Mutations in a peptidylprolyl-cis/trans-isomerase gene lead to a defect in 3'-end formation of a pre-mRNA in Saccharomyces cerevisiae.

In a genetic screen aimed at the identification of trans-acting factors involved in mRNA 3'-end processing of budding yeast, we have previously isolated two temperature-sensitive mutants with an apparent defect in the 3'-end formation of a plasmid-derived pre-mRNA. Surprisingly, both mutants were rescued by the essential gene ESS1/PTF1 that encoded a putative peptidylprolyl-cis/trans-isomerase (PPIase) (Hani, J., Stumpf, G., and Domdey, H. (1995) FEBS Lett. 365, 198-202). Such enzymes, which catalyze the cis/trans-interconversion of peptide bonds N-terminal of prolines, are suggested to play a role in protein folding or trafficking. Here we report that Ptf1p shows PPIase activity in vitro, displaying an unusual substrate specificity for peptides with phosphorylated serine and threonine residues preceding proline. Both mutations were found to result in amino acid substitutions of highly conserved residues within the PPIase domain, causing a marked decrease in PPIase activity of the mutant enzymes. Our results are suggestive of a so far unknown involvement of a PPIase in mRNA 3'-end formation in Saccharomyces cerevisiae.

Inverse splicing of a discontinuous pre-mRNA intron generates a circular exon in a HeLa cell nuclear extract.

We have recently reported the first example of inverse splicing of a eukaryotic pre-mRNA intron using a whole cell extract from the yeast Saccharomyces cerevisiae. The concomitant circularization of the exon in the course of this splicing reaction gave rise to the hypothesis that the circular RNA species, which had been recently discovered in vivo in mammalian cells, were generated by inverse splicing. Here we report the formation of a circular exon in HeLa cell nuclear extracts by an inverse splicing reaction of the second intron of the human beta-globin gene from a pre-mRNA transcript in which the two intron halves flanked an artificially fused, single exon. Our data demonstrate that the mammalian pre-mRNA splicing system has indeed an intrinsic capability of aligning splice sites in reverse order and that this alignment can be followed by a complete splicing reaction, whereby the discontinuous intron sequences are removed. Thus we propose that circular exons in vivo arise as a result of an inverse splicing reaction following the pairing of a 5' splice site with an upstream 3' splice site and that the frequency of this event is influenced by the presence and strength of other, competing splice sites.

Cloning and expression of human G/T mismatch-specific thymine-DNA glycosylase.

Hydrolytic deamination of 5-methylcytosine leads to the formation of G/T mismatches. We have shown previously that these G/T mispairs are corrected to G/C pairs by a mismatch-specific thymine-DNA glycosylase, TDG, which we subsequently purified from human cells. Here we describe the cloning of the human cDNA encoding TDG. We have identified two distinct cDNA species that differ by 100 nucleotides at the 3'-untranslated region. These cDNAs contain a 410-amino acid open reading frame that encodes a 46-kDa polypeptide. The G/T glycosylase, expressed both in vitro and in Escherichia coli, migrated in denaturing polyacrylamide gels with an apparent size of 60 kDa. The substrate specificity of the recombinant protein corresponded to that of the cellular enzyme, and polyclonal antisera raised against the recombinant protein neutralized both activities. We therefore conclude that the cDNA described below encodes human TDG. Data base searches identified a serendipitously cloned mouse cDNA sequence that could be shown to encode the murine TDG homologue. No common amino acid sequence motifs between the G/T-specific enzyme and other DNA glycosylases could be found, suggesting that TDG belongs to a new class of base-excision repair enzymes.

Localization of a 34-amino-acid segment implicated in dimerization of the herpes simplex virus type 1 ICP4 polypeptide by a dimerization trap.

The herpes simplex virus type 1 immediate-early protein ICP4 plays an essential role in the regulation of the expression of all viral genes. It is the major trans activator of early and late genes and also has a negative regulatory effect on immediate-early gene transcription. ICP4 is a sequence-specific DNA-binding protein and has always been purified in a dimeric form. The part of the protein that consists of the entire highly conserved region 2 and of the distal portion of region 1 retains the ability to specifically associate with DNA and to form homodimers in solution. In an attempt to map the dimerization domain of ICP4, we used a dimerization trap assay, in which we screened deletion fragments of this 217-amino-acid stretch for sequences that could confer dimerization properties on a heterologous cellular transcription factor (LFB1), which binds to its cognate DNA sequence only as a dimer. The analysis of these chimeric proteins expressed in vitro ultimately identified a stretch of 34 amino acids (343 to 376) that could still confer DNA-binding activity on the LFB1 reporter protein and thus apparently contained the ICP4 dimerization motif. Consistent with this result, a truncated ICP4 protein containing amino acids 343 to 490, in spite of the complete loss of DNA-binding activity, appeared to retain the capacity to form a heterodimer with a longer ICP4 peptide after coexpression in an in vitro translation system.

