Search

During the development of an embryo, various processes occur which induce the formation of differentiated tissues. Some of these processes can be observed at a molecular level, for example, myosin is synthesized in muscle cells, crystaline in lens tissue, keratin in skin, immunoglobulin in lymphocytes and hemoglobin in reticulocytes (1,2).

We show that c-myc is an inducible gene that is regulated by specific growth signals in a cell-cycle-dependent manner. Specifically, agents that initiate the first phase of a proliferative response in lymphocytes (lipopolysaccharide or Concanavalin A) and fibroblasts (platelet-derived growth factor) induce c-myc mRNA. Within one to three hr after the addition of these mitogens to the appropriate cells, c-myc mRNA concentration is increased between 10- and 40-fold. This induction of c-myc mRNA occurs in the presence of cycloheximide and, therefore, does not require the synthesis of new protein species. Consequently, the induction of c-myc mRNA is not secondary to growth. In addition, c-myc mRNA is "superinduced" by the combination of cycloheximide and mitogen, a finding consistent with a model that a labile protein may regulate c-myc levels in these cells. Further, this work suggests a regulatory linkage between the function of two oncogenes--c-myc and c-sis--the latter being the putative structural gene for PDGF.

We have cloned and determined the entire nucleotide sequence of cDNAs corresponding to the putative alpha subunits of the human and rat mast cell high-affinity IgE receptors. Both human and rat cDNAs encode an NH2-terminal signal peptide, two immunoglobulin-like extracellular domains (encoded by discrete exons), a hydrophobic transmembrane region, and a positively charged cytoplasmic tail. The human and rat alpha subunits share an overall homology with one another and the immunoglobulin gene family, suggesting that they arose from a common ancestral gene and continue to share structural homology with their ligands. In addition, the rat gene is transcribed into at least three distinct forms, each of which yields a somewhat different coding sequence.

The characteristic chromosomal translocations that occur in certain human malignancies offer opportunities to understand how two gene systems can affect one another when they are accidentally juxtaposed. In the case of Burkitt lymphoma, such a translocation joins the cellular oncogene, c-myc, to a region encoding one of the immunoglobulin genes. In at least one example, the coding sequence of the rearranged c-myc gene is identical to that of the normal gene, implying that the gene must be quantitatively, rather than qualitatively, altered in its expression if it is to play a role in transformation. One might expect to find the rearranged c-myc gene in a configuration that would allow it to take advantage of one of the known immunoglobulin promoters or enhancer elements. However, the rearranged c-myc gene is often placed so that it can utilize neither of these structures. Since the level of c-myc messenger RNA is often elevated in Burkitt cells, the translocation may lead to a deregulation of the c-myc gene. Further, since the normal allele in a Burkitt cell is often transcriptionally silent in the presence of a rearranged allele, a model for c-myc regulation is suggested that involves a trans-acting negative control element that might use as its target a highly conserved portion of the c-myc gene encoding two discrete transcriptional promoters.

Transgenic mice carrying a fusion gene in which the mouse immunoglobulin enhancer has been inserted into an otherwise normal human c-myc gene develop a narrow spectrum of pre-B-cell lymphomas. Tumor occurrence is correlated with expression of the transgene in organs in which large numbers of pre-B cells predominate. These tumors, which arise stochastically, are virtually all lymphoblastic lymphomas of the pre-B-cell type. Evidently the isolated enhancer targets oncogene expression and tumorigenesis to the early B-cell population in preference to more mature B-cell populations. The transgene also confers enhanced in vitro growth properties on nontransformed pre-B cells as observed in bone marrow cultures derived from transgenic animals. These cultured cells represent a population in which the activating function of c-myc can be uncoupled from secondary oncogenic events occurring in vivo.

In BALB/c-3T3 cells, expression of the c-myc gene is stimulated by platelet-derived growth factor (PDGF). Using mouse mammary tumour virus promoter: c-myc recombinant plasmids, 3T3 sublines were constructed in which hydrocortisone was the primary determinant of myc mRNA content. The c-myc gene product is an intracellular mediator of the growth response to PDGF though probably not the only one. Both the human and the mouse c-myc genes stimulate clonal growth of 3T3 cells in PDGF-free medium suggesting new strategies for analysis of oncogenes which do not function in focus formation assays.

