| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
*
Center for Neurologic Diseases, Brigham and Women’s Hospital and Harvard Medical School, Boston, MA 02115; and
Neuroimmunology Unit, Montreal Neurological Institute, McGill University, Montreal, Quebec, Canada
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
A renewed interest in the immunoregulatory roles of B cells has been sparked in recent studies of tolerance and autoimmunity (5, 6) and in studies suggesting that B cells can act as antigenic reservoirs, thereby providing an important potential source for amplifying immune responses in both health and disease (7, 8). Activated B cells as well as autoreactive B cells in models of autoimmunity have been shown to function as effective APCs to naive T cells (9, 10). In contrast, resting B cells are viewed as poor APCs (11, 12), attributed at least in part to their low levels of expression of costimulatory molecules. The capacity of memory B cells to act as APCs may be particularly important, since, unlike dendritic cells or monocytes, B cells are able to interact with T cells in an Ag-specific manner (13, 14). Nonetheless, very little is known about human B cell subsets and their contributions to immunological memory.
The most definitive marker of memory B cells identified to date is the presence of somatically mutated, high affinity Ag receptors (15). While individual surface markers rarely distinguish perfectly between functionally distinct cell subsets, accumulating evidence has identified surface CD27 as a useful marker of human memory B cells (15, 16, 17, 18, 19). In particular, recent single-cell studies of circulating B cells in humans directly confirmed that essentially all circulating CD27+ B cells displayed variable Ig gene region somatic mutations, while no mutations were identified in the CD27-B cells (15, 19, 20). We studied the costimulatory profile of circulating CD27+ memory B cells in humans, with particular attention to the B7 pathway, which is known to modulate the threshold of activation of both naive and memory T cells (21, 22). It is well established that altering the interactions between the B7.1 (CD80) and B7.2 (CD86) molecules and their T cell counter-receptors, CD28 and CTLA-4 (CD152), can have profound effects on immune responses. Inhibition of B7/CD28 engagement results in enhanced allograft survival, reduced autoantibody production, and amelioration of autoimmunity in both animal models and human disease (23, 24). Unlike professional APCs such as dendritic cells and monocytes that express high constitutive levels of CD86, resting B cells have previously been shown to express low levels of CD86 and no CD80. It has generally been accepted that B cell activation is required to up-regulate both CD80 and CD86 expression, at which time CD86 levels typically rise more rapidly and to a higher extent than CD80 (25, 26, 27, 28).
Given the prevailing dogma, we were surprised to identify in normal adult human blood a high frequency of quiescent memory B cells expressing significant levels of CD80, yet negligible levels of CD86. We demonstrate that this novel cell population represents a phenotypically and functionally distinct human memory B cell subset. Although in a resting state, these CD80+ memory B cells have a lower threshold of activation and can be stimulated to secrete very large amounts of class-switched Igs. Moreover, they are able to efficiently present Ag to and activate T cells. The expression of costimulatory molecules in the resting state and the propensity to mediate vigorous humoral and cellular responses, provide a mechanism by which this novel CD27+CD80+ memory B cell subset could contribute to the rapid and robust immune responses that constitute the defining features of immunological memory.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
To phenotype cells as closely as possible to the in vivo circulating state and to avoid changes in levels of activation markers that may be incurred through in vitro processing, we studied whole blood samples directly ex vivo. Pediatric blood samples (from noninflammatory or infectious cases) as well as postpartum cord blood and adult samples were obtained in accordance with departmental protocols from the Boston Children’s Hospital and the Brigham and Women’s Hospital (Boston, MA), respectively. Triple-color immunofluorescent staining of fresh samples was performed within 20 min of phlebotomy. Whole blood samples were incubated with predetermined optimal concentrations of the appropriate mAbs or isotype controls (see below) for 30 min at 4°C, followed by lysis of RBC (FACS lysing solution 349202, Becton Dickinson, San Jose, CA). Samples were washed twice in staining buffer (2% FCS in PBS), immediately acquired by flow cytometry using FACSort (Becton Dickinson), and subsequently analyzed by CellQuest FACStation software. For staining of PBMCs or purified T and B cells, an identical approach was used, except for the lysing step. DNA-based cell cycle analysis was performed using the Vybrant kit (Molecular Probes, Eugene, OR) (29). Purified B cell subsets stained for CD27 and B7 were costained with propidium iodide (to exclude dead cells) and Hoechst 33342 (to measure cellular DNA content with UV source). To identify the cell cycle kinetic status of the B cell subsets, we used Ki-67 analysis, which distinguishes quiescent DNA 2n Ki-67-negative (G0) cells from DNA 2n Ki-67-positive (G1) cells (30). Purified B cells were first surface stained for expression of CD27 and CD80, then fixed in 4% paraformaldehyde and permeabilized with 0.1% saponin. Samples were incubated with anti-Ki-67 Ab or the appropriate isotype control (see below), then washed twice in permeabilization buffer and once in staining buffer before acquisition. For positive controls in staining experiments, B cells were freshly purified (see below) and incubated with or without activating CD40 ligand (CD40L)-transfected L cells (below) for 40 h before staining and FACS analysis.
