| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
*
Department of Microbiology and Immunology, University of Maryland School of Medicine, Baltimore, MD 21201; and
Department of Laboratory Medicine and Section of Immunobiology, Yale University School of Medicine, New Haven, CT 06510
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
1 L-chain
(1+). The Ig H-chains of these early populations of B
cells are encoded by a variety of VH and D exons
undiversified by hypermutation while later, oligoclonal populations are
dominated by mutated rearrangements of the VH186.2 and
DFL16.1 gene segments. To assess directly Ab affinities within these
defined splenic microenvironments, representative VDJ rearrangements
were recovered from B cells participating in the early immune response
to NP, inserted into Ig H-chain expression cassettes, and transfected
into J558L (H-; 1+) myeloma cells. These
transfectoma Abs expressed a remarkably wide range of measured
affinities (Ka = 5 x
104-1.3 x 106 M-1) for NP.
VDJs recovered from both foci and early GCs generated comparable
affinities, suggesting that initial differentiation into these
compartments occurs stochastically. We conclude that Ag normally
activates B cells bearing an unexpectedly wide spectrum of Ab
affinities and that this initial, promiscuous clonal activation is
followed by affinity-driven competition to determine survival and
clonal expansion within GCs and entry into the memory and bone marrow
plasmacyte compartments. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Models for selective entry into the GC compartment have generally focused on lineage commitment (5) or the affinity of the BCR (6, 7). Lineage theories predict intrinsic commitment to the early AFC or GC pathways (3, 8, 9). In contrast, initial BCR affinity may determine the fate of B cells. For example, the intrinsic affinity of the BCR may translate into different degrees of activation, resulting in alternative cellular fates (10, 11). This model, in analogy to the development of thymocytes responding to agonist and antagonist peptides (12, 13, 14), suggests that very low affinity B cells might be capable of forming AFCs but unable to enter the GC/memory cell compartment (15, 16, 17, 18, 19, 20).
Early humoral immune responses are usually characterized by Abs bearing
a wide range of affinities (15, 16, 17, 18, 21, 22, 23, 24). Indeed, a significant
fraction of these early Abs have very low or even undetectable affinity
for the immunizing Ag (17, 18, 25, 26, 27, 28, 29, 30, 31). While clonal selection theories
allow for receptor heterogeneity (32, 33), production of
"unspecific" Ab is unexpected. Very low affinity Abs may result
from nonspecific or bystander effects; several studies of B cell
repertoires and early Ab production have discounted the physiologic
relevance of Abs with association constants below the micromolar range
(Ka < 106 M-1) (10, 19, 34, 35, 36). On the other hand, the presence of early, unspecific Abs
could indicate that the initial affinity requirement for B cell
activation is extremely permissive. Indirect support for this comes
from in vitro work demonstrating the ability of very low affinity
(Ka 
105 M-1) BCRs to
transduce activation signals (29, 37, 38). Additionally, the
participation of low affinity B cells in immune responses is suggested
by the recovery of VDJ rearrangements from GCs that likely encoded Ab
Ka values less than 106
M-1 (39, 40); however, a direct physical assessment of
these affinities has not been performed.
Here we investigate the Ka of BCRs expressed by AFC and early GC B cells specific for the (4-hydroxy-3-nitrophenyl)acetyl (NP) hapten. Authentic VDJ rearrangements, recovered from AFC and GC B cells, were reexpressed as IgG1 transfectoma Abs, and their affinities were measured by fluorescence quenching. These Abs had Ka values for NP ranging from 5.0 x 104 M-1 to 1.3 x 106 M-1; a subset encoded Abs of such low affinity that specific binding could be discerned only when they were expressed as IgM pentamers. Early AFC and GC B cell populations possess equivalent and, often, very low affinities for Ag. The presence of very low affinity B cells in nascent GCs reveals the effective absence of affinity thresholds for cell entry into the differentiation pathway leading to humoral memory. Very low affinity B cells and Ab are physiologically relevant and specifically elicited by Ag. Nevertheless, affinity-driven selection becomes evident later in the GC reaction (41, 42, 43, 44, 45, 46, 47). Together, these observations suggest that the great specificity of humoral immune responses is not the consequence of highly selective clonal activation but of competitive survival and proliferation of higher affinity B cells.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
C57BL/6 mice (6–8 wk of age) were purchased from The Jackson Laboratory (Bar Harbor, ME) or Taconic Farms (Germantown, NY). All mice were maintained in sterile microisolator cages on a 12-h day/night cycle and provided with sterile food and water ad libitum.
