Immunoglobulin double-isotype expression by trans-mRNA in a human immunoglobulin transgenic mouse

Channel catfish immunoglobulins: repertoire and expression

The channel catfish, Ictalurus punctatus, is widely recognized as an important model for studying immune responses in ectothermic vertebrates. It is one of the few fish species for which defined viable in vitro culture systems have been established and is currently the only fish species from which a variety of functionally distinct clonal leukocyte lines are available. Moreover, there is a large basis of biochemical and molecular information on the structure and function of catfish immunoglobulins (Igs). Catfish, as other teleosts, have a tetrameric homolog of IgM as their predominant serum Ig plus a homolog of IgD. They also have genetic elements basically similar to those of mammals, which encode and regulate their expression. The catfish Ig heavy (H) chain locus is a translocon-type locus with three Igdelta genes linked to an Igmu gene or pseudogene. The catfish IgH locus is estimated to contain approximately 200 variable (V) region genes representing 13 families as well as at least three diversity (D) and 11 joining (JH) genes. The catfish has two light (L) chain isotypes, F and G, both encoded by loci organized in multiple cassettes of VL-JL-CL with the VL in the opposite transcriptional orientation. Hence, all requisite components for encoding antibodies are present in the catfish, albeit with certain variations. In the future, whether or not additional unique features of Ig function and expression will be found remains to be determined.

Pre-mRNA trans-splicing: from kinetoplastids to mammals, an easy language for life diversity

Since the discovery that genes are split into intron and exons, the studies of the mechanisms involved in splicing pointed to presence of consensus signals in an attempt to generalize the process for all living cells. However, as discussed in the present review, splicing is a theme full of variations. The trans-splicing of pre-mRNAs, the joining of exons from distinct transcripts, is one of these variations with broad distribution in the phylogenetic tree. The biological meaning of this phenomenon is discussed encompassing reactions resembling a possible noise to mechanisms of gene expression regulation. All of them however, can contribute to the generation of life diversity.

DNA deamination in immunity

A functional immune system is one of the prerequisites for the survival of a species. Humans have one of the most complicated immune systems, with the ability to learn from and adapt to pathogens. At first, a primary repertoire of antibodies is generated, which, upon antigen encounter, will diversify and adapt to produce a highly specific and potent secondary response, part of which is kept in memory to fight off future infections. In this review, the mechanism as well as the specificities of the key protein in the secondary immune response, activation-induced cytidine deaminase (AID), are highlighted, as well as its role in the DNA deamination model of immunoglobulin diversification. The review also highlights aspects of AID's regulation on both the transcriptional as well as post-translational level and its potential molecular mechanism and specificity. Furthermore, it expands outside the involvement of AID in somatic hypermutation, class switching, and gene conversion to discuss the implications of DNA deamination in epigenetic modifications of DNA (as a potential demethylase), the induction of mutations during oncogenesis, and includes an evolutionary comparison to the DNA deaminase family member APOBEC3G, a key protein in human immunodeficiency virus pathogenesis.

In-cell generation of antibody single-chain Fv transcripts by targeted RNA trans-splicing

The humoral immune response propels the production of a diversified pool of antibodies with high affinity and selectivity for the eliciting antigen. Their isolation entails either B-cell cloning or the linking of authentic pairs of variable region genes encoding them. We hypothesized that targeted RNA trans-splicing (TS) inside the B-cell nucleus could be harnessed as a novel means to link both variable region genes and reconstitute genuine immune B-cell specificities. This could be accomplished by a special targeting gene harboring a peptide linker exon flanked by sequences capable of targeting both heavy (HC) and light chain (LC) transcripts. Following sequential trans-splicing reactions, the resulting RNA in each cell would encode the two variable regions, joined by the peptide linker. In this study, we examined genetic components and configurations required for the separate trans-splicing steps and for the combined two-step reactions. Using a model antibody, we show that in transiently transfected cells, we can target variable region exons through both their acceptor and donor splice sites, precisely joining an exon encoding a synthetic linker and the complementary variable region so as to form a single-chain Fv. We also demonstrate the accurate formation of single-chain Fv transcript as a result of trans-splicing of RNA synthesized from two chromosomal genes expressed by a stably transfected B-cell hybridoma. Our attempts to link the two variable region genes via a synthetic linker exon through sequential trans-splicing events were only successful with regard to both ends of the linker and to the 3' end of the light chain, but repeatedly resulted in a deletion at the 5' end of the joined heavy chain transcript. The implications of our findings on the potential application of trans-splicing for the isolation of useful antibodies are discussed.

Development of spliceosome-mediated RNA trans-splicing (SMaRT) for the correction of inherited skin diseases

Gene therapy of large genes (e.g. plectin and collagen genes) is hampered by size limitations for insertions of the currently used viral vectors. To reduce the size of these insertions spliceosome-mediated RNA trans-splicing (SMaRT), which provides intron-specific gene-correction at the pre-RNA level, can be an alternative approach. To test its applicability in skin gene therapy, SMaRT was used in the context of the 4003delTC mutation in the collagen XVII gene (COL17A1) causing generalized atrophic benign junctional epidermolysis bullosa. A beta-galactosidase (beta-gal) trans-splicing assay system was established using intron 51 of COL17A1 as the target for trans-splicing. In this system, intron 51 is flanked by the 5'exon and the 3'exon of the beta-gal gene, the latter containing two in-frame stop codons. Cotransfection of a pre-trans-splicing molecule consisting of the binding domain of intron 51 and the 3'exon of beta-gal without the stop codons resulted in a 300-fold increase of beta-gal activity compared to controls. A 2-3-fold increase in efficiency was obtained through an elongation of the binding domains. Replacement of the complete 3'end of the COL17A1 gene was shown using a collagen XVII mini-gene construct. The beta-gal assay was used in human keratinocytes to evaluate the influence of a keratinocyte-specific spliceosome background. Reverse transcription polymerase chain reaction and beta-gal activity assay showed functional correction of the stop-codons in cultured human keratinocytes and in an immortalized GABEB cell line harbouring the 4003delTC mutation. These results demonstrate that SMaRT is feasible in a keratinocyte-specific context and therefore may be applied in skin gene therapy.