We have used cytogenetic and recombinational analysis to determine the position of a transgene integrated into the mouse genome. The transgene maps to band F on the physical map of mouse chromosome 15 by in situ analysis and is tightly linked genetically to a cluster of loci that include the mutations caracul (Ca) and microcytic anemia (mk). Genetic analysis of the offspring of noninbred animals carrying the transgene and marker loci demonstrates a significant deficiency of homozygous progeny at weaning. When inbred mice heterozygous for the transgene are mated, about one-quarter of their offspring are homozygous; none of these animals survives more than 1 day after birth. It appears likely that a recessive insertional mutation has occurred as a result of transgene integration into a locus required for postnatal viability. We call this mutation transgenic perinatal lethality (Tg.ple).

The human lambda immunoglobulin locus displays a series of restriction fragment length polymorphisms that are readily detected in small populations of normal individuals. Similar polymorphisms appear in populations of wild mice, suggesting that the lambda locus is subject to rapid variation within a single species. Here we show that the polymorphisms seen in the human lambda locus seem to have arisen from unequal meiotic crossing over, altering the number of lambda from as few as six to as many as nine per haploid genome. This expansion and contraction in the number of human lambda genes is significant in that it may affect an individual's capacity to produce variation among lambda light chain genes.

Cultured MPC 11 mouse myeloma cells synthesize not only gamma2b heavy and kappa light chains but also a carboxyl terminal (constant region) fragment of kappa light chain. In vitro translational analysis of total cytoplasmic and microsomal RNA indicates that these cells contain RNA which directs synthesis of both a light chain precursor and a light chain fragment precursor. Variant clones which do not synthesize either heavy or light chains continue to synthesize the light chain fragment. One such "nonproducing" variant was studied in detail. It does not contain translatable mRNA for the intact light chain but does contain RNA which is translated into the light chain fragment precursor. Nucleic acid hybridization analysis with a cDNA probe specific for the constant region of kappa light chains revealed that microsomal RNA from the wild-type cell contains both 14S and a 10S species of kappa specific RNA, whereas the variant contains only the 10S species. Translational analysis of these same RNAs indicates that the 14S species codes for the light chain precursor, while the 10S RNA codes for the light chain fragment precursor.

We have constructed two SV40 recombinants carrying a complete mouse α-globin gene with its presumptive promoter region. In one recombinant the globin gene can be transcribed either from its own promoter or from the adjacent viral late region promoter. In the other efficient globin expression should depend only upon the promoter carried by the chromosomal globin gene. We show that both viruses direct the synthesis of functional globin mRNA in infected monkey kidney cells and that this mRNA has a 5′ terminus indistinguishable from that of authentic globin mRNA. These results suggest that the cloned globin gene contains a functional promoter that is accurately recognized in monkey kidney cells.

A family of germ-line immunoglobulin D-region genes has been cloned and mapped at regular intervals along a 33-kilobase length of human chromosomal DNA. Each member of the family varies slightly in sequence, but precisely conserves the recombinational signals and spacing that flank each gene. This region seems to have been formed by the tandem amplification of large and still well conserved segments of genomic DNA. Further, structural comparisons of germ-line and rearranged D segments suggest that D segments may recombine with each other.

We have derived and mated separate strains of transgenic mice that carry either the v-Ha-ras or the c-myc gene driven by the mouse mammary tumor virus (MMTV) promoter/enhancer. Mice carrying the MMTV/v-Ha-ras transgene manifest two distinct disturbances of cell growth. The first, a benign hyperplasia of the Harderian lacrimal gland, is diffuse, involves the entire gland, and likely requires only the abnormal action of the v-Ha-ras gene. The second involves the focal development of malignancies of mammary, salivary, and lymphoid tissue and likely requires additional somatic events. When the MMTV/v-Ha-ras and MMTV/c-myc strains are crossed to yield hybrid mice, their joint action results in a dramatic and synergistic acceleration of tumor formation. Since these tumors arise stochastically and are apparently monoclonal in origin, additional somatic events appear necessary for their full malignant progression, even in the presence of activated v-Ha-ras and c-myc transgenes.

