| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Unité du Développement des Lymphocytes, Institut Pasteur, Paris, France
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
5 and V-preB, thus forming a pre-B
cell receptor (BCR).3
BCR formation occurs later and results from the association of this
heavy chain with a 
or light chain. According to the ordered
model of IgH rearrangement (2), somatic recombination is a
hierarchical two-step process in which D to JH
rearrangement precedes VH recruitment. This model
was based on the observation that Abelson murine leukemia
virus-transformed pro-B cells usually carry two
D-JH rearrangements. Some of these lines could
further rearrange VH-DJH in culture.
These observations were also consistent with the description of
rearrangements carried by individual cells corresponding to different
stages of B cell development in mice (3, 4). Unlike D to
JH rearrangement, which occurs on both alleles,
VH to DJH rearrangement is
thought to be under a control that determines the phenotypic allelic
exclusion of the IgH, a phenomenon first described in 1965
(5). The mechanistic basis of such control has been under
intense debate over the last 20 years. Consensus has been reached
around the feedback or regulated model of Ig gene rearrangement.
According to this model, initially proposed for Ig light chain
(6), but soon adopted for the Ig heavy chain, the
expression of a rearranged Ig heavy chain at the cell surface is
thought to inhibit somatic rearrangement in the heavy chain locus.
Chronologically, this hypothesis is sustained by the following
observations: 1) Alleles with D-JH rearrangements
are frequent, both in cell lines and in normal B lymphocytes
(2). 2) In transgenic mice for an Ig heavy chain (for a
review, see Ref. 7), rearrangement of the endogenous
alleles is decreased (8, 9, 10); however, when the transgenic
molecule lacks the transmembrane domain (11, 12) or can
only be secreted (13), allelic exclusion of the endogenous
alleles is no longer observed. 3) In mice carrying one allele with a
disrupted transmembrane exon, dual-expressing cells were detected,
indicating that allelic exclusion is no longer observed
(14). 4) In pre-B cells carrying two productive
rearrangements, only one of the chains is able to pair with the
surrogate light chains to form a pre-BCR (15). Notwithstanding the compelling evidence in favor of the feedback and ordered models of allelic exclusion, a number of conflicting observations made in Ig transgenic models should be considered. It was clear that the endogenous Ig alleles were not completely excluded, and this was observed both at the cell surface (16, 17) and at the DNA levels (8, 9, 10, 12). The proportion of cells expressing more than one heavy chain at the cell surface was a function of the strain, the transgene, the specific transgenic line, and the type of B cell population analyzed. These observations can be explained by the down-regulation of transgene expression, loss of the transgene itself due to sister chromatid recombination (18), or transgene expression after endogenous IgH recombination. Cells with endogenous IgH expression could then be selected for the diversification of the repertoire (19). Therefore, the Ig transgenic models may give a skewed readout of the dynamics of rearrangement throughout B cell ontogeny.
In this study, we developed a complementary approach that could bypass these drawbacks. We challenged the system by introducing an extra Ig heavy chain locus in germline configuration. This was achieved by obtaining embryos trisomic for the chromosome 12 and therefore triallelic for the IgH locus (20). Fetal liver cells from trisomic embryos restore hemopoiesis, including lymphopoiesis of irradiated RAG2-/- mice, as previously described (21, 22). In these mice, there is no evidence of cytogenetic instability, and the trisomic cells are morphologically and functionally similar to control euploid hemopoietic cells (21, 22). In this study, we analyze chromosome 12 trisomic (Ts12) reconstitution chimeras with different IgH combinations for the ratio of allotype expression in B cells, the degree of IgH allelic exclusion, and the recombination status of the three alleles.