Gene conversion-like sequence transfers between transgenic antibody V genes are independent of RAD54

Homology-based Ig gene conversion is a major mechanism for Ab diversification in chickens and the Rad54 DNA repair protein plays an important role in this process. In mice, although gene conversion appears to be rare among endogenous Ig genes, Ab H chain transgenes undergo isotype switching and gene conversion-like sequence transfer processes that also appear to involve homologous recombination or gene conversion. Furthermore, homology-based DNA repair has been suggested to be important for somatic mutation of endogenous mouse Ig genes. To assess the role of Rad54 in these mouse B cell processes, we have analyzed H chain transgene isotype switching, sequence transfer, and somatic hypermutation in mice that lack RAD54. We find that Rad54 is not required for either transgene switching or transgene hypermutation. Furthermore, even transgene sequence transfers that are known to require homology-based recombinations are Rad54 independent. These results indicate that mouse B cells must use factors for promoting homologous recombination that are distinct from the Rad54 proteins important in homology-based chicken Ab gene recombinations. Our findings also suggest that mouse H chain transgene sequence transfers might be more closely related to an error-prone homology-based somatic hypermutational mechanism than to the hyperconversion mechanism that operates in chicken B cells.

The IgH locus of the channel catfish, Ictalurus punctatus, contains multiple constant region gene sequences: different genes encode heavy chains of membrane and secreted IgD

The delta-chain of catfish IgD was initially characterized as a unique chimeric molecule containing a rearranged VDJ spliced to C micro 1, seven C domain-encoding exons (delta1-delta7), and a transmembrane tail. The presence of cDNA forms showing splicing of delta7 to an exon encoding a secretory tail was interpreted to indicate that membrane (deltam) and secreted (deltas) forms were likely expressed from a single gene by alternative RNA processing. Subsequent cloning and sequence analyses have unexpectedly revealed the presence of three delta C region genes, each linked to a micro gene or pseudogene. The first (IGHD1) is located 1.6 kb 3' of the functional C micro (IGHM1). The second (IGHD3) is positioned immediately downstream of a pseudo C micro (IGHM3P), approximately 725 kb 5' of IGHM1. These two delta genes are highly similar in sequence and each contains a tandem duplication of delta2-delta3-delta4. However, IGHD1 has a terminal exon encoding the transmembrane region, whereas IGHD3 has a single terminal exon encoding a secreted tail. The occurrence of IGHD3 immediately downstream of a micro pseudogene indicates that the putative deltas product may not be expressed as a chimeric micro delta molecule. Western blots and protein sequencing data indicate that an IGHD3-encoded protein is expressed in catfish serum. Thus, catfish deltam transcripts appear to originate from IGHD1, whereas deltas transcripts originate from IGHD3 rather than, as previously inferred, from a single expressed delta gene. The third delta (IGHD2) is associated with a pseudo C micro (IGHM2P); its presence is inferred by Southern blot analyses.

IgG2a and igA co-expression by the natural autoantibody-producing murine B lymphoma T560

The T560 B lymphoma produces polyreactive IgG2a with the features of natural autoantibody. All T560 cells bear and secrete IgG2a but a small fraction spontaneously co-express IgA. Cells secreting IgA alone cannot be detected. IgA secretion is enhanced by interaction of T560 cells either with activated T cells and cognate antigen, or with LPS, but not with cytokines, including IL-5 and TGF-beta. IgA and IgG2a mRNAs have identical V186.2. DFL 16 and JH1 sequences from framework 2 through JH1. PCR analysis reveals that previous recombination events have led to deletion of the mu, gamma3, gamma1, gamma2b constant region genes from both the productive and the unproductive chromosome but the former has retained gamma2a, epsilon and alpha, the latter only alpha. Digestion-circularization (DC)-PCR experiments provide formal proof of DNA recombination between Ca and the intron upstream of Cmu. Evidently, the productive chromosome has switched only as far as gamma2a, the unproductive all the way to the alpha constant region gene. The unproductive allele is transcriptionally active as evidenced by the presence of mRNA encoding Calphal inappropriately spliced to a cryptic splice site in the downstream intron of DQ52 (eliminated from the productive chromosome). A specific RT-PCR using oligonucleotide primers derived from the upstream initiation site of the Ialpha exon and from Calpha1 discloses that T560 cells contain alpha-germ line mRNA, presumably transcribed from the Ialpha-region of the productive chromosome, spliced to Calpha. Treatment with LPS stops production of these spliced transcripts suggesting that it may promote either DNA recombination in cells spontaneously transcribing Ialpha or a change in splicing such that Ialpha sequence is no longer joined to Calpha. Verification of the DC-PCR product by sequencing reveals that the T560 and B10.A IgA (Ig2b allotype) hinge is different from the BALB/c IgA (Ig2a allotype) hinge: it has two extra Cys and has eliminated the first Thr, a potential glycosylation site in BALB/c IgA.

