Antibody Repertoire Development in Fetal and Neonatal Piglets. VI. B Cell Lymphogenesis Occurs at Multiple Sites with Differences in the Frequency of In-frame Rearrangements

Marek Šinkora, Jishan Sun, Jana Šinkorová, Ronald K. Christenson, Steven P. Ford and John E. Butler
J Immunol February 15, 2003, 170 (4) 1781-1788; DOI: https://doi.org/10.4049/jimmunol.170.4.1781

Abstract

B cell lymphogenesis in mammals occurs in various tissues during development but it is generally accepted that it operates by the same mechanism in all tissues. We show that in swine, the frequency of in-frame (IF) VDJ rearrangements differs among yolk sac, fetal liver, spleen, early thymus, bone marrow, and late thymus. All VDJ rearrangements recovered and analyzed on the 20th day of gestation (DG20) from the yolk sac were 100% IF. Those recovered at DG30 in the fetal liver were >90% IF, and this predominance of cells with apparently a single IF rearrangement continued in all organs until approximately DG45, which corresponds to the time when lymphopoiesis begins in the bone marrow. Thereafter, the proportion of IF rearrangements drops to ∼71%, i.e., the value predicted whether VDJ rearrangement is random and both chromosomes were involved. Unlike other tissues, VDJs recovered from thymus after DG50 display a pattern suggesting no selection for IF rearrangements. Regardless of differences in the proportion of IF rearrangements, we observed no significant age- or tissue-dependent changes in CDR3 diversity, N region additions, or other characteristics of fetal VDJs during ontogeny. These findings indicate there are multiple sites of B cell lymphogenesis in fetal piglets and differences in the frequency of productive VDJ rearrangements at various sites. We propose the latter to result from differential selection or a developmentally dependent change in the intrinsic mechanism of VDJ rearrangement.

Differentiation of B cell precursors from multipotent stem cells to immature B cells is characterized by the status of Ig gene rearrangement and the expression of certain genes (1, 2, 3, 4, 5). The onset of B cell lymphogenesis is associated with combinatorial joining of VH, DH, and JH gene segments in the Ig H chain (HC)3 locus, i.e., VDJ rearrangement. Precursor pro-B cells that have productively rearranged their VDJs further differentiate to the early pre-B cell stage (preB-I) in which the pre-B cell receptor (pre-BCR), consisting of VDJ spliced to the IgM HC (μHC) is expressed on the cell surface in association with surrogate L chain (LC) (λ5 plus VpreB) and the signaling components Ig-α and Ig-β (reviewed in Ref.6). This pre-BCR is thought to deliver signals that stop further rearrangement in the HC locus. Differentiation to the late pre-B cell stage (preB-II) is characterized by Ig LC gene segment rearrangement (Vλ or Vκ jointed to Jλ or Jκ) and replacement of the surrogate LC with authentic LC on the cell surface (2, 3). This authentic BCR is the defining marker of immature B cells that are the direct predecessors of the transitional B cells designated T-1 and T-2 B cells (7).

Development of a functional B cell requires that the VDJ rearrangement is productive, i.e., the generated reading frame is open or in-frame (IF), and contains no stop codon(s). Only such VDJ rearrangements can be thereafter spliced and translated, and resultant μHC can be consequently expressed on the cell surface. Unlike the LC locus, the HC locus in mammals so far studied can be rearranged only once because of the nontandem arrangement of VH, DH, and JH segments resulting in the lost of recombination signal sequences during VDJ rearrangement. This is particularly demonstrated in chickens and swine that have only a single JH (8, 9, 10). B cell precursors that do not succeed in making a productive rearrangement using the first chromosome (theoretically two-thirds of all cells if contribution of reading frame is only counted and the process of rearrangement is random) have a second chance to generate a translatable μHC by using the second chromosome, again with a ∼67% chance to failure. B cell precursors that do not attain pre-B-I due to lack of a productive rearrangement are presumably eliminated. Thus, <55% of all precursor B cells that start the recombination process can express a μHC, and the survivors will have frequencies of IF and out-of-frame (OF) rearrangements of ∼71 and ∼29%, respectively, if the process is random.

We report in this study that certain B cell populations from fetal tissues do not fulfill the textbook description of B cell lymphogenesis that yields 71% IF rearrangements. Rather, fetal B cells generated before B cell lymphogenesis in the bone marrow have ∼100% of their VDJs IF and appear to carry no rearrangement on the second allele. We also show that most B cells found in the thymus during late fetal life carry VDJ rearrangements that show no selection for IF rearrangement, whereas those arising from bone marrow follow the established paradigm. We hypothesize that differences in the microenvironment at different sites of B cell lymphogenesis and at different times in these sites control the frequency of IF rearrangement either through selection or differences in the rearrangement program.

Materials and Methods

Source of animal tissue, RNA, and DNA

White cross-bred gilts (one-quarter Yorkshire, one-quarter Chester White, one-quarter Large White, one-quarter Landrace) from the Roman L. Hruska U.S. Meat Animal Research Center (Clay City, NE) and Yorkshire and Meshian F1 cross-bred animals from Iowa State University (Ames, IA) were used in the study. Animals were hand-mated and scheduled for slaughter and collection of fetuses at different ages. Gestation age was calculated from the day of mating. Gestation in swine is 114 days. All gilts were healthy and normal at slaughter, and fetuses were immediately removed from the gravid uterus. Different organs were removed from the fetuses, immediately frozen in liquid nitrogen, and stored at −70°C until preparation of RNA, cDNA, and DNA as described previously (11, 12). In the case of fetal liver, yolk sac, and bone marrow, cell suspensions were prepared before freezing. All animal experiments were approved by the Animal Care and Use Committees of the Roman L. Hruska U.S. Meat Animal Research Center and Iowa State University, according to guidelines in the Animal Protection Act.