Ag and immunizations
The succinic anhydride ester of NP (Cambridge Research
Biochemicals, Cambridge, U.K.) was reacted with chicken 
-globulin
(CG; Sigma, St. Louis, MO). Mice were immunized with a single i.p.
injection of 50 µg NP-CG precipitated in alum. Cohorts of mice were
killed by cervical dislocation at various times after primary
immunization; their spleens were removed, frozen, and processed for
histology and DNA amplification, as described (1, 48). Six immunized
mice provided the VDJ rearrangements expressed as transfectoma Abs
(Table I
).
|
Goat anti-mouse Ig conjugated with horseradish peroxidase
(anti-IgG1-HRP), streptavidin-HRP (SA-HRP), and
streptavidin-alkaline phosphatase (SA-AP) were purchased from Southern
Biotechnology Associates (Birmingham, AL). HRP-labeled peanut
agglutinin (PNA-HRP) was purchased from E-Y Laboratories (San Mateo,
CA). The anti-allotypic Ab RS-3.1 (anti-IgMa) (49)
was a gift from Dr. T. Imanishi-Kari, Tufts University School of
Medicine). The anti-mouse 1 L-chain Ab Ls136 (50) was used to
detect NP-specific serum Ab and B cells in situ (1). The Ls136 and
RS-3.1 hybridomas were grown in culture, and their Ab products were
purified over protein G-Sepharose (Pierce, Rockford, IL). BSA was
purchased from United States Biochemical (Cleveland, OH). The succinic
anhydride esters of NP and its derivative
(4-hydroxy-5-iodo-3-nitrophenyl)acetyl (NIP) (Cambridge Research)
were coupled to BSA as described (48, 51). Substitution ratios for
NP-BSA or NIP-BSA were determined by absorbance at 430 nm. NIP-BSA,
Ls136, and RS-1.3 were biotinylated with
biotin-N-hydroxysuccinamide (Vector Laboratories,
Burlingame, CA) using the manufacturer’s protocol.
Histologic staining and recovery of VDJ rearrangements from individual GC and foci
Serial, 6-µm-thick frozen sections of spleen were cut in a
cryostat microtome (International Equipment, Needham Heights, MA), thaw
mounted onto poly-L-lysine-coated slides, fixed in ice-cold
acetone for 10 min, and stored at -20°C (1, 48). Rehydrated sections
were stained in tandem with PNA-HRP and Ls136-biotin followed by SA-AP.
The identity of NP-reactive AFC and GC cells was confirmed by staining
adjacent sections with NIP-BSA-biotin (data not shown and 48 .
Endogenous peroxidase activity was blocked by a 10-min incubation in
3% H2O2 before staining. HRP and AP activities
were visualized using 3-aminoethylcarbazole and naphthol AS-MX
phosphate/Fast Blue BB (Sigma), respectively (48). NP-specific GCs
(surface +, PNA+) and AFC foci (cytoplasmic
+, PNA-) were identified by bright field
microscopy. Focus and GC B cells were recovered by micromanipulation as
previously described (1, 48).
Cellular material from single GCs or foci were processed, and the
recovered genomic VDJ rearrangements were amplified by PCR (40, 51).
Briefly, an initial round of 40 amplification cycles used primers
complementary to DNA 5' of the transcription start site of V186.2 and
to a region in the JH2-JH3 intron.
Two-microliter aliquots of this reaction mixture were reamplified for
an another 40 cycles using nested primers complementary to the first 20
nucleotides of the V186.2 gene segment and to the terminal portion of
JH2. To facilitate the recovery of B cells carrying
analogue VH gene rearrangements, both 5'-primers share
significant homology (
90%) with other (V186.2 and V3 subgroups) J558
VH gene family members, and the internal 3'-primer binds to
both JH1 and JH2 segments. Amplified DNA was
extracted, cloned into plasmid, and sequenced (40, 48).
Creation of transfectomas and site-directed mutants
Recovered VDJ segments were excised from the sequencing plasmid
pBSK+ (pBluescript; Stratagene, La Jolla, CA) and inserted
into a shuttle vector (pBSK.JJ) to obtain the 5' regulatory region
necessary for Ig H- chain expression (52). The modified VDJ segments
were then excised, purified, and placed into an Ig1 expression
vector carrying the gpt selection marker, pEC1-gpt (46), as
described (52). The resulting constructs were linearized and
transfected into J558L myeloma cells
(H-;1+) by electroporation (53). Growth
under mycophenolic acid selection and subcloning by limiting dilution
was followed by screening for the production of 1+,
1+ Abs by ELISA; positive clones were further scored for
their ability to bind NP and NIP. Some 1+,
1+ transfectoma proteins did not bind NP or NIP in
ELISA, even when highly concentrated and purified.
The B1-8 hybridoma line (50) secretes a canonical, NP-specific
1+/IgM Ab encoded by a V186.2/DFL16.1/JH2
rearrangement (41, 46) The B1-8 VDJ sequence was recovered and cloned
into pEC1; transfection and subcloning into J558L was performed to
make an IgG1 Ab, B1-81. Site-directed mutagenesis (United States
Biochemical) of the B1-81 sequence generated specific base pair
mutations that significantly alter affinity for the NP hapten 1) in
CDR1 at codon 33 (TGG
TTG) that causes a Trp-to-Leu exchange
(H33L1 and 2) in CDR2 at codon 50 (AGGGGG) that yields an
Arg-to-Gly replacement (H50G1). The replacement at position 33
produces a 10-fold increase in Ab affinity (40, 41, 54), whereas the
position 50 exchange dramatically reduces the strength of NP binding
(45, 55, 56). These single substitution VDJ sequences were transfected
into J558L and subcloned.
