| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
The Walter and Eliza Hall Institute of Medical Research, Royal Melbourne Hospital, Victoria, Australia
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsReferences |
|---|
Memory and GC B cells can be distinguished on the basis of several
physical characteristics. GC B cells are activated and as such possess
the cell surface characteristics of activated B cells; they are larger,
bind higher levels of the lectin peanut agglutinin (PNA) (5), and
express higher levels of CD24 (heat-stable Ag) (6) and the
costimulatory Ag CD86 (B7.2) than both naive and memory B cells (7).
Memory B cells and GC B cells have other characteristics in common,
such as mutated V genes and isotype-switched Ig heavy chains (reviewed
in 2 . Histology can also be used to locate Ag-specific GCs (8)
and consequently determine the duration of the GC reaction. Using such
an approach, Bachmann et al. (9) have observed GCs that are specific
for a viral Ag for 
100 days after primary immunization, albeit at
decreasing levels. These authors postulated that this long-term GC
reaction was part of the process of maintaining memory or serum Ab
titers and could be unique to the particular Ag used (9). In histologic
studies, however, the quantification of cell populations is
difficult.
Previously we and others have reported the kinetics with which
Ag-specific B cells appear and disappear during the primary response to
model T cell-dependent Ags (10, 11, 12, 13, 14, 15). These studies revealed that the
frequency of B cells that are specific for a haptenic Ag peaked in the
spleen at around day 14 and then declined rapidly during the next 5 wk
before stabilizing over the next 200 to 300 days to around 1% of the
level seen at day 14 (15, 16). Although not formally proven, it was
assumed that the memory B cell population was established during the
phase of rapid decline. Since it was not possible to distinguish
between memory and GC B cells in these experiments, these data do not
resolve the two possible models of generating a memory B cell
population. In one model, the GC reaction ceases when Ag-specific Ab is
in excess of Ag. This Ab could compete with GC B cells for Ag and
consequently block access to Ag on the follicular dendritic cells
and/or deliver a negative signal through Fc
RII (reviewed in 17 , thereby terminating the GC reaction. In this way, the memory
population could be generated as a result of the synchronous
dissolution of the GCs. In a second model, memory B cells are
continually produced by the GC. During the early stages of the
response, when Ag is in excess, emigres may reenter the GC for
additional rounds of mutation and selection; however, at later stages,
they remain outside, leading to a memory population that accumulates in
an inverse relationship with the decline of the GC B cell
population.
We have used a flow cytometry-based system to study the association between the expression of the membrane glycoprotein CD38 (18, 19) and the generation of B cell memory during the primary response to the hapten (4-hydroxy-3-nitrophenyl)acetyl (NP) in C57BL/6 mice. First, we confirmed that CD38 expression is down-regulated on Ag-specific GC B cells (20, 21). Subsequently, we found that it was reexpressed on a progressively increasing fraction of the memory B cell population. At all times, the CD38low isotype-switched Ag-specific B cells exhibited the phenotype of activated cells and had reduced survival potential in vitro, which was typical of GC B cells. Their CD38high counterparts, however, displayed characteristics of resting memory B cells. By determining the proportion of Ag-specific IgG1+ B cells that were CD38high at different times after immunization, we were able to plot the development of the memory population in relation to the decline of the GC reaction. Our results indicate that the formation of the memory population is a gradual process that occurs over many weeks.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsReferences |
|---|
C57BL/6 mice that had been bred and maintained at our institute were used when they reached 8 to 12 wk of age. Hemizygous transgenic mice of the Eµ-bcl-2–22 strain (22) that had been backcrossed onto the C57BL/6 background for more than 18 generations were provided by Dr. A. W. Harris. Mice were immunized i.p. with 100 µg of alum-precipitated NP that was conjugated to keyhole limpet hemocyanin (KLH) (NP/KLH conjugation ratio = 13:1) as previously described (12).
Preparation of cell suspensions
Mice were killed by cervical dislocation on the stated days after immunization, spleens were removed, and single-cell suspensions prepared as previously described (14). Briefly, spleens were forced through a metal mesh and suspended in PBS containing 1% FCS. Viable cells were counted using a light microscope and trypan blue exclusion.