Preparation of cell suspensions

Cell suspensions from fetal liver and spleen were prepared in cold PBS by carefully teasing the tissues using a forceps and then by passage through a 70-μm mesh nylon membrane. Cells from excised whole yolk sacs were isolated by collagenase digestion as described previously (13, 14). Bone marrow cells were directly flushed from the tibia and/or femur. Erythrocytes and most erythroblasts were removed from the pelleted cells using hypotonic lysis (13, 14). Cell suspensions were washed twice in cold PBS and filtered through a 70-μm mesh nylon membrane, and cell numbers were determined by hemacytometer.

PCR and RT-PCR for VDJ, LC-κ, LC-λ, TdT, recombinase activation gene-1 (RAG-1), β-actin, and IgA HC (CHα)

First-strand cDNAs synthesis was performed by reverse transcription of 2 μg of total RNA primed with 200 ng of random hexamer primer using 0.5 mM dNTP, 1 mM DL-DTT, 100 ng BSA, 20 U of RNasin (Promega, Madison, WI), and 200 U of Moloney murine leukemia virus reverse transcriptase (Life Technologies, Rockville, MD) in a recommended buffer in a total volume of 20 μl. As a negative control, reactions were performed replacing the Moloney murine leukemia virus polymerase by H2O. Genes or gene segments of interest were amplified from cDNA and DNA preparations using a sense and anti-sense primer pairs given in Table I⇓. In the case of DG30 fetal liver and DG20 yolk sac, second round-nested or seminested PCR was used for the recovery of VDJs (Table I⇓). As controls for determining relative transcript expression and efficiency of amplification, β-actin and CHα were amplified from cDNA and DNA, respectively. All PCR were performed in a 25 μl volume using ∼200 ng of template DNA or 10% of cDNA preparations and 2.5 μl of 10× PCR buffer, 2 mM MgCl2, 0.4 mM dNTP, 1 U of thermostable KlenTaq1 DNA polymerase (Ab Peptides, St. Louis, MO), 18 pmol of a sense primer and 18 pmol of a anti-sense primer. After 2 min of initial denaturation at 94°C, the samples were subjected to 30–45 cycles of amplification (30 s of denaturing at 94°C, 30 s of annealing at 55°C for amplification of TdT and RAG-1, 61°C for amplification CHα or 58°C for all other amplification, and 60 s extension at 72°C) in a PTC-200 DNA Engine thermal cycler (MJ Research, Waltham, MA). PCR products were analyzed in 1.5% agarose gels that were stained by ethidium bromide. In some cases, products were excised from the gel and purified using the Wizard Plus Miniprep DNA Purification System (WIZARD kit, Promega).

View this table:
Table I.

Sequences of oligonucleotides used as primers

Complementarity-determining region (CDR)3 length analysis (CDR3 spectratyping and CDR3 fingerprinting) and analysis of IF and OF status

CDR3 length analysis referred to in this study as spectratyping was performed on uncloned VDJs amplified by PCR from total DNA isolated from whole organs or cell suspensions and on VDJs amplified from randomly selected clones for analysis of CDR3 length and IF vs OF status. Technically, the CDR3 segments of the amplified VDJs were re-amplified using an FR3 up-primer (Table I⇑) that anneals to all porcine VH genes and a 32P-labeled JH down-primer. Use of this primer set yields PCR products of uniform length for each VDJ in >90% of the amplifications, which together with prolonged final extension time and reduced elongation time (initial denaturation at 94°C for 1 min, 20 cycles of denaturation at 94°C for 40 s, annealing at 55°C for 30 s, and final extension at 72°C for 30 min), reduced shadow band formation. The re-amplified CDR3s were then placed in 100% formamide containing bromphenol blue and xylene cyanol and separated on sequencing gels after denaturation at 95°C for 5 min (15). Images were either directly obtained using an Instant Imager (Packard Instrument, Meriden, CT) or indirectly using autoradiography. Length standards were prepared by amplification of a mixture of individual cloned VDJs of known length (66, 57, 48, 39, and 30 bp). Furthermore, VDJs amplified from total genomic DNA were isolated from DG110 spleen and used as size references for determining IF and OF rearrangements. The 3 bp difference between IF rearrangements of different lengths could be easily identified because of their higher band intensity. As an independent confirmation of the accuracy of determining IF and OF assignments, 86 clones were sequenced to reconfirm their reading frame. Their initial assignment based on spectratyping was confirmed in 100% of the cases using sequence data.

Cloning of VDJs and hybridization

Cloning was performed as described previously (15). Briefly, amplified and purified VDJ rearrangements were re-amplified by PCR using primers for second-round VDJ amplification (Table I⇑), and Taq polymerase was replaced by Pfu polymerase (Stratagene, La Jolla, CA). The use of Pfu eliminates adenine overhang and facilitates efficient ligation. The PCR product was isolated using a WIZARD kit and cloned into EcoRV digested pBluescript phagemids by blunt-end ligation. The ligation mixture was used to transform DH5α competent cells, and the positive recombinant clones that were identified by blue/white selection were transferred to nylon membranes and the membranes hybridized with a 32P end-labeled FR2 pan-specific oligonucleotide probe (cgccaggctccagggaag). These membranes were wrapped in plastic film and the positive clones were identified using an Instant Imager (Packard Instrument).

Sequence analysis

Sequencing of VDJs was done using the SequiTherm EXCEL II DNA Sequencing kit (Epicentre Technologies, Madison, WI) designed for cycle sequencing as described previously (15). Specifically, the amplified polynucleotides were analyzed on 10% polyacrylamide sequencing gel using a POKER FACE II SE 1600 (Hoefer Scientific Instruments, San Francisco, CA) or Sequi-Gen GT apparatus (Bio-Rad, Hercules, CA). Gels were dried in a 583 gel dryer (Bio-Rad), and Kodak X-Omat Blue XB-1 film was used for autoradiography. Alternatively, samples were sent to the DNA core facility of University of Iowa where they were sequenated using the fluorescent automated sequencer 373A or 377 (Applied Biosystems, Foster City, CA). Sequences of VDJs were analyzed for reading frame continuity, CDR3 length, N region additions, DH usage, and mutation frequency.

