Search

The human genome is rich in sequences which are structurally related to the 7SL RNA component of the signal recognition particle. The 7SL DNA sequence family consists of four 7SL genes, 500 7SL pseudogenes (which are truncated at one or both ends of the 7SL sequence) and 500,000 Alu sequences. Both 7SL genes and Alu elements are transcribed by RNA polymerase III, and we show here that the internal 7SL promoter lies within the Alu-like part of the 7SL gene. Why then does RNA polymerase III transcribe the few 7SL genes so efficiently, while transcripts from the far more abundant Alu elements are not readily detectable? We find that a human 7SL gene and a synthetic Alu sequence derived from it are expressed 50-100-fold more efficiently in vitro than either a representative Alu element or two 7SL pseudogenes. 5' Deletion and insertion mutants of the 7SL gene demonstrate that, in conjunction with the internal promoter, the first 37 nucleotides upstream from the transcription start site are essential for efficient and accurate initiation in vitro. We suggest that the genomic sequences upstream from most Alu elements and 7SL pseudogenes do not contain this element, and consequently that only a small subset of such sequences can be transcribed in vivo. This may help to explain the homogeneity of the Alu family within each mammalian genome, as well as the species-specific differences between mammalian Alu families.

We found that the genes for human U2 small nuclear RNA (snRNA) are organized as a nearly perfect tandem array of 10 to 20 copies per haploid genome. Although the coding region for the mature form of U2 RNA was only 188 base pairs (bp) long, the basic repeating unit of the tandem array was 6 kilobase pairs in length. Comparison of DNA sequences immediately upstream from human U1 and U2 genes revealed two regions of strong homology: region I (15 bp long) lay upstream of region II (20 bp long) and was separated from it by about the same distance in U1 genes (25 bp) as in U2 genes (21 bp); however, region I and region II were located 174 bp further upstream from the 5' end of the snRNA coding sequence in U1 genes than in U2 genes. Homologs of region II were also found upstream of the snRNA coding region in a mouse U2 gene and two rat U1 genes. Murphy et al. (Cell 29:265-274, 1982) have found that sequences within region II may function as the equivalent of a TATA box for initiation by RNA polymerase II in vitro at a position 183 bp upstream from the 5' end of the human U1 snRNA coding region. In light of the data reported here, this result suggests that region II does indeed play a role in transcription but that its position relative to the actual initiation site can vary.

We report here the isolation and primary structure of two genes encoding rat U3 small nuclear RNA. One of the genes encodes U3B RNA; the other encodes an RNA which is almost identical to U3A RNA. Both genes are expressed after microinjection into the nuclei of Xenopus laevis oocytes and can direct the accumulation of mature U3 RNA as well as longer transcripts which may be the U3 precursors. We have also isolated and sequenced four other regions of the rat genome homologous to U3 RNA. One of these almost certainly represents a second U3B gene; the other three are pseudogenes which appear to have been generated by the reverse flow of genetic information from U3 RNA back into the genome. Using genomic blotting techniques, we show that the rat U3 genes are present in only a few copies per haploid genome and are probably not closely linked to one another.

We present the sequences of five additional human U2 pseudogenes which are very similar to the U2.13 pseudogene reported previously [Van Arsdell et al. (1981) Cell 26, 11-17]. All six U2 pseudogenes preserve the 5' end of the mature U2 snRNA sequence, and all six are flanked by nearly perfect direct repeats that differ in sequence and range in length from 16 to 21 base pairs. The 3' ends of the six U2 pseudogenes are truncated at five different sites between position 33 and 82, and in two cases the 3' end of the pseudogene overlaps the downstream direct repeat by 5 or 6 base pairs. The structure of these six U2 pseudogenes contrasts with that of four human U3 pseudogenes [Bernstein et al. (1983) Cell 32, 461-472] all of which are identically truncated at position 69 or 70, and appear to be derived from a self-primed 74 base reverse transcript of U3 snRNA. Comparison of the U2 and U3 pseudogenes suggests a model for their generation in which the 3' end of the pseudogene is always truncated relative to the initial cDNA template.

Xenopus oocytes injected with human U2 snRNA genes synthesize mature U2 as well as a U2 precursor with about 10 extra 3' nucleotides (human pre-U2 RNA). Formation of the pre-U2 3' end requires a downstream element located between position +16 and +37 in the U2 3'-flanking sequence. The distance between this element and the U2 coding region can be increased without affecting formation of the pre-U2 3' end. When the natural sequence surrounding the pre-U2 3' end is changed, novel 3' ends are still generated within a narrow range upstream from the element. The 3' terminal stem-loop of U2 snRNA is not required for pre-U2 3' end formation. A sequence within the 3' element (GTTTN0-3AAAPuNNAGA) is conserved among snRNA genes transcribed by RNA polymerase II. Our results suggest that the 3' ends of pre-U2 RNA and histone mRNA may be generated by related but distinct RNA processing mechanisms.