To gain insight into the role of the limb deformity (ld) gene in limb morphogenesis, we examined the morphologic details of early embryonic limb formation in the mutant ld/ld mouse. Initial morphological differences between wild-type and homozygous ld embryos are apparent during early gestational day 10, a time period during which anteroposterior limb morphogenesis occurs. As a result of a shortened anteroposterior axis, the mutant limb bud appears more pointed than its wild-type counterpart. In addition, the apical ectodermal ridge (AER), a structure crucial to both proximodistal and anteroposterior limb development, fails to differentiate properly in mutant ld embryos. Consistent with these observations, molecular analysis of the limb promordia shows that the limb ectoderm contains a level of ld transcripts fivefold higher relative to its mesenchyme. Furthermore, expression of ld transcripts in other parts of the developing embryo and in primitive streak embryos (gestational day 7) suggests possible roles for this gene in the earliest determinative events of morphogenesis. These data lead us to conclude that ld gene products are required for both proper AER differentiation and anteroposterior pattern formation in limb mesenchyme.

The human immune system offers special advantages for study of the development and evolution of the immune response. A variety of human cell lines are available that are arrested at various stages of development, and human genes provide a convenient evolutionary point of comparison with the already well characterized genes of the mouse. In this paper, we describe the procedure we have used to clone the human kappa chain genes in both germline and rearranged configurations. We have taken advantage of distantly related probes derived from the mouse and nonstringent conditions of hybridization to find the human genes among phage lambda recombinants formed with partially purified genomic restriction fragments of human DNA. In addition to establishing a physical map of the human kappa C and J regions, we have determined the entire sequence of a germline human constant region gene (the Inv3 allele) and two of its J segments, as well as the V/J recombination site of an active human kappa chain gene. For purposes of comparison, we also determined the sequence of the chromosomal mouse constant region gene and its flanking sequences. Although mouse and human sequences have changed extensively during the 70 million years since the two species diverged. significant blocks of homology have been conserved selectively. Some of these have obvious significance in terms of DNA and RNA splicing reactions. By forming heteroduplex structures between mouse and human genes we were able to identify four human J regions that are much more stringently conserved throughout their coding sequences than are the C region genes. In addition, the middle j structure (J3) of the mouse (which is thought to be inactive) appears to be missing from the human genome.

To determine whether RNA splicing plays an obligatory role in gene expression, we have constructed a series of SV40-transducing viruses carrying various combinations of splice junctions derived from the viral genome and a mouse globin gene. All of the viruses that retain at least one functional splice junction, derived from either the viral or the mouse genome, encode stable hybrid RNAs. In contrast, a virus from which all the splice junctions have been removed fails to produce any detectable stable RNA. These results suggest that splicing is a prerequisite for stable RNA formation.

The solution of the problem of how immunoglobulin genes produce antibody molecules is the result of extraordinary developments in the fields of immunology and molecular biology. The immunochemists, of course, discovered the interesting features of the structure of antibody molecules and proposed a variety models to account for the structural and organizational features of this remarkable class of proteins. The molecular biologists, on the other hand, set out to develop the genetic approaches that would—in the end—put these theories to the test. Six or seven years ago neither of these groups could have anticipated the spectacular success that the development of recombinant DNA technology has made possible. Many of the questions raised by immunologists are now answered in concrete terms. We know how immunoglobulin genes are encoded in germline DNA and how this structure is altered in antibody producing cells. We know that several powerful mechanisms exist to shuffle bits of DNA and RNA in somatic cells and we know how these mechanisms act to create diversity. We also suspect that we are viewing mechanisms that have significance beyond the immune system itself.