Enrichment of IgG1+ spleen cells
For cell-sorting experiments, surface IgG1+ spleen cells were preenriched according to the following procedure. Live cells were separated by density centrifugation using 1.09 g/cm3 of Nycodenz (Nycomed Pharma, Oslo, Norway) and subsequently stained with rat anti-mouse IgG1 magnetic cell separation beads (Miltenyi Biotec, Bergisch Gladbach, Germany) in the presence of 10 µg/ml mouse IgM myeloma protein (MOPC-104E, Bionetics, Kensington, MD) to reduce background staining. Magnetically stained cells were isolated on a mini magnetic cell separation column (MiniMACS, Miltenyi Biotec) according to the manufacturer’s protocol. The isolated fraction was highly enriched for IgG1-expressing cells, such that between 20 and 50% of cells were positive depending upon the time point after immunization. The cells were still stainable with fluorochrome-labeled goat anti-mouse IgG1 after isolation.
Culture conditions
B220+ spleen cells were sorted into flat-bottom microculture plates (Falcon, Becton Dickinson, San Jose, CA) at a concentration of 105 cells/well and cultured in RPMI 1640 supplemented with 5% heat-inactivated FCS and 10-4 M 2-ME. Some cultures were stimulated either with goat anti-mouse IgM F(ab')2 fragments (Jackson ImmunoResearch, West Grove, PA) at a final concentration of 20 µg/ml or with a culture supernatant from the hybridoma FGK-45 (a gift from A. Rolink, Basel Institute of Immunology, Basel, Switzerland), which produces rat anti-mouse CD40 Abs. Both of these stimulations were performed in the presence of murine IL-4 that had been added in the form of a culture supernatant from transfected X63 cells (kindly provided by Dr. Fritz Melchers) (23). In activation experiments, cultures were harvested at days 1 and 3, and cells were analyzed on the FACScan (Becton Dickinson) (see below).
Cell viability assay
Between 1000 and 5000 sorted cells/well were placed in Terasaki plates (Nunc, Naperville, IL) in a final volume of 10 µl medium (RPMI 1640 containing 5% FCS and 2-ME). The cultures were incubated at 37°C. At the indicated time points after the initiation of the culture, the plates were removed from the incubator, and propidium iodide (PI) was added to a final concentration of 0.5 µg/ml. Thereafter live (intact) and dead (PI-stained) cells were counted using a UV microscope. At each time point, PI was added to fresh cultures that had not previously been removed from the incubator, and between 150 and 700 cell bodies were counted.
Immunofluorescence and flow cytometry
NP-specific GC or memory cells, defined as
IgM-IgD-IgG1+ NP-binding
B220+, were resolved exactly as previously described (14).
When a seventh parameter was needed for analysis, the same cells were
revealed with 11–26C (anti-IgD) biotin, 331-12 biotin,
streptavidin-tricolor (Caltag, San Francisco, CA), anti-B220
phycoerythrin (PE) (RA3-6B2, PharMingen, San Diego, CA), NP
allophycocyanin, anti-IgG1 Texas Red (Southern Biotechnology
Associates, Birmingham, AL), and either NIMR5/18 (anti-CD38)
fluorescein or PNA fluorescein (Vector, Burlingame, CA). Dead cells
were excluded with PI staining. These PI+ dead cells were
highly fluorescent and were therefore detectable in both the tricolor
and PE channels (see Figs. 2
and 5). Ab-forming cells (AFCs) were
detected using 281.2 (anti-syndecan) fluorescein or
allophycocyanin. Staining was conducted as previously described (16).
Cells were finally resuspended in PBS containing FCS (3%) and PI (1
µg/ml) to exclude dead cells. Analysis and sorting were preformed on
a FACStarPlus (Becton Dickinson). B cells activated in
vitro were stained with anti-B220 (RA3-6B2, PharMingen) PE, GL-1
(anti-B7-2/CD86) fluorescein or NIMR5/18 fluorescein and analyzed
on a FACScan (Becton Dickinson).
