Isolation of an integrated provirus of Moloney murine leukemia virus with long terminal repeats in inverted orientation: integration utilizing two U3 sequences.

We have previously described the construction of a mutant of Moloney murine leukemia virus, in594-2, which carries a 2-base-pair insertion in the U5 region of the genome and is partially defective in forming the integrated proviral DNA. We have now recovered a cloned copy of an unusual provirus from rat cells infected with this mutant. The viral genome is flanked by long terminal repeats in inverted orientation, with U3 sequences joined to cellular DNA at both of the outer edges. In addition, the provirus is a recombinant, containing a segment of a VL30 element in inverted orientation in place of the Moloney murine leukemia virus env region. The recovery of this provirus indicates that two U3 regions can be used for viral integration and suggests that there may be no absolute requirement in the reaction for those U5 sequences outside the 13-base-pair inverted repeats.     

Homologous sequences in the Campoletis sonorensis polydnavirus genome are implicated in replication and nesting of the W segment family.

Polydnaviruses (PDVs) are double-stranded DNA viruses with segmented genomes that replicate only in the oviducts of some species of parasitic wasps and are required for the successful parasitization of lepidopteran insects. PDV DNA segments are integrated in the genomes of their associated wasp hosts, and some are nested; i.e., smaller segments are produced from and largely colinear with larger segments. To determine the internal structure of nested viral segments, the first complete nucleotide sequence of a PDV genome segment and its integration locus was determined. By restriction mapping, Southern blot, and sequence analyses, we demonstrated that the Campoletis sonorensis PDV segment W is integrated into wasp genomic DNA. DNA sequence analysis revealed that proviral segment W terminates in two 1,185-bp direct long terminal repeats (LTRs) in the wasp chromosome, while only one LTR copy is present in the extrachromosomal (viral) W. The results suggest that terminal direct repeats are a general feature of PDV DNA segment integration but that the homology and size of the repeats can vary extensively. Segment W contains 12 imperfect direct repeats of six different types between 89 bp and 1.9 kbp with 65 to 90% homology. The orientation and structure of the repeats suggest that W itself may have arisen through sequence duplication and subsequent divergence. Mapping, hybridization, and sequence analyses of cloned R and M demonstrated that these segments are nested within segment W and that internal imperfect direct repeats of one type are implicated in the homologous intramolecular recombination events that generate segments R and M. Interestingly, segment nesting differentially increases the copy number of genes encoded by segment W, suggesting that the unusual genomic organization of PDVs may be directly linked to the unique functions of this virus in its obligate mutualistic association with parasitic wasps.     

Machinery to support genome segment inversion exists in a herpesvirus which does not naturally contain invertible elements

In many herpesviruses, genome segments flanked by inverted repeats invert during DNA replication. It is not known whether this inversion is a consequence of an inherently recombinagenic replicative mechanism common to all herpesviruses or whether the replication enzymes of viruses with invertible segments have specifically evolved additional enzymatic activities to drive inversion. By artificially inserting a fusion of terminal sequences into the genome of a virus which normally lacks invertible elements (murine cytomegalovirus), we created a genome composed of long and short segments flanked by 1,359- and 543-bp inverted repeats. Analysis of genomic DNA from this virus revealed that inversion of both segments generates equimolar amounts of four isomers during the viral propagation necessary to produce DNA for analysis from a single viral particle. We conclude that a herpesvirus which naturally lacks invertible elements is able to support efficient segment inversion. Thus, the potential to invert is probably inherent in the replication machinery of all herpesviruses, irrespective of genome structure, and therefore genomes with invertible elements could have evolved simply by acquisition of inverted repeats and without concomitant evolution of enzymatic activities to mediate inversion. Furthermore, the recombinagenicity of herpesvirus DNA replication must have some importance independent of genome segment inversion.     

Isolation and characterization of human DNA fragments with nucleotide sequence homologies with the simian virus 40 regulatory region.

