Regulation of genome rearrangement events during lymphocyte differentiation

Stem cell origins of leukaemia and curability

It is suggested that most childhood acute lymphoblastic leukaemias and some other paediatric cancers are chemo-curable because they arise in stem cell populations that are functionally transient, chemosensitive and programmed for apoptosis. Most adult acute leukaemias are chemo-incurable at least in part because they originate in relatively drug resistant stem cells with extensive self-renewal capacity. The latter property in turn increases the probability of clones evolving with multi-drug resistance. Particular mutations may superimpose additional adverse features on leukaemic cells.

Germ line maintenance of the pseudogene donor pool for somatic immunoglobulin gene conversion in chickens

Somatic immunoglobulin diversity is generated in avian species by sequential gene conversion of variable (V) gene segments of the immunoglobulin heavy- and light-chain loci during B-cell development. The germ line pools of donor sequence information for somatic V-region gene conversion are found in families of V pseudogenes, located 5' of the single functional V gene of each locus. The sequence relationships among the pseudogenes (psi VL) and functional VL1 gene of the chicken light-chain alleles in three inbred strains were compared to determine the extent of diversity within the germ line pseudogene cluster. Numerous differences were observed. For example, compared with the previously reported CB allele and the G4 allele, the S3 allele contains two intact pseudogenes between psi VL16 and psi VL18. These two adjacent psi VL gene segments (psi VL17a and psi VL17b) could have given rise to the psi VL17 segment of the G4 and CB alleles by homologous recombination. The majority of other sequence polymorphisms among the psi VL alleles appear to be the result of meiotic gene conversion. The incidence of untemplated mutations within psi VL segments is significantly lower than the incidence of mutation within the pseudogene flanking regions. Together with the observations that most psi VL segments have open reading frames and lack stop codons, these data support the hypothesis that the psi VL cluster resembles a functional multigene family maintained by evolutionary selection for its functional role in generating somatic antibody diversity. Meiotic gene conversion events within the psi VL cluster serve both to introduce diversity by the exchange of short segments between family members and to prevent the accumulation of random mutations.

Germ line maintenance of the pseudogene donor pool for somatic immunoglobulin gene conversion in chickens

Somatic immunoglobulin diversity is generated in avian species by sequential gene conversion of variable (V) gene segments of the immunoglobulin heavy- and light-chain loci during B-cell development. The germ line pools of donor sequence information for somatic V-region gene conversion are found in families of V pseudogenes, located 5' of the single functional V gene of each locus. The sequence relationships among the pseudogenes (psi VL) and functional VL1 gene of the chicken light-chain alleles in three inbred strains were compared to determine the extent of diversity within the germ line pseudogene cluster. Numerous differences were observed. For example, compared with the previously reported CB allele and the G4 allele, the S3 allele contains two intact pseudogenes between psi VL16 and psi VL18. These two adjacent psi VL gene segments (psi VL17a and psi VL17b) could have given rise to the psi VL17 segment of the G4 and CB alleles by homologous recombination. The majority of other sequence polymorphisms among the psi VL alleles appear to be the result of meiotic gene conversion. The incidence of untemplated mutations within psi VL segments is significantly lower than the incidence of mutation within the pseudogene flanking regions. Together with the observations that most psi VL segments have open reading frames and lack stop codons, these data support the hypothesis that the psi VL cluster resembles a functional multigene family maintained by evolutionary selection for its functional role in generating somatic antibody diversity. Meiotic gene conversion events within the psi VL cluster serve both to introduce diversity by the exchange of short segments between family members and to prevent the accumulation of random mutations.

Available lambda B cell repertoire in the mouse: evidence of positive selection by environmental factors

We have recently shown that, from two BALB/c mice treated with rabbit anti-C lambda 2/C lambda 3 antibodies coupled to lipopolysaccharide, variable heavy chain (VH) family repertoires associated with lambda 2 or lambda 3 light chains can differ from one lambda subtype to another and from one individual mouse to another. Indeed, 4 out of 6 lambda 2 (VxJ2) hybridomas from one mouse preferentially expressed the VH10 family while 3 out of 8 lambda 2 (V2J2) and 5 out of 8 lambda 2 (VxJ2) hybridomas from a second mouse preferentially expressed the S107 and VGAM3.8 VH families, respectively. In this report, we describe the structural basis of such preferential pairings by sequence analysis of the 12 lambda 2 hybridomas. The sequence comparison of their VH regions show that each preferential association of a VH family to one V lambda region is restricted to the use of a single member or very closely related members inside a VH family and that a great variability of CDR3 of heavy chain is observed. We, therefore, suggest that environmental factors can modify the available lambda B cell repertoire through a positive selection of particular VH/V lambda pairings. Moreover, our data support that this selection does not require clonal expansion and punctual somatic mutation.