|
|
Splenic tissues were prepared as previously described (14) and
stained using either PNA biotin (Vector) or NIMR5/18 (anti-CD38)
biotin followed by streptavidin-alkaline phosphatase (AP) (Southern
Biotechnology Associates). Purified RA3-6B2 (anti-B220) was
followed by mouse anti-rat Ig-
-horseradish peroxidase (HRP)
(PharMingen). The protocol used for staining sections was as previously
described (14). HRP staining was detected using the Vectastain kit
(Vector), and streptavidin staining was visualized with the Vectastain
AP kit III, fast blue (Vector), in the presence of levamisole.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsReferences |
|---|
The original aim of this work was to determine whether CD38 had a
role in either GC reaction or memory B cell generation. Preliminary
results indicated that isotype-switched, Ag-specific B cells isolated
from the spleens of C56BL/6 mice at 14 days postimmunization did not
proliferate in response to stimulation with anti-CD38 Abs in the
presence of IL-4 and IL-5, although
IgM+IgD+ B cells from the same mice did
proliferate (data not shown). Since the majority of Ag-specific B cells
in the spleen are isotype-switched GC B cells at this time after
immunization (11, 24), we assessed whether the lack of proliferation
was due to the absence of CD38 surface expression. We determined that
this was indeed the case, as indicated by immunohistologic staining of
sequential splenic sections from mice at 14 days after primary
immunization with NP13-KLH (Fig. 1, A and B)
using anti-B220 in combination with either anti-CD38 or the
lectin PNA (5). Strong CD38 staining of the B cell follicles was
apparent, but there was virtually no overlap with the PNA+
GCs. The observation that murine GC B cells are CD38-
confirmed the recent report of Oliver et al. (21).
|
.
Ag-specific B cells, defined as
IgM-IgD-B220+IgG1+
NP-binding, were gated and analyzed for CD38 expression. At days 6 and
12 postimmunization, NP-specific IgG1+B220high
cells are mainly GC cells (6, 12), and the majority of those GC cells
had down-regulated CD38 expression relative to the level on naive B
cells. However, at day 34 postimmunization, we found a dichotomy in
CD38 expression among NP-specific IgG1+ B cells. Indeed, at
successively later time points, an increasing fraction of NP-specific B
cells were CD38high. Figure 3
illustrates the decrease in the number of splenic NP-specific
IgG1+ cells with time (Fig. 3A) as well
as the simultaneous increase in the proportion which
CD38high (Fig. 3B). A quantification of
these data is given in Table I. After day
85, the percentage of NP-binding IgG1+ cells that were
CD38high plateaued at <100%, indicating the persistence
of the GC B cell phenotype (see below).
|
|
The persistence of B cells with a GC phenotype for many weeks
after primary immunization could represent either the persistence of
GCs themselves or the generation of two phenotypically distinct types
of memory B cells. To distinguish between these possibilities, we
examined a number of characteristics of the CD38high
and CD38low Ag-specific B cells. Comparing the forward
light scatter (FSC) distribution of the CD38high and
CD38low cell populations at 21, 29, and 35 days
postimmunization revealed that CD38low cells always had a
higher FSC distribution; this higher distribution is indicative of a
larger cell size and compatible with these cells being activated (Fig. 4). Furthermore, 7-parameter staining
protocols using either PNA or anti-CD38 were used in parallel at
certain time points. This allowed a direct comparison of these two
methods in determining the proportion of NP-binding IgG1+ B
cells that were of GC or memory phenotype. Both approaches gave
essentially the same result (Fig. 5),
validating the use of the CD38low phenotype as a marker of
GC B cells. Furthermore, these CD38low cells showed
diminished proliferation in response to CD40 stimulation in vitro (data
not shown), which is typical of GC B cells and in agreement with the
findings of Oliver et al. (21).