The mammalian alpha- and beta--globin genes are thought to have evolved from a common ancestral sequence by a duplication event that occurred over 500 million years ago. We have now determined the entire nucleotide sequence of a cloned mouse alpha--globin gene, including regions that flank and interrupt the coding sequence, and have compared this sequence with the sequences of the two mouse beta--globin genes (Konkel, Tilghman and Leder, 1978; Konkel, Maizel and Leder, 1979). Like the two beta genes, the alpha gene is interrupted by two intervening sequences at precisely homologous positions, suggesting that these interruptions were present and have been preserved throughout vertebrate evolution. While the alpha and beta genes conserve considerable (approximately 55%) sequence homology in their coding regions, this homology--with certain interesting exceptions--is lost in the highly divergent flanking and intervening sequences. These exceptions are short preserved sequences positioned in such a way that they might encode signals for transcriptional initiation, poly(A) addition and RNA splicing. Furthermore, a comparison of the recently divererged beta genes and the long separate alpha gene allows us to distinguish two clearly different modes of nucleotide sequence change in evolution: a fast mode which is characterized by drastic sequence alterations involving deletions and insertions, and a slow mode which preserves sequence homology to a large extent and involves mainly point mutations.

We have constructed two strains of bacteriophage λ to meet the requirements of an EK2 host vector system as established by the NIH Advisory Committee on Recombinant DNA Research. Both strains rely on the λgt·λC system devised originally by Davis and his colleagues (1). Here we describe some of the properties of these vectors and their use in cloning a specific Eco Rl fragment of mouse DNA which encodes extensive portions of the mouse ribosomal genes.

A dispersed immunoglobulin pseudogene carries two hallmarks of RNA processing—spliced J and C regions and a poly (A)-rich tail. Its discovery strengthens the notion that processed genes are a significant feature of the mammalian genome and that genetic information can return to the genome via an RNA intermediate.

We have studied immunoglobulin double-isotype expression in a transgenic mouse (TG.SA) in which expression of the endogenous immunoglobulin heavy chain locus is almost completely excluded by a nonallelic rearranged human mu transgene. By flow-cytometric analyses, we have shown that a small, but significant, portion (about 4%) of transgenic spleen cells expresses human mu (transgene) and mouse gamma (endogenous) chains when cultured in vitro with bacterial lipopolysaccharide and interleukin 4. By using amplification of cDNA by the polymerase chain reaction, followed by cloning and sequencing of the amplified cDNA fragment, we have demonstrated expression of trans-mRNA consisting of the transgenic variable and endogenous constant (gamma 1) region sequences. Such trans-mRNA could be produced by either switch recombination or trans-splicing between the transgene and endogenous sterile gamma 1-gene transcripts. These results indicate that trans-splicing might be a possible mechanism for the immunoglobulin double-isotype expression in normal B lymphocytes that have not rearranged the second expressed constant region gene.

Publisher Summary This chapter discusses the organization and evolution of cloned globin genes. The genes responsible for the production of mouse hemoglobin are subject to at least two interesting and different forms of regulation. One form operates in trans and controls the expression of alpha and beta genes located on different chromosomes. The other operates in cis and regulates the expression of individual alpha or beta genes that are located close to one another on the same chromosome. The globin genes represent a complex set of sequences that are expressed in a coordinate fashion during the development of red blood cells. While this gene family can consist of as many as 10–14 members, three of these genes have been cloned and their entire nucleotide sequence has been determined. The globin system, both in mouse and other species, depends upon the development of recombinant DNA technology and rapid nucleotide sequencing techniques. Recombinant DNA technology provides a means of selecting a particular segment of the genome from among a million or so similar segments. This segment can then be amplified and made amenable to detailed structural studies and to genetic manipulation and assay.

A cultured, murine erythroleukemic cell line, which initially contains no detectable hemoglobin, can be induced to synthesize hemoglobin in quantities comparable to those found in normal red blood cells. In order to distinguish between several molecular mechanisms which might explain this induction, radioactive DNA complementary to mouse globin messenger RNA was used as a hybridization probe to measure globin genes and mRNA quantitatively during this form of differentiation. The results indicate that the number of globin genes does not change as these cells accumulate hemoglobin and that there are less than five copies of the globin genes per haploid genome. On the other hand, differentiated cells accumulate, on the average, 7000 to 8000 molecules of globin mRNA per cell, compared with less than ten in each undifferentiated cell. The accumulation of globin mRNA ceases in the presence of actinomycin D, suggesting that it is dependent on de novo RNA synthesis. It is also inhibited by cycloheximide and puromycin, suggesting a requirement for continued protein synthesis. Although a mechanism involving the post-transcriptional stabilization of newly synthesized globin mRNA can not be ruled out, these results are most simply explained on the basis of transcriptional activation of globin genes. Further data suggest that the globin mRNA synthesized in these erythroleukemic cells is relatively stable (chemical half-life more than 10 h) and is indistinguishable from authentic reticulocyte globin mRNA. The amount of globin mRNA, synthesized at the rate of approximately 20 nucleotides per second, has been measured and found comparable to the amount in mouse reticulocytes.