The following fluorophore-labeled mAbs were used for staining CD80 and
CD86: PE-L307.4 (mIgG1 anti CD80/B7.1, Becton Dickinson) and FITC- or
PE-2331/FUN1 (mIgG1 anti-CD86/B7.2). Both L307.4 (anti-CD80)
and FUN1 (anti-CD86) were of the same isotype (mIgG1), excluding
the possibility that nonspecific binding was responsible for the
differences in staining patterns. Staining for the B7 molecules on B
cells and monocytes in the same whole blood samples provided an
additional measure of comparison. The pattern of CD80 and CD86
expression was further confirmed with Cy-L307.4 (mIgG1
anti-CD80/B7.1, PharMingen, San Diego, CA) and FITC-BB1 (mIgM
anti-CD80/B7.1,), and with PE-IT2.2 (mIgG2b anti-CD86/B7.2),
respectively. Other Abs used for staining cells (purchased from
PharMingen unless otherwise noted) were: Cy-UCHT1 (mIgG1 anti-CD3),
FITC-RPA-T4 (mIgG1 anti-CD4), PE-ICRF44 (mIgG1 anti-CD11b),
Cy-B43 (mIgG1, anti-CD19), FITC-2H7 (mIgG2b anti-CD20),
FITC-M-A251 (mIgG1 anti-CD25/IL-2R), FITC- or PE-M-T271 (mIgG1
anti-CD27), FITC-5C3 (mIgG1 anti-CD40), Cy-10.1 (mIgG1
anti-CD64/Fc
RI), FITC-J4–117 (mIgG1, anti-CD72),
PE-G46–2.6 (mIgG1 anti-HLA-A,B,C), FITC-TU39 (mIgG2a
anti-HLA-DR,DP,DQ), and FITC-ki-67 (mIgG1 anti-Ki-67, DAKO).
The isotype controls used were: FITC-, PE-, or Cy-MOPC-21 (mIgG1);
FITC-G155-178 (IgG2a); FITC- or PE-27-35 (IgG2b); and FITC-G155-228
(IgM). FITC-X0927 (mIgG1, DAKO) was used as the control Ab for
intracytoplasmic Ki-67 staining.
Cell separation
Peripheral blood leukocytes were obtained by leukapheresis from healthy adult platelet donors, and PBMCs were separated by Ficoll/Hypaque density gradient centrifugation (Pharmacia Biotech, Uppsala, Sweden). B cells were freshly purified using MACS CD19+ magnetic microbeads (503-01, Miltenyi Biotec, Auburn, CA), as previously described (31). Purities were consistently >99%. For the isolation of B cell subsets, purified B cells were sorted on the basis of surface expression of CD27 and CD80. Briefly, B cells at 20 x 106/ml staining buffer were incubated with predetermined optimal concentrations of the appropriate mAbs for 20 min at 4°C, then washed twice in staining buffer. Stained B cells were then immediately sorted (FACSort, Becton Dickinson) into CD27-CD27+/CD80-, and CD27+/CD80+ subsets. Highly purified (>97%) CD4+ T cells were isolated from fresh PBMCs by a two-step negative selection process using T cell subset enrichment columns (R&D Systems, Minneapolis, MN) followed by magnetic bead depletion of unwanted cells, as previously described (32).