|
The reduced proliferation of CD38low Ag-specific
B cells in response to CD40-mediated stimulation could result from
either an intrinsic difference in the responsiveness of the two B cell
types or the reduced survival of CD38low B cells in
culture. This latter possibility was tested by performing in vitro
survival experiments. Naive B220+, CD38high,
and CD38low Ag-specific B cells
(IgM-IgD-IgG1+NP+)
were sorted from the spleens of C57BL/6 mice around 30 days
postimmunization and cultured without deliberate stimulation. The
fraction of viable cells in each culture was determined after 24 and
48 h on the basis of PI exclusion. In two independent experiments,
CD38high NP-binding IgG1+ cells were found to
have in vitro survival characteristics that were essentially the same
as those seen in naive B220+ B cells. The survival
potential of CD38low NP-binding IgG1+ cells, on
the other hand, was significantly reduced (Fig. 6). It is likely that this difference
reflects the reduced survival potential of CD38low
IgG1+ B cells and is not a result of the cell-sorting
procedure, since naive B cells stained with anti-B220 only or with
anti-B220 plus anti-CD38 showed identical survival without
stimulation in vitro. As an additional control, purified B cells
expressing a bcl-2 transgene were cultured without
stimulation and were found to survive much longer than any of the other
sorted populations. The reduced survival potential of
CD38low Ag-specific B cells at this time point after
immunization is a further indication that these cells are in an
activated state and is consistent with the phenotype revealed by FACS
(Fig. 4).
|
We subsequently asked whether the down-regulation of CD38 seen on
GC B cells also occurred on the AFCs that develop during the response
to T cell-dependent Ags. Splenic AFCs were identified on the basis of
syndecan (CD138) expression (6, 12) at 8 days postimmunization with
NP13-KLH, at which time the vast majority were
contained within extrafollicular foci (11, 25). CD38 was observed to be
down-regulated on all AFCs (syndecan+) relative to its
level on naive B cells
(IgM+IgD+syndecan-) (Fig. 7). Interestingly, the level of CD38
expression on AFCs appeared to decrease in parallel with the level of
surface IgM and IgD. This observation is consistent with the hypothesis
that CD38 down-regulation is associated with Ig isotype switch (21).
Thus, in the early stages of the immune response to T cell-dependent
Ags, CD38 down-regulation is common to both the intra- and
extra-follicular pathways of B cell differentiation.
|
An analysis of Ag-specific B cells indicated that CD38
down-regulation was a normal outcome of T cell-dependent B cell
activation. Therefore, we determined whether CD38 down-regulation could
be mimicked by in vitro stimulation. B220+ splenic B
cells were purified by FACS and stimulated with anti-IgM or
anti-CD40 Abs in the presence and absence (data not shown) of IL-4.
At the initiation of culture, all B220+ cells expressed
CD38 (data not shown). After 1 day, CD86 (B7-2) was up-regulated on B
cells that had been stimulated with either anti-CD40 or
anti-IgM, whereas there was no change in CD38 expression (data not
shown). At day 3 of culture, however, both anti-CD40 and
anti-IgM stimulated B cells had down-regulated surface CD38
compared with unstimulated B cells (Fig. 8). Thus, both B cell receptor
crosslinking and stimulation through CD40 induced the down-regulation
of CD38, indicating that the down-regulation seen in vivo was a
consequence of B cell activation and was not due to changes in the
environment of the B cells. In repeated experiments, stimulation
through CD40 induced a less marked down-regulation of CD38 than did
anti-IgM stimulation, indicating that these two stimuli have a
somewhat different impact on CD38 expression. Stimulating B cells with
anti-CD40 or anti-IgM in the absence of IL-4 resulted in a
modulation of CD38 expression that was identical with that seen in the
presence of IL-4 (data not shown), indicating that IL-4 is not critical
to this process.
|
In this report we have demonstrated that memory B cells can be distinguished from their GC precursors on the basis of the level of expression of CD38, and that this distinction can be used to monitor the development of the Ag-specific memory B cell compartment. That the CD38low and CD38high phenotypes are indicative of GC and memory B cells, respectively, was confirmed by several criteria: Ag-specific isotype-switched CD38low cells were found to have cell surface characteristics that were similar to GC B cells, such as the expression of high levels of B220 (6) and the ability to bind high levels of PNA. Additionally, the CD38low cells were larger than their CD38high counterparts as determined by FSC but had poor survival in vitro, both of which are characteristic of GC B cells. CD38 down-regulation, however, is not an exclusive property of GC B cells, since a similar alteration in expression was observed on the AFCs found in the spleen during the early stages of the immune response.