Mechanisms that generate human immunoglobulin diversity operate from the 8th week of gestation in fetal liver

The repertoire of immunoglobulin expressed very early in human development was approached by cloning and sequencing 55 rearranged and 11 germ-line VH transcripts, after amplification by polymerase chain reaction of cDNA libraries derived from two fetal livers at 8 and 13 weeks of gestation. All families with the exception of VH2, were expressed as soon as 8 weeks, with preferential usage of certain germ-line genes. Very few somatic mutations, randomly localized, were identified. By contrast, in a series of clones derived from the same VDJ rearrangement using the VH6 family, extensive mutations had taken place, mostly accumulated in the third complementarity-determining region (CDR3) suggesting that the specialized enzymatic machinery was at hand very early during human development. Some other characteristics of the fetal repertoire also emerged, namely increased usage of JH3 and JH2, as compared to the adult pattern, where JH4 is dominant and reduced length of the D/CDR3 regions. All D gene families were identified, and their usage frequently involved D-D fusions. N diversity was present very early, and increased with age. Identification of germ-line transcripts pertaining to all six VH families including pseudogenes, in the E55 library, revealed a population very different as compared to rearranged gene transcripts. This suggests that a large portion of VH locus is accessible for transcription, bringing no evidence of correlation between preferential rearrangement of a given VH gene and its localization in the locus.

Transcription enhances intrachromosomal homologous recombination in mammalian cells

The influence of transcription on homologous intrachromosomal recombination between direct and inverted repeats has been examined by using Chinese hamster ovary cells. Recombination was monitored between two integrated neomycin (neo) genes, including one silent allele and a second allele regulated by the inducible mouse mammary tumor virus promoter. Transcription of mouse mammary tumor virus neo alleles was regulated with the glucocorticoid hormone dexamethasone. Alleles transcribed at high levels recombined about two- to sevenfold more frequently than identical alleles transcribed at low levels. Direct repeats recombined primarily by a gene conversion mechanism; inverted repeats produced a variety of rearranged products. These results are discussed in relation to recombinational processes that regulate gene expression, influence gene family structures, and mediate genomic instability associated with cellular transformation and tumorigenesis.

Mutagen-induced recombination in mammalian cells in vitro

It is now clear from in vitro studies that mutagens induce recombination in the cell, both homologous and nonhomologous exchanges. The recombination events induced are extrachromosomal events, exchanges between extrachromosomal DNA and chromosomes, and inter- as well as intrachromosomal exchanges. However, not all types of DNA damage can induce recombination. The mechanisms involved in the induction process are not known but may involve activation of DNA repair systems. In addition, stimulation of mRNA transcription by mutagens, different recombination pathways and how the assay system is constructed may affect the frequency and characteristics of the observed recombination events.

Transcription enhances intrachromosomal homologous recombination in mammalian cells

The influence of transcription on homologous intrachromosomal recombination between direct and inverted repeats has been examined by using Chinese hamster ovary cells. Recombination was monitored between two integrated neomycin (neo) genes, including one silent allele and a second allele regulated by the inducible mouse mammary tumor virus promoter. Transcription of mouse mammary tumor virus neo alleles was regulated with the glucocorticoid hormone dexamethasone. Alleles transcribed at high levels recombined about two- to sevenfold more frequently than identical alleles transcribed at low levels. Direct repeats recombined primarily by a gene conversion mechanism; inverted repeats produced a variety of rearranged products. These results are discussed in relation to recombinational processes that regulate gene expression, influence gene family structures, and mediate genomic instability associated with cellular transformation and tumorigenesis.

Immunoglobulin kappa light chain germ-line transcripts in human precursor B lymphocytes

B lymphoblastoid cell lines (BLCL), established from bone marrow and peripheral blood mononuclear cells from two severe combined immunodeficiency (SCID) patients, manifested a complete absence of genomic rearrangements of the immunoglobulin (Ig) heavy (H) and light (L) chain loci. The BLCL contained germ-line transcripts of the Ig kappa region locus of approximately 1.2 kilobase (kb). By cDNA cloning and sequence analysis the transcripts were shown to consist of a C kappa segment, a J kappa 1 gene segment, 160 base pairs (bp) of J kappa 1 5' intervening sequence, containing the heptamer/nonamer recombination recognition sequences and at the 5' end a 523-bp segment designated human kappa zero, The first 206 bp of this 5' segment were homologous to the reported murine kappa zero region. Genomic restriction mapping and DNA sequence analysis demonstrated that the human kappa zero segment is located approximately 4 kb upstream of J kappa 1. The kappa zero segment contains a putative promoter region with an OCT2 binding site, and has a splice donor site to accomplish splicing to an acceptor site 160 bp upstream of J kappa 1. Expression of the kappa zero gene segment was found in BLCL derived from normal fetal bone marrow, in which both Ig kappa loci were in the germ-line configuration. These findings indicate that the described transcripts are not only present in SCID, but also in normal developing pre-B lymphocytes. The expression of germ-line Ig kappa L chain transcripts may be associated with the locus becoming accessible to gene rearrangement.

Creation of immunoglobulin diversity by intrachromosomal gene conversion

Not all vertebrates create an immunoglobulin repertoire through the recombination of individual members of variable (V), diversity (D) and joining (J) gene segment families. In chickens, for example, a diverse set of immunoglobulins is created by intrachromosomal gene conversion of the single variable gene segments of the immunoglobulin heavy and light chain genes. Recent evidence from other species such as the rabbit suggests that gene conversion may be a more widespread mechanism for the creation of immunologic diversity than previously supposed.

IGM kappa/lambda EBV human B cell clone: an early step of differentiation of fetal B cells or a distinct B lineage?