B cell subset activation and Ig secretion
Immediately following separation, B cell subsets were incubated in U-bottom 96-well plates, with irradiated (3000 rad) CD40L-transfected L cells (provided by Y.-J. Liu, DNAX, Palo Alto, CA) and IL-4 (100 U/ml; R&D Systems) and IL-2 (10 U/ml; provided by Teceleukin, National Cancer Institute, Frederick, MD) in complete medium with 10% FCS (modified from Ref. 18). For assays of proliferation kinetics, 2.5 x 104 sorted B cells were incubated with 5.0 x 103 irradiated L cells in triplicate wells. [3H]Thymidine (1 µCi/well; NEN/DuPont, Boston, MA) was added for 18 h at the indicated times. Plates were then harvested (Harvestor 96; Tamtec, Orange, CT), and B cell proliferation was assessed by measuring thymidine incorporation in a beta scintillation counter (Betaplate 1205; Wallac, Gaithersburg, MD). To assess the activation propensity of the different B cell subsets, 2.5 x 104 sorted B cells were incubated with a decreasing titration of irradiated L cells. [3H]Thymidine was added at time zero, and B cell proliferation was assessed at 18 and 48 h.
For Ig measurements, 1.0 x 105 sorted B
cells were incubated with 2.0 x 104
irradiated L cells or anti-Ig
and anti-Ig
light chain Abs
in triplicate, and day 7 or day 10 supernatants were analyzed by ELISA.
Control conditions included triplicate wells with irradiated L cells
alone, B cell subsets alone, stained but unsorted purified B cells, and
whole PBMCs. When assessing the effects of IL-10 on B cells subsets,
triplicate wells were set up as described above, either with or without
IL-10 (100 ng/ml; R&D Systems). Sandwich ELISAs were developed for Ig
measurements with linear range sensitivities as follows: IgA, 500 pg/ml
to 
100 ng/ml; IgM, 2–2000 ng/ml; total IgG, 300 pg/ml to 100 ng/ml;
and IgG subtypes: IgG1, 20–5000 ng/ml; IgG2, 1–750 ng/ml; IgG3,
1–500 ng/ml; and IgG4, 1–5000 ng/ml. Briefly, capture Abs were coated
on Immunlon plates (Dynex, Chantilly, VA) in Tris high salt
(THS)3 buffer at 50
µl/well, kept overnight at 4°C, then blocked with 2% BSA (Sigma,
St. Louis, MO) in THS (100 µl/well) for 2 h at 37°C. Plates
were then washed twice, and supernatant samples and standards (diluted
in culture medium) were added at 50 µl/well and kept overnight at
4°C. Plates were washed twice, and biotin-conjugated secondary Abs
were added at the indicated concentrations in 50 µl/well THS with
Tween 20 (Sigma, St. Louis, MO) and incubated for 45 min. Plates were
then washed three times and incubated with 50 µl/well
avidin-peroxidase (Sigma) diluted as indicated in THS with Tween for 30
min at room temperature. Finally, plates were washed four times,
developed with TMB (Kirkegaard & Perry, Gaithersburg, MD), and stopped
with ELISA stop solution (100 µl/well each) before reading at
450 nm.
Assessing APC function
For MLR assays, the capacity of B cell subsets to activate allogeneic CD4+ T cells was assessed. B cell subsets or control cells were freshly isolated, treated with mitomycin C (100 µg/ml for 2 h) or irradiated (600 rad), and titrated into wells containing 1.0 x 105 freshly purified allogeneic CD4+ T cells in 200 µl complete medium with 10% FCS. Irradiated responders were used as fillers to normalize total cell numbers across wells. Twenty-four-hour [3H]thymidine incorporation was measured on day 5. For Ag-dependent experiments, 1.0 x 105 mitomycin C-treated or irradiated B cell subsets pulsed with glutamate, lysine, alanine, and tyrosine (GLAT; glatiramer acetate, 4 µg/ml, provided by TevaMarion Partners Petach-Tikva, Israel). were incubated with 2.0 x 105 autologous CD4+ T cells in complete medium with 10% pooled human AB serum (PelFreeze, Brown Deer, WI). Twenty-four-hour [3H]thymidine incorporation was measured at the indicated times. In initial experiments no differences in proliferative responses were found between mitomycin C treatment and irradiation of B cells. Irradiation was chosen as the preferred approach for subsequent experiments to avoid further loss of cells incurred by the additional wash required with mitomycin C treatment. Nonirradiated B cell subsets were also used as APCs to further control for the possibility that the B cell subsets differed in their radiosensitivity. For CD80 blocking experiments, F(ab')2 were generated from mIgG1 anti-human CD80 mAb (provided by Mary Collins and Beatriz Carreno, Genetics Institute, Cambridge, MA), using the ImmunoPure F(ab')2 preparation kit (Pierce, Rockford, IL). Protein recovery was determined by using absorbance at 280 nm, and fragment purity was confirmed by gel electrophoresis. Control F(ab')2 were similarly prepared from mouse IgG1 (MOPC-21, Sigma). Where indicated, anti-CD80 or control F(ab')2 were added at 5 µg/ml to triplicate wells at the onset of the cultures.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We studied the phenotype of circulating human B cells using
triple-color immunofluorescent staining of whole blood samples obtained
directly ex vivo. Surprisingly, a substantial frequency of normal adult
B cells (CD19+) expressed significant levels of
CD80, but not CD86 (Fig. 1
A).