Previous studies on the formation of the memory B cell compartment
determined that an immune response could be transferred adoptively to a
naive recipient at 2 wk postimmunization with a hapten-carrier
conjugate (3, 4), and that this response resided primarily in the
PNA-binding fraction (4). After 4 wk, immunity was predominantly seen
among the IgD- (3), PNA-nonbinding (4) fraction. This
observation presumably reflects the formation of a recirculating memory
compartment during this interval. Another measure of the generation of
a memory compartment is provided by a study in which the frequency of
proliferating, isotype-switched, Ag-specific B cells was determined at
various times after immunization with a protein Ag (10). It was found
that the initial 4 wk of the response was a period of extensive
proliferation of isotype-switched Ag-specific B cells; after this, only
a small fraction of such cells was proliferating. The end of the period
of extensive B cell proliferation presumably corresponds to the
formation of the stable, resting memory B cell population. It is of
interest that a small but detectable fraction of Ag-specific B cells
was proliferating at up to 140 days postimmunization, a phenomenon also
observed in an anti-hapten response (8) and an anti-viral
response (9). In these latter cases, the proliferating cells were
localized by histology. Bachmann and colleagues suggested that the
persisting GC B cells were a consequence of the particular Ag used, a
viral-coat protein (9). The results presented in this report indicate
that such a phenomenon is not unique to viral Ags and may be a more
general property of the B cell response to Ag. Using the reappearance
of CD38 on IgG1+ NP-binding B cells as a marker of the
memory phenotype, we have determined that memory B cells become a
discernible population by day 21. From this point on, the proportion of
Ag-specific B cells displaying a memory phenotype continues to
increase, although the fraction of total spleen that these cells
represent changes little (Fig. 3 and Table I). Interestingly, this time
interval coincides with that seen in a study showing that the ablation
of CD4+ T cells before day 21 resulted in a failure to
generate adoptively transferable B cell memory in an anti-hapten
response, while development of memory was unaffected after this time
point (26).
A comparison of the kinetics of memory B cell generation with that of the decline of GC B cells gives some insight into the manner in which memory is generated. Our data, showing that GC and memory B cells coexist, indicate that the generation of memory is not a sudden event triggered by an environmental signal such as the titer of Ag-specific Ig. The memory population is formed reasonably early in the response and then maintained at an approximately constant level while the GC population continues to decline. The question arises as to how the size of the memory population remains unchanged despite an ongoing GC reaction. That is, why is the output of the persistent GCs not reflected in an increase in the size of the memory compartment? There are three possible explanations as to why this might occur. The first is that the GC and memory populations are in equilibrium, so that any input into the memory population is balanced by an equal output that can take the form of death, recirculation back into the GC, or migration out of the spleen. The second possibility is that the output of these GCs is not apparent in the memory population because they have a different fate, such as becoming AFCs. The third explanation is that these are not true GCs in the sense of continuing the process of affinity maturation, but rather are structures associated with, for example, the maintenance of B cell memory. While our data do not address these possibilities directly, some observations can be made. For example, the level of V gene somatic mutation seen among secondary response B cells is equal to that found 21 days into the primary response (15, 27), suggesting that the composition of the memory compartment is fixed at more or less this time point. Thus, the GCs that persist beyond this point either do not contain somatically mutating B cells or their output does not enter the memory population. Do these cells give rise to high affinity AFCs in the bone marrow? Again the answer would appear to be no, since such AFCs are apparent by about day 10 of the response and are all high affinity by day 21, (6, 15) precluding the need for further affinity maturation. Although little is known concerning the life span of bone marrow AFCs after primary immunization, existing evidence suggests that such cells live for extensive periods without replenishment (15, 28). Collectively. these data would argue against the persistent GCs in the spleen contributing to either the memory or the long-term AFC populations. Of the possibilities raised previously, the last is that these structures are important in the maintenance of B cell memory as has been suggested on the basis of histologic studies in the rat (8) and mouse (9). The application of the cell separation techniques outlined in this report will assist in resolving this important question.