Isolation of the murine ribonuclease gene Rib-1: structure and tissue specific expression in pancreas and parotid gland.

The mouse pancreatic ribonuclease gene Rib-1 was isolated from a library of mouse genomic DNA and sequenced. This small gene contains a nontranslated exon of 52 base pairs, an intron of 791 base pairs, and a coding exon of 741 base pairs. Rib-1 transcripts were detected in parotid gland as well as in pancreas. The abundance of the transcripts were approximately 200-fold greater in pancreatic RNA than in parotid RNA. The sites of transcription initiation were mapped by primer extension and ribonuclease protection assays. One major initiation site and several minor initiation sites were identified in pancreatic RNA. Transcription in parotid appears to be initiated from the same sites. Parotid-specific transcripts were not detected. The data suggest that Rib-1 is transcribed in pancreas and parotid from the same promoter. This is in contrast with the mechanism for production of amylase in pancreas and parotid, which is accomplished by tissue specific expression of different gene copies.

Studying DNA modifications and DNA-protein interactions in vivo. A window onto the native genome.

The study of native genomes has been greatly facilitated by the use of direct genomic sequencing and footprinting strategies. This review provides an overview of the techniques involved and presents some highlights of the biological results obtained with these methods.

Mismatch-specific thymine DNA glycosylase and DNA polymerase beta mediate the correction of G.T mispairs in nuclear extracts from human cells.

To avoid the mutagenic effect of spontaneous hydrolytic deamination of 5-methylcytosine, G.T mispairs, arising in DNA as a result of this process, should always be corrected to G.C pairs. We describe here the identification of a DNA glycosylase activity present in nuclear extracts from HeLa cells, which removes the mispaired thymine to generate an apyrimidinic (AP) site opposite the guanine. We further show, using a specific antibody and inhibitors, that the single nucleotide gap, created upon processing of the AP site, is filled in by DNA polymerase beta. This finding substantiates the proposed role of this enzyme in short-patch DNA repair.

Retroviral and pseudogene insertion sites reveal the lineage of human salivary and pancreatic amylase genes from a single gene during primate evolution.

We have analyzed the junction regions of inserted elements within the human amylase gene complex. This complex contains five genes which are expressed at high levels either in the pancreas or in the parotid gland. The proximal 5'-flanking regions of these genes contain two inserted elements. A gamma-actin pseudogene is located at a position 200 base pairs upstream of the first coding exon. All of the amylase genes contain this insert. The subsequent insertion of an endogenous retrovirus interrupted the gamma-actin pseudogene within its 3'-untranslated region. Nucleotide sequence analysis of the inserted elements associated with each of the five human amylase genes has revealed a series of molecular events during the recent history of this gene family. The data indicate that the entire gene family was generated during primate evolution from one ancestral gene copy and that the retroviral insertion activated a cryptic promoter.

In vitro correction of G.T mispairs to G.C pairs in nuclear extracts from human cells.

In differentiated cells, only a specific subset of genes is expressed. Recently, several genes have been shown to be transcriptionally inactivated by methylation of cytosine residues, mainly within their promoter sequences. Spontaneous hydrolytic deamination of 5-methylcytosine to thymine, which has been estimated to generate up to 12 G.T mismatched base pairs in the human genome per day, could have a deleterious effect on the expression of such genes. We recently reported that mammalian cells possess a specific repair pathway, which counteracts the mutagenic effects of this deamination by correcting G.T mismatches almost exclusively to G.C pairs. We show here that, in nuclear extracts from HeLa cells, this repair is mediated by excision of the aberrant thymidine monophosphate residue, followed by gap-filling to generate a G.C pair. We also provide preliminary evidence that the initial step of this process involves a DNA glycosylase.

Expression of the human amylase genes: recent origin of a salivary amylase promoter from an actin pseudogene.

The human genes encoding salivary amylase (AMY1) and pancreatic amylase (AMY2) are nearly identical in structure and sequence. We have used ribonuclease protection studies to identify the functional gene copies in this multigene family. Riboprobes derived from each gene were hybridized to RNA from human pancreas, parotid and liver. The sizes of the protected fragments demonstrated that both pancreatic genes are expressed in pancreas. One of the pancreatic genes, AMY2B, is also transcribed at a low level in liver, but not from the promoter used in pancreas. AMY1 transcripts were detected in parotid, but not in pancreas or liver. Unexpected fragments protected by liver RNA led to the discovery that the 5' regions of the five human amylase genes contain a processed gamma-actin pseudogene. The promoter and start site for transcription of AMY1 are recently derived from the 3' untranslated region of gamma-actin. In addition, insertion of an endogenous retrovirus has interrupted the gamma-actin pseudogene in four of the five amylase genes.