Staining of cells, immunoreagents, and flow cytometry

Staining of cells for flow cytometry analysis was performed as described previously (13, 14). Briefly, cell suspensions were incubated with a combination of two primary mAbs of different subisotypes for 30 min and subsequently washed twice. The following mouse anti-pig mAbs were used as primary immunoreagents: anti-CD2 (MSA4, IgG2a), anti-CD3ε (PPT3, IgG1 or PPT6, IgG2b), anti-TCRγδ (PPT17, IgG1 or PPT16, IgG2b), anti-CD25 (K231.3B2, IgG1), anti-CD45 (K252.1E4, IgG1, pan-leukocyte Ag), anti-SWC3a (74-22-15, IgG2b, pan-myelomonocytic Ag), anti-SLA-DR (MSA3, IgG2a, MHC class II type Ag), and anti-IgM (LIG4, IgG1). Mixtures of goat polyclonal Abs specific for mouse Ig subclasses (Southern Biotechnology Associates, Birmingham, AL) labeled with FITC or R-PE were then added to the cell pellets in appropriate combinations. After 30 min, cells were washed three times and analyzed on a standard FACSort flow cytometer (BD Immunocytometry Systems, Mountain View, CA). Electronic compensation was used to eliminate residual spectral overlaps between individual fluorochromes. Damaged and dead cells were excluded from analysis using propidium iodide fluorescence.

Results

Early B cell lymphogenesis begins in the yolk sac and then moves to the liver

All swine VH genes are members of a single, highly homologous family, share a common leader sequence and several FR sequences, and are rearranged to the single JH segment (8, 16). Thus all VDJ rearrangements can be amplified from DNA or cDNA using a single primer set. Using this principle, VDJ rearrangements were first detected in the yolk sac DNA at DG20 (Fig. 1⇓, first row, left panel). Spectratypic analysis of the CDR3 in these rearrangements indicated that the preimmune B cell repertoire of the yolk sac is very small (Fig. 2⇓A, column I), and that <10 B cells are present in a given sample (Fig. 2⇓C). Oligoclonal patterns were a common feature of all samples but prominent CDR3s of the same length were not shared among samples and VDJ rearrangement could only be recovered from half of all yolk sac samples prepared (data not shown). The yolk sac is apparently the first site of VDJ rearrangement because amplification of VDJs from other parts of the embryo at DG20, DG23, DG26, and DG28 was unsuccessful (data not shown). Because the yolk sac is actually extraembryonic, VDJ rearrangement within the embryo was not seen until DG30 in the fetal liver (Fig. 1⇓, first row, middle panel). Based on spectratype, the VDJ repertoire from fetal liver was also oligoclonal with few prominent bands (Fig. 2⇓B, column I) and large interanimal variation (data not shown).

FIGURE 1.
FIGURE 1.

Detection of transcripts for VDJ, LC, TdT, and RAG-1 in the yolk sac at DG20 (left panel) and fetal liver at DG30 (middle panel) and DG40 (right panel). β-actin was used as a control for determining relative transcript expression. cDNA synthesis was performed with the reverse transcriptase (+RT), whereas (−RT) indicates that cDNA synthesis was performed without reverse transcriptase.

FIGURE 2.
FIGURE 2.

Length analysis of CDR3 in representative samples from yolk sac (A) and early fetal liver (B). VDJ rearrangements were recovered form DNA (columns I) or cDNA (columns II and III), using an FR1 up- and JH down-primer set (columns I and II) or FR1 up- and Cμ down-primer set (column III). The CDR3 segments were then amplified using FR3 up-primer and 32P-labeled JH down-primer set and run on sequencing gel. Note that use of FR1 up-/Cμ down-primer set yielded no detectable product from yolk sac cDNA (A, column III) and less CDR3 product from fetal liver cDNA (B, column III) than when the FR1 up/JH down-primer set was used (A, column II and B, column II, respectively). The right panel (C) is a model spectratype from data obtained when different numbers of B cells were analyzed, i.e., 10, 20, 30, 50, and 100 cells (data are based on randomly picked IF VDJ clones recovered from spleen at DG110). The intensity of the bands in the model depicts the number of clones with the same length of CDR3. CDR3 size markers are indicated on the extreme right margin.

The extent of B cell lymphogenesis in yolk sac and fetal liver was assessed using different sets of primers to amplify the rearranged VDJs by RT-PCR. Use of the FR1 up-/Cμ down-primer set (Table I⇑) yielded no detectable product in the yolk sac (Fig. 2⇑A, column III), and significantly fewer CDR3 bands were recovered from the fetal liver (Fig. 2⇑B, column III) than when the FR1 up-/JH down-primer set was used (Fig. 2⇑, A and B, columns II). This finding indicates that spliced mature VDJ transcripts are absent or infrequent in the yolk sac, whereas such product are recovered in fetal liver although in fewer number than nonspliced primary transcripts.

TdT is active from earliest VDJ rearrangements, and detection of TdT, RAG-1, LC-λ, and LC-κ transcripts discriminates between yolk sac and fetal liver

Although VDJ rearrangements could be detected in the yolk sac at DG20, we were unable to amplify transcripts for TdT, RAG-1, and Cλ- or Cκ-LC (Fig. 1⇑). Nevertheless, sequence analysis of cloned VDJs (see below) showed that TdT had been active at the time when the first VDJ rearrangements occurred in the yolk sac. Surprisingly, the number of N region additions and overall average CDR3 length were similar in the DG20 yolk sac and in lymphoid tissues at DG110 (see below) (17). However, transcripts for TdT and both types of LC were recovered from the DG30 and DG40 fetal liver (Fig. 1⇑, middle and right panel, respectively). Both TdT and RAG-1 were present in cDNAs prepared from thymus and bone marrow at DG40 and DG50, respectively (Fig. 3⇓). Furthermore, TdT transcripts were present in fetal liver at DG30 and DG40 and again from DG70 until birth (Fig. 3⇓, left panel). TdT transcripts could also be recovered in the spleen of fetuses from DG60 until birth (Fig. 3⇓, left panel).

FIGURE 3.
FIGURE 3.