Cloned segments of mouse chromosomal DNA provide direct evidence for the somatic rearrangement of kappa variable and constant region genes in antibody producing cells. This rearrangement apparently affects only one member of an allelic pair of light chain genes.

Abnormal T cell function is a feature of a spectrum of inherited and acquired diseases. We have detected a frequent restriction fragment length polymorphism in the human T cell antigen receptor beta-chain locus that may aid in the analysis of these disorders. A study of a panel of 18 normal individuals, testing for the presence of the polymorphism, showed it to account for 36% of the alleles in that group. In view of the fact that the T cell receptor beta-chain locus has been mapped to chromosome 7, and that the disease ataxia telangiectasia (AT) is associated both with abnormal T cell function and with chromosomal abnormalities of the same region of chromosome 7, we investigated the possibility that the polymorphism could demonstrate linkage of the T cell receptor locus to the gene for that disease. We demonstrated that the mutation causing AT did not lie within the beta-chain locus itself, and that there was preliminary evidence that the two loci were not closely linked. This polymorphism may provide a useful tool for the study of other genetic disorders associated with abnormalities of T cell function, as well as disorders associated with inherited or acquired abnormalities of chromosome 7.

c-myc mRNA is a cell-cycle associated transcript whose induction is regulated by agents that initiate the first phase of a proliferative response. Specifically, c-myc mRNA levels are increased between 10 and 40 fold within one to three hours after incubation of lymphocytes with 1ipopolysaccharide or Concanavalin A or fibroblasts with platelet-derived growth factor (PDGF). This induction of c-myc does not require the synthesis of new protein species and therefore is regulated directly by the biochemical events that immediately follow PDGF binding in fibroblasts and mitogen binding in lymphocytes. In fibroblasts that have been synchronized with respect to the cell-cycle by a short incubation in PDGF, c-myc mRNA levels are maximal in early G0/G1 and diminish to a quiescent level by the time the cells enter the S phase. Thus, c-myc RNA is a labile transcript with a maximal half-life of three hours. Finally, our experiments provide data that places the action of two oncogenes in a hierarchy: c-sis, the putative structural gene for PDGF, regulates the expression of c-myc.

c-myc mRNA is a cell-cycle associated transcript whose induction is regulated by agents that initiate the first phase of a proliferative response. Specifically, c-myc mRNA levels are increased between 10 and 40 fold within one to three hours after incubation of lymphocytes with lipopolysaccharide or Concanavalin A or fibroblasts with plateletderived growth factor (PDGF). This induction of c-myc does not require the synthesis of new protein species and therefore is regulated directly by the biochemical events that immediately follow PDGF binding in fibroblasts and mitogen binding in lymphocytes. Thus, the c-myc oncogene encodes a product which may function as an intracellular mediator of the growth response to mitogens. PDGF is the putative translation product of the sis gene. The product of the v-sis gene, obtained as a supernatant from v-sis transfected NRK cells, induces c-myc expression in quiescent BALB/c 3T3 cells. Also, v-sis transfected BALB/c 3T3 cells display induced levels of c-myc mRNA following prolonged incubation in platelet poor plasma. Thus, oncogenes may be linked in functional hierarchies: the product of one oncogene, sis, regulates the expression of a cellular proto-oncogene, c-myc.

