High-density hapten labeling and HRP conjugation of oligonucleotides for use as in situ hybridization probes to detect mRNA targets in cells and tissues.

Oligonucleotides that carry a detectable label can be used to probe for mRNA targets in in situ hybridization experiments. Oligonucleotide probes (OPs) have several advantages over cDNA probes and riboprobes. These include the easy synthesis of large quantities of probe, superior penetration of probe into cells and tissues, and the ability to design gene- or allele-specific probes. One significant disadvantage of OPs is poor sensitivity, in part due to the constraints of adding and subsequently detecting multiple labels per oligonucleotide. In this study, we compared OPs labeled with multiple detectable haptens (such as biotin, digoxigenin, or fluorescein) to those directly conjugated with horseradish peroxidase (HRP). We used branching phosphoramidites to add from two to 64 haptens per OP and show that in cells, 16-32 haptens per OP give the best detection sensitivity for mRNA targets. OPs were also made by directly conjugating the same oligonucleotide sequences to HRP. In general, the HRP-conjugated OPs were more sensitive than the multihapten versions of the same sequence. Both probe designs work well both on cells and on formaldehyde-fixed, paraffin-embedded tissues. We also show that a cocktail of OPs further increases sensitivity and that OPs can be designed to detect specific members of a gene family. This work demonstrates that multihapten-labeled and HRP-conjugated OPs are sensitive and specific and can make superior in situ hybridization probes for both research and diagnostic applications.

Identification of differentially expressed mRNAs in human fetal liver across gestation.

Differential gene expression, with its precise start and stop times, is believed to be critical for the programmed development of new cells and tissues. Within the developing fetus, one tissue of particular interest is fetal liver. This organ undergoes rapid changes in the pathway toward liver development in utero since it is also the major site of hematopoiesis, until bone marrow hematopoiesis predominates. Believing that patterns would emerge from the bi-weekly large-scale inspection of expressed genes in the fetal liver, we employed differential display reverse transcription-polymerase chain reaction (DDRT-PCR) as ourprimary inspection tool. Using DDRT-PCR, we isolated cDNAs differentially expressed throughout fetal liver development and in adult liver. We displayed approximately 25 000 cDNAs from 10 and 24 week fetal liver and adult liver. From this initial screen, we determined that approximately 0.1-1% of the mRNA population undergoes expression changes. We extracted, purified and sequenced 25 differentially displayed cDNA bands. Fourteen cDNAs had similarities to known genes, while 11 cDNAs were not similar to any characterized gene. The differentially expressed cDNAs from known genes present in fetal liver include alpha-fetoprotein, stem cell factor, erythroid alpha-spectrin, 2,3-bisphosphoglycerate mutase, insulin-like growth factor-2, porphobilinogen deaminase and Mac30. The differentially expressed cDNAs present in adult liver but not in 10 week fetal liver were nicotinamide deaminase, human fibrinogen-related protein and alpha-acid glycoprotein. The majority of differentially expressed genes found during this effort appear to be turned on during organogenesis, however, some genes were found that are apparently turned off completely.

Analysis of differential display RT-PCR products using fluorescent primers and GENESCAN software.

Differential display reverse transcription PCR (DDRT-PCR) is a procedure used to identify the induction or repression of gene expression. In most DDRT-PCR protocols, radioisotopes are incorporated during PCR and the cDNA products are detected by autoradiography. This report describes the fluorescent labeling of cDNAs and their detection on automated sequencers from PE Applied Biosystems. A fluorescent tag can be incorporated into the PCR product by using either a labeled primer or a labeled dUTP. The fluorescent signals are analyzed by GENESCAN software. Fluorescent DDRT-PCR increases throughput and obviates the handling of hazardous radioisotopes. A PCR cycling profile, expected to give improved reproducibility, is also described. Because amplified cDNAs can't be recovered from the automated sequencer gel, suggestions are given for the identification and recovery of differentially expressed cDNAs.

Intron creation and polyadenylation in maize are directed by AU-rich RNA.

Intron recognition in Angiosperms is hypothesized to require AU-rich motifs within introns. In this report we examined the role of AU-rich motifs in pre-mRNA processing. AU-rich segments of maize introns inserted near the single intron of the maize Bronze-2(Bz2) gene result in alternative splicing. Other insertions of AU-rich sequence in the Bz2 cDNA resulted in de novo intron creation using splice junctions at the edges of the AU-rich region. Surprisingly, the five AU-rich inserts that we tested also caused polyadenylation, even though none had been selected for that function in plants. Insertions of GC-rich sequence into Bz2 did not cause either splicing or polyadenylation. We propose that AU-rich motifs are a general signal for RNA processing in maize and that in the absence of a 5' splice site, polyadenylation is the default pathway.