TdT (left panel) and RAG-1 (right panel) expression during fetal ontogeny in fetal liver, spleen, bone marrow (BM), and thymus. DG are indicated on the extreme left margin for each panel. N.A., not applicable.

There are no significant age- or tissue-dependent changes in VDJ repertoire and CDR3 diversity during fetal life

Sequence analysis of randomly sampled VDJ rearrangements recovered from the yolk sac at DG20 and the fetal liver at DG30 showed that N region additions were similar in number as found in bone marrow and spleen at DG110 (Fig. 4⇓A). Very few recognizable P nucleotide additions were detected (Fig. 5⇓). The frequency of somatic point mutations within VDJ remained low during fetal life and did not increase with age (Figs. 4⇓B and 5⇓). There was also no age-dependent change in the mean length of CDR3 (Fig. 4⇓C). Consistent with earlier studies (8, 11), only a single JH and only two DH segments (DHA and DHB) were recovered. The overall length of the utilized DH segments was comparable at DG20, DG30, and DG110 (Fig. 4⇓D). There was also no difference between mean length of the DHA and DHB segments used in CDR3 despite the fact that DHA is longer (36 bp germline) than DHB (28 bp germline) (Fig. 5⇓). We found no evidence for D-D rearrangements, for mini-DH sequences, or that short homology segments had been inserted (Fig. 5⇓). The individual sequences did not show close clonal relationship, and no significant reading frame preferences were found (Fig. 5⇓).

FIGURE 4.
FIGURE 4.

Analysis of VDJ rearrangements recovered form yolk sac (YS) at DG20, fetal liver (FL) at DG30, bone marrow (BM) at DG110, and spleen (SP) at DG110. Number of N-additions (A), somatic hypermutations within VDJ (B), CDR3 length (C), and length of DH segments (D) are indicated. Actual data points are indicated by black dots (•) and error bars represent ± SEM. Proportional VH and DH usage (E) is also shown for YS (▪), FL (▦), BM (□), and SP (▤).

FIGURE 5.
FIGURE 5.

The DNA nucleotide sequence of CDR3s recovered from the yolk sac at DG20 (upper 10 sequences) and fetal liver at DG30 (lower 10 sequences). Five representative sequences for each type of DH segment used (DHA or DHB) are shown. Germline sequence of DHA, DHB, and JH are indicated in each headline. Bold nucleotides in “DH ” column indicate point mutations and underlined nucleotides in “additions” columns indicate P nucleotides. IF and OF status and reading frame (RF) preferences are indicated on the extreme right.

Approximately two-thirds of early VDJs used DHB, and proportional usage of DHB during fetal ontogeny was always slightly higher than usage of DHA (Fig. 4⇑E). Analysis of VH usage in early VDJ rearrangements was also consistent with our earlier reports (11) in showing that four nonmutated VH genes (VHA, VHB, VHC, and VHE) accounted for 75–90% of all VH usage (Fig. 4⇑E). Although the frequency of particular VH segments used in VDJ rearrangements varies during fetal ontogeny (probably due to differences between individual animals and litters), we did not observe preferential usage of one particular VH segment at the time of the earliest VDJ rearrangements (Fig. 4⇑E).

VDJ rearrangements in early fetal development are nearly all IF, whereas VDJs recovered from thymus after DG50 displayed a pattern suggesting no selection for IF rearrangements

The oligoclonal spectratype of VDJs from yolk sac DNA and fetal liver DNA suggested that selection and proliferation of B cells carrying a single IF rearrangement was preferred (Fig. 2⇑). This contrasts with the spectratype seen later in the bone marrow, thymus, and other lymphoid organs in which OF rearrangements were routinely observed (Fig. 6⇓). This tendency was further confirmed by examining the CDR3 length of randomly selected VDJ clones recovered from yolk sac (50 clones), fetal liver (50 clones for DG30 and 90 clones for other DG), spleen (90 clones for each DG), bone marrow (90 clones for each DG), and thymus (90 clones for each DG) at different time points during fetal ontogeny (Fig. 7⇓). This analysis showed that randomly selected VDJ clones recovered from the yolk sac at DG20 were 100% IF, and those recovered from the fetal liver at DG30 were >90% IF (Fig. 7⇓). Furthermore, CDR3 length analysis of randomly selected VDJ clones recovered from fetal liver, spleen, and thymus before DG45 gave a value of ∼90% IF rearrangements as exemplified in DG30 fetal liver (Fig. 7⇓). These percentages are significantly greater than the expected ∼71% of IF VDJ rearrangement based on the assumption that cell survival and VDJ rearrangements are random events. After DG45, values approximating 71% IF rearrangements were observed for VDJ clones from all studied organs, suggesting random VDJ rearrangement involving both chromosomes (Fig. 7⇓). The most striking effect of development on proportion of IF rearrangement was seen in the fetal thymus. At DG40 ∼93% of VDJs were IF, at DG50 ∼75%, but only ∼33% after DG50. Note, that VDJ rearrangements from 86 clones were sequenced as independent confirmation of the accuracy of determining IF and OF assignments by CDR3 length analysis. Their initial assignment based on spectratyping was confirmed in 100% of the cases using sequence data (data not shown). Moreover, the frequencies of IF and OF rearrangements determined by CDR3 length analysis of randomly selected clones were again reconfirmed by sequencing analysis of randomly selected VDJ clones for yolk sac at DG20 (13 sequences; 100% IF), fetal liver at DG30 (48 sequences; 92% IF), bone marrow at DG110 (35 sequences; 68% IF), and spleen at DG110 (34 sequences, 76% IF).

FIGURE 6.
FIGURE 6.

Length analysis of CDR3 in yolk sac (YS), fetal liver, thymus, spleen, and bone marrow at different time points during fetal ontogeny (DG indicated in the upper line). VDJ rearrangements recovered from DNA were initially amplified using FR1a up- and JH down-primer set, and the resulting PCR product secondarily amplified using FR3 up-primer and 32P-labeled JH down-primer set. CDR3 size markers are indicated on the left.

FIGURE 7.
FIGURE 7.

