per GC B cell division has been argued to approach that which would prevent the clonal expansion of B cells during an immune response (Shlomchik et al., 1989). This suggests that if individual GCs are founded by only small numbers of B cell precursors, subsequent generation of effector populations of sufficient size, expressing BCRs of the affinity and specificity necessary for rapid antigen (pathogen) clearance, might not occur. Therefore, proliferation of pre-GC B cells at the perimeter of the follicle may allow nucleation of the GC reaction by antigen-specific clones large enough to ensure a productive outcome of the subsequent GC response. This scenario would not be inconsistent with the fact that the GC reaction is oligoclonal (Kroese et al., 1987; Jacob et al., 1991; Liu et al., 1991), if the number of B cell precursors to the proliferative phase of the response at the follicular perimeter was limited, or if affinity and specificity-based clonal selection took place in this microenvironment.

Limited CSR during this phase of the response may sub serve a similar function. As discussed above, IgG forms of the BCR appear to have enhanced ability to stimulate proliferation as compared to IgM and IgD forms (Martin and Goodnow, 2002; Pape et al., 2003). Therefore, CSR to IgG isotypes could interfere with the clonal selection of GC B cells with increased affinity and specific for the driving antigen due to SHM. Moreover, delaying the irreversible decision regarding which of the H chain isotypes to express at the level of secreted antibody until after the mutation-selection process is well underway may maximize the probability that highly specific antibodies with effector functions most suited to the nature of the infecting pathogen are generated during TD immune responses. Testing these ideas will require the development of experimental approaches that selectively perturb the migration to and proliferation of B cells at the perimeter of follicles. In this regard, chemokine receptors expressed by responding B cells seem particularly obvious targets. Indeed, Cyster and colleagues (Reif et al., 2002) have observed lodging of CCR7-deficient anti-HEL transgenic B cells at the follicular perimeter following transfer into HEL immunized, CCR7-sufficient recipient mice.

Experimental Procedures


HKI65 VH knock-in mice have been previously described (Heltemes-Harris et al., 2004). To produce the double transgenic HKI65/Vk10 mouse line, HKI65 mice were bred to a line of conventional Vk10A light chain transgenic mice termed Vk1060 (a kind gift of Dr. Lawrence Wysocki). Vk1060 transgenic mice bear a functional Vk10A-Jk1 gene encoding the light chain of canonical anti-Ars antibodies. Tg(IghelMD4)4Ccg/J mice were purchased from the Jackson Laboratory (Bar Harbor, ME). C57BL/6 and B6.SJL-Ptprca Pepcb/BoyJ (B6.CD45.1) mice were originally obtained from the Jackson Laboratory and were bred in house. Mice were housed under pathogen-free conditions and were given autoclaved food and water. The use of mice in these studies was approved by the University's Institutional Animal Care and Use Committee.

Adoptive transfers and immunizations

B cells were enriched from HKI65/Vk10 or Tg(IghelMD4)4Ccg/J spleens via anti-CD4, anti-CD8 and anti-Thy1.2 and C-mediated lysis. Alternatively, non B-cells were depleted by incubation with biotinylated anti-CD3, anti-Thy1.2 and anti-CD11b (Mac-1), or anti-CD43 followed by MACS (Milteny Biotec, Auburn, CA). In some experiments enriched B cells were labeled with CFSE (Molecular Probes, Eugene, OR) as previously described (Lyons and Parish, 1994). B cells (2×104-107) were injected into the tail vein or retro-orbital sinus of syngeneic recipients. Three immunization protocols were used. Recipient mice were either injected with donor cells and immunized 12 hours later with 100 μg of Ars-KLH (or HEL) in alum i.p. (post-immunization protocol), or received 100 μg of Ars-KLH (or HEL) in alum i.p. one week or three days before cell transfer with an additional injection of 50 μg of Ars-KLH (or HEL) in PBS i.p. at the time of adoptive transfer (pre-immunization protocols).