Class switch recombination of the chicken IgH chain genes: implications for the primordial switch region repeats

In mammals and the amphibian, Xenopus, isotypes of antibodies have been shown to be changed through class switch recombination within the IgH chain gene locus. Here, we identified switch (S) repetitive sequences in the 5' introns of the Ig C(mu) and C(gamma) genes of the chicken. The S(mu) region is composed of two homologous regions, S(mu)1 and S(mu)2. The S(mu)1 region is an upstream 3.7 kb sequence composed of 37 repeats of a consensus sequence containing tandem repeats of the decamer ACCAGTATGG. The S(mu)2 region is a downstream 1.4 kb sequence consisting of simple tandem repeats of a decamer CCCAGTACAG. The S(gamma) region contains repeats of the decamer TATGGGGCAG. Analysis of chicken IgG-producing hybridomas revealed that the C(mu) gene was deleted from the chromosome by the recombination occurring between the S(mu) and S(gamma) regions. Recombination breakpoints at the C(mu) gene of splenocytes from an immunized chicken were scattered around the S(mu) region and two such breakpoints, the precise position of which were determined, were located within possible hairpin loop structures at the palindromic sequence of S(mu)1. A primordial palindromic sequence from which the prevalent switch repeat motifs of mammals, chickens and amphibians may have diverged is presented.

RNA maturation of the rice SPK gene may involve trans-splicing

A gene encoding a calcium-dependent seed-specific protein kinase (SPK) is abundantly expressed in developing rice seeds (Kawasaki, T et al. Gene (1993) 129, 183-189). Rice genomic clones encoding SPK were isolated using the entire cDNA fragment as a probe. Physical mapping of these genomic clones indicated that the genomic region corresponding to the entire cDNA was divided into two different regions, SPK-A and SPK-B, located on different rice chromosomes. The results of RACE-PCR analyses showed that the respective transcripts from SPK-A and SPK-B contained additional sequences which were not found in the SPK cDNA, and that these sequences were removed like introns during maturation of the SPK mRNA. These results suggest that two different RNAs were independently transcribed from SPK-A and SPK-B and joined, possibly by trans-splicing.

Spliceosome-mediated RNA trans-splicing as a tool for gene therapy

We have developed RNA molecules capable of effecting spliceosome-mediated RNA trans-splicing reactions with a target messenger RNA precursor (pre-mRNA). Targeted trans-splicing was demonstrated in a HeLa nuclear extract, cultured human cells, and H1299 human lung cancer tumors in athymic mice. Trans-splicing between a cancer-associated pre-mRNA encoding the beta-subunit of human chorionic gonadotropin gene 6 and pre-trans-splicing molecule (PTM) RNA was accurate both in vitro and in vivo. Comparison of targeted versus nontargeted trans-splicing revealed a moderate level of specificity, which was improved by the addition of an internal inverted repeat encompassing the PTM splice site. Competition between cis- and trans-splicing demonstrated that cis-splicing can be inhibited by trans-splicing. RNA repair in a splicing model of a nonfunctional lacZ transcript was effected in cells by a PTM, which restored significant beta-galactosidase activity. These observations suggest that spliceosome-mediated RNA trans-splicing may represent a general approach for reprogramming the sequence of targeted transcripts, providing a novel approach to gene therapy.

Rapid and efficient cloning of cDNAs encoding Krüppel-like zinc finger proteins by degenerate PCR

To isolate cDNAs encoding Krüppel-like zinc finger proteins consisting of several hundred members, most of which are yet to be identified, from a limited number of available cells, we developed a rapid and efficient zinc finger gene cloning method based on reverse transcription-polymerase chain reaction (RT-PCR) using tagged, degenerate oligonucleotide primers corresponding to the conserved H/C link followed by the reverse blue selection to identify clones containing properly amplified fragments. More than 5x103 blue colonies were obtained from only 1ng of total RNA. Eighty-eight out of 89 clones, which were randomly picked up from blue colonies and sequenced, encoded 60 different zinc fingers with the expected structure, and among them, only four have been previously described. Furthermore, it was possible to rapidly select clones that were differentially expressed in a tissue and stimulation-specific manner by a differential screening of the zinc-finger cDNA library using probes consisting of distinct sets of the zinc-finger PCR products. These results indicate that our PCR-based method is quite efficient and suitable for analyzing not only zinc finger genes but also other large gene families, especially when the available cells are very limited.

Immunoglobulin double isotype-producing hybridomas isolated from an autoimmune (NZB x NZW)F1 mouse

In the sera of (NZB x NZW)F1 (B/W) mice that develop a lupus-like syndrome, increased levels of IgG antibodies (Ab) reacting with TNP have been detected before the appearance of IgG anti-DNA Ab and clinical symptoms. A single injection of trinitrophenyl-bovine serum albumin (TNP/BSA) in physiological saline into a young B/W mouse (3 months old), followed by fusion of its splenocytes 3 days later, gave rise to hybridomas simultaneously secreting IgM and IgG Ab with anti-TNP reactivity. Both mu and gamma chains were detected in culture supernatants by ELISA, and double isotype-producing cells were labeled by immunofluorescence. Molecular analysis of two of these double isotype-producing hybridomas showed the presence of mRNA coding for both mu and gamma chains of Ig, and this gamma mRNA could be translated in vitro into a gamma heavy (H) chain. Comparison of the H chain variable-region sequences of IgM and IgG revealed 100% homology between mu and gamma V(H) genes in one clone, while mu and gamma V(H) genes showed only 80% homology in the other clone. Both clones produced a single kappa light (L) chain. These two hybridomas, isolated from a B/W mouse, thus represent two different mechanisms of double isotype expression: the first one corresponds to an IgM to IgG switch, while the second one reflects a lack of allelic exclusion.

In vitro synthesized SV40 cRNA is trans-spliced after microinjection into the nuclei of mammalian cells

We present for the first time experimental evidence that in vitro synthesized RNA (cRNA) is trans-spliced after microinjection into the nuclei of mammalian tissue culture cells. The template used for cRNA synthesis was the early SV40 BstXl/BamHI DNA fragment. This DNA fragment encodes exclusively for the second T-antigen exon and contains the intact small t-antigen intron. To generate the corresponding mRNA (T1-mRNA) by trans-splicing, the cells utilize a 5' cryptic splice site located within the second T-antigen exon of one cRNA molecule which is spliced to the small t-antigen 3' splice site of another cRNA molecule. Formation of the T1-mRNA by trans-splicing was confirmed by RT-PCR analysis and DNA sequencing. Efficient trans-splicing required that competitive small t-antigen cis-splicing be inhibited by deletion of the small t-antigen 5' splice site. The T1-mRNA was not generated when the cryptic 5' splice site was mutated.