In marked contrast, circulating monocytes (CD64+)
expressed, as expected, high levels of CD86, while only a small
fraction (not more than 2%) expressed CD80 (Fig. 1, A and C). The circulating
CD19+CD80+ B cells were
predominantly memory (CD27+) B cells (Fig. 1B, lower right). Naive
(CD27-) B cells expressed equally low levels of
CD80 and CD86. Thirty-four to 57% (mean, 45%; n = 9)
of circulating human adult memory B cells were
CD80+ (Fig. 1C, far right),
representing approximately 12–25% (mean, 17%; n = 9)
of the total circulating B cell pool. We predicted that if these
CD80+ cells developed as a subset of memory B
cells (as opposed to merely representing the subset of
recently activated memory B cells at any given point in time), then
their frequency would increase gradually over time, as has been
previously shown for CD27+ B cells
(17). Indeed, the CD80+ B cell
subset was essentially absent in cord blood (Fig. 1D), and
its frequency in the circulation increased gradually with age, always
representing approximately half the circulating memory
(CD27+) B cell pool (Fig. 1E).
|
Since CD80 has been viewed as a marker of B cell activation, it
was possible that the
CD27+CD80+ cells simply
represented an activated subset of circulating memory cells. However,
the observations that the frequency of these cells increases gradually
over time (Fig. 1
E) and that in the adult they constitute
approximately half the circulating memory B cell pool suggested that
the expression of CD80 on these memory B cells did not merely reflect
recent activation. To directly compare the activation profiles of the
memory B cell subsets and naive B cells, freshly purified B cells from
normal adults were phenotyped by three-color FACS analysis for a range
of activation markers. We found no differences in the activation states
of the CD27+CD80+ and
CD27+CD80- memory B cells.
Both subsets expressed identical levels of CD86, CD25, CD40, and class
II (Fig. 2A). DNA-based cell
cycle analysis using Hoechst 33342 and propidium iodide revealed
equivalent proportions of viable
G0/G1, S phase, and
G2 cells in both memory B cell subsets
(G0/G1: 92 ± 4% for
CD27+CD80+, 88 ± 5%
for CD27+CD80-,
p > 0.3; S phase: 3.3 ± 1.8% for
CD27+CD80+, 5.2 ±
3.6% for CD27+CD80-,
p > 0.4; G2: 5.1 ± 2% for
CD27+CD80+, 6.6 ±
1.8% for CD27+CD80-,
p > 0.4; mean ± SD for four independent
experiments; significance determined by Student’s unpaired
t test). Furthermore, there were no differences between both
memory subsets and the naive CD27- B cells with
respect to the intracytoplasmic expression of Ki67 (Fig. 2A), a nuclear proliferation marker used to distinguish
between G0 and G1 phases of
the cell cycle (30). Together our findings are consistent
with a prior report that circulating human memory B cells are in a
resting, nondividing state (33). We conclude that the
circulating
CD19+CD27+CD80+
cell subset represents the first description of the expression of the
CD80 costimulatory molecule on a population of quiescent memory B
cells.
|
We next wished to study the capacity of the memory B cell subsets
to proliferate and secrete Igs. Since variable expression of
costimulatory molecules on the B cell subsets would induce different
degrees of T cell help, we chose to stimulate B cells in the absence of
T cells. Given the observation that the levels of CD40 were identical
on all B cell subsets (Fig. 2A), we used CD40L-transfected L
cells (provided by Y.-J. Liu, DNAX) to stimulate highly purified B cell
subsets. The kinetics of B cell proliferation were determined by
[3H]thymidine incorporation at different times
following stimulation. Significantly greater proliferative responses
were measured from the
CD27+CD80+ memory B cells
compared with either the
CD27+CD80- memory subset
or the CD27- (naive) B cells during the first 2
days following stimulation (Fig. 3A). In contrast, by the fifth
day following stimulation, proliferation of the
CD27+CD80+ cells reached a
plateau, while CD27+CD80-
cells were proliferating more rapidly. CD27-
(naive) B cells were the slowest to initiate proliferation, but, unlike
the memory subsets, followed an exponential growth curve such that by 5
days poststimulation they were incorporating significantly more
[3H]thymidine than either of the memory cell
subsets. This observation is consistent with prior reports that
CD40-mediated stimulation of naive B cells primarily induces
proliferation, while similar stimulation of memory B cells primarily
promotes differentiation and Ig secretion, a mechanism thought to
prevent B cell repertoire freezing (34).