Addition of A- and U-rich sequence increases the splicing efficiency of a deleted form of a maize intron.

Plant introns are generally short (< 200 nt) and AU-rich, and an elevated AU content is necessary for efficient splicing. Further, an intron in some plant genes enhances gene expression by a post-transcriptional mechanism that results in an increase of cytoplasmic mRNA. The specific intron features responsible for efficient splicing and enhancement are not well characterized in plants. Internal deletions of up to 80% of two maize introns, Adh1 intron 1 and maize actin 3, indicate that large segments of these introns are dispensable for normal function. However, extensive deletion (> 75%) of Adh1 intron 1 diminishes both intron enhancement and splicing efficiency. This finding suggests that there are internal sequence motifs required for intron function, and that these motifs are redundant. We attempted to repair a deletion-impaired Adh1 intron 1 variant by adding back either oligomers of defined sequence content or fragments of maize internal intron sequence. The addition of AU-rich oligomers improved splicing efficiency and in one example, a U-rich oligomer activated a cryptic 3' splice acceptor. We also found that replacing the region proximal to the Adh1 intron 1 3' acceptor with U-rich sequence improved splicing. We found that adding G- and C-rich oligomers did not improve intron function, but a C-rich oligomer activated a cryptic 3' acceptor. The addition of internal intron sequence to an impaired intron improved splicing, and in one case, resulted in the activation of a cryptic 3' acceptor. We present evidence that U-rich sequence immediately upstream of the 3' splice junction increases splicing efficiency and contributes to, but does not uniquely specify, 3' acceptor AG choice.

The impact of AUG start codon context on maize gene expression in vivo.

In mammals, the sequence context surrounding an AUG start codon can alter the efficiency at which translation is initiated. Less is known about the AUG context requirements for translation initiation in plants. Using a maize transient assay, we present evidence that the naturally occurring AUG start codon of the Alcohol dehydrogenase-1 is efficiently used in vivo. We have also tested the effects of upstream, out-of-frame AUGs on the translation of firefly luciferase reporter gene mRNAs. The presence of an upstream out-of-frame AUG, even when surrounded by a 'poor' context, eliminated most luciferase expression, suggesting efficient translation initiation at the upstream AUG. The relaxed requirements for AUG context in maize suggest that plants and mammals may differ in their requirements for efficient translation initiation.

Insertion of non-intron sequence into maize introns interferes with splicing.

Transposable element (TE) insertion into or near plant introns can cause intron skipping and alternative splicing events, resulting in reduced expression. To explore the impact of inserted sequences on splicing, we added non-intron sequence to two maize introns and tested these chimeric introns in a maize transient expression assay. Non-intron sequence inserted into Adh1-S intron 1 and actin intron 3 decreased expression from the luciferase reporter gene; the insertion sites tested were not in intron regions thought to be essential for splicing. Alternatively spliced mRNAs were not observed in transcripts derived from the insertion variants. In contrast, addition of an internal segment of an intron to Adh1-S intron 1 resulted in normal splice site selection and efficient processing. Because the normal intron sequence (including the conserved splice junctions) was retained in all constructs, we hypothesize that added non-intron sequence can interfere with intron recognition and/or splicing.

Intron enhancement of gene expression and the splicing efficiency of introns in maize cells.

The inclusion of the alcohol dehydrogenase 1-S(Adh 1-S) intron 1 in the transcription unit of maize gene constructs has been shown to increase gene expression in cultured maize cells. We have extended these studies with Adh1-S intron 1 using the firefly luciferase, Escherichia coli beta-glucuronidase and chloramphenicol acetyltransferase reporter genes adjoined to different plant promoters and find enhancement of transient gene expression in all cases but one. We also show that the enhancement phenomenon can be mediated by the third intron of the maize actin gene. In all cases tested, the inclusion of an intron results in increased levels of steady-state RNA. The degree of enhancement depends on the exon sequences flanking the intron; flanking exons also influence the efficiency of intron splicing. Unexpectedly, unspliced RNAs accumulate during the transient assay.

Insertion of Mu1 elements in the first intron of the Adh1-S gene of maize results in novel RNA processing events.