The proportion of IF (▦) and OF (□) VDJ rearrangements in different organs at different time points during fetal ontogeny (DG indicated in the lower line). Results are based on examining the CDR3 length of randomly selected VDJ clones recovered from yolk sac (YS), fetal liver, spleen, bone marrow (BM), and thymus. Based on the assumption that cell survival and VDJ rearrangements are random events and that VDJ rearrangement may proceed on both chromosomes, expected value of 71.4% of IF VDJ rearrangements is expected. A value of 33.3% of IF VDJ rearrangement would be expected in the absence of selection. Errors of 10% from expected values are indicated by hairlines. The values are based on analysis of at least 50 clones for yolk sac and fetal liver at DG30 and 90 clones for each remaining bar.

Putative B cell precursors can be detected by flow cytometry in fetal liver

Cells expressing IgM or μHC could not be clearly detected by flow cytometry in yolk sac at DG20 (data not shown) and fetal liver at DG30 (Fig. 8⇓A), although VDJ rearrangements could be detected by PCR (Fig. 1⇑). Analysis of other cell surface markers revealed cells with a phenotype characteristic of putative B cell precursors. These cells expressed high levels of MHC class II (SLA-DR; Fig. 8⇓A, column II) and low levels of CD2 (Fig. 8⇓A, column I) and CD25 (Fig. 8⇓A, column III). Neither T cells (Fig. 8⇓A, column V) nor true lymphoid cells (CD45+SWC3a−; Fig. 8⇓A, column IV) were detected in the fetal liver at DG30. By comparison, a mixture of B cell precursors (IgM−/lowSLA-DRhighCD2lowCD25low) and immature B cells (IgM+SLA-DRhighCD2lowCD25low) were seen in DG110 bone marrow (Fig. 8⇓B), whereas predominantly immature B cells (IgMhighSLA-DRhighCD2lowCD25low) were seen in DG110 spleen (Fig. 8⇓C).

FIGURE 8.
FIGURE 8.

Flow cytometric analysis of cells isolated from fetal liver at DG30 (A), bone marrow at DG110 (B) and spleen at DG110 (C). Cells were stained with anti-CD2, anti-SLA-DR (MHC-II), anti-CD25, and anti-IgM in appropriate combinations (columns I–III) and analyzed in a mononuclear gate defined by forward/side scatter. Control staining of anti-CD45 (pan-leukocyte marker) and anti-SWC3a (pan-myelomonocytic marker) allowed identification of different leukocyte populations (column IV). Anti-CD3ε and anti-TCRγδ (column V) were used to identify αβ and γδ T cells.

Discussion

The earliest source of multipotent hematopoietic cells in the mouse embryo is the para-aortic splanchnopleura (18) or aorta-gonad-mesonephros (19, 20). Although cells derived from this tissue can generate lymphocytes and other cell types when cultured under appropriate condition, B cell lymphogenesis and VDJ rearrangements could not be detected in these tissues (18, 19, 20). It is generally regarded that development of the preimmune BCR repertoire in humans and mice initially begins in the yolk sac, then continues in fetal liver, and finally shifts to the bone marrow (5). We show in this study that B cell lymphogenesis and the appearance of B cells in the pig fetus follows the same pattern. Specifically, VDJ rearrangement in fetal piglets was first detected at DG20 in the yolk sac and then 10 days later in the fetal liver, although RAG-1 was not detected by RT-PCR (Figs. 1⇑ and 3⇑). The yolk sac in the fetal pig involutes after DG24 (21), so termination of its role in B lymphogenesis after DG20 would be expected. Thus it appears that fetal liver is the major site of B cell lymphogenesis in the porcine embryo before the bone marrow becomes active at about DG45 (14, 22).

Sequence analysis showed N region additions in the earliest VDJ rearrangements recovered from the yolk sac, although like RAG-1, TdT was not detected by RT-PCR. The frequency of N region additions did not change during fetal life, which is in sharp contrast with rodents and humans in which TdT is less active early in fetal ontogeny (23, 24, 25). Our failure to detect TdT and RAG-1 expression at DG20 suggests perhaps that the number of cells involved in VDJ rearrangement was extremely low. The alternative explanation for our failure to detect TdT and RAG-1 transcripts in the yolk sac is that the VDJ rearrangements detected were from immigrant B cells. This explanation, however, seems unlikely because: 1) VDJ rearrangements were not found elsewhere in the fetus at DG20; 2) LC transcripts were not recovered; and 3) no mature VDJ transcripts were recovered. Furthermore, the absence of somatic mutation and the preferential use of VHB and DHB in the VDJs recovered from yolk sac are not characteristic of those from the mother (12, 17).

An intriguing observation in these studies was the nearly exclusive recovery of IF rearrangements before DG50, whereas later in ontogeny, 56–78% of recovered rearrangements were IF as would be predicted by random recombination on both chromosomes. One might expect to find only IF rearrangements if: 1) only germline gene segments with no nucleotide additions or segment trimming occurred; 2) all OF rearrangements are edited to yield open reading frames; 3) production of an OF rearrangement triggers an apoptotic pathway or activation of a deletion element; or 4) cells that produce an OF rearrangement on their first attempt die from lack of selection before an IF rearrangement occurs on the other chromosome. The first explanation does not fit the data presented because unlike mice and early human fetuses, N region additions and segment trimming are evident from analysis of DG20 and DG30 sequences (Figs. 4⇑ and 5⇑).