We found that all three alleles are available for recombination and that the allotypic ratio of Ig-expressing B cells reflects the allelic ratio. Although trisomic cells are able to rearrange all three alleles, the levels of IgH phenotypic allelic exclusion are not altered in these animals, as compared with euploid mice. Likewise, most VHDJH rearrangements of the silenced allele are unproductive, indicating that the majority of these cells are also genetically excluded. These results provide further support for the feedback model of allelic exclusion. However, we also observed that the reading frame (RF) II usage in DJH from the IgH Ts12 splenic B cells is higher than in controls, suggesting that some D-JH rearrangements might have occurred after productive VHDJH rearrangement.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Rb(4.12)5Bnr/C57BL/6, Rb(6.12)Sic/BALB/c, Rb(8.12)5Bnr/C57BL/6, and Rb(8.12)Sic/BALB/c wild-derived Robertsonian (Rb) mice were purchased from the Medizinische Universität zu Lübeck (Lubeck, Germany). A/J females were purchased from Charles River Breeding Laboratories (Clion, France); BALB/c and C57BL/6 females were purchased from Iffa Credo (L’Arbresle, France). RAG2-/-/129/Sn animals (23) and C57BL/6.Ly-5.1-congenic animals were purchased from the Centre de Développement des Techniques Avancées pour l’Experimentation Animale-Centre National de la Recherche Scientifique (Orleans, France). Experiments were performed with animals bred in the Pasteur Institute animal facilities.
Reconstitution chimeras
Doubly heterozygous males Rb(4.12)5Bnr/Rb(6.12)Sic, Rb(4.12)5Bnr/Rb(8.12)Sic, and Rb(6.12)Sic/Rb(8.12)Sic were obtained and mated with A/J, BALB/c, or C57BL/6 females. Pregnant females were killed by cervical dislocation between days 14 and 17 of gestation, the plug day considered to be day 1. Ts12 fetuses were identified by their exencephalic phenotype and by cytogenetics. Ts12-nucleated fetal liver cells suspended in RPMI 1640 complete culture medium (10% FCS) were transplanted i.v. (>5 x 106cells/animal) into 600 rad irradiated RAG2-/-/129/Sn mice (8–16 wk old). Ts12 cell hemopoietic reconstituted chimeras were identified on the basis of the presence of circulating IgGs 1 mo following injection. To generate mice with mixed Ts12 and euploid hemopoietic systems, Ts12 IgHabb and euploid IgHab (BALB/c x C57BL/6.Ly-5.1+) bone marrow cells from reconstitution chimeras (both in RAG2-/-/129Sv) were mixed at different ratios and used as before to reconstitute RAG2-/-/129/Sn mice (8–16 wk old). Euploid and Ts12 cells were distinguished using a homemade anti-Ly-5.1, PE-labeled Ab. Throughout our study, chimeras were perpetuated by repeated transplantation of bone marrow cells into RAG2-/- recipients, for more than 1 year.
Cytogenetics
To obtain a karyotype of the trisomic embryos, a fraction (<10%) of fresh total fetal liver cell suspension in RPMI 1640 complete medium was incubated for 1 h at 37°C in the presence of PWM (Life Technologies, Gaithersburg, MD). Further processing involved standard techniques. Trisomic fetuses were readily identifiable by the presence of two Rb translocations (37 acrocentric and 2 metacentric chromosomes or 41 chromosome arms) (24), whereas euploid fetuses displayed 38 acrocentric chromosomes and one Rb translocation.
Fluorescence surface staining and flow cytometry
Abs used for common FACS stainings of bone marrow, spleen, and FACS Lysing Solution (Becton Dickinson, Sunnyvale, CA)-treated peripheral blood cells were the following: FITC anti-IgMa, anti-IgDa, PE anti-B220, anti-IgMb, anti IgDb, biotin anti-B220, anti-IgDe, and streptavidin cychrome tricolor (all from PharMingen, San Diego, CA). Dead cells were eliminated by propidium iodide exclusion, and 10,000–200,000 events were recorded per sample. Fluorescence was measured with FACScan flow cytometer (Becton Dickinson) using the CellQuest 3.1 software (Becton Dickinson). In stainings with propidium iodide-, cychrome tricolor, and PE-labeled Abs, exclusion of dead cells did not decrease the percentage of PE or cychrome tricolor-positive cells in more than 5% of the population absolute number.