Three species of small nuclear RNA from the lower eucaryote Dictyostelium discoideum have been isolated and characterized with regard to size, cellular abundance, modified nucleotide content, and 5'-end structures. Previous studies had shown that the nuclei of mammalian cells contain a number of discrete low molecular weight, nonribosomal, nontransfer RNA molecules known as small nuclear RNAs. The mammalian small nuclear RNAs range in size from approximately 100 to 250 nucleotides and are quite abundant, in some cases approaching ribosomal RNA in number of copies/cell. Some of these molecules have an unusual cap structure at their 5'-ends similar to that found on eucaryotic messenger RNAs, and a number contain a characteristic set of internal modifications as well. Our results indicate that the small nuclear RNAs of Dictyostelium resemble their counterparts in higher eucaryotic cells structurally, but are present in significantly fewer copies/cell. The implications of these findings for small nuclear RNA function are discussed.

We find that both human and rat U3 snRNA can function as self-priming templates for AMV reverse transcriptase in vitro. The 74 base cDNA is primed by the 3' end of intact U3 snRNA, and spans the characteristically truncated 69 or 70 base U3 sequence found in four different human U3 pseudogenes. The ability of human and rat U3 snRNA to self-prime is consistent with a U3 secondary structure model derived by a comparison between rat U3 snRNA and the homologous D2 snRNA from Dictyostelium discoideum. We propose that U3 pseudogenes are generated in vivo by integration of a self-primed cDNA copy of U3 snRNA at new chromosomal sites. We also consider the possibility that the same cDNA mediates gene conversion at the 5' end of bona fide U3 genes where, over the entire region spanned by the U3 cDNA, the two rat U3 sequence variants U3A and U3B are identical.

Multiple copies of a gene that encodes human U1 small nuclear RNA were introduced into mouse C127 cells with bovine papilloma virus as the vector. For some recombinant constructions, the human U1 gene copies were maintained extrachromosomally on the viral episome in an unrearranged fashion. The relative abundance of human and mouse U1 small nuclear RNA varied from one cell line to another, but in some lines human U1 RNA accounted for as much as one-third of the total U1. Regardless of the level of human U1 expression, the total amount of U1 RNA (both mouse and human) in each cell line was nearly the same relative to endogenous mouse 5S or U2 RNA. This result was obtained whether measurements were made of total cellular U1 or of only the U1 in small nuclear ribonucleoprotein particles that could be precipitated with antibody directed against the Sm antigen. The data suggest that the multigene families encoding mammalian U1 RNA are subject to some form of dosage compensation.

We examined the structure of the promoter for the human U2 snRNA gene, a strong RNA polymerase II transcription unit without an obvious TATA box. A set of 5' deletions was constructed and assayed for the ability to direct initiation of U2 snRNA after microinjection into Xenopus oocytes. Sequences between positions -295 and -218 contain an activator element which stimulates accurate initiation by 20- to 50-fold, although as few as 62 base pairs of 5' flanking sequence are sufficient to direct the accurate initiation of U2 RNA. When the activator was recloned in the proper orientation at either of two different upstream locations, the use of the normal U2 start site was stimulated. Inversion of the element destroyed the stimulation of accurate U2 initiation, but initiation at aberrant upstream start sites was enhanced by the element in both orientations. A 4-base-pair deletion that destroyed the activity of the element lies within a sequence (region III) which is highly conserved among U2 genes from different organisms. Mutations in the activator also affected the ability of the U2 template to compete with a wild-type U1 gene in coinjection experiments. We propose that the element enhances the efficiency of transcription in part by facilitating the association of a limiting factor with transcription complexes. Human U1 snRNA genes possess a region homologous to U2 region III, and we suggest that upstream activator elements may be a general feature of snRNA promoters.

We have cloned and partially characterized 24 loci from the human genome which are complementary to U1, U2, or U3, the three major species of small nuclear RNA (snRNA) in HeLa cells. When compared to the known U1 (human) and U2 (rat) snRNA sequences, the DNA sequences we report here for the complementary regions from two of the clones, U1.11 and U2.7, reveal the presence of truncated and divergent gene copies. Furthermore, most if not all of the 24 cloned loci contain gene copies that are significantly divergent from the homologous HeLa snRNA species because DNA from every recombinant phage except U1.7 and U1.15 proved unable to form snRNA.DNA hybrids which protect full-length HeLa snRNA from ild digestion with ribonuclease T1. Hence, we refer to these loci as snRNA pseudogenes. In both clones U1.11 and U2.7, an element of the dominant middle repetitive DNA sequence family in the human genome, the Alu family, is located upstream from the snRNA pseudogene and in the same orientation. Alu elements in the same location and orientation relative to bona fide genes have previously been found in the human beta-globin gene cluster [Duncan, C. H., Biro, P. A., Choudary, P. V., Elder, J. T., Wang, R. C., Forget, G. B., deRiel, J. K. & Weissman, S. M. (1979) Proc. Natl. Acad. Sci. USA 76, 5095-5099]. We discuss the significance of these findings in relation to the nature of snRNA multigene families and other reported examples of pseudogenes.