Receptor editing of VDJ rearrangements by successive rearrangement of the type seen in the loci encoding LCs and the TCR is impossible in swine because they have only a single JH (8). Although receptor revision by gene conversion has been suggested to operate in some species (26, 27), evidence that this nonreciprocal exchange can alter the reading frame in CDR3 has not been shown. Although intriguing, our recovery of nearly 100% IF rearrangements is not unique and was previously reported in chickens (9) and rabbits (28). The simultaneous DJ rearrangements on both alleles prior to full VDJ rearrangement on one allele, as known for mice (29), appears not occur in chickens (9) and rabbits (28), and VD rearrangements can precede VDJ rearrangements in rabbits (28). Therefore, the majority of B cells in rabbits and chickens have JH in germline configuration and number of IF VDJ rearrangements approximates 100%. Unfortunately, we currently have no data on the rearrangement status of the second allele in pre-DG50 piglet B cells that display circa 100% IF rearrangements. Because the porcine VH locus has not been mapped, the possibility remains that its organization allows rearrangements or contains regulatory elements that could explain our observations in the similar way as reported for LC locus of chickens (30), mice (31), or humans (32). Currently we lack mAb to pan- and stage-specific B cell CD markers that would allow us to examine the status of VDJ rearrangement on both alleles in B cells from different sources at various stages of B cell lymphogenesis. In any case, it is notable that B cells in chickens and rabbits with 100% IF VDJ rearrangements are generated during a narrow window in fetal life (33). These B cells are thereafter maintained through life and their somatic diversification occurs in the chicken bursa (33) and the rabbit appendix (27), respectively. These observations in chickens and rabbits, together with our data on pre-DG50 piglet B cells, hint at the possibility that B cells generated early in ontogeny and/or in non-bone marrow sites are limited in progressive VDJ rearrangement on second chromosome compared with B cells generated in the bone marrow.

The possibility of a deletion element, as previously cited, could also explain our results as well as the activation of a apoptotic pathway for cells that fail to generate an IF VDJ in reasonable time. This idea is similar to our previous hypothesis, which is a kinetic selection hypothesis. Realizing that B cell lymphogenesis is under cytokine control and surrounding microenvironment, a different milieu of cytokines and/or surrounding cells may be present in yolk sac and fetal liver than in bone marrow. This could translate into a slower rate of VDJ rearrangement such that pro-B cells unable to produce a pre-BCR on their membrane die before a productive rearrangement is made on the second allele and the product can be expressed on the cell membrane. Feedback regulation in cells with a pre-BCR would prevent subsequent rearrangements on the second allele. This would result in a scenario whereby all B cells selected for expansion in yolk sac and liver would be those with a single IF rearrangement. Small amounts of OF rearrangements in non-bone marrow tissues before DG50 (Figs. 6⇑ and 7⇑) might come from developing B-lineage cells that are just in a selection step for IF rearrangement.

B cell lymphogenetic activity in bone marrow begins at about DG45 (13, 14, 22), and thereafter the proportion of IF rearrangements reach a value of ∼70% indicating that the rearrangement process is random, occurs sequentially on both chromosomes, and follows the mouse paradigm. Because we observed no significant age- or tissue-dependent changes in VDJ diversity during fetal life, i.e., VH usage and N region additions, the IF/OF ratio is the only difference that distinguishes early B cells from yolk sac and fetal liver from those later recovered from bone marrow on DNA level. Whether these represent developmentally distinct B cell subsets as described for B-1 and B-2 cells in mice (34) or the differential influence of the environment in yolk sac and fetal liver vs the bone marrow, remains to be determined. Regarding the former, the self-generating B-1 population has been traditionally recognized by its expression of CD5 and its comparatively low expression of IgD (34, 35). However, only rodents and primates express IgD (8) and CD5 expression can change with age and antigenic stimulation (36, 37). Moreover, the studies in different species showed discrepancy between expression CD5 and fetal vs conventional B cell lineage commitment (34, 35, 38, 39). Regarding environmental factors, stromal ligands involved in positive (40) or negative (41) B cell selection may differ among the sites where B cells develop. For example, the occurrence of only 33% IF VDJ rearrangements in thymus after DG50 (Fig. 6⇑) may indicate the lack of either positive or negative selection.

Data presented in this report indicate the complexity of B cell lymphogenesis during fetal life and the very real possibility that this process differs in regard to the age of the fetus and the tissue/organ involved. Because liver, bone marrow, and thymus have significantly different cellular architectures, microenvironmental factors are more likely to be responsible than intrinsic progenitor factors. Whether B cells developed at different sites represent discrete subsets with discrete functions remains to be determined. However, because VH usage, DH usage, and TdT activity is similar in fetal liver and bone marrow, certain intrinsic or selective factors are conserved across tissues/organs and the stage of ontogeny.

We previously presented evidence for the unusual spectratype of late-term fetal thymus (17). Here we expand these observations by showing that this unusual pattern is not present before DG70. Because all polynucleotides differing by one nucleotide are equally represented after DG70, data suggest lack of B cell selection. Because this is the dominant spectratype of thymus DNA, it could not be derived from the scattered mature B cells found in this organ (38) but must be due to pro-B cells that do not express a BCR because we cannot detect or recover this population using an anti-Ig reagent (M. Šinkora, unpublished data). Thus we consider this unusual spectratype must be derived from a sizeable unselected pro-B cell population that resides in the thymus but which does not develop further to B cell status. Data in this report indicate that this population is not present in early thymus. Rather, the spectratype from early thymus is probably derived from the small numbers of mature B cells in this organ that are most likely immigrants from other lymphoid tissues.

The phenomena described in this report support the value of studying fetal B cell lymphogenesis in nonrodent models. These observations provide a basis for hypotheses that can test for differences in B cell ontogeny that may be relevant to all species.

Acknowledgments

We are grateful to Patrick Weber for technical assistance and the following researchers for the gifts of mAbs: Dr. H. Yang (Institute of Animal Health, Pirbright, U.K.) for anti-CD3ε (PPT3 and PPT6) and anti-TCRγδ (PPT16 and PPT17); Dr. J. K. Lunney (Animal Parasitology Institute, Beltsville, MD) for anti-CD2 (MSA4), anti-SLA-DR (MSA3), and anti-SWC3a (74-22-15); and Dr. C. R. Stokes (University of Bristol, U.K.) for anti-CD25 (K231.3B2) and anti-CD45 (K252.1E4).

Footnotes

  • ↵1 This work was supported by Grant 524/01/0919 from the Grant Agency of the Czech Republic and National Science Foundation award MCB00-77237 to Universty of Iowa.