Intracytoplasmatic stainings
A total of 2–4 x 104/ml bone marrow or spleen cells was cultured for 5 days at 37°C in RPMI 1640 complete medium (5% FCS; 25 µg/ml LPS), with a feeder layer of S17 stromal cells. After 3–5 days, cells were then fixed in 95% ethanol on slides according to standard procedures and incubated with FITC anti-IgMa and/or biotin anti-IgMb, followed by a streptavidin-Texas Red incubation when necessary. Buffered glycerol-mounted slides were analyzed on a Axiophot Zeiss (Zeiss, Oberkochen, Germany) fluorescence microscope.
Cell sorting
IgM allotype-specific B220+-expressing splenocytes were isolated in a FACStarPlus cell sorter (Becton Dickinson). Dead cells in sorted samples were scored by propidium iodide and/or trypan blue exclusion. The purity of the sorted population was measured in a FACScan flow cytometer (Becton Dickinson) using CellQuest 3.1 software (Becton Dickinson).
Semiquantitative PCR conditions
Detection of IgH in the germline state was performed on Ts12 IgHabb splenic cells sorted as IgMb B220+ (purity >95%; about 50,000 cells/animal; n = 3), which were put in a well with 6000 irradiated S17 cells and were stimulated with LPS, as described. After 4–5 days, 1.5 x 106 cells were harvested and DNA was prepared by standard phenol:chloroform extraction and precipitation. As control, DNA from S17 cells was also prepared. Detection of the IgH germline-specific amplicon was obtained in a PCR (35 cycles; 95°C, 30 s; 63°C, 30 s; 72°C, 1 min; final elongation at 72°C for 10 min) with the forward primer 5'-JH1 5'-CCCGGACAGAGCAGG-3' and the reverse primer 3'-JH1 5'-GGTCCCTGCGCCCCA-3' (25) under the following conditions: 2 mM dNTPs; 1.25 MgCl2; 12.5 pmol each oligo; 2.5 U Taq polymerase (Life Technologies); total volume, 25 µl. As a control for the amount of genomic template DNA, a RAG2 amplicon was amplified, under the same PCR conditions, with the forward primer 5'-TAGGGAACGGGAGGTGAGAGT-3' and the reverse primer 5'-TGACAGCAGTATGAAAGGAGG-3'. Serial dilutions covered a >700-fold dilution, starting from 200 ng of DNA.
PCR amplification of IgH rearrangements, cloning, and sequencing
(VH)DJH rearrangements were amplified from IgM-expressing B220+-sorted splenic B cells, which were further expanded in culture, as described. DNA was prepared by standard phenol:chloroform extraction, followed by precipitation. PCRs were performed starting with 5–200 ng of DNA, and the standard conditions were the following: 2 mM dNTPs, 2.5 mM MgCl2, 12.5 pmol of each oligonucleotide, and 1.25 U Taq polymerase (Life Technologies) in a final volume of 25 ml; final elongation at 72°C for 10 min.
To distinguish between BALB/c and C57BL/6 IgH alleles, we focused on (VH)DJ1H rearrangements due to the presence of an intronic single-base polymorphism between these two strains, which is very close to and 3' of JH1 (26). For VH DJH1 amplifications, a first reaction (30 cycles; 95°C, 30 s; 60°C, 30 s; 72°C, 1 min) was performed using the following VH-specific forward primers in separate reactions: VHA (J558) 5'-GCGAAGCTTARGCCTGGGRCTTCAGTGAAG-3'; VHB (Q52) 5'-GCGAAGCTTCTCACAGAGCCTGTCCATCAC-3'; VHE (7183; DNA4) 5'-GCGAAGCTTGTGGAGTCTGGGGGAGGCTTA-3'; and the reverse primer JH2/3 5'-CCAGTAAGCAAACCAGGCACA-3'. Then, a second PCR (30 cycles; 95°C, 30 s; 58°C, 30 s; 72°C, 1 min) was performed on 1 µl of the first PCR, using the same VH-specific primers and the reverse primer JH1 5'-AAAAAAGCCAGCTTACCTGA-3'.