In agreement with the clonal theory, one B lymphocyte synthesizes one antibody due to allelic and isotypic exclusion. We analyzed an EBV B-cell clone, E29.1, derived from an 11 week-old embryo, and secreting both IgM kappa and IgM lambda. Structural analysis of produced IgM, indicated that lambda-containing pentamers could be considered hybrid molecules, expressing both the kappa and lambda. chains, with a kappa/lambda ratio between 5 and 10. It was also found that 60% of the lambda chains were secreted in free form, presumably as a result of a better affinity of mu chains for kappa chains. The sequence of the three transcripts had an entirely ORF (Open Reading Frame), and were very close to germline sequences, with, however, an additional codon between V kappa and J kappa gene which has never been described in adult myeloma protein or cDNA human sequence. This observation is suggestive of N diversity taking place in kappa chains. The possible role of Kde (kappa deleting element) recombination onto kappa/lambda locus activation was analyzed on a collection of 23 lambda clones. The status of rearrangement of kappa genes indicated that 35% of these clones had retained, at least, one kappa allele without the Kde recombination, four lambda clones had one kappa allele in germline configuration. Different hypotheses of maturation from pre-B cell to B cell with activation of light chain genes are discussed.

Ectopic recombination within homologous immunoglobulin mu gene constant regions in a mouse hybridoma cell line

We have transferred a pSV2neo vector containing the wild-type constant region of the immunoglobulin mu gene (C mu) into the mutant hybridoma igm482, which bears a 2-bp deletion in the third constant-region exon of its haploid chromosomal mu gene (C mu 3). Independent igm482 transformants contain the wild-type immunoglobulin C mu region stably integrated in ectopic chromosomal positions. We report here that the wild-type immunoglobulin C mu region can function as the donor sequence in a gene conversion event which corrects the 2-bp deletion in the mutant igm482 chromosomal C mu 3 exon. The homologous recombination event restores normal immunoglobulin M production in the mutant cell.

MONOMORPHIC ORGANIZATION OF HUMAN DIVERSITY (DH) HEAVY CHAIN VARIABLE GENES

While recent evidence indicates that human immunoglobulin heavy chain variable (VH) genes exhibit a high degree of heterogeneity, little is known concerning the polymorphism of the diversity (DH) gene complex. This locus comprises two clusters, major and minor, which are physically linked with the VH locus. In assessing the variability of the human major and minor DH clusters, we found no evidence for a substantial restriction site polymorphism. We also noted that, in contrast to what was found in the Japanese population, the DH1 gene is not deleted in an appreciable proportion of the European population. We propose that, because DH genes impart the critical functions associated with the third complementarity-determining region, the genomic organization of the human DH locus has been evolutionarily conserved through selection pressure mechanisms.

Ectopic recombination within homologous immunoglobulin mu gene constant regions in a mouse hybridoma cell line

We have transferred a pSV2neo vector containing the wild-type constant region of the immunoglobulin mu gene (C mu) into the mutant hybridoma igm482, which bears a 2-bp deletion in the third constant-region exon of its haploid chromosomal mu gene (C mu 3). Independent igm482 transformants contain the wild-type immunoglobulin C mu region stably integrated in ectopic chromosomal positions. We report here that the wild-type immunoglobulin C mu region can function as the donor sequence in a gene conversion event which corrects the 2-bp deletion in the mutant igm482 chromosomal C mu 3 exon. The homologous recombination event restores normal immunoglobulin M production in the mutant cell.

Regulation of TCR alpha and beta gene allelic exclusion during T-cell development

Early in their development, most T cells become committed to the expression of one, and only one, TCR alpha beta combination. How do T cells achieve this TCR allelic exclusion? This article discusses the configuration and expression of TCR alpha and beta genes in mature T-cell lines and TCR alpha beta transgenic mice, and proposes three nonexclusive models to account for the significant occurrence of T cells with two productive alpha gene rearrangements.

Cell interaction molecules and cytokines which participate in B lymphopoiesis

A molecular and cellular definition of the bone marrow microenvironment is rapidly contributing to our understanding of lymphohaemopoiesis. While lineage specific genes and their protein products are being identified, information is accumulating about mechanisms which may regulate their expression. Stimulation of B lymphocyte precursor replication, and other discrete functions, are being attributed to cytokines such as interleukin 7 (IL-7). The activity of these factors may be controlled at the level of synthesis, local concentration and interaction with extracellular matrix. Extremely small amounts of IL-7 are made by stromal cells, which are themselves being thoroughly studied as cloned cell lines. This in vitro characterization suggests that stromal cells can make at least 12 cytokines, that they can respond to some of those cytokines themselves, and that they retain differentiation potential. Several molecules have been identified which are probably required for recognition between cells in marrow. It is noteworthy that they belong to several previously described families of adhesion molecules and none is unique to that tissue. VCAM-1 is constitutively expressed on stromal cells in marrow and can be recognized by pre-B cells which bear the integrin VLA-4. The same pair of molecules is probably responsible for extravasation of leukocytes in other tissues during inflammation. Cell adhesion molecules are likely to work in a carefully coordinated and cooperative fashion. Their activity can be controlled by expression or, in some cases, modulated after display on the cell surface. For example, while most haemopoietic cells bear CD44, only certain cells utilize it for recognition of the ligand hyaluronate. The affinity for hyaluronate can be experimentally regulated and depends on the cytoplasmic domain of CD44. This capability for dynamic change may be important for transient interactions between cells, permitting movement of maturing precursors within and from marrow.

