No abstract

No abstract

Clones homologous to the 5' region of the Ultrabithorax gene of Drosophda melanogaster have been isolated from D. pseudoobscura, D. funebris and Musca domestica. Regions that encode most of the Ubx protein have been sequenced in all three of these species, and the 5' upstream region has been sequenced in D. funebris to a point -1000 bases upstream of the probable mRNA start site. Here we compare these sequences with those described elsewhere for D. melanogaster. Deduced amino acid sequences of the Ubx protein show 8% (D. pseudoobscura), 15% (D. funebris) and 22% (M. domestica) divergence from D. melanogaster. However, these figures mask very different rates of evolution in different regions of the protein. A glycine-rich ('hinge') region is conserved in each of these species, although its length is variable. Comparison of D. funebris and D. melanogaster sequences in the long 5' untranslated leader region of the mRNA, and in the region immediately upstream of the start point of transcription, reveals tightly conserved elements embedded in an otherwise non-homologous sequence. These conserved elements include a 118-bp region that spans the mRNA start site, an internally repetitive (TAA), region in the untranslated leader and a short repeated motif immediately upstream of the ATG codon that initiates the major open reading frame of the Ubx protein. Two other conserved elements were identified upstream of the transcription start site; both elements have structural features consistent with a role as recognition sites for regulatory proteins.

A chicken 3-tubulin cDNA probe has been used to screen two independently generated human genomic libraries. Of 13 EcoRI fragments detectable in a human genomic Southern blot experiment, 7 correspond in size to EcoRI fragments isolated from recombinant bacteriophage. The location of (8-tubulin-specific regions and the direction of transcription were determined within each cloned fragment. One clone (503) contained a (-tubulin-specific region of6.8 kilobase pairs (kbp) that included three intervening sequences as well as a number of inverted repeat structures. The remaining clones contained (3-tubulin-specific sequences that were close to or, in two cases, substantially less than 1.9 kbp long. Because mature human .0-tubulin mRNA is approximately 1.9 kbp long, these short DNA regions cannot on their own encode a functional (3-tubulin mRNA. Analysis using 3'-and 5'-specific probes derived from the chicken cDNA clone showed the presence ofboth ofthese end regions within one truncated tubulinlike sequence. A second short tubulin-specific region failed to hybridize with a 3'-specific probe. These short sequences are therefore likely to be examples of pseudogenes that have arisen by loss of a portion of DNA essential to the production of functional human f-tubulin mRNA.Microtubules are constituent parts of a diverse variety of eukaryotic subcellular structures. An integral part of the mitotic apparatus, cilia, flagella, and elements ofthe cytoskeleton, they consist principally of two soluble proteins, named a-and ,B-tubulin, each with a molecular weight ofabout 55,000. Work performed in this laboratory (1) has demonstrated that, in chickens, separate mRNAs encode a-and ,B-tubulin; thus,-these proteins are transcribed from distinct genes. Construction of cDNA clones from isolated chicken a-and /&tubulin mRNAs and use of these probes in Southern blotting experiments of whole genomic DNAs (2) suggested the presence of about 4 copies each per genome of a-and /3-tubulin genes in the chicken; the corresponding value in experiments using human DNA is about 14 copies per genome.The expression of a number of tubulin genes might account for the reported appearance of multiple a-and ,B-tubulins isolated from various sources (3-6). Alternatively, such heterogeneity may reflect posttranslational modifications of a protein transcribed from one or a few genes. The extent to which protein heterogeneity is reflected at the genetic level can only be adequately revealed by a direct examination of the structure and organization ofthe genes themselves. In addition, such an analysis is a necessary prerequisite to addressing questions relating to tubulin gene expression. Several lines of evidence-including protein sequence data (7), blot hybridization experiments (2), and the ability of a-and /-tubulins from different species to copolymerize (8)-strongly suggest that tubulins are highly evolutionarily conserved. This paper describes the isolation and characterization of cloned DNA fragments containing 3tubulin-specific sequences from...

The sequence of two human beta-tubulin pseudogenes is described. One contains an intervening sequence but lacks sequences encoding the 55 N-terminal amino acids of the polypeptide chain. A second has no introns but has a polyadenylate signal and an oligoadenylate tract at its 3' end, and it is flanked by a short direct repeat. These sequences have arisen by different mechanisms, including one that probably involves reverse transcription of a processed messenger RNA and reintegration of the complementary DNA copy into the genome.