Spleens were snap frozen in OCT compound and cryosections (5-6 μm) made as previously described (Vora et al., 1998). Immunofluorescence staining was also performed as previously described (Heltemes-Harris et al., 2004). The following mAbs and reagents were used: E4-biotin, 5Ci-biotin (both made in house), anti-CD1d-biotin and anti-TCRβ (H57-597)-biotin (both from BD Biosciences) followed by SA-PE (Molecular Probes) or SA-Alexa 633 (Molecular Probes); rat anti-CD169 (MOMA1) (Serotec Inc, Raleigh, NC), rat anti-mouse Ki67 (DakoCytomation, Glostrup, Denmark), FDC-M1 and FDC-M2 (both from ImmunoKontact, Abingdon, United Kingdom), all followed by goat anti-rat IgG-Alexa 633 (Molecular Probes); GL7-FITC, MOMA1-FITC (Serotec Inc, Raleigh, NC); anti-CD11c-FITC; anti-B220 (RA3-6B2)-FITC, anti-TCRβ (H57-597)-PE and FITC, anti-IgMa-PE, and anti-CD35-PE (all from BD Biosciences); and anti-CD45.2 (104)-PE (eBioscience).

5-Bromo-2′-deoxyuridine (BrdU) labeling and analysis

BrdU labeling was performed using the BrdU Labeling and Detection Kit II from Roche Diagnostics Corporation (Indianapolis, IN) and BrdU (Sigma-Aldrich, St. Louis, MO). Briefly, mice were injected i.p. with 1 mg of BrdU and sacrificed five or 12 h later. Spleens were snap frozen in OCT compound and cryosections (5-6 μm) made as previously described (Vora et al., 1998). BrdU detection was done as described in the kit manual. Alternatively, BrdU labeled cells were revealed with rat anti-BrdU antibodies (Serotec, Raleigh, NC) followed by mouse anti-rat IgG-AP (Jackson Immunoresearch Laboratories).

Flow cytometry

Single-cell suspensions were prepared and stained with combinations of the following Abs and reagents: E4-biotin (made in house), PNA-biotin (Vector Laboratories), GL7, anti-B220 (RA3-6B2)-PE, anti-CD45.2 (104)-PE, anti-CD138 (281.2)-PE (all from BD Biosciences), anti-CD45.2-PE and biotin (eBiosciences), anti-IgD (11-26)-PE (Southern Biotech, Birmingham, AL), and anti-mouse IgG-biotin (Jackson Immunoresearch). Staining with GL7 was followed by biotin-mouse anti-rat IgG (Jackson Immunoresearch Laboratories). Whole mouse Ig (Jackson Immunoresearch Laboratories) was used to block Fc receptors. CyChrome-SA (BD Biosciences), SA-PerCP-Cy5.5 or SA-PE-Cy5.5 (both from eBioscience) were used to detect biotinylated Abs and other reagents.

Cells were either sorted immediately using an EPICS Elite-ESP or a MoFlo high-performance cell sorter (DakoCytomation), or analyzed on an EPICS XL-MCL (Coulter, Hialeah, FL). If not analyzed immediately, cells were fixed in 2% formaldehyde. Data were analyzed using the FLOWJO software (Treestar, San Carlos, CA).

Molecular genetics

B cells in peaks of CFSE fluorescence intensity corresponding to different cell divisions were sorted into 1X PBS. Genomic DNA was isolated using the DNeasy Blood & Tissue Kit (Qiagen). For laser capture microdissection of GCs, frozen spleen sections (7-8 μm) were mounted on PEN-Membrane slides (Leica) and immunohistochemistry was performed as previously described (Vora et al., 1998). Briefly, sections were labeled with anti-CD45.2 (104)-biotin (eBioscience), followed by SA-AP (Vector) and PNA-HRP (Sigma). Slides were developed using the Alkaline Phosphatase Substrate Kit III and NovaRed Substarte kit for peroxidase (both from Vector). Tissue from individual PNA+, CD45.2+ GCs was microdissected using the LMD6000 laser microdissection system (Leica) and genomic DNA was released by Proteinase K treatment. The HKI65 knock-in locus was amplified via nested PCR performed on DNA from sorted cells and microdissected tissue samples using Pfx50 DNA polymerase (Invitrogen). A first round of 35 cycles was performed using the primers 5′-CAACCTATGATCAGTGTCCTC-3′ (hybridizing 5′ of the leader exon) and 5′-GGACTCCACCAACACCATCAC-3′ ((hybridizing to the JH2-JH3 intron). A small aliquot of the first round was amplified an additional 32 cycles using the primers 5′-CAGGTGTCCACTCTGAGGTTC-3′ (hybridizing to the end of the leader exon and the beginning of the HKI65 VH gene) and 5′-GTGTCCCTAGTCCTTCATGACC-3′ (hybridizing to the JH2-JH3 intron 5′ to the first round reverse primer). PCR products of the expected size were cloned into plasmid vectors, sequenced at the Kimmel Cancer Center Nucleic Acids Facility and analyzed using CLUSTALW multiple sequence alignment program.