Nuclear pre-mRNA processing in plants: distinct modes of 3'-splice-site selection in plants and animals.

The report that human growth hormone pre-mRNA is not processed in transgenic plant tissues (A. Barta, K. Sommergruber, D. Thompson, K. Hartmuth, M.A. Matzke, and A.J.M. Matzke, Plant Mol. Biol. 6:347-357, 1986) has suggested that differences in mRNA splicing processes exist between plants and animals. To gain more information about the specificity of plant pre-mRNA processing, we have compared the splicing of the soybean leghemoglobin pre-mRNA with that of the human beta-globin pre-mRNA in transfected plant (Orychophragmus violaceus and Nicotiana tabacum) protoplasts and mammalian (HeLa) cells. Of the three introns of leghemoglobin pre-mRNA, only intron 2 was correctly and efficiently processed in HeLa cells. The 5' splice sites of the remaining two introns were faithfully recognized, but correct processing of the 3' sites took place only rarely (intron 1) or not at all (intron 3); cryptic 3' splice sites were used instead. While the first intron in human beta-globin pre-mRNA was not spliced in transfected plant protoplasts, intron 2 processing occurred at a low level, indicating that some mammalian introns can be recognized by the plant intron-splicing machinery. However, excision of intron 2 proved to be incorrect, involving the authentic 5' splice site and a cryptic 3' splice site. Our results indicate that the mechanism of 3'-splice-site selection during intron excision differs between plants and animals. This conclusion is supported by analysis of the 3'-splice-site consensus sequences in animal and plant introns which revealed that polypyrimidine tracts, characteristic of animal introns, are not present in plant pre-mRNAs. It is proposed that an elevated AU content of plant introns is important for their processing.

Concerted evolution of human amylase genes.

Cosmid clones containing 250 kilobases of genomic DNA from the human amylase gene cluster have been isolated. These clones contain seven distinct amylase genes which appear to comprise the complete multigene family. By sequence comparison with the cDNAs, we have identified two pancreatic amylase genes and three salivary amylase genes. Two truncated pseudogenes were also recovered. Intergenic distances of 17 to 22 kilobases separate the amylase gene copies. Within the past 10 million years, duplications, gene conversions, and unequal crossover events have resulted in a very high level of sequence similarity among human amylase gene copies. To identify sequence elements involved in tissue-specific expression and hormonal regulation, the promoter regions of the human amylase genes were sequenced and compared with those of the corresponding mouse genes. The promoters of the human and mouse pancreatic amylase genes are highly homologous between nucleotide -160 and the cap site. Two sequence elements thought to influence pancreas-specific expression of the rodent genes are present in the human genes. In contrast, similarity in the 5' flanking sequences of the salivary amylase genes is limited to several short sequence elements whose positions and orientations differ in the two species. Some of these sequence elements are also associated with other parotid-specific genes and may be involved in their tissue-specific expression. A glucocorticoid response element and a general enhancer element are closely associated in several of the amylase promoters.

Evolution of the amylase multigene family. YBR/Ki mice express a pancreatic amylase gene which is silent in other strains.

The two isozymes of pancreatic amylase in mouse strain YBR/Ki are encoded by closely linked genes which are independently regulated. We have isolated these two pancreatic amylase genes, Amy-2.1 and Amy-2.2, from a cosmid library of YBR/Ki genomic DNA and compared the nucleotide sequences of coding regions with the amino acid sequences of the protein isozymes. Transcripts of both genes were also isolated from a pancreatic cDNA library and partially sequenced. The results demonstrate that Amy-2.1 encodes the A1 isozyme of YBR/Ki pancreatic amylase, while Amy- 2.2 encodes the insulin-dependent B1 isozyme. Similarities of restriction maps and nucleotide sequences suggest that Amy-2.1 is closely related to the active Amy-2a gene previously isolated from strain A/J (Schibler, U., Pittet, A.-C., Young, R. A., Hagenbüchle, O., Tosi, M., Gellman, S., and Wellauer, P. K. (1982) J. Mol. Biol. 155, 247-266). Expression of Amy-2.2 may be limited to strain YBR/Ki. The inactive Amy-X gene from A/J (Schibler, U., Pittet, A.-C., Young, R. A., Hagenbüchle, O., Tosi, M., Gellman, S., and Wellauer, P. K. (1982) J. Mol. Biol. 155, 247-266) is apparently a null allele of Amy-2.2. An additional amylase gene from YBR/Ki has been identified as a pancreatic amylase pseudogene which diverged between sixteen and thirty-two million years ago. The pancreatic amylase subfamily in strain YBR/Ki thus consists of two active genes and one pseudogene. The low rate of amylase production in YBR/Ki pancreas, relative to that of other inbred strains, can be accounted for by the lower number of gene copies in this strain. Comparison of pancreatic amylase genes from different inbred strains provides evidence for several duplication and deletion events during the recent evolution of this chromosome region.