The 5' and 3' splice sites come together via a three dimensional diffusion mechanism

We present evidence that the splice sites in mammalian pre-mRNAs are brought together via a three dimensional diffusion mechanism. We tested two mechanisms for splice site pairing: a lateral diffusion ('scanning') model and the currently favored three dimensional diffusion ('jumping') model. Two lines of evidence that distinguish between these two models are presented. The first utilized bipartite splicing substrates tethered by double-stranded RNA stems predicted to provide either a moderate or severe block to splice site pairing via a scanning mechanism. Splice site pairing via a jumping mechanism was expected to be unaffected or affected minimally. The second approach utilized a flexible poly(ethylene glycol) moiety within the intron. This insertion was predicted to reduce scanning efficiency but not the efficiency of a three dimensional diffusion mechanism. The best explanation for the data with the bipartite RNAs is that splice site pairing occurs through three dimensional diffusion. Kinetic analysis of the poly(ethylene glycol) containing substrate showed that neither the lag phase nor the initial rates of mRNA production and spliceosome assembly were affected by this insertion. Therefore, both experimental approaches supported the three dimensional diffusion model of splice site pairing.

IgA class switch in I alpha exon-deficient mice. Role of germline transcription in class switch recombination

Studies have implicated defective Ig class switch in the pathogenesis of IgA deficiency. To understand better the molecular events that regulate IgA class switch, a 1.4-kb region of the IgA locus containing the I alpha exon was replaced with a human hypoxanthine phosphoribosyltransferase minigene by gene targeting in murine embryonic stem cells. The I alpha exon-deficient mice derived from these embryonic stem cells had normal IgA levels in serum and secretions and normal numbers of IgA B cells in Peyer's patches and spleen. Further, I alpha exon-deficient B cells efficiently underwent IgA class switch in vitro, despite the absence of I alpha exon-containing germline transcripts. Notably, I alpha exon-deficient B cells did not require TGF-beta for IgA class switch since stimulation with LPS alone led to IgA expression. Nonetheless, whereas I alpha exon-deficient B cells constitutively expressed human hypoxanthine phosphoribosyltransferase transcripts, they did not produce IgA in the absence of LPS stimulation. These results demonstrate that the I alpha exon or transcripts containing the I alpha exon are not required for IgA class switch. Further, the effects of TGF-beta on I alpha locus transcription can be supplanted by expression of a heterologous minigene at that locus, but a second signal is required for the induction of IgA class switch.

Enhancer-dependent interaction between 5' and 3' splice sites in trans

Splice-site selection and alternative splicing of nuclear pre-mRNAs can be controlled by splicing enhancers that act by promoting the activity of upstream splice sites. Here we show that RNA molecules containing a 3' splice site and enhancer sequence are efficiently spliced in trans to RNA molecules containing normally cis-spliced 5' splice sites or to normally trans-spliced spliced leader RNAs from lower eukaryotes. In addition, we show that this reaction is stimulated by (Ser + Arg)-rich splicing factors that are known to promote protein-protein interactions in the cis-splicing reaction. Thus, splicing enhancers facilitate the assembly of protein complexes on RNAs containing a 3' splice site, and this complex is sufficiently stable to functionally interact with 5' splice sites located on separate RNAs. This trans-splicing is mediated by interactions between (Ser + Arg)-rich splicing factors bound to the enhancer and general splicing factors bound to the 5' and 3' splice sites. These same interactions are likely to play a crucial role in alternative splicing and splice-site selection in cis.

Germ-line human epsilon heavy chain gene RNA transcripts utilize the full range of alternative 3' splicing seen in productive epsilon mRNA

An early step in immunoglobulin isotype switching involves the initiation of active transcription of downstream heavy chain loci while they are still in germ-line configuration. This results in the production of 'sterile' germ-line RNA transcripts that do not contain VDJ sequences, the role of which in isotype switching is under intense scrutiny. We investigated whether such human epsilon germ-line transcripts employ the full complement of complex alternative 3' RNA splicing, splicing we have recently reported occurring with productive epsilon mRNA transcripts. Using a reverse transcriptase polymerase chain reaction (RT-PCR) strategy, in which the 5' primer was located in the I region of the epsilon gene, a region expressed in germ-line but not productive (VDJ containing) epsilon transcripts, we showed that the full range of alternative 3' epsilon splices occur with germ-line transcripts. These results demonstrate that epsilon 3' splicing events are independent of 5' isotype DNA switching. Additionally, we showed that, just as with mature epsilon mRNA, the relative production of the various epsilon germ-line mRNA isoforms was responsive to modulation by stimuli such as interleukin-10 (IL-10). Thus B cells, when stimulated to produce epsilon germ-line transcripts, generate a family of germ-line mRNA that differ not only in their initiation sites but, more importantly, also differ in their 3' sequences, sequences that could be important in regulation of the parent gene itself. Furthermore, by discontinuous or trans-splicing, cells could utilize these various epsilon germ-line transcripts to produce the full range of mature IgE proteins prior to undergoing deletional recombination of isotype switching.