We have cloned two rearranged kappa immunoglobulin genes from the mouse myeloma cell line S107, and find that both are expressed. One gene, designated S107A, encodes the secreted kappa chain that participates in phosphocholine binding and expression of the T-15 idiotype. The other gene, designated S107B, as described here, contains an unusual junction between a V region unrelated to that of S107A and a different J region. The V-J junction preserves the triplet reading frame, but 6 nucleotides have been deleted at the recombination site. Nucleotide sequence analysis of the germline V-region precursor of S107B in comparison with other germline kappa-variable sequences reveals an "extra" 2 nucleotides in S107B between codon 95 and the palindromic heptanucleotide CACAGTG previously implicated in V-J recombination; this difference may be relevant to the 6 nucleotide deletion. Both S107A and S107B genes are expressed in the S107 cell as protein products, but unlike the S107A kappa chain, the S107B protein product is not secreted into the medium. The expression of these two kappa genes in the S107 cell has implications for theories of allelic exclusion.

Hybridization kinetic analyses with synthetic DNA indicate that there are only two to three copies of the κ constant region gene per haploid genome. This result lends weight to the argument that the immunoglobulin light chain is encoded by more than one continuous gene sequence.

Transgenic mice carrying an immunoglobulin enhancer-driven c-myc oncogene develop rapid-onset pre-B cell lymphomas. The incidence of these malignancies is greatly reduced when an additional transgene encoding the membrane-bound form (but not the secreted form) of human Ig/ji is bred into the susceptible strain. This suppressive effect correlates with a subtle alteration in B-cell development induced by the immunoglobulin transgene.

We have devised a sensitive means to assess the anti-tumor effect of cytokines that act via the mobilization of host-mediated defenses. This assay involves transfecting malignant cells to produce a specific cytokine (in this case, IL-4) and measuring the ability of transfectants to form tumors alone and when mixed with a variety of nontransfected tumor cells. In this way, we have identified a potent, non-cell autonomous, anti-tumor effect of IL-4 which is effective against a wide range of tumor cell types in vivo. The effect is reversed by anti-IL-4 antibody, correlates closely with levels of IL-4 production, and is evident in nu/nu mice. The anti-tumor effect seems to be mediated by an inflammatory infiltrate composed of eosinophils and macrophages.

We have purified rabbit globin mRNA using oligo(dT)-cellulose and sucrose gradient centrifugation. Both α- and β-globulin mRNA molecules behave heterogeneously with respect to their elution properties during chromatography on oligo(dT)-cellulose. Those fractions eluted at the lowest ionic strength are most active in directing cell-free globin biosynthesis. By making use of hybridization with synthetic [3H]DNA complementary to globin mRNA, we have shown that this technique can be used to quantitate the extent of mRNA purification. Thus, globin mRNA is approximately 90-fold purified from reticulocyte polysomal RNA and originally constituted slightly more than 1% of the polysomal RNA. Since more than 98% of the globin mRNA sequences are bound to oligo(dT)-cellulose, we suggest that most polysomal globin mRNAs contain a poly (A)-rich region and that this region is not of uniform length nor preponderately associated with either the α- or β-globin mRNAs. In addition, we observe that the 9S globin mRNA most resistant to dissociation from oligo (dT)-cellulose is most active in directing globin biosynthesis.

The human IgE gene encodes a polypeptide chain that is involved in allergic reactions and in the immune response to parasitic disease in man. We have cloned three chromosomal regions corresponding to this sequence and find that two of them derive from curiously duplicated gene segments that also encode IgA constant-region genes. One of the IgE sequences corresponds to the active gene, and its structure defines a complete amino acid sequence of the human IgE constant region. The other cloned segment is a pseudogene from which the first two IgE coding domains have been deleted and replaced by a switch-like sequence that also occurs close to the normal IgE gene. The third IgE segment remains unlinked to the other heavy-chain genes. Evidently, the epsilon-alpha locus has been the site of several complicated genetic rearrangements during recent evolutionary time.

Several DNAase I hypersensitive sites mark the putative regulatory region immediately 5′ of the myc gene. A sequence near at least one site binds to a protein(s) from nuclear extracts in vitro. Three patterns of chromatin structure exist, one associated with the translocated myc allele in a Burkitt lymphoma (BL 31), one associated with the non-translocated (germline) allele in the same Burkitt cell and one associated with the germline myc allele in non-malignant B cells. The non-translocated and transcriptionally silent myc allele in BL 31 shows only one strong hypersensitive site, a site which may mediate negative control over myc. The heavy chain immunoglobulin enhancer that is juxtaposed with myc on the translocated allele in BL 31 may be responsible for the chromatin structure on this deregulated allele. These data have novel implications for the activation of the myc oncogene by translocations.