Structure and exon to protein domain relationships of the mouse carbonic anhydrase II gene.

We have isolated a cosmid clone containing the entire mouse (YBR strain) carbonic anhydrase (CA) II gene in 38 kilobase pairs of genomic DNA. The gene was found to be composed of seven exons and six introns. A TATA box (TATAAAA) and a possible CCAAT box (CCACT) have been located beginning 92 and 142 base pairs, respectively, upstream from the initiation codon ATG. When the regions encoded by exons and protein domains are examined, all but 1 of the 30 putative active site residues are encoded by four exons: exons 2 and 3 mainly code for hydrophilic residues and exons 4 and 6 mostly hydrophobic residues. Two intron splice positions, one between the codons for Glu-116 and Leu-117 and the other interrupting the codon for Gly-143, are located at the bottom of the active site cavity, and the former separates two of the three histidine residues forming ligands to the active site zinc ion. The other four splice sites map to the exterior of the molecule. Thus, except for the possible association of the 29 active site residues encoded by four exons, no obvious correspondence is seen between the regions coded by exons and the functional or secondary structural domains of the mouse CA II molecule. During this study, the possible basis for the two electrophoretic types, CA IIa and CA IIb, of inbred mouse strains was detected as a Gln/His interchange at position 38.

A 78-kilobase region of mouse chromosome 3 contains salivary and pancreatic amylase genes and a pseudogene.

Genetic studies have demonstrated that salivary and pancreatic amylase genes are closely linked in human and mouse. To analyze the arrangement of genes within the amylase cluster, a library of YBR mouse genomic DNA was cloned in the cosmid vector pJB8. Clones containing amylase genes were identified by hybridization with amylase cDNA probes. Salivary and pancreatic amylase genes were isolated on separate cosmid clones, but no overlapping clones were evident from the initial screening. A strategy for the rapid isolation of terminal noncoding fragments from the cosmid clones was developed. By using these terminal fragments for chromosome "walking," a map of 78 kilobases of the amylase gene region was constructed. The salivary and pancreatic amylase genes are present within this region in the same 5'-to-3' orientation, separated by 22 kilobases of genomic DNA. A truncated amylase pseudogene is located 10 kilobases downstream from the pancreatic amylase gene.

Isolation and characterization of the chicken cardiac myosin light chain (L-2A) gene. Evidence for two additional N-terminal amino acids.

The contractile proteins of striated muscle are encoded by multigene families and constitute an excellent system to investigate differentiation and developmental control of gene expression. Different forms of myosin light chains are expressed in skeletal muscle as well as in the myocard. To study the gene structure and molecular mechanisms underlying differential gene expression, the structural cardiac myosin light chain 2 (MLC-2A) gene was isolated from a chicken genomic DNA library. Restriction enzyme mapping, electron microscopic analysis, and partial sequencing revealed that the gene coding for the MLC mRNA of 700 nucleotides in length extends over 4.2 kilobases of DNA and is interrupted by 5 introns. Sequence analysis led to the detection of two codons for additional amino acids at the N terminus which were not reported to be present in the mature protein and are presumably removed post-translationally. These two amino acids, methionine and alanine, are coded on two separate exons split by the largest intron of the entire gene. Southern blot analysis of genomic chicken DNA indicates the presence of one MLC-2A gene per haploid chicken genome.

Sequence analysis of the cloned mRNA coding for glyceraldehyde-3-phosphate dehydrogenase from chicken heart muscle.

Using a cloned cDNA (pGAP30) the nucleotide sequence for chicken glyceraldehyde-3-phosphate dehydrogenase mRNA has been determined. The cDNA insert contains 1051 nucleotides representing the amino acid coding sequence, with the exception of 49 NH2-terminal amino acids, and includes the entire 3'-noncoding region. Sequence information on the missing 5' terminus of the mRNA, not represented in the clone pGAP30, was obtained by extension of the cDNA using an 85-nucleotide-long internal fragment as a primer. Thus the sequence of 310 amino acids of chicken glyceraldehyde-3-phosphate dehydrogenase representing 93% of the complete primary structure could be derived. The coding portion exhibits non-random utilization of synonymous codons with a strong bias for codons with G or C at the third position. The non-coding region contains several octanucleotides which are repeated and shows a potentially stable stem-and-loop structure located towards the end of the mRNA. Hypothetical functional implications of the putative secondary structure are discussed.