|
B demonstrates this at the
earliest time point tested (18 h) after stimulation. The
CD27+CD80+ memory cells
were readily induced to proliferate with small numbers of stimulators,
while measurable induction of proliferation in the
CD27+CD80- memory cells
and in the CD27- naive cells required greater
numbers of stimulator cells. Consistent with the observation that the
CD27+CD80+ cells can be
more readily activated is the demonstration that CD72 is expressed at
lower levels on the surface of these memory B cells (Fig. 2B). CD72 has been shown to increase the threshold of B cell
activation by negatively regulating B cell receptor signaling
(35). Lower levels of surface expression of CD72 are
therefore associated with a lower threshold of B cell activation.
To study the Ab-producing capacity of the B cell subset, we developed a
sensitive sandwich ELISA assay to quantify the levels of secreted IgA,
IgM, total IgG, and the IgG subclasses (IgG1–4) in culture
supernatants. Following CD40-mediated activation,
CD27- cells secreted essentially no Igs,
consistent with their naive state and previous reports
(16, 17, 18). While both memory subsets secreted considerable
amounts of IgA, IgG, and IgM, the CD80+ subset
consistently secreted substantially greater levels of all isotypes
measured, which was also true for all IgG subclasses (Fig. 4A). These increases in Ab
secretion were reproduced with Pokeweed mitogen and with more
physiological stimuli such as B cell Ag receptor cross-linking with
anti-Ig light chain Abs or stimulation of the freshly isolated B
cell subsets with CD3-activated autologous CD4+ T
cells. In all cases, between 3.5- and 24-fold more Igs were secreted by
the CD27+CD80+ subset
compared with the
CD27+CD80- subset across
all isotypes and subclasses (data not shown). In these experiments
essentially no Ig secretion was measured when the freshly isolated B
cell subsets were cultured in the absence of stimulation (e.g., using
mock-transfected L cells or unstimulated CD4+ T
cells). These observations confirm that while poised to secrete large
amounts of Igs, the
CD27+CD80+ B cells
circulate as a unique subset of memory B cells and not a population of
cells signaled in vivo to differentiate into Ab-producing plasma cells.
This is consistent with prior reports that CD27+
B cells are not terminally committed to plasma cell differentiation
(36).
|
B). Thus,
independent of their potential to costimulate T cells (e.g., via
CD80/CD28 interactions) and thereby recruit additional T cell help, the
CD80+ memory B cell subset secreted
class-switched Igs much more readily and to significantly higher levels
than the CD80- subset, stimulated in the
same way. CD27+ CD80+ memory B cells are efficient activators of T cells in keeping with their distinct phenotype
We next examined the relative capacity of the B cell subsets to
function as APCs. In initial experiments freshly sorted B cell subsets
(CD27-,
CD27+CD80-,
CD27+CD80+) or whole B
cells from the same normal donors were either treated with mitomycin C,
or irradiated and then incubated with highly purified allogeneic
CD4+ T cells. T cell proliferation was assessed
at 5 days by [3H]thymidine incorporation. A
clear hierarchy of Ag presenting capacity was identified (Fig. 5A). As predicted, the
CD27+CD80+ memory B cell
subset induced significantly greater proliferation than the
CD27+CD80- subset, even at
the low stimulator:effector ratio of 1:50 (2,000 B cells:100,000 T
cells). Naive B cells were relatively poor APCs, consistent with
previous reports (11, 12). Since it is known that compared
with naive B cells murine memory B cells can acquire variable degrees
of radioresistance (11), we included nonirradiated B cell
subsets as control APCs. The proliferative response to the each B cell
subset was not significantly different whether the subset was
irradiated or not (Fig. 5A), confirming that proliferation
by B cells had a negligible contribution to the measured proliferation
and that, importantly, the higher proliferation induced by the
CD80+ memory cells was due to their enhanced APC
capacity, and not to their relative radioresistance.
|
B). The engagement of CD27 on B cells and of CD70 on T
cells is known to play an important role in mediating productive
responses of both B cells and T cells (37, 39). Thus,
despite their resting state, the
CD27+CD80+ memory B cells
are distinguished by several features that promote their ability to
interact with and efficiently activate T cells.