We have characterized a variant Burkitt lymphoma in which translocation joins the immunoglobulin κ locus on chromosome 2 to the c-myc gene on chromosome 8. This Burkitt lymphoma is especially interesting because, in contrast to the more common lymphomas that carry 8;14 translocations, it carries a translocation that involves a light chain locus and occurs 3′ to and at least 20 kb downstream of the c-myc gene. Furthermore, the c-myc gene from the translocated chromosome is abnormally expressed in that there is a characteristic shift in c-myc promoter utilization and an increase in c-myc transcript. These disturbances could be explained by novel structural alterations that occur in the c-myc gene and include a duplication of a 2.5 kb segment of DNA containing the two c-myc promoters and their untranslated leader exons. Interestingly, these alterations arise at a considerable distance from the translocation breakpoint.

We have created a transgenic mouse strain in which an autosomal transgene bearing elements of the RSV LTR and a translocated c-myc gene obeys very unusual rules. If the transgene is inherited from the male parent, it is expressed in the heart and no other tissue. If it is inherited from the female parent, it is not expressed at all. This pattern of expression correlates precisely with a parentally imprinted methylation state evident in all tissues. Methylation of the transgene is acquired by its passage through the female parent and eliminated during gametogenesis in the male. These observations provide direct molecular evidence that autosomal gene expression can depend upon the sex of the parent from which the gene is inherited. They also provide a plausible mechanism for understanding parental imprinting that may be relevant to the failure of parthenogenesis in mammals, the apparent non-Mendelian behavior of some autosomal genes, and the role of methylation in gene regulation.

The c-Jun and c-fos proto-oncogenes encode proteins that form a complex which regulates transcription from promoters containing AP-1 activation elements. c-Jun has specific DNA binding activity, while c-Fos has homology to the putative DNA binding domain of c-Jun. Following in vitro translation, c-Jun binds as a homodimer to the AP-1 DNA site, while c-Fos fails to dimerize and displays no apparent affinity for the AP-1 element. Cotranslated c-Jun and c-Fos proteins bind 25 times more efficiently to the AP-1 DNA site as a heterodimer than does the c-Jun homodimer. These experiments suggest that in growth factor-stimulated cells c-Jun binds DNA as a dimer with c-Fos as its natural partner. However, overexpression of c-Jun protein in the absence of c-Fos may result in formation of aberrant homodimeric transcription complexes, which could abrogate the normal mechanisms controlling gene expression.

Immunoglobulin kappa light-chain diversity arises, in large part, from an array of germ-line V-region genes that undergo somatic recombination with one of four active J-region segments. The diversity provided by this combinational system is increased by a recombination mechanism that allows variation of crossover points so as to generate additional diversity at a critical region of the light chain. The elaborate mechanism for generating diversity is accompanied not only by considerable waste, in terms of unused V and J regions in a given cell, but also by a range of aberrant recombinants that fail to produce active immunoglobulin genes.

A chromosomal myc gene displays one of three patterns of activity depending upon the arrangement of the gene and its allelic partner. In nonmalignant B cells both myc alleles are normally expressed. In Burkitt lymphoma cells carrying both a translocated and a nontranslocated myc allele, the translocated allele is inappropriately expressed, while the nontranslocated allele is virtually inactive. Here we examine the chromatin structure of these genes using DNAase I hypersensitivity in nonmalignant lymphoblastoid cells and in the Burkitt lymphoma, BL31 . Three hypersensitivity patterns emerge that correlate with the state of the gene and reveal sites associated with putative regulatory structures. One region is associated with the two myc promoters, one with a specific nuclear protein binding site, and one--which is markedly enhanced in the inactive germline gene in the Burkitt cell--with a putative negative control region. The perturbation of the normal pattern in this particular Burkitt cell may be due to the action of an immunoglobulin enhancer.