To create low affinity NP-specific IgM molecules, the position 50
mutant, H50G, and the VDJ rearrangement present in transfectoma T1
(V23/DFL16.1; Table I
) were subcloned into a vector that contains a
11.6-kb Cµ fragment including the intronic enhancer and membrane and
secretory exons (57). The VDJ rearrangements were inserted into this
vector as 4-kb EcoRI fragments that included their
5'-regulatory regions. Construct integrity was confirmed by restriction
digest mapping and sequencing of the VH region. IgM
expression constructs were transfected into J558L by electroporation.
Transfectants secreting 1+ IgM Abs were identified by
ELISA and cloned.
Transfectoma affinity measurements
Abs were purified by concentration of transfectoma supernatants
grown in miniPerm bioreactors (Heraeus, South Plainfield, NJ), followed
by affinity chromatography over protein G-Sepharose. Ab concentrations
and purity were determined by spectrophotometry and
anti-IgG1/anti-1 sandwich ELISA, using B1-81/1 as a
reference standard. All purified proteins were found to consist of
>90% IgG1 activity. Their Ka values for NP and
NIP haptens were measured by fluorescence quenching in a Shimadzu
RF-5301 fluorospectrophotometer (Shimadzu Scientific Instruments,
Columbia, MD) (42, 58, 59). Excitation and emission wavelengths were
280 and 340 nm, respectively; temperature (25°C) and pH (7.4) were
held constant. Titration was conducted by adding NP- or NIP-caproate
(Cambridge Research) over a three-log range
(10-8-10-5 M) to a known concentration of Abs
(33 µM in 2.5 ml PBS) in quartz cuvettes. To correct for nonspecific
quenching, MOPC 21, an anti-dextran monoclonal mouse IgG1/
, and
MOPC 104e, an anti-trinitrophenol monoclonal mouse IgM/1
(Sigma), were used. Emission signal loss, or quench, vs Ag
concentration was plotted according to the Scatchard equation to derive
association constants (Ka values) for
half-maximal binding (42, 58). All samples were measured in triplicate,
and affinities were expressed as mean Ka values
(±SD).
Measurement of functional avidity of IgG1 and IgM Abs
B1-8, H50Gµa/1, and
T1(V23)µa/1 were purified by affinity chromatography
over goat anti-mouse IgM-Sepharose 4B (Sigma) and compared with
IgG1 Abs (B1-81/1, H50G1/1, H33L1/1, and
T1(V23)1/1) to analyze avidity-dependent NP binding by IgG1 and
IgM. In brief, 96-well ELISA plates (Falcon, Becton Dickinson, Oxnard,
CA) were coated with 50 µg/ml NP22-BSA,
NIP21-BSA, NP5-BSA, or CG in 0.1 M carbonate
buffer (pH 8.6) at 4°C overnight. Plates were washed in PBS
containing 0.05% Tween 20 (Sigma) and blocked with 5% FCS in
carbonate buffer at 37°C for 3 h. Abs were serially titrated in
ELISA plates from 1 ng/ml to 200–400 µg/ml and incubated for
2 h at 22°C. Plates were then washed, and Ls136-biotin was
applied for 2 h; a subsequent wash was followed by SA-HRP. The TMB
peroxidase substrate kit (Bio-Rad, Hercules, CA) was used to
detect bound enzyme activity. Optical densities at 430 nm were
determined on an ELISA reader (SoftMAX, Molecular Devices, Sunnyvale,
CA) with values reported as the mean OD of three independent
experiments.
The T1 and H50G IgG1 and IgM Abs showed enhanced binding if sample incubations were performed at 4°C (data not shown); this enhancement is characteristic of natural, low affinity cryoglobulin Abs (60).
Avidity/affinity determinations by 2-ME treatment
Abs were assessed for NP reactivity by ELISA after treatment
with 2-ME. Mild reduction with 2-ME treatment causes 19S pentameric IgM
to dissociate into 7S H2L2 subunits that remain
capable of binding Ag (61). Briefly, equivalent
(H2L2) concentrations of IgM and IgG1 Abs (0.02
mg/ml) were diluted in 0.1 M sodium phosphate/4 mM EDTA (PB/EDTA) and
1.0 M 2-ME was added to a final concentration of 0.01 M (61). Reduced
and mock-treated samples (receiving an equal volume of PB/EDTA) were
incubated at 37°C for 2 h and serially diluted in
NP22-BSA-coated ELISA plates, followed by Ls136 to measure
Ab binding (55). Relative NP reactivity was measured as a percentage of
the B1-81 end point. NP binding was identical for reduced,
mock-treated, and untreated IgG1 Abs (data not shown).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The Ab response of C57BL/6 mice to NP-CG is dominated by B
cells bearing the 1 L-chain, whereas immunization with carrier
protein alone elicits virtually no 1+ Ab or cells (20, 24, 45, 48, 62). By day 10 after immunization, the majority of these
1+ B cells also express VH V186.2-to-DFL16.1
rearrangements that encode a tyrosine-rich CDR3 region with a consensus
motif, YYGS (45, 48, 56, 63), a combination referred to as the
canonical anti-NP BCR (43, 50). However, earlier in the response
(days 4–8 postimmunization), a large proportion of activated,
1+ B cells express noncanonical combinations of
VH and D gene segments (Fig. 1) (1, 48, 64). To investigate the
affinity of these NP-dependent, noncanonical B cells, VDJ
rearrangements were recovered from 1+ focus and GC B
cell populations by microdissection, amplified by a specific PCR, and
cloned into plasmids for sequence analysis (40, 48). At the times
examined (2–10 days postimmunization), H-chain genes recovered from
GCs have few mutations (Table I) and the 1 L-chain genes from GCs
were generally unmutated (data not shown, and Refs. 1, 40). This latter
finding is consistent with previous studies demonstrating a
10-fold reduction in the frequency of mutations in 1
compared with H-chain genes (62, 65, 66).