While the poor survival of the GC B cells in vitro was expected (29),
it was interesting to find that purified memory B cells survived only
as well as naive B cells. The reason why immunologic memory is long
lived is still not clearly understood. There is debate over whether the
persistence of memory is intrinsic to the memory B cells or is the
result of the interaction of these cells with some component of their
environment. Various groups have reported the up-regulation of the cell
survival-promoting genes bcl-2 and
bcl-XL in memory B cells (30, 31), suggesting
that their expression could confer a degree of longevity to memory B
cells that is not available to GC B cells. While this appears to be the
case (Fig. 6), memory B cells themselves have no particular survival
advantage over naive B cells. Gray and Skarwell (32) found that
transferred memory was short lived in the absence of Ag. Similarly, it
has been reported that the Ab-mediated depletion of nerve growth factor
in a mouse resulted in the almost immediate loss of memory phenotype B
cells (33). Both these latter experiments argue that factors extrinsic
to the memory B cells are important in the maintenance of memory. Our
data support this concept, since memory B cells survive to the same
extent as naive B cells but not nearly as well as B cells
constitutively expressing high levels of Bcl-2. The requirement of
extrinsic signals for the maintenance of B cell memory and the
observation of persistent GC phenotype B cells reported here and
elsewhere (8, 9) may consequently be directly linked.
Our analysis of CD38 expression on mouse B cells during an immune response to a T cell-dependent Ag allows a direct comparison between mouse and human B cells. At first inspection, it would appear that CD38 expression is regulated in a diametrically opposite manner in these two systems. Human naive B cells are CD38- and become CD38+ upon both entry into the GC and after differentiation into AFCs, while memory B cells are again CD38- (see 20 . If human B cell types are grouped according to levels of CD38 expression, then naive and memory B cells constitute one group and GC B cells and AFCs constitute another. Interestingly, the same groupings can be made for mouse B cells, although their CD38 levels are inverted. The inconsistent expression pattern on human and mouse B cells raises questions about the role of CD38 in B cell activation and differentiation. CD38 is an ectoenzyme that catalyzes both the conversion of nicotinamide adenine dinucleotide to cyclic ADP-ribose, which can mobilize Ca2+, and also the hydrolysis of cyclic ADP-ribose to form adenosine diphosphoribose (34, 35). Abs to CD38 induce a number of effects on B lymphocytes including proliferation, protection from apoptosis, and the inhibition of B lymphopoiesis (19, 35). How these activities relate to the in vivo role of CD38 is currently unclear, although there has been some speculation that the down-regulation may result from the internalization of CD38 along with the B cell receptor (36). A consequence of this internalization may be that the second messengers generated by the enzymatic activity of CD38 could operate intracellularly. One final point to note concerning the low level of CD38 on mouse GC B cells is the support this lends to the recent observation of neoteny by GC lymphocytes (37, 38). These investigators noted that GC B cells display a number of phenotypic characteristics that are otherwise restricted to immature lymphocytes. The level of CD38 found on pro-B and pre-B cells is lower than that found on naive recirculating B cells and is similar to that of GC B cells (Ref. 18 and our unpublished observations).
In conclusion, our results have shown that CD38 is selectively down-regulated on splenic GC B cells and AFCs. However, it is reexpressed on B cells that are derived from the GC, namely memory B cells. On the basis of this observation, we have shown that the memory B cell compartment is formed during wk 3 of the response to a nonreplicating hapten-protein Ag, but that B cells with a GC phenotype persist for many weeks after the initial immunization. The resolution of GC and memory B cells is an important step in determining the criteria necessary for the transition from one differentiation state to another.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. David Tarlinton, The Walter and Eliza Hall Institute of Medical Research, P.O. Royal Melbourne Hospital, Victoria 3050, Australia. E-mail:
3 Abbreviations used in this paper: GC, germinal center; AFC, Ab-forming cell; KLH, keyhole limpet hemocyanin; NP, (4-hydroxy-3-nitrophenyl)acetyl; PI, propidium iodide; PNA, peanut agglutinin; PE, phycoerythrin; AP, alkaline phosphatase; HRP, horseradish peroxidase; FSC, forward light scatter.