RNA from sorted cells was extracted using the RNeasy mini kit (Qiagen), according to the manufacturer's instructions and first strand cDNA was synthesized using TaqMan reverse transcription reagents (Applied Biosystems). Quantitative real-time RT-PCR was performed using the following gene expression assays: Mm 99999055_m1 (Cxcr4), Mm00432086 (Blr1), Mm01301785_m1 (Ccr7), Mm00507774_m1 (Aicda), Mm01703609_g1 (Ighg), Mm00477633_m1 (Bcl6), Mm00476128 (Prdm1), Mm99999915_g1 (Gapdh) and universal master mix (Applied Biosystems). Data were collected using the ABI Prism 7000 sequence detection system (Abbott Diagnostics). Individual gene expression was normalized to GAPDH.

Supplementary Material


Supplemental Figure 1. Analysis of extent of proliferation and cell surface phenotype of HKI65/Vk10 B cells using the pre immunization protocol

Chimeric mice were generated as described in the legend to Figures 1 and and22 using CFSE labeled HKI65/Vk10 splenic B cells (107 injected per recipient mouse). (A) At the indicated time points, splenic B cells were stained with GL7 and analyzed for degree of CFSE dilution via flow cytometry. Boxes indicate GL7+ and negative gates and dashed lines demark various CFSE division peaks. (B) Quantitation of cells in various CFSE division peaks at day 3 of the response; (C) Spleen cells obtained from mice three days after donor cell injection were stained with reagents specific for the indicated markers and levels of these markers expressed on cells in various CFSE division peaks were evaluated by flow cytometry. Data are representative of those from 12 mice in seven independent experiments.

Supplemental Figure 2. Proliferating HKI65/Vk10 B cells are abundant in the spleen but not the blood during the early stages of the Ars-KLH response

HKI65/Vk10 CFSE labeled splenic B cells (107 per recipient in (A) and 3×106 per recipient in (B)) were injected into recipients that either had not been immunized (naïve); had been immunized one week early with Ars-KLH (A), or were immunized 12 hours later with Ars-KLH (B) as described in the Legends to Figures 1 and and33 and Experimental Procedures, and chimeric mice sacrificed at the indicated time points. Lymphocyte populations from blood and spleen of the same mice were analyzed for proliferation via flow cytometry. A total of 105 events in the lymphocyte gate were collected in every case. The data in (A) are representative of at two independent experiments and the data in (B) are representative of those obtained from two naive and two immunized mice.

Supplemental Figure 3. Location of proliferating canonical B cells after direct immunization of HKI65+/- mice

HKI65+/- mice were directly immunized with 1 mg of Ars-KLH in alum i.p. and sacrificed at the indicated time points and spleen analyzed via immunohistology using Abs specific for the indicated markers. Image scale is indicated in the lower left panel. Data are representative of images obtained from two independent experiments where two or three mice were sacrificed at each time point. (B) Quantitative analysis of the percentage of E4+ B cells at the perimeter of follicles (defined as describe in the Legend to Figure 2) that were Ki67+ on day six of the response. Each triangle, and circle reflects data collected from an individual field. Data were obtained from at least three mice for each day. In this experiment, HKI65+/- mice were immunized with 100 μg of Ars-KLH in alum i.p.. (C) Location of E4+ canonical B cells in HKI65+/- mice sacrificed at day six after immunization. Images are from adjacent sections. Image scale is indicated in the left panel.