Maize transposable elements, when inserted in or near genes, alter expression by several transcriptional and post-transcriptional mechanisms. Three independent, unstable insertions of the transposable element Mutator (Mu) into the first intron of the Alcohol dehydrogenase-1 (Adh1) gene have been shown to decrease expression [Strommer et al. (1982). Nature 300, 542-544]. We have developed an approach to elucidate the underlying molecular mechanisms responsible for the mutant phenotypes. Mu1 elements were inserted into Adh1-S intron 1 in vitro to create plasmid facsimiles of the mutant alleles. The Mu1 element was also inserted at novel positions within intron 1 to create new mutations. The Mu1/intron constructions were placed between the Adh1-S promoter/exon 1 segment and a reporter gene (firefly luciferase or beta-glucuronidase), and these chimeric gene constructs were tested in transient assays in maize protoplasts. When compared with the appropriate control, the Mu1 insertions decreased reporter gene expression to levels approximating the alcohol dehydrogenase enzyme activities observed for the Adh1-S mutants in vivo. The Mu1 insertions also showed a polarity effect with luciferase expression increasing as the insertions were placed nearer the 3' splice junction. In addition, Mu1 insertions within a different intron, actin intron 3, also significantly reduced luciferase expression, indicating that Mu1 insertions within introns are likely to diminish expression in many genes. The presence of the Mu1 sequences was correlated with decreased levels of steady-state luciferase transcript. Deletion analysis of the Mu1 element and RNase mapping indicate that the transposable element contains RNA processing signals in its central region that are largely responsible for the decrease in expression.

Bronze-2 gene of maize: reconstruction of a wild-type allele and analysis of transcription and splicing.

The maize Bronze-2 (Bz2) gene, whose product acts late in the anthocyanin biosynthetic pathway, has been cloned and its transcript has been mapped. We have developed a general procedure for reconstructing wild-type alleles from transposable element-induced mutants. An existing transposon-containing clone, bz2::mu1 [McLaughlin, M., and Walbot, V. (1987). Genetics 117, 771-776], was modified by replacing the region of bz2::mu1 containing the transposon with the corresponding polymerase chain reaction-amplified sequence from the progenitor allele that has no Mu insertion. Particle gun delivery of the reconstructed Bz2 gene to embryonic scutellar tissue lacking a functional Bz2 gene complemented the bz2 mutant phenotype, as demonstrated by the production of purple spots. Having cloned the wild-type allele, we then analyzed the Bz2 transcript, whose features include an 82-nucleotide 5'-untranslated leader, one small intron (78 base pairs) within the coding region, and multiple polyadenylation sites. Four Mutator transposon insertions that eliminate gene function were mapped within the 850-nucleotide transcription unit. We found that variable levels of unspliced Bz2 RNA are present in purple husk tissue; this finding may indicate that the expression of Bz2 is regulated in part at the level of transcript processing.

Developmental and genetic aspects of Mutator excision in maize.

The regulation of excision of Mu elements of the Mutator transposable element family of maize is not well understood. We have used somatic instability of Mu receptor elements from the Bronze 1 and Bronze 2 loci to monitor the frequency and the timing of excision of Mu elements in several tissues. We show that spot size in the aleurone of a bz2::mu1 stock varies between one to approximately 256 cells. This indicates that excision events begin eight divisions prior to full aleurone differentiation and end after the last division of the aleurone. We show that excision is equally biased for late events in all other tissues studied. A locus on chromosome 5 has been identified that affects spot size, possibly by altering the timing of Mu excision. Using somatic excision as an assay of Mutator activity, we found that activity can change in small sectors of the tassel; however, there are no overall activity changes in the tassel during the period of pollen shedding. We also report the recovery of germinal revertants for the bz1::mu1 and bz2::mu1 alleles. One of these revertant alleles was characterized by Southern blot analysis and found to be similar to the progenitor of the mutable allele.

Comparison of Tetrahymena ARS sequence function in the yeasts Saccharomyces cerevisiae and Schizosaccharomyces pombe.

We have isolated several Tetrahymena thermophila chromosomal DNA fragments which function as autonomously replicating sequences (ARS) in the heterologous Saccharomyces cerevisiae and Schizosaccharomyces pombe selection systems. The Tetrahymena ARS sequences were first isolated in S. cerevisiae and were derived from non-ribosomal micro- and macronuclear DNA. Sequence analysis of the ARS elements identified either perfect or close matches with the 11 bp S. cerevisiae ARS core consensus sequence. Subcloning studies of two Tetrahymena ARS elements defined functional regions ranging in size from 50 to 300 bp. Testing of the ARS elements in S. pombe revealed that most of the T. thermophila inserts confer ARS function in both yeasts, at least in the sense of promoting a high transformation frequency to plasmids which contain them. However, the actual sequences responsible for ARS activity were not always identical in the two yeasts.

A restriction fragment length polymorphism in the 5' non-transcribed spacer of the rDNA of Tetrahymena thermophila inbred strains B and C3.