A recombinant library of human DNA sequences was screened with a segment of simian virus 40 (SV40) DNA that spans the viral origin of replication. One hundred and fifty phage were isolated that hybridized to this probe. Restriction enzyme and hybridization analyses indicated that these sequences were partially homologous to one another. Direct DNA sequencing of two such SV40-hybridizing segments indicated that this was not a highly conserved family of sequences, but rather a set of DNA fragments that contained repetitive regions of high guanine plus cytosine content. These sequences were not members of the previously described Alu family of repeats and hybridized to SV40 DNA more strongly than do Alu family members. Computer analyses showed that the human DNA segments contained multiple homologies with sequences throughout the SV40 origin region, although sequences on the late side of the viral origin contained the strongest cross-hybridizing sequences. Because of the number and complexity of the matches detected, we could not determine unambiguously which of the many possible heteroduplexes between these DNAs was thermodynamically most favored. No hybridization of these human DNA sequences to any other segment of the SV40 genome was detected. In contrast, the human DNA segments isolated cross-hybridized with many sequences within the human genome. We tested for the presence of several functional domains on two of these human DNA fragments. One SV40-hybridizing fragment, SVCR29, contained a sequence which enhanced the efficiency of thymidine kinase transformation in human cells by approximately 20-fold. This effect was seen in an orientation-independent manner when the sequence was present at the 3' end of the chicken thymidine kinase gene. We propose that this segment of DNA contains a sequence analogous to the 72-base-pair repeats of SV40. The existence of such an "activator" element in cellular DNA raises the possibility that families of these sequences may exist in the mammalian genome.     

Role of specific simian virus 40 sequences in the nuclease-sensitive structure in viral chromatin.

A nuclease-sensitive region forms in chromatin containing a 273-base-pair (bp) segment of simian virus 40 DNA encompassing the viral origin of replication and early and late promoters. We have saturated this region with short deletion mutations and compared the nuclease sensitivity of each mutated segment to that of an unaltered segment elsewhere in the partially duplicated mutant. Although no single DNA segment is required for the formation of a nuclease-sensitive region, a deletion mutation (dl45) which disrupted both exact copies of the 21-bp repeats substantially reduced nuclease sensitivity. Deletion mutations limited to only one copy of the 21-bp repeats had little, if any, effect. A mutant (dl135) lacking all copies of the 21- and 72-bp repeats, while retaining the origin of replication and the TATA box, did not exhibit a nuclease-sensitive region. Mutants which showed reduced nuclease sensitivity had this effect throughout the nuclease-sensitive region, not just at the site of the deletion, indicating that although multiple determinants must be responsible for the nuclease-sensitive chromatin structure they do not function with complete independence. Mutant dl9, which lacks the late portion of the 72-bp segment, showed reduced accessibility to BglI, even though the BglI site is 146 bp away from the site of the deletion.     

The structure of the termini of the DNA of Epstein-Barr virus.

We have studied the DNA of Epstein-Barr virus (EBV) isolated from the B95-8 strain of that virus (Miller and Lipman, 1973). When EBV DNA is partially digested with lambda-exonuclease and allowed to reanneal, up to 50% of the full-length molecules circularize. The arrangements of nucleotide sequences containing the terminal repeats identified in this circularization experiment have been determined. Those fragments of viral DNA generated by digestion with restriction endonucleases which are terminal and contain the terminal repeats have been identified by their sensitivity to digestion of full-length DNA by lambda-exonuclease and by virtue of their being partially homologous to one another. The population of DNA molecules in the B95-8 strain of EBV was found to be nonuniform. The nonuniformity results from different molecules having different numbers of a 0.37 megadalton terminal repeat at each end. About 70% of molecules have four terminal repeats at one end, while four equal classes, each comprising approximately 25% of the population, have one, two, three or four repeats at the other end. The arrangements of nucleotide sequences identified as being terminal in virion DNA were studied in the intracellular circular viral DNA of cells transformed by a single particle on EBV. All fragments produced by digestion with endonucleases and scored as being terminal in virion DNA were absent from intracellular circular DNA. An additional fragment was identified in the digests of intracellular DNA of each transformed clone. The molecular weights of the new fragments equal the sum of the molecular weights of two terminal fragments which are joined upon intracellular circularization of viral DNA.     

Copy number of tandem direct repeats within the inverted repeats of Marek's disease virus DNA

We previously reported that DNA of the oncogenic strain BC-1 of Marek's disease virus serotype 1 (MDV1) contains three units of tandem direct repeats with 132 base pair (bp) repeats within the inverted repeats of the long regions of the MDV1 genome, whereas the attenuated, nononcogenic viral DNA contains multiple units of tandem direct repeats (Maotani et al., 1986). In the present study, the difference in the copy numbers of 132 bp repeats of oncogenic and nononcogenic MDV1 DNAs in other strains of MDV1 was investigated by Southern blot hybridization. The main copy numbers in different oncogenic MDV1 strains differed: those of BC-1, JM and highly oncogenic Md5 were 3, 5 to 12 and 2, respectively. The viral DNA population with two units of repeats was small, but detectable, in cells infected with either the oncogenic BC-1 or JM strain. The MDV1 DNA in various MD cell lines contained either two units or both two and three units of repeats. The significance of the copy number of repeats in oncogenicity of MDV1 is discussed.     