Supplemental Table 1. Analysis of mutations in HKI65 knockin loci recovered from sorted and microdissected cellsa


We thank Scot Fenn for technical assistance and all members of the Manser laboratory for their indirect contributions to this work. We also thank Wojciech Jankowski of the Kimmel Cancer Center Bioimaging Facility, the Kimmel Cancer Center Flow Cytometry Facility and Dr. Mark Curtis of the Department of Pathology at Thomas Jefferson University for technical assistance with microdissection. This study was supported by grants from the NIH to T. Manser.


Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.


  • Allen CD, Ansel KM, Low C, Lesley R, Tamamura H, Fujii N, Cyster JG. Germinal center dark and light zone organization is mediated by CXCR4 and CXCR5. Nat Immunol. 2004;9:943–952. [PubMed]
  • Allen CD, Okada T, Tang HL, Cyster JG. Imaging of germinal center selection events during affinity maturation. Science. 2007;5811:528–531. [PubMed]
  • Allman D, Jain A, Dent A, Maile RR, Selvaggi T, Kehry MR, Staudt LM. BCL-6 expression during B-cell activation. Blood. 1996;12:5257–5268. [PubMed]
  • Berek C, Berger A, Apel M. Maturation of the immune response in germinal centers. Cell. 1991;67:1121–1129. [PubMed]
  • Berney C, Herren S, Power CA, Gordon S, Martinez-Pomares L, Kosco-Vilbois MH. A member of the dendritic cell family that enters B cell follicles and stimulates primary antibody responses identified by a mannose receptor fusion protein. J Exp Med. 1999;190:851–860. [PMC free article] [PubMed]
  • Camacho SA, Kosco-Vilbois MH, Berek C. The dynamic structure of the germinal center. Immunol Today. 1998;19:511–514. [PubMed]
  • Casson L, Manser T. Random mutagenesis of two CDR amino acids yields an unexpectedly high frequency of antibodies with increase affinity for both cognate antigen and autoantigen. J Exp Med. 1995;182:743–750. [PMC free article] [PubMed]
  • Cattoretti G, Chang CC, Cechova K, Zhang J, Ye BH, Falini B, Louie DC, Offit K, Chaganti RS, Dalla-Favera R. BCL-6 protein is expressed in germinal-center B cells. Blood. 1995;1:45–53. [PubMed]
  • Clark EA, Lane PJ. Regulation of human B-cell activation and adhesion. Annu Rev Immunol. 1991;9:97–127. [PubMed]
  • Cohn M. Immunology: What are the rules of the game? Cell Immunol. 1972;5:1–20. [PubMed]
  • Cyster JG. Chemokines, sphingosine-1-phosphate, and cell migration in secondary lymphoid organs. Annu Rev Immunol. 2005;23:127–159. [PubMed]
  • Flenghi L, Bigerna B, Fizzotti M, Venturi S, Pasqualucci L, Pileri S, Ye BH, Gambacorta M, Pacini R, Baroni CD, et al. Monoclonal antibodies PG-B6a and PG-B6p recognize, respectively, a highly conserved and a formol-resistant epitope on the human BCL-6 protein amino-terminal region. Am J Pathol. 1996;5:1543–1555. [PubMed]
  • Garside P, Ingulli E, Merica RR, Johnson JG, Noelle RJ, Jenkins MK. Visualization of specific B and T lymphocyte interactions in the lymph node. Science. 1998;281:96–99. [PubMed]
  • Goodnow CC, Crosbie J, Adelstein S, Lavoie TB, Smith-Gill SJ, Brink RA, Pritchard-Briscoe H, Wotherspoon JS, Loblay RH, Raphael K. Altered immunoglobulin expression and functional silencing of self-reactive B lymphocytes in transgenic mice. Nature. 1988;334:676–682. [PubMed]
