Amount of RNA secondary structure required to induce an alternative splice.

We set up an alternative splicing system in vitro in which the relative amounts of two spliced RNAs, one containing and the other lacking a particular exon, were directly proportional to the length of an inverted repeat inserted into the flanking introns. We then used the system to measure the effect of intramolecular complementarity on alternative splicing in vivo. We found that an alternative splice was induced in vivo only when the introns contained more than approximately 50 nucleotides of perfect complementarity, that is, only when the secondary structure was much more stable than most if not all possible secondary structures in natural mRNA precursors. We showed further that intron insertions containing long complements to splice sites and a branch point inhibited splicing in vitro but not in vivo. These results raise the possibility that in cells most pre-mRNA secondary structures either are not maintained long enough to influence splicing choices, or never form at all.     

Expression of rat transforming growth factor alpha mRNA during development occurs predominantly in the maternal decidua.

Previous studies have shown that transforming growth factor alpha is expressed during rodent development. To establish the site(s) of transforming growth factor alpha mRNA expression during rat embryogensis, we performed in situ hybridization and Northern blot analyses on samples of embryonic and maternal tissues at various gestational ages. Our results indicate that the high levels of transforming growth factor alpha mRNA that are observed during early development are the result of expression in the maternal decidua and not in the embryo. Decidual expression appears to be induced after implantation, peaks at day 8, and then slowly declines through day 15 at which time the decidua is being resorbed. Expression of transforming growth factor alpha mRNA is highest in that region of the decidua adjacent to the embryo and is low or nondetectable in the uterus, placenta, and other maternal tissues. The developmentally regulated expression of transforming growth factor alpha mRNA in the decidua, together with the presence of epidermal growth factor receptors in this tissue, suggests that transforming growth factor alpha stimulates proliferation locally through an autocrine mechanism. Since epidermal growth factor receptors are present in the embryo and placenta, transforming growth factor alpha produced in the decidua may also act on these tissues through paracrine or endocrine mechanisms.     

Eucaryotic RNA polymerase conditional mutant that rapidly ceases mRNA synthesis.

We have isolated a yeast conditional mutant which rapidly ceases synthesis of mRNA when subjected to the nonpermissive temperature. This mutant (rpb1-1) was constructed by replacing the wild-type chromosomal copy of the gene encoding the largest subunit of RNA polymerase II with one mutagenized in vitro. The rapid cessation of mRNA synthesis in vivo and the lack of RNA polymerase II activity in crude extracts indicate that the mutant possesses a functionally defective, rather than an assembly-defective, RNA polymerase II. The shutdown in mRNA synthesis in the rpb1-1 mutant has pleiotropic effects on the synthesis of other RNAs and on the heat shock response. This mutant provides direct evidence that the RPB1 protein has a functional role in mRNA synthesis.     

Expression of cell type-specific markers during pancreatic development in the mouse: implications for pancreatic cell lineages

Summary The islet cells of the mammalian pancreas are comprised of four different endocrine cell types, each containing a specific hormone. Islet cells also contain two enzymes of the catecholamine biosynthetic pathway: tyrosine hydroxylase (TH) and aromatic L-amino acid decarboxylase (AADC). The cell lineage relationships of these different cell types have not been examined and it is not known whether, during development, they originate from the same or from different precursor populations. In this study we used immunocytochemical procedures to determine whether developing pancreatic cells express markers common to endocrine and exocrine cell types. We found that acinar cell precursors express AADC prior to the appearance of an exocrine marker and that the expression of AADC in acinar cells persists throughout embryogenesis to the first month of postnatal life. At this time, acinar cells do not contain AADC. We also found that exocrine cells containing AADC never express other islet-cell markers. These findings suggest that while acinar and islet cells both arise from precursor cells containing AADC, these progenitor cells do not express a combined endocrine-exocrine phenotype.     

The chicken carbonic anhydrase II gene: evidence for a recent shift in intron position.

The complete nucleotide sequence of the coding region of the chicken carbonic anhydrase II (CA II) gene has been determined from clones isolated from a chicken genomic library. The sequence of a nearly full length chicken CA II cDNA clone has also been obtained. The gene is approximately 17 kilobase pairs (kb) in size and codes for a protein that is comprised of 259 amino acid residues. The 5' flanking region contains consensus sequences commonly associated with eucaryotic genes transcribed by RNA polymerase II. Six introns ranging in size from 0.3 to 10.2 kb interrupt the gene. The number of introns as well as five of the six intron locations are conserved between the chicken and mouse CA II genes. The site of the fourth intron is shifted by 14 base pairs further 3' in the chicken and thus falls between codons 147 and 148 rather than within codon 143 as in the mouse gene. Measurements of CA II RNA levels in various cell types suggest that CA II RNA increases in parallel with globin RNA during erythropoiesis and exists only at low levels, if at all, in non-erythroid cells.     

The alternating conformation of analogues of poly[d(AT)].

Synthetic DNAs were prepared containing 6-methyl adenine (m6A) in place of adenine and 5-ethyl uracil (Et5U) or 5-methoxymethyl uracil (Mm5U) in place of thymine. All three modifications destabilized duplex DNAs to varying degrees. The binding of ethidium was studied to analogues of poly[d(AT)]. There was no evidence of cooperative binding and the "neighbour exclusion rule" was obeyed in all cases although the binding constant to poly[d(m6AT)] was approximately 6 fold higher than to poly[d(AT)]. 31P NMR spectra were recorded in increasing concentrations of CsF. Poly[d(AEt5U)] showed two well-resolved signals separated by 0.55 ppm in 1 M CsF compared to 0.32 ppm for poly[d(AT)] under identical conditions. In contrast, poly[d(AMm5U)] and poly[d(m6AT)] showed two signals separated by 0.28 ppm and 0.15 ppm respectively, only when the concentration of CsF was raised to 2 M. The signals for poly[d(AT)] in 2 M CsF were better resolved and were separated by 0.41 ppm. These results suggest that minor modifications to the bases may have conformational effects which could be recognized by DNA-binding proteins.