|
1 and transfected into J558L myeloma cells (Fig. 2A). Transfectomas producing
1+ IgG1 Abs were cloned, and their monoclonal Ig
products were purified for determination of specificity and affinity.
Table I shows the affinities of transfectoma Abs for NP- and
NIP-caproate as determined by fluorescence quenching (42, 58, 59, 67).
Three control IgG1 Abs, B1-81, H33L1, and H50G1 (Fig. 2B), were generated for affinity standards (see
Materials and Methods). Representative quenching
curves for the control Abs (Fig. 3)
illustrate that a wide range of affinities (Ka)
for NP, 1.2 x 105 M-1 (H50G1) to
2.0 x 107 M-1 (H33L1), may be
accurately determined. This range includes affinities reported for
early Ab (41, 54, 55, 56, 68). To exclude the possibility that
1+ B cells recovered from the plasmacytic foci and GCs
of immunized mice were specific for the carrier protein, transfectoma
Abs were assessed for reactivity to CG. No CG binding was detected by
fluorescence quenching or ELISA for any transfectoma Ab (data not
shown).
|
|
1+ foci and GCs ranged from 9.0 x
104–1.3 x 106 M-1. This
10-fold range illustrates that significant CDR3 diversity can be
generated even by joining identical V,D, and J gene segments. The
average affinity of these canonical Abs was 6.0 (± 3.3) x
105 M-1, a value virtually identical to
primary anti-NP serum Ab (42, 46, 50, 59). The average affinity
encoded by noncanonical VDJ fragments (1.4 (± 0.8) x 105
M-1) was modestly, but significantly
(p 
0.05), lower than that of the
VH186.2 set. To facilitate comparison, unmeasurably low
affinities from the noncanonical set were arbitrarily assigned a value
equal to the lowest Ka detectable by
fluorescence quenching (5 x 104 M-1).
This is a conservative estimate of their true affinities (see below).
Of the 13 transfectoma Abs studied, only 2, T1 and T2, did not exhibit
detectable binding for NP-caproate (Table I). The absence of NP
binding by the T2 Ab is likely due to replacement (R) mutations at
codons 16 and 80 of the C1H4 gene segment, since its unaltered clonal
relative T6 binds NP. The T1 Ab, however, is encoded by an unmutated
V23 VH gene segment combined with an atypical DFL16.1
sequence and represents a true germline affinity. Avidity-dependent binding of low affinity IgM Abs
To ensure that the VDJ rearrangement present in the T1
transfectoma (Table I) was obtained from an NP-specific B cell, we
constructed IgM-secreting transfectant lines. The high valency of IgM
enhances weak binding to Ag by permitting multiple, concurrent
interactions (i.e., increased avidity) (37, 38, 69, 70). Fig. 2C shows the IgM constructs for H50Gµa and
T1(V23)µa. Transfection of these constructs into the
J558L myeloma line yielded µa+/1+ Abs.
Canonical IgG (B1-81), IgM (B1-8), and the high affinity mutant
H33L1 Abs were used as controls. The relative avidities of
transfectoma IgG1 and IgM Abs were assessed in ELISA by measuring
binding to highly (NP22-BSA) or sparsely
(NP5-BSA) substituted substrates over a wide range of Ab
concentrations. Avidity-dependent heteroclitic binding was determined
on NIP21-BSA.
The 1+ IgG1 Abs encoded by the H50G and T1(V23) VDJ
rearrangements did not bind to solid phase NP- or NIP-BSA (Fig. 4). However, H50G bound NP in ELISAs as
an IgM, emphasizing the limitation of solid phase assays for detection
of low affinity, bivalent Abs (71, 72, 73). Indeed, fluorescence quenching
revealed specific NP binding by the H50G1 Ab (Fig. 3; Table I),
which could be detected in ELISA only with IgM proteins.