Failure of IgG production due to a defect in the opening of the chromatin structure of I gamma 1 region in a patient with IgG and IgA deficiency

Patients with common variable immunodeficiency (CVID) display reduced levels of two or all three of the major immunoglobulin isotypes, and the deficiency is characterized by failure of B cells to differentiate into plasma cells in many cases. A patient (14 years old, female) showed normal serum IgM levels and low serum IgG and IgA levels, including low levels of all IgG subclasses. Northern blot analysis suggested that the patient's B cells may be defective at the immunoglobulin heavy chain isotype switch. The germ-line C gamma 1 transcript was amplified from cDNA of healthy controls by the addition of recombinant IL-2 (rIL-2) to pokeweed mitogen-stimulated peripheral mononuclear cells or Staphylococcus aureus Cowan I (SAC)-stimulated IgM-producing lymphoblastoid cell lines (LCL) transformed by Epstein-Barr virus, while it was not amplified from cDNA of the patient. In the I gamma 1 region of LCL cultured with SAC plus rIL-2, the inner cytosine in the 5' C-C-G-G 3' sequence nearest the 3' site of the I gamma 1 region, at least, was not completely unmethylated in the patient. Moreover, the DNase I hypersensitive site was not induced in the patient's LCL by SAC plus rIL-2. These results indicate that the defects of the immunoglobulin heavy chain isotype switch in the patient's B cells are due to failure in the synthesis of germ-line C gamma transcripts, and this may be caused by defects in opening of the chromatin structures of specific switch regions.

DNA rearrangement can account for in vitro switching to IgG1

During immune responses, B lymphocytes may switch from the expression of immunoglobulin M (IgM) to the expression of another isotype (e.g., IgG, IgE, IgA). In stable hybridomas and myelomas expressing a "switched" (S) isotype, DNA deletions between S mu and a "downstream" S region (S region recombination) have been found. In primary B cells, studies of the molecular basis of switching have been limited by the ability to sensitively quantitate the amount of DNA deletion; such studies would be of interest because other nondeletional mechanisms (trans-splicing, alternative processing of a long transcript) have been proposed to account for isotype switching in certain circumstances. We have applied the digestion-circularization polymerase chain reaction (DC-PCR) technique to measure the amount of S region recombination that occurs in the course of class switching in primary B lymphocytes. Resting B cells were cultured in lipopolysaccharide (LPS) and interleukin 4 (IL-4) to stimulate switching to IgG1. These cells begin to express membrane IgG1 at day 2.5 of culture and reach maximum expression by day 4.5. DNA was prepared from cultured cells and analyzed for S mu-S gamma 1 rearrangement by DC-PCR. Chimeric switch regions, indicating S mu-S gamma 1 recombination, were detected in amounts that, in most cases, correlated with surface expression. Furthermore, when cells were sorted on the basis of surface IgG1 expression, a mean of at least one S mu-S gamma 1 rearrangement per cell was seen in five out of seven experiments. In general, the IgG1+ cells obtained at 4.5 and 5.5 d of culture had close to 2 S mu-S gamma 1 rearrangements per cell. In IgG1- cells, S mu-S gamma 1 rearrangements were detectable, but at frequencies substantially lower that in IgG1+ cells. Thus, these results indicate that DNA deletion accompanies class switching in normal B cells stimulated with LPS and IL-4.

B-cell apoptosis induced by antigen receptor crosslinking is blocked by a T-cell signal through CD40

In mice transgenic for an autoantibody, self-reactive B cells have been shown to be eliminated upon interaction with membrane-bound self-antigens in the periphery as well as in the bone marrow, suggesting that both immature and mature B cells are eliminated by multimerization of surface immunoglobulins (sIg). Activation of mature B cells by antigens may thus require a second signal that inhibits sIg-mediated apoptosis. Such a second signal is likely to be provided by T helper cells, because B-cell tolerance is more easily induced in the absence of T helper cells. To assess the molecular nature of the signal that inhibits sIg-mediated apoptosis, we used anti-IgM-induced apoptotic death of WEHI-231 B lymphoma cells as a model system. Here we report that the signal for abrogating sIg-mediated apoptosis is generated by association of the CD40L molecule on T cells with the CD40 molecule on WEHI-231 cells. T-cell help through CD40 may thus determine whether B cells are eliminated or activated upon interaction with antigens.

Replacement of germ-line epsilon promoter by gene targeting alters control of immunoglobulin heavy chain class switching

Recent work has shown that the ability of cytokines to direct immunoglobulin heavy chain class-switch recombination to particular heavy chain constant (C) region (CH) genes correlates with the induction of specific germ-line CH transcripts. To test the role of germ-line transcripts in class switching, we have used homologous recombination to mutate the immunoglobulin heavy chain locus of the 18.81A20 murine pre-B-cell line. In the parent cell line, the combination of interleukin-4 (IL-4) and lipopolysaccharide (LPS) induces germ-line epsilon locus transcription prior to class switching to epsilon. The heavy chain locus of the mutated cell line contains the immunoglobulin heavy chain enhancer and variable region gene promoter in place of the LPS/IL-4-responsive germ-line epsilon promoter. The mutant cell line constitutively transcribes the epsilon locus in the absence of IL-4. Strikingly, the mutant cell line also switches to epsilon in the absence of IL-4. This result demonstrates that, at least in the 18.81A20 cell line, germ-line epsilon transcription plays a direct role in class switching to the epsilon locus. In addition, the ability to change the pattern of class switching by altering transcriptional activity indicates that transcription of germ-line CH is mechanistically important in regulation of class switching.