For DJ4H amplifications, a PCR (35 cycles; 95°C, 30 s; 62°C, 30 s; 72°C, 30 s) was performed using a DFS 5'-ACGTCGACTTTTGTSAAGGGATCTACTACTG-3' forward primer that recognizes all D elements except Q52 and the reverse primer JA 5'-GGGTCTAGACTCTCAGCCGGCTCCCTCAGGG-3'. For the amplification of DJH1 rearrangements, a first PCR (35 cycles; 95°C, 30 s; 60°C, 30 s; 72°C, 1 min) was performed using DFS and J2/3 as forward and reverse primers; 1 µl of the PCR was then used on a second PCR (30 cycles; 95°C, 30 s; 58°C, 30 s; 72°C, 1 min) with DFS and JH 2/3' 5'-GACAATAAATGATCCTTGGC-3' as forward and reverse primers, respectively.
(VH)DJH amplicons were cloned with the Topo TA cloning kit according to the instructions of the manufacturer (Invitrogen, San Diego, CA). The Dye Terminator Sequencing Kit (Perkin-Elmer, Foster City, CA) was used to determine the sequence of the cloned amplicons following the instructions of the manufacturers. VH-specific primers were used for the VHDJH sequences; JA was used to sequence the DFS-JA-amplified D-JH4 amplicons. Sequences were obtained in an ABI 370A DNA sequencer (Applied Biosystems, Foster City, CA). For the analysis of RF usage in DJH rearrangements, DJH1 and DJH4 rearrangements were pooled.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
For the generation of Ts12 B cells, we made use of male mice
double heterozygous for two balanced Rb metacentric translocations
carrying chromosome 12 as one of their arms: Rb(4.12;8.12) and Rb(4.12;
6;12) mice. In these animals, the nonsegregation of the chromosome 12
pair is frequent, as each copy is fused to a different chromosome (Fig. 1
A) (24). Upon
fecundation, a third copy of the chromosome 12 is added, and these
embryos, although with a limited life span, reach the 16 days
postcoitum (dpc) stage. Additionally, they can be easily distinguished
from euploid siblings by their exencephalic phenotype
(24). Fetal liver cell Ts12 animals were used to
reconstitute the hemopoietic system of irradiated
RAG2-/- mice. In a Ts12 hemopoietic system, not
only is the aneuploid karyotype stable over time, but the Ts12
hemopoietic cells (including lymphoid cells) are similar to the euploid
ones as to morphology and function (21, 22). We initially
tried to obtain embryos bearing three different allotypes for which
specific Abs are available. Rb(4.12;8.12) and Rb(4.12; 6;12) double
heterozygous males in a F1 (BALB/c x
C57BL/6) background (IgHa and IgHb allotypes) were crossed with IgHe
A/J females. The low breeding efficiency and the low frequency (<2%)
of 13–17 dpc Ts12 embryos observed allowed the generation of one
single Ts12 chimera (Fig. 1B) and prompted us to switch to
other strains of females. C57BL/6 and BALB/c females were used instead
of A/J. The frequency of trisomic embryos was then increased. This
allowed us to establish RAG2-/-
mice-reconstituted Ts12 fetal liver cells from five different embryos,
from which the allotypic combinations are shown in Table I. These Ts12 hemopoietic systems were
maintained by successive reconstitution of
RAG2-/- mice with bone marrow cells from other
RAG2-/- mice previously reconstituted with Ts12
fetal liver cells.