Similarly, IgG1 Abs bearing the T1(V23) V region had no detectable NP
binding by fluorescence quenching and demonstrated only slight
heteroclitic binding to NIP (Table I). However, Fig. 4 shows that the
T1(V23)µa Ab bound appreciably to NIP and marginally to
NP in ELISA. Neither H50Gµa nor T1(V23)µa
exhibited detectable binding to NP5-BSA; at low hapten
densities, multivalency could not compensate for the very low affinity
of these Abs. Interestingly, H50Gµa Abs remained unable
to bind NIP, suggesting that Arg at residue 50 is crucial for
heteroclicity (Fig. 4). The H50Gµa and
T1(V23)µa Abs do not bind CG in ELISA (data not shown),
confirming the specificity of avidity-dependent hapten binding.
|
). This
reduction dissociates 19S IgM into 7S subunits and resulted in the
complete loss of hapten binding in ELISA by H50Gµa and
T1(V23)µa (see Materials and
Methods). In contrast, the ELISA titer of the higher
affinity control IgM B1-8 was only slightly diminished by 2-ME
treatment, and reduction did not affect the binding of IgG1 Abs.
|
While differences in affinity existed between canonical and
noncanonical Abs, our sampling did not reveal differences in the
initial affinities of Ig produced by those B cells committed to early
Ab production (AFC foci) or to the GC/memory pathway. Transfectoma Abs
encoded by H-chain genes recovered from plasmacytic foci had an average
affinity of 5.7 (± 3.9) x 105 M-1, a value
not significantly different from the Ig receptors present in early
1+ GCs, 3.9 (± 2.2) x 105 M-1
(Table I).
Three pairs of VDJ rearrangements (T23 and T21; T30 and T25; and T6 and
T2 (Table I)), represent clonally related sets that were recovered from
the same GCs and share identical CDR3 junctions. These related
rearrangements differ from their partners by 1- to 5-point mutations
representing their independent somatic evolution (40, 51, 74, 75).
Although this sample of paired mutants is small, it is striking that
mutations in each pair resulted in decreased affinity for NP. For
example, the additional position 81 and 82 R mutations in the T21
rearrangement diminish its Ka for NP to a log
below that of its progenitor, T23 (Table I). Likewise, two R mutations
present in T2 completely abolish the relatively robust NP binding
(2.12 x 105 M-1) present in its
precursor, T6. Even the effect of a single R mutation (position 41) in
T25 lowers the affinity for NP compared with its progenitor, clone T30.
Thus, mutations that reduce affinity, even within a single clonal
genealogy, are frequent (54, 76, 77, 78, 79, 80). It may be that most V(D)J
mutations, even those present in mature GCs, do not contribute to
affinity maturation. Alternatively, the small size of the NP
determinant may limit the mutations that lead to improved affinity
(11, 81). Nevertheless, these lower affinity BCRs seem competent to
survive within GCs, at least initially, and may even enter the memory
compartment (44, 56).
The 1 Abs characteristic of the anti-NP response generally
demonstrate heteroclitic, or higher affinity, binding to NIP (24, 43, 50). With the exception of the mutated clone T21 and the position 50
mutant (H50G1), canonical transfectoma Abs exhibited heteroclitic
binding to NIP (Table I). In contrast, Abs encoded by VH186.2-analogue
gene segments were generally not heteroclitic (Table I). Only the
noncanonical T1 transfectoma, which had undetectable NP binding,
demonstrated enhanced reactivity to NIP, albeit at the lowest threshold
of detection (Ka 5.0 x 104
M-1).
The H50G mutation in GC B cells
Our conclusion that mutation commonly reduces affinity and that
lower affinity mutants can survive in GCs is supported by studies on
the population genetics of NP-specific B cells in AFC foci and GCs (1, 40, 48, 51). We have previously recovered three independent examples of
the ArgGly exchange present in the H50G1 transfectoma (Table I).
This high frequency suggests that position 50 may represent a
mutational hotspot, especially since the Gly replacement severely
reduces affinity (55). Nonetheless, this exchange does not appear to
preclude clonal survival in GCs. Fig. 6
illustrates the genealogy of a canonical B cell clone recovered from a
single GC 14 days after immunization with NP-CG. In this GC, an
ArgGly mutation at position 50 has occurred but allowed subsequent
mutations in daughter cells (Fig. 6). These findings and the recovery
of an NP-specific hybridoma containing Gly at position 50 (54, 82)
demonstrate that B cells bearing this deleterious mutation and its
consequent low affinity (Table I) are capable of activation,
proliferation, and differentiation in vivo.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The exquisite specificity of serum Ab responses has led to the notion
that only B cell receptors with Ka values well
above 105 M-1 are specifically activated by Ag
(10, 19, 34, 81). Low affinity Abs that bind Ag in an avidity-dependent
manner are often dismissed as physiologically irrelevant (19, 83). Yet,
almost 50% (6/13) of the recovered VDJ sequences we studied encoded
1+ Abs with affinities that fell below this hypothetical
threshold (Ka 3 x 105
M-1). A few exhibited only avidity-dependent binding and
by the usual convention would have been considered nonspecific or
insignificant. Nonetheless, each VDJ rearrangement was recovered by
precise microdissection from collections of 1+ AFCs and
GCs that depend upon immunization with NP (48). Indeed, while only
about 40% (5 of 13) of these rearrangements contained the V186.2
VH gene segment and the YYGS CDR3 motif that dominate the
primary response to NP (45, 48, 56, 63), all but the mutated T2
VDJ sequence generated 1 Abs that bound NP or NIP
specifically.