Finally, we wished to confirm the enhanced APC capacity of the
CD27+CD80+ B cells in an
Ag-dependent system. Because assessment of Ag-dependent interactions
between human B cells and T cells is often limited by the low precursor
frequencies of responding cells, we chose to use a random copolymer of
GLAT, a well-characterized Ag shown to induce high frequency
CD4+ T cell responses that are mediated through
the TCR in an MHC class II-restricted fashion (40, 41). We
used freshly sorted B cell subsets to present GLAT to autologous
CD4+ T cells and defined the kinetics of T cell
proliferation. Fig. 5B demonstrates that
CD27+CD80+ B cells elicited
significantly stronger and more rapid Ag-dependent T cell responses
than CD27+CD80- cells.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In a recent report McHeyzer-Williams et al. (42) elegantly described a novel memory B cell subset in the murine system, defined as surface B220-. The authors tracked the development of the B220- memory B cells following recall antigenic exposure and identified them as major constituents of the non-Ab-secreting, quiescent memory B cell pool. Upon antigenic rechallenge, these B220- cells displayed a lesser degree of proliferation, but a more robust Ab response compared with the B220+ memory subset, reminiscent of the differential responses of our CD80+ and CD80- human memory subsets, respectively. While the profile of costimulatory molecule expression on the murine memory subsets was not reported, it is also interesting to note that a subpopulation of the B220- memory cells was shown to express high levels of CD11b, similar to the CD80+ memory subset in the current study. In contrast to the circulating human CD27+CD80+ memory B cell subset, however, the B220- murine memory subset, studied in the bone marrow and spleen, reportedly expressed very low levels of CD19 and did not include IgM-secreting cells. While these murine and human B cell subsets are unlikely to represent equivalent memory B cell subpopulations, the study by McHeyzer-Williams et al. (42) and our current study both provide novel insights into the phenotypic and functional heterogeneity that exists within the memory B cell compartment. Our findings are underscored by the recent demonstration in the murine system by Harris et al. (43) of distinct effector B cell subsets (Be1 and Be2) that are able to differentially regulate T cell responses.
Several theories have supported an important role for B cells in propagating immune responses, in particular their ability to participate in the activation of naive or primed T cells in an Ag-dependent fashion. Janeway and Mamula (44) proposed a model in which B cells are viewed only as second line APCs after the more professional dendritic cells. More recently, Bretscher (45) extended his original two-signal model of T cell activation to the two-step, two-signal model (8) in which he advocates an important role for memory B cells in the second step of the activation of precursor Th cells. Shared by all prevailing models of precursor T cell activation is the general assumption that any effective B cell contribution requires preactivation of the B cells with the consequent up-regulation of inducible costimulatory molecules (46). Our demonstration of the expression of high levels of costimulatory molecules on a subset of quiescent circulating memory B cells may help to reconcile some of the differences between existing models. These CD27+CD80+ B cells are able to readily contribute to rapid and productive B cell-T cell interactions and stimulate efficient Ag-dependent CD4+ T cell responses without requiring an immediate preactivation step.
Taking advantage of the identical levels of CD40 expression on the B cell subsets, our CD40L-mediated activation assays demonstrate that CD27+CD80+ memory B cells are capable of secreting very large amounts of Igs compared with the CD80- memory B cells. This is independent of the additional T cell help that the CD27+CD80+ memory B cells may be able to recruit by virtue of their costimulatory molecule expression. Moreover, these differences are further amplified with the addition of the regulatory cytokine IL-10, which preferentially and powerfully augments Ab secretion from the CD80+ subset. The enhanced Ig secretion from the CD80+ memory B cells is not restricted to particular isotypes or IgG subclasses, suggesting that the capacity to mount robust Ab effector responses is a broadly defining characteristic of this memory subpopulation.
Although circulating in a resting state, the CD80+ memory B cells possess a cell surface phenotype that predicts differential activation requirements and an enhanced capacity to interact with their environment. Decreased levels of surface CD72 expression are consistent with a lower threshold of activation, as is illustrated in our findings of the enhanced responsiveness to suboptimal stimuli and the more rapid proliferation kinetics of the CD27+CD80+ memory subset compared with the CD27+CD80- subset and the naive B cells. The high levels of expression of the integrin family adhesion molecule/type 3 complement receptor, CD11b, coexpressed on CD80+ cells, may reflect preferential migration patterns or enhanced complement fixing capacity. CD11b has also been described as a coreceptor for the B cell Ag receptor (47). The increased levels of CD27 enable more efficient interactions with T cells through CD27-CD70 interactions that may further contribute to the ability of these CD80+ memory B cells to efficiently stimulate T cells as well as differentiate into powerful Ab-secreting cells.