|
|
One month after reconstitution, Ts12 chimeras were analyzed for
the presence of serum IgM and of B cells in the bone marrow and
periphery. Levels of serum IgM and B cell numbers were similar to those
of chimeras reconstituted with euploid fetal liver cells (data not
shown). FACS analysis of bone marrow, spleen, and blood cells with
allotypic-specific anti-IgM (IgD) Abs revealed that, both in the
bone marrow and in the periphery, the ratio of IgMa (IgDa)- vs IgMb
(IgDe)-expressing cells reflects the allotypic ratio at the level of
the karyotype. In other words, abb trisomic chimeras have approximately
one-third of the IgMa and two-thirds of IgMb B cells, whereas aab
trisomic chimeras exhibit two-thirds of IgMa+ and
one-third of IgMb+ B cells (Fig. 2). Similar results were obtained in
chimeras with different allotypic combinations, since the ratio of
IgDa- vs IgDe-expressing cells also changes accordingly to the ratio of
IgHa and IgHe alleles.
|
B). This later chimera resulted from a crossing-over
event between chromatids from chromosome 12 homologues in the
Rb(4.12;8.12)/F1 male, leading to the
cosegregation of both IgHa alleles during meiosis II. The observation
that in these two IgHaab chimeras the IgMa:IgMb ratio is similar
indicates that the rearrangement of the IgH alleles is not altered by
paternal or maternal disomy in chromosome 12. Furthermore,
reconstituted chimeras with euploid (Ly-5.1+) and
Ts12 (Ly-5.1-) cells show a ratio of
allotype-specific IgM-expressing cells, which reflects the allelic
combination, suggesting that this ratio is solely determined by the
genetic characteristics of the B cells (Fig. 2) and independent of
external factors. Levels of Ig heavy chain allelic exclusion in Ts12 mice are similar to those of normal mice
In euploid IgHab mice, one-half of B cells expresses IgMa and the
other half expresses IgMb. By flow cytometric criteria, B cells
expressing both allotypes are below 2%. These observations can be
extended to IgHab Rb(4.12;8.12) and IgHab Rb(4.12; 6;12) mice,
indicating that the fusion of chromosome 12 with another chromosome at
the level of the centromere does not interfere with somatic
rearrangement or with the expression of the Ig heavy chain (data not
shown). Ts12 chimeras were compared with euploid
F1 animals or chimeras reconstituted with euploid
F1 fetal liver cells. The frequency of putative
dual expressors in chromosome 12 chimeras is not different from that of
euploid cells (Fig. 2 and Table I). The same comparison was also
performed by intracytoplasmic staining of B cells stimulated ex vivo by
LPS. As before, no significant difference was observed between Ts12 and
euploid plasma cells (Table I), and the allotypic ratio follows what
has been observed at the surface level. We recently developed an assay
to estimate the frequency of B cells escaping allelic exclusion.
Putative dual expressors are sorted, expanded in culture by LPS
stimulation for 2–4 days, and reanalyzed by flow cytometry or by
intracytoplasmic staining. We found that the frequency of dual
expressors in euploid IgHa/b mice is in the order of
1:104 (27). The same assay performed
on cells from three F1 animals and three
Ts12-reconstituted chimeras revealed no significant differences in the
numbers of dual expressor B cells (data not shown). We conclude that
Ts12 reconstitution chimeras have levels of allelic inclusion
comparable with the ones found in euploid mice.
Ts12 B cells can rearrange the three alleles
The ratio of Ig allotype-expressing cells in the Ts12
reconstitution chimeras does not provide any information concerning the
rearrangement status of the three alleles in each cell. If within each
cell one chromosome is randomly excluded from rearrangement, the
allotypic ratio would be the same as if in all cells the three alleles
are competent to rearrange. To distinguish between these two scenarios,
we performed a semiquantitative PCR using a set of primers that
amplifies a region between the most 3' D element and
JH1 (Fig. 3). Since
this region is deleted upon D to JH
rearrangement, the assay quantifies alleles in germline configuration.