Evidence for failure of V(D)J recombination in bone marrow pre-B cells from X-linked agammaglobulinemia

X-linked agammaglobulinemia (XLA) results from a failure of B lymphoid development. We have previously examined pre-B cell hybrids from three patients with XLA and found them to be limited to production of a novel germ line transcript of the Ig H chain locus composed of a leader sequence (LS) spliced to the constant region of mu chain (C mu) as mRNA and polypeptide. These transcripts result from transcriptional activation of the germ line heavy chain locus from an LS exon upstream of the embryonic JH locus. Germ line LS-C mu transcripts are produced by pre-B cells from normal bone marrow and fetal liver, indicating that they are products of normal pre-B cell development, as part of the process of transcriptional activation to provide access for the recombinase. Bone marrow from three patients with XLA has been examined directly by polymerase chain reaction amplification to determine whether the exclusive production of LS-C mu by XLA pre-B cell hybrids is representative of XLA pre-B cells. I report that LS-C mu is the predominant Ig molecule produced by XLA pre-B cells, with limited production of the D mu product of DJH intermediate stage of V(D)J recombination. Mature VHDJH recombinations were not detected with a variety of primers that amplify VH sequences. I conclude that XLA is associated with a limitation in V(D)J recombination that may cause the failure of pre-B cell development.

Germ line transcription of the immunoglobulin heavy chain locus directs production of mu chain without VDJ

Immunoglobulin VDJ recombination is associated with transcriptional activation of the Ig variable region elements. We have previously described a novel Ig mu chain protein and mRNA produced by pre-B cell hybrids from normal and X-linked agammaglobulinemic bone marrow. We have now characterized the mRNA encoding this protein and find that it is composed of a 5' leader sequence spliced to C mu (LS-C mu), lacking the variable (V), diversity (D), and joining (J) gene sequences. The leader sequence is encoded by a novel exon 16 kb upstream of the JH locus. Transcription of the germ line heavy chain locus from this LS exon results in transcriptional activation of the JH locus, apparently the initial step in commitment to B lymphoid development. Polymerase chain reaction amplification of normal bone marrow shows that these germ line LS-C mu transcripts are a product of bone marrow pre-B cells. Production of LS-C mu commences a sequential process of transcriptional activation, with concordant translation of Ig rearrangement intermediates, in the process of creating a productive VDJ rearrangement.

Restricted expression of recombination activating gene (RAG-1) in mouse lymphoid tissues

In an attempt to determine the distribution of recombinase activity in the mouse thymus, spleen and lymph nodes, we used the in situ hybridization method to examine the expression of the recombination activating genes RAG-1 and RAG-2. Expression of RAG-1 was found in most cortical thymocytes but not in the majority of medullary thymocytes. Although hybridization signals of RAG-2 were not as intense as those of RAG-1, the localization of RAG-2 transcripts was similar to that of RAG-1. In the spleen, expression of RAG-1 was found only in limited cells near the sinus, and the majority of the cells within the follicle were negative for RAG-1 transcript. In nude mice, RAG-1-expressing cells were detected in the same regions, which suggests that in situ hybridization signals of RAG-1 in the spleen are due to the cells of B cell origin. In the lymph nodes, expression of RAG-1 was found only in the medullary region. Expression of RAG-2 transcript in the spleen and the lymph nodes, if any, was too faint to determine the specific localization. These results suggest that most of the cortical thymocytes and some cells in the spleen are capable of rearranging T cell receptor genes and immunoglobulin genes, respectively, but the possibility of some other explanation could not be ruled out in RAG-1 expressing cells of the spleen and the lymph nodes.

Function and control of recombination-activating gene activity

The RAG-1 and RAG-2 genes synergistically confer VDJ recombinase activity to nonlymphoid cell lines. To unequivocally test RAG gene function, we created lines of mice that lack functional copies of these genes. Consistent with the possibility that RAG gene encode the tissue-specific components of VDJ recombinase, RAG-2-deficient mice are viable but have a severe combined immune deficiency due to inability to initiate VDJ recombination and thereby generate mature lymphocytes. RAG-2-deficient mice have no obvious defect in any tissue or lineage other than lymphocytes, indicating that VDJ recombinase activity and RAG-2-gene function is required only for lymphocyte development. Levels of RAG-1 and RAG-2 expression in primary murine lymphoid tissues and lymphoid bone marrow cultures generally are much higher than those of transformed precursor B-cell lines. Low-level RAG gene expression in permanent cell lines results from a decline during propagation due to outgrowth of cells with lower RAG expression levels. The low and variable level of RAG gene expression in transformed pre-B cell lines correlates with low and variable rates of endogenous VDJ recombination; therefore, such lines are not reliable models for experiments aimed at studying mechanisms that target this activity to particular variable region gene segments. To generate such a system, we introduced RAG genes into B-lineage lines under the control of a heat shock-inducible promoter; heat-shock treatment induces extremely high-level but transient RAG expression accompanied by parallel induction of VDJ recombinase activity. Such cells efficiently rearrange transfected VDJ recombination substrates in a regulated manner that is dependent on the activity of transcriptional control elements associated with the target V gene segments.