Spliced leader RNAs from lower eukaryotes are trans-spliced in mammalian cells

Exon sequences present on separate RNA molecules can be joined by trans-splicing in trypanosomatids, Euglena, and in the nematode and trematode worms. Trans-splicing involves an interaction between a 5' splice site present in a spliced leader RNA and a 3' splice site located near the 5' end of pre-messenger RNAs. In vitro trans-splicing of artificial mammalian pre-mRNAs has been reported, but the efficiency of splicing appears to depend on sequence complementarity between the two substrates. There has been speculation that some natural pre-mRNAs can be trans-spliced in mammalian cells in vivo, but alternative interpretations have not been ruled out. Here we show that spliced leader RNAs can be accurately trans-spliced in mammalian cells in vivo and in vitro. Both nematode and mammalian 3' splice sites can function as acceptors for trans-splicing in vivo. These results reveal functional conservation in the splicing machinery between lower eukaryotes and mammals, and they directly demonstrate the potential for trans-splicing in mammalian cells.

Anomalous rearrangements of the immunoglobulin heavy chain genes in human leukemias support the loop-out mechanism of class switch

Discrete rearrangements of immunoglobulin genes are characteristic of lymphoproliferative diseases of B cells and provide direct evidence of their clonal nature. In addition, because leukemic transformation and growth may amplify B cell clones regardless of selection by antigen, analysis of rearranged Ig genes in leukemic clones may give insight into molecular events taking place during the ontogenesis of normal B cells. We have tested DNA samples from patients with chronic B cell leukemias in search for abnormal rearrangements of the Ig heavy chain gene region. By Southern blot analysis we found an unexpected break in the JH-C mu region in 7 out of 118 cases. Two of these cases were investigated in detail by constructing from each a phage genomic library and isolating the phage clones containing the break points. In both cases the JH-C mu separation was confirmed. Further analysis demonstrated that in both cases the abnormality was an inversion of the Ig heavy chain gene between C mu and one of the C gamma segments. This inversion structure strongly suggests that, as has been demonstrated in murine cell lines and in splenocytes stimulated in vitro, class switching in human B lymphocytes occurs in vivo via a loop-out deletion mechanism. The frequency of abnormal events may be as high as 15%. Our data indicate that a proportion of cases of chronic B cell leukemia arise from a cell which has attempted an Ig class switch.

Common variable immunodeficiency with increased surface IgM-positive double-bearing B cells

We report a case of common variable immunodeficiency (CVI) that shows low levels of IgG and IgA, but a normal quantitative or qualitative level of IgM. T-cell functions were not disturbed. Increased numbers of surface IgM (sIgM) and sIgD, sIgM and sIgG, sIgM and sIgA double-bearing B cells were observed as compared with a control. No IgG and IgA induction upon stimulation with Staphylococcus aureus Cowan I (SAC) and recombinant interleukin-2 (rIL-2), or pokeweed mitogen (PWM) and rIL-4 or rIL-6 was observed, although there was proliferation. Although mu mRNA was expressed as much as in a healthy control, transcription of gamma mRNA and alpha mRNA was very low. Furthermore, no enhanced effects of gamma mRNA and alpha mRNA were recognized upon stimulation with rIL-4 and rIL-6. These results suggest that the patient's B cells might be defective at the switching process from mu, mu and delta, mu and gamma to gamma or mu and alpha to alpha.

Interallelic V(D)J trans-rearrangement within the beta T cell receptor gene is infrequent and occurs preferentially during attempted D beta to J beta joining

Previous work has demonstrated that intergenic V(D)J rearrangement, a process referred to as trans-rearrangement, occurs at an unexpectedly high frequency. These rearrangements generate novel V(D)J combinations which could conceivably have some role in the normal immune system, and since they probably arise through chromosomal rearrangements akin to those associated with lymphoid neoplasia, they may also serve as a model for investigating recombinational events which underlie oncogenesis. In view of the existence of a mechanism that permits relatively frequent intergenic trans-rearrangements, it seems reasonable that interallelic trans-rearrangements involving segments belonging to each of the two alleles of a single antigen receptor gene might also occur. To determine the frequency of such rearrangements, we examined thymocytes of F1 progeny of a cross between SWR mice, which have a deletion spanning 10 of the known V beta segments, and NZW mice, which have a deletion involving all J beta 2 segments. Rearranged TCR-beta genes containing V beta segments from the NZW chromosome and J beta segments from the SWR chromosome were amplified from the DNA of F1 thymocytes with the polymerase chain reaction. Using this approach, we found that such rearrangements are relatively uncommon, being present in about 1 in 10(5) thymocytes, a frequency lower than that of V gamma/J beta intergenic trans-rearrangements. The ratio of conventional cis-rearrangement to interallelic trans-rearrangement for any particular V beta segment appears to be about 10(4):1. The structure of the junctions in all trans-rearrangements analyzed closely resembles conventional cis-rearrangements, indicating involvement of V(D)J recombinase in the ultimate joining event. However, in contrast to cis-rearrangements, a strong bias for inclusion of D beta 1 segments over D beta 2 segments was noted, suggesting that interallelic trans-rearrangement may occur preferentially during attempted D-J joining. J beta 2 segment usage in trans-rearrangements also appeared to differ from that expected from previously studied cis-rearrangements. The results have implications with respect to the events and timing of conventional cis-rearrangement during thymocyte differentiation, and the prevalence of various types of trans-rearrangements.

Chromosomal position of rearranging gene segments influences allelic exclusion in transgenic mice

Formation of a complete immunoglobulin heavy-chain transcription unit involves the ordered rearrangement of variable (V), diversity (D), and joining (J) region gene segments. In antibody-producing cells, this process is regulated such that only one of two antibody genes is expressed. Experiments with transgenic mice suggest that this mechanism, known as allelic exclusion, is mediated through the membrane-bound form of the immunoglobulin heavy chain. However, in all transgenic lines produced to date exclusion of the endogenous genes by the transgene is incomplete. To characterize the molecular basis for this escape from regulation, we have examined the rearrangements of endogenous immunoglobulin heavy-chain genes. We find that a transgene that encodes the membrane-bound form of human IgM efficiently inhibits rearrangements of endogenous gene segments located at the 5' end of the heavy-chain locus. However, recombining elements found at the 3' end of the locus escape and continue to undergo recombination. A transgene that encodes the secreted form of the same immunoglobulin protein has no effect on recombination, regardless of position of the recombining segment in the chromosome. These results have important implications for our understanding of the control of allelic exclusion.