Numerous Length Polymorphisms at Short Tandem Repeats in Human Cytomegalovirus

We show the presence of numerous short tandem repeats in the human cytomegalovirus (HCMV) genome and assess their usefulness as molecular markers. The genome is shown to contain at least 24 microsatellite regions that exhibit length polymorphisms. Insertion-deletion polymorphisms at these short tandem repeats are common (80% of repeats examined are polymorphic among two laboratory strains and 10 clinical isolates). This is the first report of widespread microsatellite length polymorphism in a viral genome. Some regions are highly polymorphic: one was revealed by DNA sequencing to contain length variants at five closely linked sites, which combined resulted in 10 variants for this region among the 12 strains and isolates examined. This study not only provides a new molecular marker system for this virus but also extends our understanding of microsatellite polymorphism in two important ways. First, variable-length repeats in HCMV can be considerably shorter than polymorphic repeats previously found in other organisms. Second, highly variable microsatellite repeats are not confined to prokaryotes and eukaryotes, as previously assumed. This variation provides a useful marker system for distinguishing viral isolates, and similar markers are also likely to be found in other large-genome DNA viruses.     

Preferred crossover sites on polyomavirus DNA.

RmI is a hybrid replicon consisting of polyomavirus (Py) and mouse sequences that yields unit-length polyomavirus DNA via recombination between two directly repeated viral sequences of 182 base pairs (S repeats). To define the contribution of the S repeats in this intramolecular recombination, we derived from RmI a series of replicons containing the original S repeats as well as additional direct viral repeats which were 1 to 2 kilobases in length (L repeats). After mouse 3T6 cells were transfected with these constructs, recombination products that displayed the physical properties of homologous recombinants were detected. The structures of these recombinants indicated that whereas repeat length influences the likelihood of recombination, crossover occurs preferentially near the S repeats, provided that one of them is proximal to the viral origin of replication. This finding suggests that recombination near the S repeats depends on a process initiated near the viral origin of replication.     

Interruption of an alpha-satellite array by a short member of the KpnI family of interspersed, highly repeated monkey DNA sequences.

We describe here the interruption of a cloned African green monkey alpha-satellite array by an 829-base-pair-long nonsatellite DNA segment. Hybridization experiments indicate that the sequences within the interruption are homologous to segments frequently found in the 6-kilobase-pair-long members of the KpnI family of long, interspersed repeats. These data confirm and extend earlier results suggesting that sequences common to the KpnI family can occur independently of one another and in segments of variable lengths. The 829-base-pair-long segment, which is termed KpnI-RET, contains a terminal stretch of adenosine residues preceded by two typical but overlapping polyadenylation sites. KpnI-RET is flanked by direct repeats of a 14-base-pair-long segment of alpha-satellite that occurs only once in the satellite consensus sequence. These structural features suggest that KpnI-RET was inserted into the satellite array as a movable element.     

Amplification of a short nucleotide sequence in the repeat units of defective herpes simplex virus type 1 Angelotti DNA

It has been shown earlier that the reiterated regions TRS and IRS bracketing the Us segment of herpes simplex virus type 1 Angelotti DNA are heterogeneous in size by stepwise insertion of one to six copies of a 550-base-pair nucleotide sequence. Considerably higher amplification of this sequence was observed in defective viral DNA: up to 14 copies were detected to be inserted in the repeat units of a major class of defective herpes simplex virus type 1 Angelotti DNA, dDNA1, which originated from noncontiguous sites located in UL and the inverted repeats of the S component of the parental genome. Physical maps were established for the cleavage sites of KpnI, PstI, XhoI, and BamHI restriction endonucleases on the repeats of dDNA1. The map position of the insertion sequence was determined. It was demonstrated that the amplified inserts were not distributed at random among or within the repeats. A given total population of dDNA1 molecules consisted of different homopolymers, each of which contained a constant number of inserts in all of its repeats. Assuming that a rolling-circle mechanism is involved in the generation of full-length defective herpes simplex virus type 1 Angelotti DNA from single repeat units, these data suggest that the 550-base-pair sequence is amplified in the repeats before the replication process.     