High-level expression of a human immunoglobulin gamma 1 transgene depends on switch region sequences

We describe that chimeric mouse-human immunoglobulin heavy chain (IgH) genes lacking a switch region and controlled by an IgH promoter and the intronic enhancer are only weakly expressed in transgenic mice. Insertion of part of the human C gamma 1 or murine Cmu switch region into the major intron of the chimeric IgH gene results in a 10(2)-to 10(3)-fold increase in transgene expression. Analysis of B cell hybridoma clones from transgenic mice suggests that switch sequences influence IgH transgene expression at the cellular level. However, the effect of switch region sequences on IgH gene expression observed in vivo is not apparent in transfected B cell lines. These results indicate that switch region sequences which are located proximal to the constant part of the gene and which are normally present in a rearranged IgH gene after class switching represent a novel type of regulatory element that plays a critical role in IgH gene expression in vivo.

Immunoglobulin heavy-chain and CD3 delta-chain gene enhancers are DNase I-hypersensitive in hemopoietic progenitor cells

Multipotential interleukin 3-dependent non-immortalized murine hemopoietic progenitor cells have DNase I-hypersensitive sites in the immunoglobulin heavy-chain and CD3 delta enhancers and transcribe germ-line T-cell antigen receptor gamma-chain (TCR gamma), but not IgM or TCR beta, genes. Induction of myeloid differentiation in these cells clones down expression and/or transcriptional accessibility of the immunoglobulin heavy-chain and TCR gamma genes. The CD3 delta enhancer region remains DNase I-hypersensitive but closes down in B cells. In embryonic stem cells and pan-mesodermal cells, these genes or enhancer regions are neither expressed nor DNase I-hypersensitive. These data suggest that lineage potential may be programmed, at least in part, by alterations in the accessibility or conformation of regulatory regions of genes and that some promiscuity of gene expression and/or accessibility can precede lineage commitment and maturation in progenitor cells induced to self-renew by interleukin 3.

RAG-1-deficient mice have no mature B and T lymphocytes

The V(D)J recombination activation gene RAG-1 was isolated on the basis of its ability to activate V(D)J recombination on an artificial substrate in fibroblasts. This property and the expression pattern in tissues and cell lines indicate that RAG-1 either activates or catalyzes the V(D)J recombination reaction of immunoglobulin and T cell receptor genes. We here describe the introduction of a mutation in RAG-1 into the germline of mice via gene targeting in embryonic stem cells. RAG-1-deficient mice have small lymphoid organs that do not contain mature B and T lymphocytes. The arrest of B and T cell differentiation occurs at an early stage and correlates with the inability to perform V(D)J recombination. The immune system of the RAG-1 mutant mice can be described as that of nonleaky scid mice. Although RAG-1 expression has been reported in the central nervous system of the mouse, no obvious neuroanatomical or behavioral abnormalities have been found in the RAG-1-deficient mice.

Autoimmune VH gene family: PCR-generated murine germline VH10 genes

A relatively large number of variable region genes (V) contribute, via gene rearrangements with smaller numbers of additional gene elements (D and J), to generate diversity in the immune response. While some VH gene families are thought to contain 100- 1000 members, the VH10 family has only two known functioning members with 99% sequence homology. Both members (monoclonal antibodies) are capable of binding DNA, and since they were derived from inbred mice afflicted with the lupus syndrome they are considered autoimmune antibodies. Relative uniqueness of the VH10 primary nucleotide sequence presents a model system with which to examine unrearranged VH genes and attempt to identify germline genes eventually expressed as autoantibodies. PCR amplified germline sequences of the VH10 family are highly conserved, with few base substitutions evenly distributed between both framework and CDR regions. It was determined that the PCR amplified germline sequences are highly similar to the DNA sequences of the two monoclonal VH10 antibodies, and a non-functional psuedo-germline gene was found that is identical to a non-functional cDNA derived from a hybridoma cell line. These findings indicate that the use of unique CDR DNA sequences for the identification and amplification of specific germline V genes via PCR can yield vital information that may answer fundamental questions about the origins of autoimmune anti-DNA antibodies in afflicted individuals. The nature of the germline gene populations and the possible microheterogeniety of these genes may prove to be important in understanding the role of autoimmune antibodies in normal and diseased individuals.

Characterization of the immunoglobulin heavy chain complementarity determining region (CDR)-III sequences from human B cell precursor acute lymphoblastic leukemia cells

Sequence analysis of the immunoglobulin heavy chain complementarity determining region (CDR)-III of B-lineage cells at various stages has provided important insights concerning B cell maturation and selection. Knowledge of human CDR-III sequences has been relatively limited compared with that of the murine system. We analyzed the CDR-III sequences of B cell precursor acute lymphoblastic leukemia (pre-B ALL) cells in 23 newly diagnosed and 10 relapsed patients, in order to elucidate the organization of CDR-III in B cell precursors. We found a very low frequency of somatic mutations in D and JH regions, preferential use of DLR, DXP, DHQ52, and DN elements, and of 3' side JH segments, and no predominant usage of D coding frames. Unusual joinings such as VH-D-D-JH and VH-JH were observed in three, and one sequences, respectively. We compared the CDR-III sequences derived from 10 patients between diagnosis and relapse. Two of them had three spots of mutated nucleotides at relapse, all of which were found in the N region near the D segments. Our data showed the possibility of somatic mutation at relapse, in addition to developmentally regulated rearrangement of the immunoglobulin gene at the stage of B cell precursors.