  • ↵2 Address correspondence and reprint requests to Dr. Marek Šinkora, Department of Immunology and Gnotobiology, Institute of Microbiology, Czech Academy of Science, Doly 183, 549 22 Nový Hrádek, Czech Republic. E-mail address: Marek.Sinkora{at}worldonline.cz

  • 3 Abbreviations used in this paper: HC, H chain; pre-BCR, pre-B cell receptor; CDR, complementarity-determining region; DG, day of gestation; μHC, IgM HC; CHα, constant portion of HC for IgA; IF, in-frame; LC, L chain; OF, out-of-frame; RAG, recombinase activation gene; SWC, swine workshop cluster.

  • Received August 28, 2002.
  • Accepted December 6, 2002.

References

  1. ↵
    Hardy, R. R., C. E. Carmack, S. A. Shinton, J. D. Kemp, K. Hayakawa. 1991. Resolution and characterization of pro-B and pre-pro-B cell stages in normal mouse bone marrow. J. Exp. Med. 173: 1213
    OpenUrlAbstract/FREE Full Text
  2. ↵
    Ehlich, A., S. Schaal, H. Gu, D. Kitamura, W. Muller, K. Rajewsky. 1993. Immunoglobulin heavy and light chain genes rearrange independently at early stages of B cell development. Cell 72: 695
    OpenUrlCrossRefPubMed
  3. ↵
    Ghia, P., E. ten Boekel, E. Sanz, A. de la Hera, A. Rolink, F. Melchers. 1996. Ordering of human bone marrow B lymphocyte precursors by single-cell polymerase chain reaction analyses of the rearrangement status of the immunoglobulin H and L chain gene loci. J. Exp. Med. 184: 2217
    OpenUrlAbstract/FREE Full Text
  4. ↵
    Rolink, A. G., E. ten Boekel, T. Yamagami, R. Ceredig, J. Andersson, F. Melchers. 1999. B cell development in the mouse from early progenitors to mature B cells. Immunol. Lett. 68: 89
    OpenUrlCrossRefPubMed
  5. ↵
    Hardy, R. R., K. Hayakawa. 2001. B cell development pathways. Annu. Rev. Immunol. 19: 595
    OpenUrlCrossRefPubMed
  6. ↵
    Melchers, F., H. Karasuyama, D. Haasner, S. Bauer, A. Kudo, N. Sakaguchi, B. Jameson, A. Rolink. 1993. The surrogate light chain in B-cell development. Immunol. Today 14: 60
    OpenUrlCrossRefPubMed
  7. ↵
    King, L. B., J. G. Monroe. 2000. Immunobiology of the immature B cell: plasticity in the B-cell antigen receptor-induced response fine tunes negative selection. Immunol. Rev. 176: 86
    OpenUrlCrossRefPubMed
  8. ↵
    Butler, J. E., J. Sun, P. Navarro. 1996. The swine Ig heavy chain locus has a single JH and no identifiable IgD. Int. Immunol. 8: 1897
    OpenUrlAbstract/FREE Full Text
  9. ↵
    Reynaud, C. A., A. Dahan, V. Anquez, J. C. Weill. 1989. Somatic hyperconversion diversifies the single Vh gene of the chicken with a high incidence in the D region. Cell 59: 171
    OpenUrlCrossRefPubMed
  10. ↵
    McCormack, W. T., L. W. Tjoelker, C. B. Thompson. 1991. Avian B-cell development: generation of an immunoglobulin repertoire by gene conversion. Annu. Rev. Immunol. 9: 219
    OpenUrlCrossRefPubMed
  11. ↵
    Sun, J., C. Hayward, R. Shinde, R. Christenson, S. P. Ford, J. E. Butler. 1998. Antibody repertoire development in fetal and neonatal piglets. I. Four VH genes account for 80 percent of VH usage during 84 days of fetal life. J. Immunol. 161: 5070
    OpenUrlAbstract/FREE Full Text
  12. ↵
    Butler, J. E., J. Sun, P. Weber, P. Navarro, D. Francis. 2000. Antibody repertoire development in fetal and neonatal piglets. III. Colonization of the gastrointestinal tract selectively diversifies the preimmune repertoire in mucosal lymphoid tissues. Immunology 100: 119
    OpenUrlCrossRefPubMed
  13. ↵
    Šinkora, M., J. Šinkora, Z. Řeháková, I. Šplíchal, H. Yang, R. M. E. Parkhouse, I. Trebichavsky. 1998. Prenatal ontogeny of lymphocyte subpopulations in pigs. Immunology 95: 595
    OpenUrlCrossRefPubMed
  14. ↵
    Šinkora, M., J. Šinkora, Z. Řeháková, J. E. Butler. 2000. Early ontogeny of thymocytes in pigs: sequential colonization of the thymus by T cell progenitors. J. Immunol. 165: 1832
    OpenUrlAbstract/FREE Full Text
  15. ↵
    Šinkora, M., J. Sun, J. E. Butler. 2000. Antibody repertoire development in fetal and neonatal piglets. V. VDJ gene chimeras resembling gene conversion products are generated at high frequency by PCR in vitro. Mol. Immunol. 37: 1025
    OpenUrlCrossRefPubMed
  16. ↵
    Sun, J., I. Kacskovics, W. R. Brown, J. E. Butler. 1994. Expressed swine VH genes belong to a small VH gene family homologous to human VHIII. J. Immunol. 153: 5618
    OpenUrlAbstract
  17. ↵
    Butler, J. E., P. Weber, M. Šinkora, J. Sun, S. J. Ford, R. K. Christenson. 2000. Antibody repertoire development in fetal and neonatal piglets. II. Characterization of heavy chain complementarity-determining region 3 diversity in the developing fetus. J. Immunol. 165: 6999
    OpenUrlAbstract/FREE Full Text
  18. ↵