Identification of a conserved lipopolysaccharide-plus-interleukin-4-responsive element located at the promoter of germ line epsilon transcripts

Treatment of splenic B lymphocytes and certain B-lineage cell lines with the mitogen lipopolysaccharide (LPS) and the lymphokine interleukin-4 (IL-4) induces expression of germ line immunoglobulin C epsilon transcripts and class switching to the C epsilon gene. We show that LPS-plus-IL-4 induction of germ line epsilon transcripts (termed I epsilon transcripts) occurs at the transcriptional level in an Abelson murine leukemia virus-transformed pre-B-cell line. A 1.1-kb region of DNA surrounding the I epsilon promoter endows inducible transcription to a heterologous reporter gene stably transfected into these cells; such inducible expression depends on combined treatment with LPS and IL-4. Analyses of constructs transiently introduced into a B-cell lymphoma line demonstrated that LPS-plus-IL-4-inducible expression can be conferred by a 179-bp segment of DNA spanning the I epsilon transcriptional initiation site. Mutational analyses demonstrated that this expression depended on DNA sequences within a conserved region directly upstream from the I epsilon transcriptional initiation region. One nuclear protein that is constitutively expressed in normal B cells binds to the downstream end of the conserved sequence; its binding specificity correlates with the functional effect of several mutations. Two additional proteins, which are induced by IL-4 treatment of splenic B cells, bind to the transcription initiation sites of I epsilon. These proteins are indistinguishable in binding assays from proteins previously shown to bind an enhancer region of the class II major histocompatibility complex gene A alpha.

Structure and expression of mouse germline immunoglobulin gamma 3 heavy chain transcripts induced by the mitogen lipopolysaccharide

Germline immunoglobulin heavy chain gene transcription is though to direct isotype switching by modulating the accessibility of specific switch regions to a recombinase. In this study, cloned cDNA copies of mouse germline Igh-8 RNAs have been used to characterize the Igh-8 transcription unit. The 5' end of these transcripts are derived from an exon denoted Ig3, located 1 kilobase 5' of the Igh-8 switch region. Sequence analysis of cDNA and genomic clones reveals that these RNAs are noncoding. In splenic B cell cultures treated with lipopolysaccharide (LPS), germline Igh-8 transcript levels are upregulated after 8 h due to increased transcription. This induction is consistent with the identification of a putative binding site for the LPS inducible transcription factor NF-kappa B approximately 150 nucleotides upstream of the sites of transcript initiation. Furthermore, nucleotide sequence comparisons reveal that the region encompassing the site of germline Igh-8 transcription initiation is highly homologous to part of the Ig2b exon, and is also conserved upstream of the Igh-1 switch region. The implications of these findings for the control of germline Igh-8 transcription is discussed.

Identification of a conserved lipopolysaccharide-plus-interleukin-4-responsive element located at the promoter of germ line epsilon transcripts

Treatment of splenic B lymphocytes and certain B-lineage cell lines with the mitogen lipopolysaccharide (LPS) and the lymphokine interleukin-4 (IL-4) induces expression of germ line immunoglobulin C epsilon transcripts and class switching to the C epsilon gene. We show that LPS-plus-IL-4 induction of germ line epsilon transcripts (termed I epsilon transcripts) occurs at the transcriptional level in an Abelson murine leukemia virus-transformed pre-B-cell line. A 1.1-kb region of DNA surrounding the I epsilon promoter endows inducible transcription to a heterologous reporter gene stably transfected into these cells; such inducible expression depends on combined treatment with LPS and IL-4. Analyses of constructs transiently introduced into a B-cell lymphoma line demonstrated that LPS-plus-IL-4-inducible expression can be conferred by a 179-bp segment of DNA spanning the I epsilon transcriptional initiation site. Mutational analyses demonstrated that this expression depended on DNA sequences within a conserved region directly upstream from the I epsilon transcriptional initiation region. One nuclear protein that is constitutively expressed in normal B cells binds to the downstream end of the conserved sequence; its binding specificity correlates with the functional effect of several mutations. Two additional proteins, which are induced by IL-4 treatment of splenic B cells, bind to the transcription initiation sites of I epsilon. These proteins are indistinguishable in binding assays from proteins previously shown to bind an enhancer region of the class II major histocompatibility complex gene A alpha.

Analysis of the immune system with transgenic mice: B cell development and lymphokines

Over the last decade transgenic mice expressing genes relevant for the immune system have been generated. Transgenic expression of immunoglobulin heavy and/or light chain genes of different isotypes and different specificities have helped to better understand phenomena relevant to B cell development such as allelic exclusion of immunoglobulins and B cell tolerance. Transgenic mice expressing interleukin genes have also been used to study the ways of action of these important growth and differentiation factors in the context of the mouse immune system.

Trans-splicing as a possible molecular mechanism for the multiple isotype expression of the immunoglobulin gene

We analyzed the molecular mechanism for the immunoglobulin (Ig) multiple isotype expression using a transgenic mouse (TG.SA) model system. Though most of the endogenous mu chain expression was excluded by the expression of the human rearranged mu transgene in the TG.SA mouse, a significant portion of splenic B lymphocytes could express the transgenic human IgM and endogenous mouse IgG simultaneously after stimulation with lipopolysaccharide and interleukin 4. The fluorescence-activated cell sorter-purified population of the human IgM+/mouse IgG+ cells expressed mRNA that consisted of properly spliced sequences of the transgenic VHDJH and the endogenous mouse C gamma genes (trans-mRNA), together with the transgenic human mu mRNA and germline transcripts of the mouse C gamma gene, without apparent rearrangement of the transgene. We also found that a lymphoma tumor, derived from the cross between the TG.SA mouse and another transgenic mouse carrying Ig H chain enhancer-driven c-myc oncogene, expressed about equal levels of the trans-mRNA and the transgenic mu mRNA without DNA rearrangement in either the transgene or the endogenous mouse switch region. These findings strongly support our previous proposal that the trans-splicing can account for the multiple isotype expression in this transgenic model and also suggest that novel molecular mechanism(s) might be involved in this reaction.