The ribosomal DNA of Tetrahymena thermophila is a 21-kb palindromic molecule which replicates autonomously in the macronuclei of this species. In addition to the rRNA coding regions, there are 5' and 3' flanking sequences which are not transcribed (non-transcribed spacer; NTS). The 5' NTS contains a bidirectional origin of DNA replication and promoter elements which direct transcription. We have identified a restriction fragment length polymorphism in the rDNA 5' NTS by comparing the B and C3 inbred strains of T. thermophila. There is a 42-bp region present in the C3 but not in the B strain rDNA; we present evidence that this difference most likely represents a deletion in the B strain rather than an insertion in the C3 strain. We also include a revised version of the nucleotide sequence of the 5' NTS DNA of inbred strain B.

Sequence of 5S ribosomal RNA gene regions and their products in the archaebacterium Halobacterium volcanii.

We show that the archaebacterium Halobacterium volcanii contains two ribosomal RNA gene clusters, in which genes for individual rRNAs lie in the order 16S-23S-5S. We have cloned the 5S rRNA genes of both clusters and present sequences of the two 5S rRNA genes and their 5' and 3' flanking regions, as well as the sequence of H. volcanii 5S rRNA. We show that a gene for a tRNACys lies downstream from one, but not the other, 5S rRNA gene, and have obtained evidence that this tRNA gene is transcribed in vivo. We discuss regions of potential secondary structure which may be involved in transcription termination. We note regions of unexpected flanking sequence conversation both within H. volcanii 5S rRNA gene regions, and between them and the corresponding 5S rRNA gene region of H. cutirubrum (Hui and Dennis 1984).

Widespread distribution of a 7S RNA in archaebacteria.

An RNA of nonribosomal origin was found in the extreme halophilic bacteria. This novel small RNA was found to be a homogeneous species by RNA fingerprinting. Analysis of the ribonuclease T1 oligonucleotides gave no evidence of the presence of posttranscriptional modifications. Comparisons of electrophoretic mobility with other RNAs of known size suggest that this is a 7S RNA containing 325-375 nucleotides. An RNA of similar mobility was found in all major divisions of the archaebacteria. Insufficient sequence information is available to determine whether these RNAs are homologs of any other known small RNA.

Phylogenetic origins of the plant mitochondrion based on a comparative analysis of 5S ribosomal RNA sequences.

The complete nucleotide sequences of 5S ribosomal RNAs from Rhodocyclus gelatinosa, Rhodobacter sphaeroides, and Pseudomonas cepacia were determined. Comparisons of these 5S RNA sequences show that rather than being phylogenetically related to one another, the two photosynthetic bacterial 5S RNA sequences show that rather than being phylogenetically related to one another, the two photosynthetic bacterial 5S RNAs share more sequence and signature homology with the RNAs of two nonphotosynthetic strains. Rhodobacter sphaeroides is specifically related to Paracoccus denitrificans and Rc. gelatinosa is related to Ps. cepacia. These results support earlier 16S ribosomal RNA studies and add two important groups to the 5S RNA data base. Unique 5S RNA structural features previously found in P. denitrificans are present also in the 5S RNA of Rb. sphaeroides; these provide the basis for subdivisional signatures. The immediate consequence of our obtaining these new sequences is that we are able to clarify the phylogenetic origins of the plant mitochondrion. In particular, we find a close phylogenetic relationship between the plant mitochondria and members of the alpha subdivision of the purple photosynthetic bacteria, namely, Rb. sphaeroides, P. denitrificans, and Rhodospirillum rubrum.

The ribonucleotide sequence of 5s rRNA from two strains of deep-sea barophilic bacteria.

Deep-sea bacteria were isolated from the digestive tract of animals inhabiting depths of 5900 m in the Puerto Rico Trench and 4300 m near the Walvis Ridge. Growth of two bacterial strains was measured in marine broth and in solid media under a range of pressures and temperatures. Both strains were barophilic at 2 degrees C (+/- 1 degrees C) with an optimal growth rate of 0.22 h-1 at a pressure 30% lower than that encountered in situ. At 1 atm they grew at temperatures ranging from 1.2 to 18.2 degrees C (+/- 0.3 degrees C), while in situ pressures increased the upper temperature limit to 23.3 degrees C. Both strains were identified as members of the genus Vibrio, based on standard taxonomic tests and mol% G + C values (47.0 and 47.1). Ribonucleotide sequences determined for 5S ribosomal RNA from each strain confirmed relationship to the Vibrio-Photobacterium group, as represented by V. harveyi and P. phosphoreum, but the barophiles were clearly distinct from these species. Secondary structure conformed to the established model for eubacterial 5S rRNA.

An unusual 5S rRNA, from Sulfolobus acidocaldarius, and its implications for a general 5S rRNA structure.

The nucleotide sequence of the 5S ribosomal RNA of the thermoacidophilic archaebacterium Sulfolobus acidocaldarius was determined. The high degree of evident secondary structure in the molecule has implications for the common higher order structure of other 5S rRNAs, both bacterial and eukaryotic.