Received for publication October 10, 1997. Accepted for publication January 9, 1998.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsReferences |
|---|
RIIb1. Curr. Opin. Immunol. 8:378.[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]
|
|
|
|
|
|
T. Onodera, J. C. Poe, T. F. Tedder, and T. Tsubata CD22 Regulates Time Course of Both B Cell Division and Antibody Response J. Immunol., January 15, 2008; 180(2): 907 - 913. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Pellegrini, N. Guinazu, M. P. Aoki, I. C. Calero, E. A. Carrera-Silva, N. Girones, M. Fresno, and S. Gea Spleen B cells from BALB/c are more prone to activation than spleen B cells from C57BL/6 mice during a secondary immune response to cruzipain Int. Immunol., December 1, 2007; 19(12): 1395 - 1402. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. F. Fischer, P. Bouillet, K. O'Donnell, A. Light, D. M. Tarlinton, and A. Strasser Proapoptotic BH3-only protein Bim is essential for developmentally programmed death of germinal center-derived memory B cells and antibody-forming cells Blood, December 1, 2007; 110(12): 3978 - 3984. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Bhattacharya, M. T. Cheah, C. B. Franco, N. Hosen, C. L. Pin, W. C. Sha, and I. L. Weissman Transcriptional Profiling of Antigen-Dependent Murine B Cell Differentiation and Memory Formation J. Immunol., November 15, 2007; 179(10): 6808 - 6819. [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]
|
|
|
|
|
|
I.-J. Kim, C. E. Burkum, T. Cookenham, P. L. Schwartzberg, D. L. Woodland, and M. A. Blackman Perturbation of B Cell Activation in SLAM-Associated Protein-Deficient Mice Is Associated with Changes in Gammaherpesvirus Latency Reservoirs J. Immunol., February 1, 2007; 178(3): 1692 - 1701. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. P. Chappell and J. Jacob Identification of memory B cells using a novel transgenic mouse model. J. Immunol., April 15, 2006; 176(8): 4706 - 4715. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Shimoda, T. Li, J. P. S. Pihkala, and P. A. Koni Role of MHC Class II on Memory B Cells in Post-Germinal Center B Cell Homeostasis and Memory Response J. Immunol., February 15, 2006; 176(4): 2122 - 2133. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Q. Chen and A. C. Ross Inaugural Article: Vitamin A and immune function: Retinoic acid modulates population dynamics in antigen receptor and CD38-stimulated splenic B cells PNAS, October 4, 2005; 102(40): 14142 - 14149. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Inamine, Y. Takahashi, N. Baba, K. Miyake, T. Tokuhisa, T. Takemori, and R. Abe Two waves of memory B-cell generation in the primary immune response Int. Immunol., May 1, 2005; 17(5): 581 - 589. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. J. Blink, A. Light, A. Kallies, S. L. Nutt, P. D. Hodgkin, and D. M. Tarlinton Early appearance of germinal center-derived memory B cells and plasma cells in blood after primary immunization J. Exp. Med., February 22, 2005; 201(4): 545 - 554. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. I. Ellyard, D. T. Avery, T. G. Phan, N. J. Hare, P. D. Hodgkin, and S. G. Tangye Antigen-selected, immunoglobulin-secreting cells persist in human spleen and bone marrow Blood, May 15, 2004; 103(10): 3805 - 3812. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Rosini, G. De Chiara, P. Bonini, M. Lucibello, M. E. Marcocci, E. Garaci, F. Cozzolino, and M. Torcia Nerve Growth Factor-dependent Survival of CESS B Cell Line Is Mediated by Increased Expression and Decreased Degradation of MAPK Phosphatase 1 J. Biol. Chem., April 2, 2004; 279(14): 14016 - 14023. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Cortes and K. Georgopoulos Aiolos Is Required for the Generation of High Affinity Bone Marrow Plasma Cells Responsible for Long-Term Immunity J. Exp. Med., January 20, 2004; 199(2): 209 - 219. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. S. Raman, R. S. Akondy, S. Rath, V. Bal, and A. George Ligation of CD27 on B Cells In Vivo during Primary Immunization Enhances Commitment to Memory B Cell Responses J. Immunol., December 1, 2003; 171(11): 5876 - 5881. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
I.-J. Kim, E. Flano, D. L. Woodland, F. E. Lund, T. D. Randall, and M. A. Blackman Maintenance of Long Term {gamma}-Herpesvirus B Cell Latency Is Dependent on CD40-Mediated Development of Memory B Cells J. Immunol., July 15, 2003; 171(2): 886 - 892. [Abstract] [Full Text] [PDF]
|
|