Isotype switching in anti-immunoglobulin-activated B lymphoblasts: differential requirements for interleukin 4 and other lymphokines to elicit membrane vs. secreted IgG1

Anti-immunoglobulin (Ig)-activated B lymphoblasts, prepared by culturing high-density B cells with anti-Ig-Sepharose for 48 h, can be induced to secrete IgM and IgG1 by a mixture of T cell-derived lymphokines containing interleukin (IL) 4. In this study we have examined the conditions required for lymphokine-mediated induction of IgG1 secretion and membrane (m)IgG1 expression in B lymphoblasts. Resting B cells exposed to IL 4 (10-100 U/ml) during anti-Ig-mediated blast transformation did not secrete IgG1 upon subsequent culture with lipopolysaccharide (LPS) regardless of whether IL 4 was present or absent during the secondary culture. In contrast, B lymphoblasts previously exposed to IL 4 did secrete IgG1 in response to T cell-derived lymphokines [EL 4 supernatant depleted of IL 4; (D)EL 4 SN]. However, optimal IgG1 secretion was obtained when B lymphoblasts were simultaneously exposed to IL 4 and other lymphokines. Pre-exposure to (D)EL 4 SN, which contains IL 5 and IL 2, failed to prepare anti-Ig blasts to secrete IgG1 in response to LPS and IL 4. Inhibition of IL 5 and IL 2 activity in EL 4 SN suppressed IL 4-mediated IgG1 secretion. Together, these data indicate that B lymphoblasts require IL 5 and IL 2 in addition to IL 4 to secrete IgG1, and that the IL 4 signal(s) must precede or accompany those provided by the other lymphokines. As a measure of the fraction of cells capable of switching to IgG1, we assessed expression of mIgG1 on B lymphoblasts by fluorescence flow cytometry. B lymphoblasts cultured for 3 days with (D)EL 4 SN and IL 4 (10-100 U/ml) were 8% to 20% mIgG1+; in the absence of IL 4 blasts did not express detectable mIgG1. Although anti-Ig blasts treated with LPS and IL 4 did not secrete appreciable IgG1, a substantial fraction of B lymphoblasts (4% - 19%) cultured with LPS and IL 4, but not LPS alone, expressed mIgG1. These results suggest that LPS and IL 4 are sufficient to commit B lymphoblasts to mIgG1 expression, but that additional signals provided by T cell-derived lymphokines are required to elicit IgG1 secretion.

Hybrid T cell receptor genes formed by interlocus recombination in normal and ataxia-telangiectasis lymphocytes

In this paper, using polymerase chain reaction (PCR), we demonstrated the occurrence of hybrid genes formed by interlocus recombination between T cell receptor gamma (TCR-gamma) variable (V) regions and TCR-beta joining (J) regions in the peripheral blood lymphocytes (PBL) from normal individuals and patients with ataxia-telangiectasia (AT). Sequence analysis of the PCR-derived hybrid genes confirmed that site-specific V gamma-J beta recombination had occurred and showed that 10 of 23 genomic hybrid genes maintained a correct open reading frame. By dilution analysis, the frequency of these hybrid genes was 8 +/- 1/10(5) cells in normal PBL and 587 +/- 195/10(5) cells in AT PBL. These frequencies and the approximately 70-fold difference between the normal and AT samples are consistent with previous cytogenetic data examining the occurrence of an inversion of chromosome 7 in normal and AT PBL. We also demonstrated expression of these hybrid genes by PCR analysis of first-strand cDNA prepared from both normal and AT PBL. Sequence analysis of the PCR-amplified transcripts showed that, in contrast to the genomic hybrid genes, 19 of 22 expressed genes maintained a correct open reading frame at the V-J junction and correctly spliced the hybrid V-J exon to a TCR-beta constant region, thus allowing translation into a potentially functional hybrid TCR protein. Another type of hybrid TCR transcript was found in a which a rearranged TCR-gamma V-J exon was correctly spliced to a TCR-beta constant region. This form of hybrid gene may be formed by trans-splicing. These hybrid TCR genes may serve to increase the repertoire of the immune response. In addition, studies of their mechanism of formation and its misregulation in AT may provide insight into the nature of the chromosomal instability syndrome associated with AT. The mechanism underlying hybrid gene formation may be analogous to the mechanism underlying rearrangements between putative growth-affecting genes and the antigen receptor loci, which are associated with AT lymphocyte clones and lymphoid malignancies.

Structure and expression of germline epsilon transcripts in human B cells induced by interleukin 4 to switch to IgE production

Interleukin 4 (IL-4)-induced IgE production coincides with the appearance of the 2.2-kb productive epsilon-mRNA, but is preceded by synthesis of a 1.7-kb epsilon-RNA. Analysis of cDNA copies of the 5' end of this RNA indicated that the 1.7-kb epsilon-RNA is a germline epsilon immunoglobulin heavy chain transcript with an exon mapping 5' to the switch region. Transcription through switch regions has been implicated in the control of class switching. However, IL-4 or cloned CD4+ T cells were able to induce germline epsilon transcripts without inducing IgE synthesis, for which both signals were required. These results indicate that induction of human germline epsilon-RNA does not necessarily result in IgE synthesis, and that additional regulatory mechanisms are involved in class switching.