  • Hauser AE, Junt T, Mempel TR, Sneddon MW, Kleinstein SH, Henrickson SE, von Andrian UH, Shlomchik MJ, Haberman AM. Definition of germinal-center B cell migration in vivo reveals predominant intrazonal circulation patterns. Immunity. 2007;26:655–667. [PubMed]
  • Haynes NM. Follicular associated T cells and their B-cell helper qualities. Tissue Antigens. 2008;71:97–104. [PubMed]
  • Heltemes-Harris L, Liu X, Manser T. Progressive surface B cell antigen receptor down-regulation accompanies efficient development of antinuclear antigen B cells to mature, follicular phenotype. J Immunol. 2004;2:823–833. [PubMed]
  • Jacob J, Kelsoe G. In situ studies of the primary immune response to (4-hydroxy-3-nitrophenyl)acetyl. II. A common clonal origin for periarteriolar lymphoid sheath associated foci and germinal centers. J Exp Med. 1992;176:679–687. [PMC free article] [PubMed]
  • Jacob J, Kassir R, Kelsoe G. In situ studies of the primary immune response to (4-hydroxy-3-nitrophenyl)acetyl. I. The architecture and dynamics of responding cell populations. J Exp Med. 1991;173:1165–1185. [PMC free article] [PubMed]
  • Jacob J, Przylepa J, Miller C, Kelsoe G. In situ studies of the primary immune respone to (4-hydroxy-3-nitrophenyl) acetyl III.The kinetics of V-region mutation and selection in germinal center B-cells. J Exp Med. 1993;178:1293–1307. [PMC free article] [PubMed]
  • Kelsoe G, Zheng B. Sites of B-cell activation in vivo. Curr Opin Immunol. 1993;5:418–422. [PubMed]
  • Kitamura D, Roes J, Kuhn R, Rajewsky K. A B cell-deficient mouse by targeted disruption of the membrane exon of the immunoglobulin mu chain gene. Nature. 1991;350:423–427. [PubMed]
  • Kroese FGM, Wubbena AS, Seijen HG, Nieuwenhuis P. Germinal centers develop oligoclonally. Eur J Immunol. 1987;17:1069–1072. [PubMed]
  • Liu YJ, Arpin C. Germinal center development. Immunol Rev. 1997;156:111–126. [PubMed]
  • Liu YJ, Zhang J, Lane PJL, Chan YT, MacLennan ICM. Sites of specific B-cell activation in primary and secondary responses to T-cell dependent and independent antigen. Eur J Immunol. 1991;21:2951–2962. [PubMed]
  • Liu X, Wysocki LJ, Manser T. Autoantigen-B cell antigen receptor interactions that regulate expression of B cell antigen receptor loci. J Immunol. 2007;178:5035–5047. [PubMed]
  • Lyons AB, Parish CR. Determination of lymphocyte division by flow cytometry. J Immunol Meth. 1994;171:131–137. [PubMed]
  • MacLennan ICM. Germinal centers. Ann Rev Immunol. 1994;12:117–139. [PubMed]
  • MacLennan IC, Toellner KM, Cunningham AF, Serre K, Sze DM, Zúñiga E, Cook MC, Vinuesa CG. Extrafollicular antibody responses. Immunol Rev. 2003;194:8–18. [PubMed]
  • Martin SW, Goodnow CC. Burst-enhancing role of the IgG membrane tail as a molecular determinant of memory. Nat Immunol. 2002;3:182–188. [PubMed]
  • McHeyzer-Williams MG, McLean MJ, Lalor PA, Nossal GJV. Antigen-driven B-cell differentiation in vivo. J Exp Med. 1993;178:295–307. [PMC free article] [PubMed]
  • Muramatsu M, Sankaranand VS, Anant S, Sugai M, Kinoshita K, Davidson NO, Honjo T. Specific expression of activation-induced cytidine deaminase (AID), a novel member of the RNA-editing deaminase family in germinal center B cells. J Biol Chem. 1999;26:18470–18476. [PubMed]
  • Notidis E, Heltemes L, Manser T. Dominant, hierarchical induction of peripheral tolerance during foreign antigen-driven B cell development. Immunity. 2002;3:317–327. [PubMed]