Two conserved essential motifs of the murine immunoglobulin lambda enhancers bind B-cell-specific factors

Two highly homologous enhancers associated with the two murine immunoglobulin lambda constant-region clusters were recently identified. In order to better understand the molecular basis for the developmental stage- and cell-type-restricted expression of lambda genes, we have undertaken an analysis of the putative regulatory domains of these enhancers. By using a combination of DNase I footprinting, electrophoretic mobility shift assay, and site-specific mutations, four candidate protein binding sites have been identified at analogous positions in both enhancers. A mutation of any of these sites decreases enhancer activity. Two of the sites, lambda A and lambda B, are essential for enhancer function, and both of these sites appear to bind both B-cell-specific and general factors. Nevertheless, isolated lambda A and lambda B sites show no evidence of inherent transactivating potential, alone or together, even when present in up to three copies. We suggest that the generation of transactivating signals from these enhancers may require the complex interaction of multiple B-cell-specific and nonspecific DNA-binding factors.

Developmentally controlled selection of antibody genes: characterization of individual VH7183 genes and evidence for stage-specific somatic diversification

Nucleotide sequence analysis of a large number of rearranged immunoglobulin heavy chain V region genes allowed the identification of six new members of the VH7183 gene family. These six new genes plus the eight previously defined genes agrees with the previously estimated complexity of this gene family. Twelve of these genes were represented among the isolated clones. A comparison of the clones, derived from 1-day- and 14-week-old BALB/c mice, suggested a biased and developmentally controlled VH7183 gene utilization. Furthermore, a developmentally controlled, non-random distribution of the functional vs. non-functional VHDJH rearrangements was observed among clones utilizing genes of this family, suggesting unsuspected regulatory aspects of Ig rearrangements in the process of B cell differentiation. Finally, a limited junctional diversity was revealed among the neonatal clones as the result of a low frequency of N-sequence addition. A similar discrepancy was also observed between neonatal and adult VHJ558 clones. In conclusion, these data suggest a programmed generation of B cell diversity similar to what has been observed for the establishment of gamma/delta T cell repertoires.

Organization and expression of the pseudo-light chain genes in human B-cell ontogeny

In pre-B cells, the mu chain is expressed at the cell surface in association with a "light chain surrogate" encoded by the V pre-B and the lambda-like genes. This mu-psi-L complex presumably triggers early steps of the B cell differentiation, possibly by controlling the Ig gene rearrangements. In the humans, the lambda-like complex contains 3 genes, located in the 22q11.2-q12.3 band, telomeric to the IGCL locus, with which they share a similar organization, pointing to a common genetic origin. Only one lambda-like gene, 14.1, is functional and specifically expressed with V pre-B in pre-B cells. This expression starts in cells which still have the IGH locus in germline configuration (pro-B stage) and ceases as soon as the IGL loci rearrange. These pre-B specific transcripts provide useful markers of cells of the B lineage in both physiological and pathological situations.

Two conserved essential motifs of the murine immunoglobulin lambda enhancers bind B-cell-specific factors

Two highly homologous enhancers associated with the two murine immunoglobulin lambda constant-region clusters were recently identified. In order to better understand the molecular basis for the developmental stage- and cell-type-restricted expression of lambda genes, we have undertaken an analysis of the putative regulatory domains of these enhancers. By using a combination of DNase I footprinting, electrophoretic mobility shift assay, and site-specific mutations, four candidate protein binding sites have been identified at analogous positions in both enhancers. A mutation of any of these sites decreases enhancer activity. Two of the sites, lambda A and lambda B, are essential for enhancer function, and both of these sites appear to bind both B-cell-specific and general factors. Nevertheless, isolated lambda A and lambda B sites show no evidence of inherent transactivating potential, alone or together, even when present in up to three copies. We suggest that the generation of transactivating signals from these enhancers may require the complex interaction of multiple B-cell-specific and nonspecific DNA-binding factors.

Regulation of immunoglobulin gene transcription

Analysis of the immunoglobulin gene suggests that their expression is controlled through the combinatorial action of tissue- and stage-specific factors (OTF-2, TF-microB, NF-kappa B), as well as more widely expressed E motif-binding factors such as E47/E12. Two basic issues cloud understanding of how these factors are involved in immunoglobulin gene regulation. First, cloning of these factors shows them to be members of families of proteins, all with similar DNA-binding specificities. OTF-2 is a member of the POU domain family, NF-kappa B is a related protein, and the microE5/kappa E2-binding factors are members of the bHLH family. Second, these binding sites and associated factors are involved in the regulation of many genes, not only the immunoglobulin genes, and in fact not only lymphoid-specific genes. These facts complicate understanding which member of a family is in fact responsible for interaction with, and activation of, a particular binding element in an enhancer/promoter. Recently, more detailed analysis of the interactions between such proteins and their related binding sites suggest that a certain level of specificity may in fact be encoded by the DNA element such that one family member of a protein is preferentially bound, or alternatively that the protein-DNA interactions that occur give subtle alterations in protein conformation that unmask an activation or protein-protein interactive domain. An additional level of regulation is imparted by combinatorial mechanisms such as adjacent DNA-binding elements and factors that may alter activity, as well as "cofactors" that, by forming a complex with the bound factor, affect its activation of a gene in a particular cell type. A third level of specificity may be obtained by factors such as NF-kappa B and the bHLH family due to their ability to create heterogeneous complexes, creating unique complexes in a tissue- or stage-specific manner. The multiple functions transcription factors such as NF-kappa B and OTF-2 play in the transcriptional regulation of multiple genes seems complex in contrast to a one factor, one gene regulation model. However, this type of organization may limit the number of factors lymphocytes would require if each lymphoid-specific gene were activated by a unique factor. Thus what appears to be complexity at the molecular level may reflect an economical organization at the cellular level. Investigation of the key factors controlling these genes suggests an ordered cascade of transcription factors becomes available in the cell during B cell differentiation.(ABSTRACT TRUNCATED AT 400 WORDS)