    Godin, I., F. Dietrlen-Lievre, A. Cumano. 1995. Emergence of multipotent hemopoietic cells in the yolk sac and in the paraaortic splanchnopleura in mouse embryos, beginning at 8.5 days postcoitum. Proc. Natl. Acad. Sci. USA 92: 773
    OpenUrlAbstract/FREE Full Text
  19. ↵
    Medvinsky, A., E. Dzierzak. 1996. Definitive hematopoiesis is autonomously initiated by the AGM region. Cell 86: 897
    OpenUrlCrossRefPubMed
  20. ↵
    Muller, A., A. Medvinsky, J. Strouboulis, F. Grosveld, E. Dzierzak. 1994. Development of hematopoietic stem cell activity in the mouse embryo. Immunity 1: 291
    OpenUrlCrossRefPubMed
  21. ↵
    Marrable, A. W.. 1969. The embryonic membranes of the pig. Vet. Res. 84: 598
    OpenUrl
  22. ↵
    Šinkora, M., J. Šinkorová, J. E. Butler. 2002. B cell development and VDJ rearrangement in the fetal pig. Vet. Immunol. Immunopathol. 87: 341
    OpenUrlCrossRefPubMed
  23. ↵
    Feeney, A.. 1990. Lack of N region in fetal and neonatal mouse immunoglobulin VDJ junctional sequences. J. Exp. Med. 172: 1377
    OpenUrlAbstract/FREE Full Text
  24. ↵
    Bangs, L. A., I. E. Sanz, J. M. Teale. 1991. Comparison of DH, JH and junctional diversity in the fetal, adult and aged B cell repertoire. J. Immunol. 146: 1996
    OpenUrlAbstract
  25. ↵
    Delassus, S., S. Darche, P. Kourilsky, A. Cumano. 1998. Ontogeny of the heavy chain immunoglobulin repertoire in fetal liver and bone marrow. J. Immunol. 160: 3274
    OpenUrlAbstract/FREE Full Text
  26. ↵
    Carlson, L. M., W. T. McCormack, C. E. Postema, E. H. Humphries, C. B. Thompson. 1990. Templated insertions in the rearranged chicken IgL V gene segment arise by intrachromosomal gene conversion. Genes Dev. 4: 536
    OpenUrlAbstract/FREE Full Text
  27. ↵
    Knight, K. L., M. A. Crane. 1994. Generating the antibody repertoire in rabbit. Adv. Immunol. 56: 411
    OpenUrl
  28. ↵
    Tunyaplin, C., K. L. Knight. 1997. IgH gene rearrangements on the unexpressed allele in rabbit B cells. J. Immunol. 158: 4805
    OpenUrlAbstract
  29. ↵
    Alt, F. W., G. D. Yancopoulos, T. K. Blackwell, C. Wood, E. Thomas, M. Boss, R. Coffman, N. Rosenberg, S. Tonegawa, D. Baltimore. 1984. Ordered rearrangement of immunoglobulin heavy chain variable region segments. EMBO J. 3: 1209
    OpenUrlPubMed
  30. ↵
    Ferradini, L., C. A. Reynaud, R. Lauster, J.-C. Weill. 1994. Rearrangement of the chicken λ light chain locus: a silencer/antisilencer regulation. Semin. Immunol. 6: 165
    OpenUrlCrossRefPubMed
  31. ↵
    Nemazee, D.. 2000. Receptor selection in B and T lymphocytes. Annu. Rev. Immunol. 18: 19
    OpenUrlCrossRefPubMed
  32. ↵
    Siminovitch, K. A., M. W. Moore, J. Durdik, E. Selsing. 1987. The human κ deleting element and the mouse recombining segment share DNA sequence homology. Nucleic Acids Res. 15: 2699
    OpenUrlAbstract/FREE Full Text
  33. ↵
    Reynaud, C. A., B. A. Imhof, V. Anquez, J. C. Weill. 1992. Emergence of committed B lymphoid progenitors in the developing chicken embryo. EMBO J. 11: 4349
    OpenUrlPubMed
  34. ↵
    Martin, F., J. F. Kearney. 2001. B1 cells: similarities and differences with other B cell subsets. Curr. Opin. Immunol. 13: 195
    OpenUrlCrossRefPubMed
  35. ↵
    Press, C. M., W. R. Hein, T. Landsverk. 1993. Ontogeny of leukocyte populations in the spleen of fetal lambs with emphasis on the early prominence of B cells. Immunology 80: 598
    OpenUrlPubMed
  36. ↵
    Wortis, H. H., R. Berland. 2001. Cutting edge commentary: origins of B-1 cells. J. Immunol. 166: 2163
    OpenUrlAbstract/FREE Full Text
  37. ↵
    Kanno, K., T. Okada, M. Abe, S. Hirose, T. Shirai. 1993. Differential sensitivity to interleukins of CD5+ and CD5− anti-DNA antibody-producing B cells in murine lupus. Autoimmunity 14: 205
    OpenUrlCrossRefPubMed
  38. ↵
    Cukrowska, B., J. Šinkora, Z. Reháková, M. Šinkora, I. Šplíchal, L. Tucková, S. Avrameas, A. Saalmuller, R. Barot-Ciorbaru, H. Tlaskalova-Hogenova. 1996. Isotype and antibody specificity of spontaneously formed immunoglobulins in pig fetuses and germ-free piglets: production by CD5− B cells. Immunology 88: 611
    OpenUrlCrossRefPubMed
  39. ↵
    Crane, M. A., M. Kingzette, K. L. Knight. 1996. Evidence for limited B-lymphogenesis in adult rabbits. J. Exp. Med. 183: 2119
    OpenUrlAbstract/FREE Full Text
  40. ↵
    Pospisil, R., R. G. Mage. 1998. B-cell superantigens may play a role in B-cell development and selection in the young rabbit appendix. Cell. Immunol. 185: 93
    OpenUrlCrossRefPubMed
  41. ↵
    Silverman, G. J., S. P. Cary, D. C. Dwyer, L. Luo, R. Wagenknecht, V. E. Curtis. 2000. A B-cell superantigen-induced persistent “hole” in the B-1 repertoire. J. Exp. Med. 192: 87
    OpenUrlAbstract/FREE Full Text
Back to top

In this issue

Print
Download PDF
Article Alerts
Email Article
Citation Tools

Jump to section