  • Onizuka T, Moriyama M, Yamochi T, Kuroda T, Kazama A, Kanazawa N, Sato K, Kato T, Ota H, Mori S. BCL-6 gene product, a 92- to 98-kD nuclear phosphoprotein, is highly expressed in germinal center B cells and their neoplastic counterparts. Blood. 1995;1:28–37. [PubMed]
  • Pape KA, Kouskoff V, Nemazee D, Tang HL, Cyster JG, Tze LE, Hippen KL, Behrens TW, Jenkins MK. Visualization of the genesis and fate of isotype-switched B cells during a primary immune response. J Exp Med. 2003;197:1677–1687. [PMC free article] [PubMed]
  • Rahman ZS, Rao SP, Kalled SL, Manser T. Normal induction but attenuated progression of germinal center responses in BAFF and BAFF-R signaling-deficient mice. J Exp Med. 2003;198:1157–1169. [PMC free article] [PubMed]
  • Reif K, Ekland EH, Ohl L, Nakano H, Lipp M, Forster R, Cyster JG. Balanced responsiveness to chemoattractants from adjacent zones determines B-cell position. Nature. 2002;416:94–99. [PubMed]
  • Rothstein TL, Eisen HN, Gefter ML. Affinity analysis of idiotype-positive and idiotype-negative Ars-binding hybridoma proteins and Ars-immune sera. Molec Immunol. 1983;20:161–170. [PubMed]
  • Schwickert TA, Lindquist RL, Shakhar G, Livshits G, Skokos D, Kosco-Vilbois MH, Dustin ML, Nussenzweig MC. In vivo imaging of germinal centres reveals a dynamic open structure. Nature. 2007;7131:83–87. [PubMed]
  • Shih TA, Meffre E, Roederer M, Nussenzweig MC. Role of BCR affinity in T cell dependent antibody responses in vivo. Nat Immunol. 2002;3:570–575. [PubMed]
  • Shlomchik MJ, Litwin S, Weigert MG. The Influence of somatic mutation onf clonal expansion. Progress in Immunology VII; Proceedings of the 7th International Congress of Immunology; Springer-Verlag, Berlin. 1989. pp. 415–423.
  • Smith KG, Hewiston TD, Nossal GJV, Tarlinton DM. The phenotype and fate of the antibody forming cells of the splenic foci. Eur J Immunol. 1996;26:444–453. [PubMed]
  • Taki S, Schwenk F, Rajewsky K. Rearrangement of upstream DH and VH genes to a rearranged immunoglobulin variable region gene inserted into the DQ52-JH region of the immunoglobulin heavy chain locus. Eur J Immunol. 1995;25:1888–1896. [PubMed]
  • Tew JG, Wu J, Fakher M, Szakal AK, Qin D. Follicular dendritic cells: beyond the necessity of T-cell help. Trends Immunol. 2001;22:361–367. [PubMed]
  • Vora KA, Tumas-Brundage K, Manser T. A PALS-associated B cell focus response is not observed during the development of the anti-arsonate germinal center reaction. J Immunol. 1998;160:728–733. [PubMed]
  • Vora KA, Tumas-Brundage K, Manser T. Contrasting the in situ behavior of a memory B cell clone during primary and secondary immune responses. J Immunol. 1999;163:4315–4327. [PubMed]
  • Wiens GD, Roberts VA, Whitcomb EA, O'Hare T, Stenzel-Poore MP, Rittenberg MB. Harmful somatic mutations: lessons from the dark side. Immunol Rev. 1998;162:197–209. [PubMed]
  • Winter DB, Gearhart PJ. Dual enigma of somatic hypermutation of immunoglobulin variable genes: targeting and mechanism. Immunol Rev. 1998;162:89–96. [PubMed]
  • Wysocki LJ, Sato VL. The strain A anti-p-azophenylarsonate major cross-reactive idiotypic family includes members with no reactivity toward p-azophenylarsonate. Eur J Immunol. 1981;11:832–839. [PubMed]