The mechanisms for the development of the distinct CD27+CD80+ and CD27+CD80- human memory B cell subsets and their respective roles remain speculative. One possibility is that the two subsets develop at the same time during the initial antigenic encounter. It is generally accepted that when a naive B cell encounters Ag during the germinal center reaction, its progeny will differentiate along either the plasma cell pathway or the memory B cell pathway (48). Little is known, however, about the potential for distinct memory B cell subsets to diverge at that time. One hypothesis is that the CD27+CD80+ and CD27+CD80- B cells develop in such a way to equip the immune system with functionally distinct memory B cells that share the same antigenic specificity, but respond to their target in different contexts, such as unique tissue microenvironments.
Alternatively, the phenotypic and functional differences between the two memory B cell subsets may reflect differences in the number of times that they encountered their unique Ags. In this context, studies of CD80 expression on T cells have shown that CD80 is not expressed on normal naive T cells or following single activation of naive T cells, but can be demonstrated on the surface of normal T cells following the many cycles of antigenic stimulation associated with T cell cloning (49). In the case of B cells, it is well accepted that both CD80 and CD86 expression can be transiently induced upon naive B cell activation (25, 26, 27). It is possible, however, that CD80 (unlike CD86) becomes chronically expressed on B cells upon sufficient repeated antigenic exposure and remains expressed even after the memory B cell has returned to a relatively quiescent state. In this way the immune system may ensure that frequently encountered Ags are met with a particularly rapid B cell response that is also capable of efficiently recruiting a cognate T cell response.
In conclusion, we define a novel CD27+CD80+ B cell subset that develops gradually after birth and comprises a substantial portion of the human adult circulating memory B cell compartment. Although in a resting state, these B cells have a lower threshold of activation and are poised to deliver vigorous memory effector responses. The unique expression of high levels of costimulatory and adhesion molecules on the surface of these resting CD27+CD80+ memory B cells provides a mechanism by which they may contribute to the efficient Ag-dependent T cell responses observed. As such, this novel memory B cell subset can mediate the rapid and robust immune responses that together constitute the hallmarks of adaptive immune memory. In this context, reports of increased frequencies of CD80+ B cells in human autoimmune disease (50, 51) and observations that autoreactive B cells may function as effective APCs (10, 52) raise the possibility that dysregulation of the CD80+ memory B cell subset may contribute to autoimmune pathogenesis. Further studies will be required to elucidate the potential roles of these cells in both health and disease, particularly at a time when novel agents are being developed to target the B7 costimulatory pathway in human clinical trials (24).
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Amit Bar-Or, Neuroimmunology Unit, Montreal Neurological Institute, McGill University, 3801 University Street, Montreal, Quebec, Canada H3A 2B4. E-mail address: abar-or{at}mni.mcgill.ca
3 Abbreviations used in this paper: THS, Tris high salt; CD40L, CD40 ligand; GLAT, glutamate, lysine, alanine, and tyrosine.