The ability of a chicken a-tubulin cDNA probe to cross-hybridize with. human DNA under stringent conditions has been exploited to screen two independently constructed human genomic libraries. Nine clones were isolated, accounting for 60% of the bands observed in a whole genomic Southern blot of human DNA. Two clones were selected for further analysis by restriction mapping, orientation experiments using 3'-or 5'-specific probes, and electron microscopy of heteroduplexes. One clone, 2a, contains an a-tubulin-specific region of 5.0 kilobases that includes three intervening sequences. The second clone, 19a, contains an a-tubulin-specific region of 5.4 kilobases and has somewhat diverged 5' and 3' ends. Clone 19a has only two intervening sequences that correspond to the first two in clone 2a. However, these intervening sequences differ in size between clones 2a and 19u and show no detectable sequence homology. The sumof the lengths of sequences in either clone that hybridize to the cDNA probe accounts for essentially the entire length ofthe cDNA molecule.A remarkable feature ofmicrotubules is the diversity of eukaryotic cellular functions with which they are associated. Included in such functions are mitosis, cell motility, intracellular transport, and secretion. The principal components of microtubules are two soluble proteins named a-and f-tubulin, each with a molecular mass of -55,000 daltons and each encoded by a separate gene or genes (1). Several observations suggest that the sequence oftubulin proteins is highly evolutionarily conserved. Hybrid microtubule structures may be formed by in vitro copolymerization of a-and (&tubulins from different species (2), NH2-terminal sequence data indicate considerable homology between sea urchin and chicken tubulins (3), and cDNA probes constructed from chicken a-and (B-tubulin mRNAs are able to cross-hybridize with genomic DNAs from a variety ofeukaryotic species under stringent conditions (4).How many genes encode a-and ,B tubulins, and what factors govern their expression? One approach to this question involves a direct examination of the genes themselves. The existence of multiple tubulins is implied by several reports of electrophoretic microheterogeneity in a-and /-tubulin proteins (5-8) and has been recently confirmed by protein sequence data that show multiple closely related forms of porcine a-tubulin (9). To isolate genomic sequences containing human a-tubulin genes, we have exploited the ability of a chicken a-tubulin cDNA probe to cross-hybridize with corresponding sequences from other eukaryotic species (4). We chose to study human tubulin gene organization because ofthe possibility that defective gene structure or expression might underlie certain disease conditions (10). We previously reported the isolation and characterization of several humanB-tubulin genes (11). This paper reports the structure of two human a-tubulin genes. MATERIALS AND METHODSScreening of Libraries, Restriction Mapping, and Southern Blot Analysis. The partial EcoRI (12) and par...

Three different approaches have been used to study nonsecreting variants of myeloma and hybrid myeloma lines. Hemolytic plaque assays were used to screen nonsecreting and light (L) chain-secreting anti-sheep red blood cell hybrid myeloma variants for reversion to specific antibody or immunoglobulin (Ig) production. No revertants were detected from among 1.5 x lo7-4.2 x lo7 cells of each line studied. In the absence of a heavy (H) chain, the P 3 (MOPC 21) x chain is not secreted but degraded intracellularly. The nature of this secretion defect was investigated in a hybrid myeloma line that produces the P 3 x chain and another L chain which is secreted without an H chain, by biochemical studies of the relative distribution of mRNA activity for the two L chains between membrane-bound and free polysomal fractions. The mutant P 3 H chain (H") produced by the line P3-NSIIII is also degraded intracellularly rather than secreted. This defect was complemented by cell fusion of P 3-NSIUl to P 1, a different Ig-secreting myeloma line. Mutant H * chains were only secreted as part of hybrid Ig molecules with P 1 L chains indicating that the defect in their secretion in the P3-NSIIil line was due to lack of a complementary Ig chain. Additionally, the preferential recovery of P 1 H chain loss variants from these hybrids that are producing two H chains and only one L chain, supports the idea that free intracellular H chains have an effect on cell viability and influence the type of variants recovered from hybrid myeloma lines.

No abstract