Approximately 3 x 105 Ts12 IgMb
B220+Ly-5.1- were sorted
and pooled from three reconstitution abb chimeras. When the PCR was
directly performed on sorted cells (>95% pure), a faint band
corresponding to IgH germline configuration could be observed (data not
shown). In addition, one-half of these cells were also set in culture
in the presence of irradiated stromal cells and LPS. After 4–5 days,
these B cells had expanded to 1.5 x 106,
and the stromal cell contribution to the DNA preparation from this
culture was estimated to be <1%. The same semiquantitative PCR assay
was then performed on DNA from these purified B cells and from stromal
cells. The virtually undetectable band corresponding to an amplicon
from an allele in germline configuration led us to conclude that the
majority of the T12 B cells had at least DJH
rearrangements on all three alleles.
|
Having established that IgH (phenotypic) allelic exclusion is
normal in Ts12 mice, we then focused on allelic exclusion at the
genetic level. By single cell PCR analysis of rearrangements in mature
B cells, it was shown that euploid cells with two productive
VHDHJ rearrangements (genetically included) are
present at low frequency (2.5%) (15). To evaluate whether
the introduction of a third allele influenced the level of genetic
allelic exclusion, we characterized
VHDJH rearrangements of the
BALB/c IgH allele in IgM+ cells expressing the
C57BL/6 allele (IgMb), from a T12 IgHabb chimera. DNA was prepared from
sorted splenic IgMb+ cells (>95% pure), which
were expanded in culture by LPS stimulation.
VHDJH rearrangements were
amplified with primers specific for the J558, Q52, and 7183
VH families. The amplicons were then cloned and
sequenced. To distinguish between BALB/c and C57BL/6 IgH alleles, we
focused on VHDJH1
rearrangements carrying a sequence polymorphism between these two
strains (26). BALB/c IgH alleles show a majority of
nonproductive rearrangements (Table II).
These results suggest that the proportion of cells with genetic allelic
inclusion in Ts12 is a minority. These data are also consistent
with the finding that a proportion of
VHDJH C57BL/6 alleles in
these cells had nonproductive rearrangements. Since there are two
C57BL/6 alleles in these cells, one is likely to be silenced in most
IgMb-expressing cells that rearranged both IgHb alleles
VH-DJH.
|
When a D-JH rearrangement is in a particular
RF, RFII, D promoter elements drive the expression of the Dµ protein,
a truncated µ-chain that lacks the V region (28). It was
proposed that this molecule signals the shut down of somatic
recombination, thus arresting cell differentiation. This would account
for the underrepresentation of D-JH elements in
RFII in mature B cells (29). In support of this model, it
has been shown that the presence of a Dµ transgene leads to a partial
block in B cell development (30). Moreover, a rearranged
heavy chain transgene suppresses the RFII counterselection in
D-JH rearrangements from the endogenous alleles,
raising that frequency from 5% to 27% (31). We performed
two independent experiments in which euploid control and Ts12 B cells
were enriched by LPS stimulation for 5 days of splenic cells (Expt. 1)
or purified by cell sorting (Expt. 2). Genomic
D-JH1 and D-JH4 fragments
were amplified, cloned, and sequenced. In the euploid control
population, we reproduce most of what is known about RF usage (Table III), i.e., we found an
overrepresentation of RFI, less of RFIII, and even less of RFII (8%).