We propose that very low affinity B cells normally participate in
responses to thymus-dependent Ags. Indeed, many investigators have
documented the production of apparently nonspecific AFCs and Ab early
in humoral responses (26, 27, 29, 30, 31). These Igs had no detectable
affinity for Ag but often shared phenotypic characteristics (e.g., Id)
with the Ag-binding Ab set(s). Genetic analysis of NP-responsive B
cells has shown that the AFC focus and early GC populations routinely
contain many noncanonical VH rearrangements (1); initially,
more than 80% of the VDJ rearrangements recovered from
1+ GCs contain noncanonical VH segments.
However, these genes are quickly replaced by the VH186.2
gene, such that, after day 10 of the response, only about 15% of VDJ
rearrangements recovered from 1+ GCs contained
VH gene segments other than V186.2 (40). Our measurements
of transfectoma Ab affinity suggest that this change in the GC
repertoire is affinity driven, since most noncanonical VDJ
rearrangements generated transfectoma Abs with two- to sixfold lower
affinity for NP than canonical VDJ sequences. In a few cases,
noncanonical Abs had no detectable NP binding by fluorescence quenching
(Ka < 5 x 104
M-1).
Promiscuous activation by Ag would increase clonal participation and diversity early in the primary immune response, perhaps providing useful humoral protection during the interval required to generate high affinity Abs. Even very low affinity Ab can augment the activation of innate (84, 85) and specific immunity (86, 87), facilitating complement fixation, opsonization, and neutralization. Indeed, the bacterium Proteus morganii elicits specific Ab with virtually unmeasurable affinity (88). In GCs, a large and diverse progenitor population would ensure multiple trajectories of clonal evolution, maximizing the chance of achieving a specific, high affinity memory B cell pool (40, 67, 74, 89). The progressive loss of low affinity B cells during the immune response, i.e., in late GCs, and from the memory B cell population (45, 90, 91), is due either to an intrinsic inability of low affinity B cells to enter the memory compartment or clonal competition for survival within these immunologic niches (91).
We found no significant difference between the mean
Ka for BCRs present in AFCs or GCs (Table I).
This supports our earlier observation of clonally related B cells in
adjacent focus and GC populations (1) and suggests that, if distinct
cell lineages give rise to these compartments (5, 8, 10), each lineage
is very similar with respect to BCR genetics and phenotype. Later in
the immune response, VH usage appears to be determined by
strong clonal selection, most notably in GCs. Although there may be
some selection for VH use in the PALS-associated foci (Fig. 1 and 48 , these AFC remain terminally differentiated and
short-lived.
Despite evidence for the overall improvement of B cell affinity in the
immune response (11, 42, 45, 46, 62, 75, 90, 92, 93, 94, 95, 96), affinity
maturation during the primary GC reaction appears to be local; i.e.,
competition takes place within but not between GCs. Previously, Vora
and Manser showed that B cells expressing a high affinity Ig transgene
and endogenous, lower affinity B cells matured concurrently within the
same spleen (67). Canonical V186.2 rearrangements are common in mature
GCs, but they are not recovered from nearby GCs that contain low
affinity, noncanonical VDJ rearrangements (Refs. 40, 51, 89; and
unpublished observations). Since our study focused on the early
primary response, changes due to somatic mutation were limited.
However, analysis of the few mutated VDJs expressed as
1+ transfectoma Abs showed that most mutations led to
neutral or deleterious effects on affinity. Additionally, the clonal
genealogy shown in Fig. 6 suggests that very low affinity B cells can
proliferate and mutate in GCs. Thus, there does not appear to be an
intrinsic affinity requirement for B cell entry or survival in GCs;
instead, each GC defines a local fitness (affinity) optimum.
Alternatively, survival of low affinity GC B cells may depend upon signals or factors that are independent of the BCR. Recent work has shown that signaling via the CD21/CD19 complex is required for the survival of even very high affinity (Ka > 1010 M-1) (97). However, a substantial body of evidence suggests that GC B cells undergo apoptosis in the complete absence of BCR stimulation (98, 99, 100). We cannot rule out the possibility that, early in the GC reaction, Ag is sufficiently abundant not to constrain AFC development and GC B cell survival (101, 102). We are currently investigating this possibility by extending our transfectoma analysis to VDJs from later GCs when clonal competition is evident (M. Shimoda and G. Kelsoe, manuscript in preparation).
A potential consequence of the presence and persistence of low affinity GC B cells would be the generation of low affinity memory cells. Low affinity memory B cells are not usual in secondary immune responses (42, 45, 90, 91, 103). This could be the result of 1) an intrinsic inability of low affinity B cells to enter the long-lived memory pool (8) or 2) the failure of low affinity memory cells to persist due to their ineffective competition for stimulation by residual Ag (92, 104). In fact, long-lived bone marrow plasmacytes undergo affinity-driven selection outside of the GC microenvironment (92). Thus, even if low affinity B cells did enter the memory compartment, continued selection/competition could reduce their survival numbers over time.