Erythroid differentiation can be readily induced by butyric acid in cultured erythroleukemic cells in the presence of inhibitors of DNA synthesis and in the absence of cell division. This result appears to rule out more complex models for globin gene expression which require gene replication or cell division (or both).

Antisera to a synthetic c-myc peptide and to c-myc antigens synthesized from various portions of the human gene expressed in Escherichia coli were used in order to characterize the protein product of the human c-myc oncogene. Although the deduced molecular weight of the human c-myc protein is 49,000, these antisera precipitate a protein from human cells that migrates in sodium dodecyl sulfate-polyacrylamide gel as if its molecular weight were 65,000. In addition, the mouse c-myc protein, whether synthesized in cells or in a cell-free system directed by pure, synthetic messenger RNA, has analogous properties and is immunoprecipitated by the antiserum to the human c-myc protein. Similar proteins are immunoprecipitated from monkey, rat, hamster, and frog cells, suggesting evolutionary conservation of antigenic structure of the c-myc protein among vertebrates. In addition, and in a manner consistent with the behavior of its messenger RNA, the immunoprecipitable c-myc protein is sharply induced by the action of mitogens on resting human T cells.

The active gene for a kappa light chain is formed by a somatic recombination event that joins one of several hundred variable region genes to one of a series of recombination sites (J-segments) encoded close to the kappa constant region gene. The nucleotide sequences of cloned germ line and somatically recombined genes define the precise organisation of these genetic segments and the site and nature of the recombination event that joined them. Apart from somatic recombination, no further alteration of ther germ line sequence has occurred. The J-segment is of special interest as it encodes signals for both DNA and RNA splicing and provides a means of generating further immunoglobulin gene diversity.

With oligo(dT)-cellulose as primer, RNA-dependent DNA polymerase catalyzes the synthesis of cellulose-bound DNA that is complementary to mouse globin mRNA. The resulting cellulose-bound or solid phase complementary DNA hybridizes specifically with globin mRNA and permits the recovery of intact globin mRNA. This simple technique for the synthesis of solid phase-bound complementary DNA provides an additional and convenient method for the purification of specific genetic sequences.

Recombinant DNA technology represents a new and powerful tool in basic and applied medical studies. In this respect it is analogous to the discovery of the medical uses of x rays. The concern and debate generated by this technique, coupled with early success in its application, serve as a lesson for those who wish to bring a rational and realistic picture of science, its advantages, and its real risks into public view.

Publisher Summary This chapter analyzes the arrangement, rearrangement, and origin of immunoglobulin genes. The immune system requires sufficient genetic information to encode approximately a million different antibody molecules, each exhibiting remarkable specificity with respect to the antigens that it binds. These antibody molecules comprised of pairs of heavy and light chains display an interesting structural feature wherein their 100 or so N-terminal amino acids differ from antibody chain to antibody chain while their C-terminal portions remain identical in primary amino acid sequence. In a study described in this chapter, the tetrasegmental structure of kappa light chain genes was examined. Immunoglobulin kappa variable region genes constitute a multiple gene family that has arisen through evolution and now consists of many small subfamilies of variable region genes that are closely related both in their structure and flanking sequences. The chapter analyzes the possible role of unequal-crossing in generating immunoglobulin gene diversity. It is illustrated in the chapter that the kappa light chain genes of the mouse are encoded as four separate genetic elements. The first two comprise a large set of split variable region genes encoding a major portion of antibody diversity.

Publisher Summary Recombinant DNA technology has fulfilled its expectations in altering the picture of the mammalian genome. However, it has been suggested that many of the promises offered by the emerging recombinant DNA technology and its safety were unrealistic and, in any case, could be kept by using other unspecified techniques. Major practical goals could be reached using recombinant DNA technology. The production of protein hormones, interferon, and useful fermentation organisms were among the immediate short-term goals. This chapter discusses the way in which a large number of laboratories using these powerful new techniques have changed the way of thinking about the mammalian genome. Evidence has been accumulated indicating that most mammalian genes are encoded in the discontinuous pieces of DNA interrupted by intervening sequences. Studies would provide an enormous insight into the molecular mechanisms that operate to regulate the expression of these genes and assure their orderly modulation.