We previously demonstrated that pseudogenes complementary to the small nuclear RNAs U1, U2 and U3 are dispersed and abundant in the human genome. Here we report that three pseudogenes (U1.101, U2.13 and U3.5) are flanked by perfect short direct repeats of 16 to 19 base pairs. In all three pseudogenes. the upstream direct repeat abuts a DNA sequence corresponding to the 5' end of the mature snRNA; in U2.13 and U3.5, the downstream direct repeat immediately follows the truncated 3' end of the snRNA sequence, whereas in U1.101, the downstream direct repeat is separated from the 3, end of the full-length snRNA sequence by a short A-rich region. We consider the direct repeats to be an indication that these three pseudogenes were created by insertion of snRNA information into a new chromosomal locus. To explain why the upstream repeat abuts a DNA sequence complementary to the 5' end of the mature snRNA, we propose a model for insertion that uses a reverse transcript of the snRNA as an intermediate. Furthermore, we note similarities between the structure of all three pseudogene loci and the Alu family of middle repetitive DNA sequences. These similarities suggest that some Alu family sequences are mobile genetic elements that can transpose to new chromosomal loci using as an intermediate a cDNA copy of an RNA transcribed from the Alu family element by RNA polymerase III.

Evidence is presented that a homogeneous cytoplasmic species known as 7S RNA is the only abundant RNA in uninfected HeLa cells which can form strong hybrids with the dominant family of middle repetitive DNA sequences in the human genome. These DNA sequences are known collectively as the Alu family, because most of them share a common Alu I restriction site. When purified 7S RNA was hybridized to three different genomic clones containing Alu family DNA sequences, a specific region (or regions) comprising at most half the RNA sequence was protected from mild digestion with T1 ribonuclease; moreover, the hybrids between 7S RNA and cloned Alu family DNA wer imperfect, since T1 RNAase was able to nick the protected 7S RNA sequences under conditions where a true RNA: DNA duplex would have been resistant. This suggests that 7S RNA is encoded either by a small subset of the 300,000 Alu family sequences in the human genome or by an entirely different family of genes. The sequence of 7S RNA has been highly conserved through recent evolution, and in both avian and murine cells the RNA is selectively incorporated into oncornavirus particles during productive infection. The cellular function of 7S RNA is unknown.

U2 RNA is one of the abundant, highly conserved species of small nuclear RNA (snRNA) molecules implicated in RNA processing. As is typical of mammalian snRNAs, human U1 and U2 are each encoded by a multigene family. In the human genome, defective copies of the genes (pseudogenes) far outnumber the authentic genes. The majority or all of the 35 to 100 bona fide U1 genes have at least 20 kilobases (kb) of nearly perfect 5' and 3' flanking homology in common with each other; these U1 genes are clustered loosely in chromosome band 1p36 (refs 5, 7) with intergenic distances exceeding 44 kb. In contrast, the 10 to 20 U2 genes are clustered tightly in a virtually perfect tandem array which has a strict 6-kb repeating unit. We report here the assignment, by in situ hybridization, of the U2 gene cluster to chromosome 17, bands q21-q22. Surprisingly, this region is one of three major adenovirus 12 modification sites which undergo chromosome decondensation ('uncoiling') in permissive human cells infected by highly oncogenic strains of adenovirus. The two other major modification sites, 1p36 and 1q21, coincide with the locations of U1 genes and class I U1 pseudogenes, respectively. We suggest that snRNA genes are the major targets of viral chromosome modification.

Of the several hundred 7SL RNA-like sequences that are dispersed in human DNA, no more than four are likely to represent genes for 7SL RNA; the majority are 7SL pseudogenes which appear to result from the reverse flow of genetic information from 7SL RNA back into genomic DNA. We present the sequence of five 7SL pseudogenes displaying an unprecedented diversity of structures. All are truncated copies of 7SL RNA, but the site of truncation can occur at either the 5' end, the 3' end or at both ends of the RNA sequence. We suggest that such diverse 7SL pseudogenes are generated by different but related pathways. In particular, we argue that two of the loci are secondary 7SL pseudogenes which derive from RNA polymerase III transcripts of primary (preexisting) 7SL pseudogenes. We also report the isolation and characterisation of a human genomic clone carrying two linked 7SL RNA coding regions, 7L30.1 and 7L30.2. The 7L30.2 locus differs by several single base changes from the known human 7SL RNA sequences and does not appear to be expressed at a detectable level in HeLa cells. The 7L30.1 locus is an authentic 7SL RNA gene encoding one of the three sequence variants of human 7SL RNA.