Sequence homologies of diverse length tandem repetitions near ends of vaccinia virus genome suggest unequal crossing over.

The 180,000 base pair (bp), covalently closed, linear duplex DNA genome of vaccinia virus contains a 10,000 bp inverted terminal repetition within which are one set of 13 and one set of 18 tandem 70 bp repeating units. A 967 bp segment containing the innermost 70 bp repeat and an adjacent region notable for a scarcity of restriction endonuclease sites has been sequenced. This was facilitated by the cloning of TaqI and partial TaqI fragments in pBR322. We found that the innermost 70 bp repeat overlaps one of two adjacent 125 bp repeats, following which are eight repeats of 54 bp, parts of 54 bp and 70 bp repeats, and four consecutive 6 to 7 bp repeats. The 70, 125, and 54 bp repeating units have extensive sequence homologies and redundancies that suggest evolution by unequal crossing over. Schemes whereby unequal crossovers of 54 bp repeats lead to a recombinant segment 86% homologous to the 125 bp repeat and unequal crossovers of 125 bp repeats lead to a recombinant segment 94% homologous to the 70 bp repeat were considered. This propensity for sequence divergence should provide a useful marker for comparing the relatedness of poxviruses.     

Optimal lengths for DNAs encapsidated by Epstein-Barr virus.

We measured the efficiency of DNA packaging by Epstein-Barr virus (EBV) as a function of the length of the DNA being packaged. Plasmids that contain oriP (the origin of latent EBV DNA replication), oriLyt (the origin of lytic EBV DNA replication), the viral terminal repeats (necessary for cleavage and packaging by EBV), and various lengths of bacteriophage lambda DNA were introduced into EBV-positive cells. Upon induction of the resident EBV's lytic phase, introduced plasmids replicated as concatemers and were packaged. Plasmid-derived concatemers of DNA with certain lengths were found to predominate in isolated virion particles. We measured the distribution of lengths of plasmid concatemers found within cells supporting the lytic phase of the viral life cycle and found that this distribution differed from the distribution of lengths of concatemers found in mature virion particles. This finding indicates that the DNA packaged into mature virions represents a selected subset of those present in the cell during packaging. These observations together indicate that the length of DNA affects the efficiency with which that DNA is packaged by EBV. Finally, we measured the length of the packaged B95-8 viral DNA and found it to be approximately 165 kbp, or 10 kbp shorter than the originally predicted size for B95-8 based on its sequence. Together with the results of other studies, these findings indicate that the packaging of DNAs by EBV is dependent on two imprecisely recognized elements: the viral terminal repeats and the length of the DNA being packaged by the virus.     

A 210-kb segment of tandem repeats and retroelements located between imprinted subdomains of mouse distal chromosome 7.

Mammalian genes subject to genomic imprinting often form clusters and are regulated by long-range mechanisms. The distal imprinted domain of mouse chromosome 7 is orthologous to the Beckwith-Wiedemann syndrome domain in human chromosome 11p15.5 and contains at least 13 imprinted genes. This domain consists of two subdomains, which are respectively regulated by an imprinting center. We here report the finished-quality sequence of a 0.6-Mb region encompassing the more centromeric subdomain. The sequence contains four imprinted genes (Ascl2/Mash2, Ins2, Igf2 and H19) and reveals previously unidentified CpG islands and tandem repeats, which may be features of imprinted genes. Most interestingly, a unique 210-kb segment consisting almost exclusively of tandem repeats and retroelements is identified. This segment, located between Th and Ins2, has features of heterochromatin-forming DNA and is highly methylated at CpG sites. The segment exhibits asynchronous replication on the parental chromosomes, a feature of the imprinted domains. We propose that this repeat segment could serve either as a boundary between the two subdomains or as a target for epigenetic chromatin modifications that regulate imprinting.     

Nuclease mechanism of the avian retrovirus pp32 endonuclease.

In vivo, the inferred circular retrovirus DNA precursor to the provirus contains two long terminal repeats (LTRs) in tandem. We studied the site-specific nicking of supercoiled DNA that contains tandem copies of avian retrovirus LTR DNA in vitro by using purified avian myeloblastosis virus pp32 endonuclease, Mg2+, and viral DNA substrates containing different LTR circle junction sequences. The results confirmed our previous observation that the pp32 protein generates two nicks, one in either viral DNA strand, each 2 nucleotides from the circle junction site. The specificity of nicking by pp32 was unchanged over an eight-fold range of protein concentration and with different avian retrovirus LTR circle junction substrates. These data are consistent with models which propose a role for the endonuclease in removal of two nucleotides from the LTR termini on integration of viral DNA in vivo.     