[Immunoglobulin genes and the origin of antibody diversity]

The genes encoding for the heavy, Kappa and Lambda chains of immunoglobulins are located on separate chromosomes. Each gene consists of many segments encoding for variable (V) and constant (C) domains of immunoglobulins. Each V gene is assembled from individual segments of germline, VL and JL for the light chains, VH, D and JH for the heavy chains. There are several hundreds VH and VL segments and a small number of D and JH segments. A complete VH gene (VH-D-JH combination) is first expressed in association with the nearest C region gene (C mu) but a class switch recombination may occur, enabling the same VH to associate with C regions of gamma, alpha or epsilon type. The generation of the antibody diversity is explained by several mechanisms: a) the rearrangements of the segments encoding for the variable domains (related to the presence of many segments and the joining variability between these segments); b) the combinations between heavy and light chains; c) the occurrence of somatic mutations in the DNA encoding for VH and VL gene segments.

Thymic nuclear matrix associated activity is not V(D)J recombinase

It was previously reported that nuclear matrix isolated from young rat thymus contained an activity that supported V(D)J recombination at a high efficiency (Dave et al., BIOCHEMISTRY 30: 4763-4767, 1991). A similar type of activity is also detected in the matrix prepared from fetal calf thymus. However, restriction enzyme mapping analyses of the recombined product clearly suggest that the double antibiotic resistance exhibited by the matrix treated plasmid substrate is not a consequence of V(D)J signal sequence recombination.

The chicken D locus and its contribution to the immunoglobulin heavy chain repertoire

Sixteen D elements were characterized from the chicken genome, 15 of which are extremely homologous. Early expression of this D repertoire was studied for both DJ and VDJ alleles. No N diversification occurs at either DJ or VD junctions. Only P additions were observed, the length of which does not appear restricted to a dinucleotide. A selection for the almost exclusive usage of the first reading frame of the D elements takes place during B cell expansion in the bursa, in parallel with the selection of productive rearrangements. All three reading frames were observed for the DJ allele at each developmental stage, although some bias for the first reading frame occurs already at the junctional stage. The high incidence of D-D junctions observed (25% among DJ sequences) might represent the major functional contribution of this multigene cluster in a system in which diversity will be generated later on by successive superimposed gene conversions. Other possible functions are discussed. The onset of D diversification through gene conversion between day 15 and day 18 of embryonic development is further documented.

A new view of prostaglandin E regulation of the immune response

Prostaglandins, particularly those of the E series, are widely regarded as immunosuppressive products of eukaryotic cells that can downregulate many aspects of B- and T-cell function. In this article, Richard Phipps and colleagues present a different concept of E series prostaglandins, based on recent evidence supporting a role for prostaglandins as potentiators of immunoglobulin class switching and of the synthesis of selected cytokines and cytokine receptors.

Biased VH gene expression in murine CD5 B cells results from age-dependent cellular selection

Flow cytometry-purified, peritoneal and splenic CD5+ and CD5- B cells from neonatal and adult C57BL/6 mice were studied for expression of VH and Vx gene families in RNA colony blot assays, and for frequencies of clones secreting antibodies to bromelain-treated mouse red blood cells (BrMRBC), single-stranded DNA, trimethyl ammonium and bovine gamma-globulin, by limiting dilution. The results show few overall differences between the two B cell subsets, which both manifest ontogenic D-proximal VH preferences that are lost with age. Biased VH11 expression in CD5 B cells is high in adult peritoneum and spleen but absent in newborns. It only partly correlates with the selection of anti-BrMRBC reactivity, which is considerably higher in peritoneum than in spleen. No particular Vx bias was observed in any of the populations studied with the possible exception of Vx22 in peritoneal CD5+ B cells. We conclude that the antibody repertoire expressed by peritoneal CD5+ B cells of adult mice is not the result of a genetic program, but rather the consequence of local, age-dependent cellular selection mechanisms.

Chicken T-cell receptor beta-chain diversity: an evolutionarily conserved D beta-encoded glycine turn within the hypervariable CDR3 domain

Unlike mammals, chickens generate an immunoglobulin (Ig) repertoire by a developmentally regulated process of intrachromosomal gene conversion, which results in nucleotide substitutions throughout the variable regions of the Ig heavy- and light-chain genes. In contrast to chicken Ig genes, we show in this report that diversity of the rearranged chicken T-cell receptor (TCR) beta-chain gene is generated by junctional heterogeneity, as observed in rearranged mammalian TCR genes. This junctional diversity increases during chicken development as a result of an increasing base-pair addition at the V beta-D beta and D beta-J beta joints (where V, D, and J are the variable, diversity, and joining gene segments). Despite the junctional hypervariability, however, almost all functional V beta-D beta-J beta junctions appear to encode a glycine-containing beta-turn. Such a turn may serve to position the amino acid side chains of a hypervariable TCR beta-chain loop with respect to the antigen-binding groove of the major histocompatibility complex molecule. Consistent with this hypothesis, the germ-line D beta nucleotide sequences of chickens, mice, rabbits, and humans have been highly conserved and encode a glycine in all three reading frames.