In the Ts12, we found a modest increase in the frequency of RF2 usage
(15%). We observe that in D-JH rearrangements
from splenic B Ts12 cells, the frequency of RFII usage is clearly below
one-third, in constrast with the data reported for an introduced
rearranged heavy chain transgene (31). We conclude that
most T12 B cell precursors rearranged all three alleles D to
JH before starting VH to
DJH. We propose that the modest increase in RFII
reflects a situation in which a minority of T12 cells rearrange D to
JH when at least one allele has or is
simultaneously undergoing productive
VH-DJH+
rearrangement, thus relieving the counterselection caused by
Dµ.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We found that the levels of Ig heavy chain allelic exclusion in Ts12 mice are similar to those of normal mice. Similar observations were reported by Du Pasquier and Hsu (34) in a pioneering study using triploid and tetraploid Xenopus animals, heterozygous for the IgH. Although multiple IgH alleles were present, each cell only produced one Ab. At the time this study was performed, the stochastic model of allelic exclusion (35) was still debated. In its pure form, this model proposed that, given a low probability of rearranging one allele productively, allelic inclusion would have to be rare, as it would result from two infrequent events. Du Pasquier and Hsu (34) argued that a stochastic model of allelic exclusion could be compatible with their results only if the frequency of multiple successful rearrangements is very low, and that this is unlikely in frogs because it would lead to a great wastage of lymphocytes. Based on this rationale, they favored some sort of feedback mechanism. We know now that the actual frequency of productive rearrangement is not sufficiently low to explain allelic exclusion. For IgH, this frequency depends on the processing of the coding ends during somatic rearrangement, the proportion of VH pseudogenes, the proportion of aberrant (abortive) rearrangements, and also on the D element sequences that, if read in a particular RFIII, contain stop codons. As was already discussed >15 years ago by Coleclough (36) and taking into account data accumulated thereafter, a strict stochastic model of allelic exclusion is untenable. The feedback model essentially solves this problem, but both are not incompatible; in fact, they are complementary. There is a stochastic component to allelic exclusion on the random choice of the allele that rearranges V to D-JH, and it is likely that simultaneous VH recruitment should have an associated probability, as low as it may be. The difficulty in evaluating such probability is illustrated by the fact that the approaches with Ts12 animals would only change it by a factor of 2 or 3. Such minor change would be difficult to detect if the frequency of simultaneous rearrangements is, indeed, low.
The work of Hsu and Du Pasquier (34) was limited to serological and cell surface analysis. It should also be pointed out that in some of their hybrids, the possibility that different sets of Ig genes do not act as normal alleles could not be formally excluded. Furthermore, in one triploid combination, the Ig type from one haploid set was absent from all cells, which was interpreted not as a manifestation of Ig allelic exclusion, but rather as a phenomenon of genome haploid set suppression in certain triploid combinations. Our data in a mammalian model, however, are clearly the end result of an active allelic exclusion mechanism of somatic rearrangement. First, the fact that Ts12 embryos are exencephalic and die in utero suggests a gene-dosage effect due to the presence and expression of alleles from the extra chromosome 12. In addition, we found that the paternal or maternal disomy is not influencing the ratio of halotype a- vs halotype b-expressing cells, which argues against an imprinting phenomenon. Finally, in the characterization of the rearrangement status of the alleles, we demonstrate that, globally and within each cell, all three loci are capable of undergoing somatic rearrangement.