Received for publication May 24, 2001. Accepted for publication September 20, 2001.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-1b therapy. J. Clin. Invest. 99:2664.[Medline]
This article has been cited by other articles:
|
|
 |
|
  M. M. Tomayko, S. M. Anderson, C. E. Brayton, S. Sadanand, N. C. Steinel, T. W. Behrens, and M. J. Shlomchik Systematic Comparison of Gene Expression between Murine Memory and Naive B Cells Demonstrates That Memory B Cells Have Unique Signaling Capabilities J. Immunol., July 1, 2008; 181(1): 27 - 38. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N.-H. Chang, T. McKenzie, G. Bonventi, C. Landolt-Marticorena, P. R. Fortin, D. Gladman, M. Urowitz, and J. E. Wither Expanded Population of Activated Antigen-Engaged Cells within the Naive B Cell Compartment of Patients with Systemic Lupus Erythematosus J. Immunol., January 15, 2008; 180(2): 1276 - 1284. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. M. Anderson, M. M. Tomayko, A. Ahuja, A. M. Haberman, and M. J. Shlomchik New markers for murine memory B cells that define mutated and unmutated subsets J. Exp. Med., September 3, 2007; 204(9): 2103 - 2114. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Duddy, M. Niino, F. Adatia, S. Hebert, M. Freedman, H. Atkins, H. J. Kim, and A. Bar-Or Distinct Effector Cytokine Profiles of Memory and Naive Human B Cell Subsets and Implication in Multiple Sclerosis J. Immunol., May 15, 2007; 178(10): 6092 - 6099. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Hausl, R. U. Ahmad, M. Sasgary, C. B. Doering, P. Lollar, G. Richter, H. P. Schwarz, P. L. Turecek, and B. M. Reipert High-dose factor VIII inhibits factor VIII-specific memory B cells in hemophilia A with factor VIII inhibitors Blood, November 15, 2005; 106(10): 3415 - 3422. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Wulff, H.-G. Knaus, M. Pennington, and K. G. Chandy K+ Channel Expression during B Cell Differentiation: Implications for Immunomodulation and Autoimmunity J. Immunol., July 15, 2004; 173(2): 776 - 786. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. E. Duddy, A. Alter, and A. Bar-Or Distinct Profiles of Human B Cell Effector Cytokines: A Role in Immune Regulation? J. Immunol., March 15, 2004; 172(6): 3422 - 3427. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Nishimura, T. Igarashi, N. L. Haigwood, R. Sadjadpour, O. K. Donau, C. Buckler, R. J. Plishka, A. Buckler-White, and M. A. Martin Transfer of neutralizing IgG to macaques 6 h but not 24 h after SHIV infection confers sterilizing protection: Implications for HIV-1 vaccine development PNAS, December 9, 2003; 100(25): 15131 - 15136. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. F. Fecteau and S. Neron CD40 Stimulation of Human Peripheral B Lymphocytes: Distinct Response from Naive and Memory Cells J. Immunol., November 1, 2003; 171(9): 4621 - 4629. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Alter, M. Duddy, S. Hebert, K. Biernacki, A. Prat, J. P. Antel, V. W. Yong, R. K. Nuttall, C. J. Pennington, D. R. Edwards, et al. Determinants of Human B Cell Migration Across Brain Endothelial Cells J. Immunol., May 1, 2003; 170(9): 4497 - 4505. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
U. Klein, Y. Tu, G. A. Stolovitzky, J. L. Keller, J. Haddad Jr., V. Miljkovic, G. Cattoretti, A. Califano, and R. Dalla-Favera Transcriptional analysis of the B cell germinal center reaction PNAS, March 4, 2003; 100(5): 2639 - 2644. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K N Potter, C I Mockridge, A Rahman, S Buchan, T Hamblin, B Davidson, D A Isenberg, and F K Stevenson Disturbances in peripheral blood B cell subpopulations in autoimmune patients Lupus, December 1, 2002; 11(12): 872 - 877. [Abstract] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
),
the CD80+CD27+ memory B cell subset represents
34–57% of the total circulating memory B cell pool (far
right). D, Consistent with their development as
a memory B cell subset, the CD80+ B cells are absent from
cord blood. E, Depicts percentage of total circulating
CD19+ B cells that are memory B cells
(CD19+CD27+, 
) or CD80+ memory B
cells (CD19+CD27+CD80+, 
) over
time. Both populations increase gradually into adulthood. Power curve
fit.
;
p < 0.01) and the CD27- naive B cells
(
; p < 0.001). In contrast,
CD27+CD80- B cell proliferation was not
detected until 48 h. Later, however,
CD27+CD80- B cell proliferation was
significantly greater than that measured for
CD27+CD80+ B cells (p <
0.02 at 120 and 168 h). CD27- naive B cells followed
an exponential proliferation kinetic, with little measured by 48
h, but significantly greater proliferation than both memory subsets at
subsequent times (p < 0.02 at 120 h;
p < 0.002 at 168 h). B,
CD27+CD80+ memory B cells have a low threshold
of activation; a hierarchy of activation is observed even at the
earliest time point assessed (18 h) poststimulation. In contrast to the
CD27- naive B cells (
) and
CD27+CD80- memory B cells (
),
CD27+CD80+ memory B cells (
) are induced to
proliferate with low numbers of stimulators (CD40L-transfected L
cells). 
, L cells alone. Significance was determined by Student’s
unpaired, two-tailed t test. Data are representative of
four independent experiments. There was no background proliferation in
cultures with L or B cells only (marked by X in
A).