V(D)J recombination: evidence that a replicative mechanism is not required

We examined a series of extrachromosomal DNA substrates for V(D)J recombination under replicating and nonreplicating conditions. Complete and partial replications were examined by monitoring the loss of prokaryote-specific adenine methylation at 14 to 22 MboI-DpnI restriction sites (GATC) on the substrates. Some of these sites are within 2 bases of the signal sequence ends. We found that neither coding joint nor signal joint formation requires substrate replication. After ruling out replication as a substrate requirement, we determined whether replication had any effect on the efficiency of V(D)J recombination. Quantitation of V(D)J recombination efficiency on nonreplicating substrates requires some method of monitoring the entry of substrate molecules into the cells. We devised such a method by monitoring DNA repair of substrates into which we had substituted deoxyuridine for 10 to 20% of the thymidine nucleotides in the DNA. The substrates which enter the lymphoid cells were repaired efficiently in vivo by the eukaryotic uracil DNA repair system. Upon plasmid harvest, we distinguished repaired (entered) from unrepaired (not entered) plasmids by cleaving unrepaired molecules with uracil DNA glycoylase and Escherichia coli endonuclease IV in vitro. This method of monitoring DNA entry does not appear to underestimate or overestimate the amount of DNA entry. By using this method, we found no significant quantitative effect of DNA replication on V(D)J recombination efficiency.

T-lymphocyte development in scid mice is arrested shortly after the initiation of T-cell receptor delta gene recombination

Scid mice lack functional lymphocytes because they carry a mutation that impairs rearrangement of immunoglobulin and T-cell receptor (TCR) genes. Rearrangement of TCR delta, but not gamma and beta genes, was routinely observed in DNA of scid thymocytes and thymocyte hybridomas. TCR delta gene rearrangements appeared to involve D delta 1, D delta 2, and J delta 1 elements only; rearrangement of elements upstream of D delta 1 (e.g., V delta 1) was not observed, and transcripts corresponding to fully assembled TCR delta genes (VDJ delta or VDDJ delta) were not detected in RNA from scid thymocytes. These findings suggest that D delta 1, D delta 2, and J delta 1 may be among the first TCR gene elements to undergo recombination and that scid T-lineage cells are developmentally arrested during or shortly after this stage of differentiation. One class of TCR delta recombination fragments (D delta 2-J delta 1) was amplified by the polymerase chain reaction (PCR) and cloned, and the recombination junctions were sequenced. Most fragments showed normal coding joints. Interestingly, five of seven coding joints that lacked N insertions showed evidence of recombination between short stretches (2-3 bp) of homologous sequence. As discussed, the general absence of V delta-, J gamma-, and J beta-associated rearrangements, despite the occurrence of normal D delta 2-J delta 1 rearrangements, raises the possibility that the scid mutation may cause premature cessation of TCR gene recombination and thereby arrest early T-cell development.

V(D)J recombination: evidence that a replicative mechanism is not required

We examined a series of extrachromosomal DNA substrates for V(D)J recombination under replicating and nonreplicating conditions. Complete and partial replications were examined by monitoring the loss of prokaryote-specific adenine methylation at 14 to 22 MboI-DpnI restriction sites (GATC) on the substrates. Some of these sites are within 2 bases of the signal sequence ends. We found that neither coding joint nor signal joint formation requires substrate replication. After ruling out replication as a substrate requirement, we determined whether replication had any effect on the efficiency of V(D)J recombination. Quantitation of V(D)J recombination efficiency on nonreplicating substrates requires some method of monitoring the entry of substrate molecules into the cells. We devised such a method by monitoring DNA repair of substrates into which we had substituted deoxyuridine for 10 to 20% of the thymidine nucleotides in the DNA. The substrates which enter the lymphoid cells were repaired efficiently in vivo by the eukaryotic uracil DNA repair system. Upon plasmid harvest, we distinguished repaired (entered) from unrepaired (not entered) plasmids by cleaving unrepaired molecules with uracil DNA glycoylase and Escherichia coli endonuclease IV in vitro. This method of monitoring DNA entry does not appear to underestimate or overestimate the amount of DNA entry. By using this method, we found no significant quantitative effect of DNA replication on V(D)J recombination efficiency.

Early occurrence of immunoglobulin isotype switching in human fetal liver

A cDNA library prepared from a human fetal liver of the first trimester of gestation was screened with Ig C mu, C gamma, C kappa and C lambda probes. Ten heavy chain clones were isolated and characterized by restriction mapping and partial sequencing. The absence of Ig light chain clone and the presence of pre-B-specific lambda-like transcripts suggest that the immune compartment of this cDNA library was mostly derived from pre-B cells. Three transcripts of mu, gamma 2 and gamma 4 isotypes contained a V-D-J-C region with an open reading frame and used members of the VHIV, VHIII and VHI families, respectively. Seven clones were derived from sterile transcripts, one C mu and six C gamma. In addition to C mu exons, the sterile mu transcript contained the 5' flanking germline region. By contrast, the gamma sterile transcripts used a 5' sequence that was spliced from the I gamma 1 region onto the first C gamma 1 exon. In addition several of these transcripts were derived from alternative splicing. The simultaneous expression of both sterile and functional gamma transcripts suggests that the switch mechanism operates in normal fetal liver very early in ontogeny.