Amplification mediated by polyomavirus large T antigen defective in replication.

The polyomavirus large T antigen promotes homologous recombination at high rates when expressed in rat cells carrying the viral replication origin and two repeats of viral DNA sequences stably integrated into the cellular genome. Recombination consists of both reciprocal and nonreciprocal events and is promoted by mutants defective in the initiation of viral DNA synthesis (L. St-Onge, L. Bouchard, and M. Bastin, J. Virol. 67:1788-1795, 1993). We have extended our studies to a rat cell line undergoing amplification of the viral insert. We show that large T antigen promotes amplification independently of its replicative function but that its origin-specific DNA binding activity is not sufficient to promote homologous recombination.     

Analysis by electron microscopy of the variable segment in the maxi-circle of kinetoplast DNA from Trypanosoma brucei

The 20.5-kbp maxi-circle from the kinetoplast DNA of Trypanosoma brucei contains a 5-kbp segment which is not cut by most restriction endonucleases and which varies in size in closely-related trypanosome strains (Borst, P., Fase-Fowler, F., Hoeijmakers, J.H.J. and Frasch, A.C.C. (1980) Biochim. Biophys. Acta 610, 197–210). We have now analysed partial denaturation maps of the linearized maxi-circles by electron microscopy and find that the variable segment is not more AT-rich than the remainder of the maxi-circle. Early denaturation begins at two separate regions of the maxi-circle outside the variable region and one of these corresponds with the position of the gene for the large (12 S) ribosomal RNA. Denaturation-renaturation of maxi-circles leads to the formation of partially mismatched duplexes that look like underwound loops in electron micrographs. These loops are only found in the variable region and they vary in size and appearance. Under our renaturation conditions single-stranded maxi-circle DNA is devoid of secondary structure and this suggests that the underwound loops arise by misalignment of straight tandem repeats in the DNA. We have also analysed heteroduplexes between maxi-circles from two closely related T. brucei strains that differ by 1 kbp in the size of their variable segment. Most molecules had no underwound loops and contained mismatched regions in the variable segment only. The appearance of these regions is diverse, varying from fully duplex with two single-stranded loops to molecules with a heterogeneous array of smaller loops. The total size of single-stranded DNA in the heteroduplexes may be as high as 1.2 μm, i.e., a factor 4 higher than the size difference between the heteroduplex partners. We conclude that the variable region consists of imperfect tandem repeats of a sequence that evolves rapidly. This region might contain the origin of maxi-circle replication.     

Bipartite structure and functional independence of adenovirus type 5 packaging elements.

Selectivity and polarity of adenovirus type 5 DNA packaging are believed to be directed by an interaction of putative packaging factors with the cis-acting adenovirus packaging domain located within the genomic left end (nucleotides 194 to 380). In previous studies, this packaging domain was mutationally dissected into at least seven functional elements called A repeats. These elements, albeit redundant in function, exhibit differences in the ability to support viral packaging, with elements I, II, V, and VI as the most critical repeats. Viral packaging was shown to be sensitive to spatial changes between individual A repeats. To study the importance of spatial constraints in more detail, we performed site-directed mutagenesis of the 21-bp linker regions separating A repeats I and II, as well as A repeats V and VI. The results of our mutational analysis reveal previously unrecognized sequences that are critical for DNA encapsidation in vivo. On the basis of these results, we present a more complex consensus motif for the adenovirus packaging elements which is bipartite in structure. DNA encapsidation is compromised by changes in spacing between the two conserved parts of the consensus motif, rather than between different A repeats. Genetic evidence implicating packaging elements as independent units in viral DNA packaging is derived from the selection of revertants from a packaging-deficient adenovirus: multimerization of packaging repeats is sufficient for the evolution of packaging-competent viruses. Finally, we identify minimally sized segments of the adenovirus packaging domain that can confer viability and packaging activity to viruses carrying gross truncations within their left-end sequences. Coinfection experiments using the revertant as well as the minimal-packaging-domain mutant viruses strengthen existing arguments for the involvement of limiting, trans-acting components in viral DNA packaging.