Is clonal competition the sole mechanism for generating specific and high affinity Ab responses and B cell memory? The removal of competing, high affinity B cells should allow us to determine whether competition is the dominant selective mechanism during the immune response. In the absence of competition, very low affinity B cells will be either unable to mount and/or sustain humoral immunity or fully competent to initiate all facets of a T-dependent immune response. We have created two lines of transgenic mice that express BCR with very low affinities for NP in an attempt to test this hypothesis (J. M. Dal Porto et al., manuscript in preparation).
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Garnett Kelsoe, Department of Immunology, Duke University Medical Center, Box 3010, Durham, NC 22710. E-mail address:
3 Abbreviations used in this paper: PALS, periarteriolar lymphoid sheath; PNA, peanut agglutinin; AFC, Ab-forming cell; CDR, complementarity-determining region; CG, chicken -globulin; GC, germinal center; NP, (4-hydroxy-3-nitrophenyl)acetyl; NIP, (4-hydroxy-5-iodo-3-nitrophenyl)acetyl; R, replacement; SA, streptavidin; HRP, horseradish peroxidase; BCR, B cell receptor; H, heavy; L, light; AP, alkaline phosphatase.
Received for publication April 2, 1998. Accepted for publication July 16, 1998.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
2a variable region. Cell. 24:625.[Medline]
light chain genes amplified from single cells is inconsistent with variable region gene conversion in germinal-center B cell somatic mutation. Eur. J. Immunol. 24:1816.[Medline]
This article has been cited by other articles:
|
|
 |
|
  M. Kuraoka, D. Liao, K. Yang, S. D. Allgood, M. C. Levesque, G. Kelsoe, and Y. Ueda Activation-Induced Cytidine Deaminase Expression and Activity in the Absence of Germinal Centers: Insights into Hyper-IgM Syndrome J. Immunol., September 1, 2009; 183(5): 3237 - 3248. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Eckl-Dorna and F. D. Batista BCR-mediated uptake of antigen linked to TLR9 ligand stimulates B-cell proliferation and antigen-specific plasma cell formation Blood, April 23, 2009; 113(17): 3969 - 3977. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. A. Luijkx, J. A. M. van Gaans-van den Brink, H. H. van Dijken, G. P. J. M. van den Dobbelsteen, and C. A. C. M. van Els Hyperproliferation of B Cells Specific for a Weakly Immunogenic PorA in a Meningococcal Vaccine Model Clin. Vaccine Immunol., October 1, 2008; 15(10): 1598 - 1605. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. W. Sohn, P. Tolar, and S. K. Pierce Membrane heterogeneities in the formation of B cell receptor-Lyn kinase microclusters and the immune synapse J. Cell Biol., July 28, 2008; 182(2): 367 - 379. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 U. Hershberg, M. Uduman, M. J. Shlomchik, and S. H. Kleinstein Improved methods for detecting selection by mutation analysis of Ig V region sequences Int. Immunol., May 1, 2008; 20(5): 683 - 694. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
I. Gonzalez-Garcia, B. Rodriguez-Bayona, F. Mora-Lopez, A. Campos-Caro, and J. A. Brieva Increased survival is a selective feature of human circulating antigen-induced plasma cells synthesizing high-affinity antibodies Blood, January 15, 2008; 111(2): 741 - 749. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. J. DiLillo, Y. Hamaguchi, Y. Ueda, K. Yang, J. Uchida, K. M. Haas, G. Kelsoe, and T. F. Tedder Maintenance of Long-Lived Plasma Cells and Serological Memory Despite Mature and Memory B Cell Depletion during CD20 Immunotherapy in Mice J. Immunol., January 1, 2008; 180(1): 361 - 371. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Aviszus, X. Zhang, and L. J. Wysocki Silent Development of Memory Progenitor B Cells J. Immunol., October 15, 2007; 179(8): 5181 - 5190. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Ueda, D. Liao, K. Yang, A. Patel, and G. Kelsoe T-Independent Activation-Induced Cytidine Deaminase Expression, Class-Switch Recombination, and Antibody Production by Immature/Transitional 1 B Cells J. Immunol., March 15, 2007; 178(6): 3593 - 3601. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. P. O'Connor, L. A. Vogel, W. Zhang, W. Loo, D. Shnider, E. F. Lind, M. Ratliff, R. J. Noelle, and L. D. Erickson Imprinting the Fate of Antigen-Reactive B Cells through the Affinity of the B Cell Receptor J. Immunol., December 1, 2006; 177(11): 7723 - 7732. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Paus, T. G. Phan, T. D. Chan, S. Gardam, A. Basten, and R. Brink Antigen recognition strength regulates the choice between extrafollicular plasma cell and germinal center B cell differentiation J. Exp. Med., April 17, 2006; 203(4): 1081 - 1091. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Han, X. Zhang, G. Wang, H. Guan, G. Garcia, P. Li, L. Feng, and B. Zheng FTY720 suppresses humoral immunity by inhibiting germinal center reaction Blood, December 15, 2004; 104(13): 4129 - 4133. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. A. Steeves and T. N. Marion Tolerance to DNA in (NZB x NZW)F1 Mice That Inherit an Anti-DNA VH as a Conventional {micro} H Chain Transgene but Not as a VH Knock-in Transgene J. Immunol., June 1, 2004; 172(11): 6568 - 6577. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Han, E. Marinova, and B. Zheng Rectification of age-related impairment in Ig gene hypermutation during a memory response Int. Immunol., April 1, 2004; 16(4): 525 - 532. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Zheng, E. Marinova, J. Han, T.-H. Tan, and S. Han Cutting Edge: {gamma}{delta} T Cells Provide Help to B Cells with Altered Clonotypes and Are Capable of Inducing Ig Gene Hypermutation J. Immunol., November 15, 2003; 171(10): 4979 - 4983. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Han, K. Yang, Z. Ozen, W. Peng, E. Marinova, G. Kelsoe, and B. Zheng Enhanced Differentiation of Splenic Plasma Cells but Diminished Long-Lived High-Affinity Bone Marrow Plasma Cells in Aged Mice J. Immunol., February 1, 2003; 170(3): 1267 - 1273. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Kanayama, T. Kimoto, K. Todo, Y. Nishikawa, M. Hikida, M. Magari, M. Cascalho, and H. Ohmori B Cell Selection and Affinity Maturation During an Antibody Response in the Mouse with Limited B Cell Diversity J. Immunol., December 15, 2002; 169(12): 6865 - 6874. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Inui, K. Maeda, D. R. Hua, T. Yamashita, H. Yamamoto, E. Miyamoto, S. Aizawa, and N. Sakaguchi BCR signal through {alpha}4 is involved in S6 kinase activation and required for B cell maturation including isotype switching and V region somatic hypermutation Int. Immunol., February 1, 2002; 14(2): 177 - 187. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y.-F. Lu, M. Singh, and J. Cerny Canonical germinal center B cells may not dominate the memory response to antigenic challenge Int. Immunol., May 1, 2001; 13(5): 643 - 655. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. G. Hannum, A. M. Haberman, S. M. Anderson, and M. J. Shlomchik Germinal Center Initiation, Variable Gene Region Hypermutation, and Mutant B Cell Selection without Detectable Immune Complexes on Follicular Dendritic Cells J. Exp. Med., October 2, 2000; 192(7): 931 - 942. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Z. Chen, S. B. Koralov, M. Gendelman, M. C. Carroll, and G. Kelsoe Humoral Immune Responses in Cr2-/- Mice: Enhanced Affinity Maturation but Impaired Antibody Persistence J. Immunol., May 1, 2000; 164(9): 4522 - 4532. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. A. Vora, K. Tumas-Brundage, and T. Manser Contrasting the In Situ Behavior of a Memory B Cell Clone During Primary and Secondary Immune Responses J. Immunol., October 15, 1999; 163(8): 4315 - 4327. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. E. Coffin, S. L. Clark, N. A. Bos, J. O. Brubaker, and P. A. Offit Migration of Antigen-Presenting B Cells from Peripheral to Mucosal Lymphoid Tissues May Induce Intestinal Antigen-Specific IgA Following Parenteral Immunization J. Immunol., September 15, 1999; 163(6): 3064 - 3070. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Kesmir and R. J. De Boer A Mathematical Model on Germinal Center Kinetics and Termination J. Immunol., September 1, 1999; 163(5): 2463 - 2469. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Takahashi, D. M. Cerasoli, J. M. Dal Porto, M. Shimoda, R. Freund, W. Fang, D. G. Telander, E.-N. Malvey, D. L. Mueller, T. W. Behrens, et al. Relaxed Negative Selection in Germinal Centers and Impaired Affinity Maturation in bcl-xL Transgenic Mice J. Exp. Med., August 2, 1999; 190(3): 399 - 410. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. M. Dal Porto, A. M. Haberman, G. Kelsoe, and M. J. Shlomchik Very Low Affinity B Cells Form Germinal Centers, Become Memory B Cells, and Participate in Secondary Immune Responses When Higher Affinity Competition Is Reduced J. Exp. Med., May 6, 2002; 195(9): 1215 - 1221. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
) represent
relative kinetics of focus and GC reactions during the primary
response. Data shown combine previous studies (1, 40, 51) and the
current work. Noncanonical VH genes include C1H4, CH10,
V23, 24.8, V102, and V583.5 (41).
), and H50G
). Fluorescence quenching was performed by measuring
the change in 280 nm excitation/340 nm emission signal of Ab samples
through a serial titration of NP-caproate (2.5 x
10-8–5.0 x 10-5 M). Curves represent
the average (±SD) of three independent measurements for each sample.
Values represent percent of maximum quench (max Q), which varied
for each Ab sampled (max QB1-8
)
and lower affinity variants H50Gµ
(
), H50G
), T1(V23)µ (
) was also tested. Curves represent
means from triplicate experiments.