We show that Ts12 chimeras exclude phenotypically two of the three IgH alleles, at the level of individual cells, probably as efficiently as euploid cells exclude one allele from expression at the cell surface. At this level, allelic exclusion is a consequence of successive constraints for the expression of both alleles. The initial constraint is the limited number of cells carrying both alleles successfully rearranged (15, 37). Furthermore, even in these few cells, only one of the allelic forms was shown to be capable of pairing with the surrogate light chain, the other being excluded from surface expression (15). At the level of the BCR expression, it is likely that the conventional light chain will also introduce a pairing constraint on heavy chains similar to that of the surrogate light chains. Nevertheless, rare mature dual-expressing cells were shown to be present in the spleen of normal mice (27). Another factor contributing to the low frequency of dual-expressing cells is a higher probability of receptor editing and/or deletion of cells expressing autoreactive receptors, in the bone marrow (38). However, it is only at the level of V to DJH rearrangement that the introduction of another allele in the germline configuration could make a difference when compared with euploid cells. We, therefore, evaluated the level of genetic allelic inclusion in the Ts12 B cells. A particular sequence polymorphism in the J region between the BALB/c and C57BL/6 alleles allowed us to analyze the rearrangement status of the silenced BALB/c allele (IgHa) from IgHb-expressing splenic B cells sorted from a IgHabb Ts12 chimera (26). Most VHDJH-recombined BALB/c alleles presented rearrangements that could not encode a full-length µ-chain, because sequences were not in frame at the level of the junctions. We concluded that there is no major alteration in the level of genetic allelic inclusion, as there is no selection for productively rearranged BALB/c alleles in the cells analyzed. Equivalent studies performed in euploid cells are somewhat contradictory. In the first evaluation of the rearrangement status of the silenced alleles in euploid cells, four of seven rearrangements had "no obvious impediments to expression" (37). More recent analyses based on single cell PCR indicate a much lower frequency of cells with genetic allelic inclusion (about 2.5%) (15). Although several cells were analyzed, the frequency of 2.5% comes from the detection of two cells, and PCR analyses do not detect rearrangements involving deletions or other alterations. Therefore, the actual frequency of allelically included cells has not been fully evaluated. Notwithstanding these cautionary notes, it seems clear that in the IgH Ts12 chimeras, most cells have nonproductive rearrangements in the silenced alleles.
Curiously, we found a modest increase in the fraction of D-JH rearrangements carrying the D element in RFII. These are usually underrepresented in euploid cells since such rearrangement leads to the expression of a truncated µ protein capable of inhibiting further rearrangements in these cells and leading to their elimination (29, 30). The finding that Ts12 cells express an increased frequency of such rearrangements suggests that DJH and VHDJH rearrangements are not as orderly separated as in euploid cells, and that some D to JH might occur at the same time or even succeed VH to DJH recombination.
Note added in proof.
The number of sequences from BALB/c IgH alleles was increased to a total of 22. Of these, 5 are + rearrangements (according to the family of the VH gene, the numbers of + to – rearrangements were 1:6 (J558), 2:5 (7183), 2:5 (Q52), and 0:1 (MRLDNA-4). The number of VH pseudogenes may be underestimated because sequence analysis of the amplicons does not detect mutations in VH regulatory regions or in coding sequences upstream of those amplified by PCR.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Vasco Barreto, Unité du Développement des Lymphocytes, Institut Pasteur, 28 rue du Docteur Roux 75 724, Paris Cedex 15, France.
3 Abbreviations used in this paper: BCR, B cell receptor; dpc, days postcoitum; Rb, Robertsonian; RF, reading frame; Ts12, chromosome 12 trisomic.
Received for publication December 13, 2000. Accepted for publication February 7, 2001.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-globulin genes in antibody-producing cells. Proc. Natl. Acad. Sci. USA 54:1765. genes: a gene for a complete light chain and a gene for a constant region polypeptide. Cell 12:453.[Medline]
genes in a transgenic mouse line. Nature 314:330.[Medline]
chain causes allelic exclusion in transgenic mice. Nature 330:482.[Medline]
genes encoding antiphosphorylcholine antibodies. J. Exp. Med. 164:627. and heavy chains to chromosomes 6 and 12 in mouse. Proc. Natl. Acad. Sci. USA 75:4494. chain genes: implications for allelic exclusion. Proc. Natl. Acad. Sci. USA 77:1937.This article has been cited by other articles:
|
|
 |
|
  G. Kalmanovich and R. Mehr Models for Antigen Receptor Gene Rearrangement. III. Heavy and Light Chain Allelic Exclusion J. Immunol., January 1, 2003; 170(1): 182 - 193. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
3. C, Anti-IgMa and anti-IgMb
FACS profile staining on splenocytes from chimeras reconstituted with
euploid F1 cells (Ly-5.1+) and Ts12 abb
(Ly-5.1-), gated on Ly-5.1 positive and negative
(n = 3). The IgMa/IgMb ratios shown in these
profiles are independent of the euploid/Ts12 cell ratio shown in the
histogram.