Genetic analysis of the tomato golden mosaic virus. II. The product of the AL1 coding sequence is required for replication.

Tomato golden mosaic virus (TGMV) belongs to the geminivirus subgroup that is characterized by a split genome consisting of two single-stranded circular DNAs. The TGMV A genome component encodes the virus coat protein as well as all of the functions necessary for viral DNA replication. Analysis of the nucleotide sequence indicates that the TGMV A component has, in addition to the coat protein encoding ORF, four overlapping open reading frames (ORFs) with the potential to encode proteins of greater than 10 kD. We have investigated the functions of these putative proteins in both symptom formation and DNA replication by creating mutations in each of the ORFs. Our results show that the AL4 ORF, which is encoded within the N-terminal region of ORF AL1, is not essential for normal virus infection. In contrast, we find that disruption of the AL3 ORF results in delay and attenuation of symptom formation. We also report that the products of the AL1 and AL2 ORFs are absolutely required for symptom formation. Studies of DNA replication show that only the AL1 open reading frame is essential for viral DNA synthesis. The significance of these results for the development of vectors from the geminiviruses is discussed.     

The nucleotide sequence of an infectious clone of the geminivirus beet curly top virus

A number of infectious clones of a Californian isolate of the leafhopper-transmitted geminivirus beet curly top virus (BCTV) have been constructed from virus-specific double-stranded DNA isolated from infected Beta vulgaris and used to demonstrate a single component genome. The nucleotide sequence of one infectious clone has been determined (2993 nucleotides). Comparison with other geminiviruses has shown that the organisation of the genome closely resembles DNA 1 of the whitefly-transmitted members. The four conserved coding regions of DNA 1 have highly homologous counterparts in BCTV with the exception of the putative coat protein which is more closely related to those of the leafhopper-transmitted geminiviruses suggesting a strong interrelationship between coat protein and insect vector. A BCTV component equivalent to DNA 2 is not required for virus infection or transmission and has not been isolated from infected plants.     

The nucleotide sequence of cloned wheat dwarf virus DNA

Restriction analysis and cloning of virus-specific double-stranded DNA isolated from plants infected with wheat dwarf virus (WDV) indicated that the virus genome, like that of maize streak virus (MSV), consists of a single DNA circle. The complete nucleotide sequence of cloned WDV DNA (2749 nucleotides) has been determined. Comparison of the potential coding regions in WDV DNA with those in the DNA of two strains of MSV suggests that these viruses encode at least two functional proteins, the coat protein read in the virion (+) DNA sense and a composite protein, formed from two open reading regions, in the complementary (-) DNA sense. Although WDV and MSV are serologically unrelated their coat proteins showed 35% direct amino acid sequence and their DNAs showed 46% nucleotide sequence homology. There was too little homology between the DNAs of WDV and those of two geminiviruses with bipartite genomes, cassava latent virus (CLV) and tomato golden mosaic virus (TGMV), to align the sequences. However comparison of the amino acid sequences of predicted proteins of WDV, MSV, TGMV and CLV revealed clear relationships between these viruses and suggested that the monopartite and the bipartite geminiviruses have a common ancestral origin. Four inverted repeat sequences which have the potential to form hairpin structures of deltaG >/= -14 kcal/mol were detected in WDV DNA. The sequence TAATATTAC present in the loop of one of these hairpins is conserved in similar putative structures in MSV DNA and in both DNA components of CLV and TGMV and may function as a recognition sequence for a protein involved in virus DNA replication.     

Independent encapsidation of tomato golden mosaic virus A component DNA in transgenic plants

Tomato golden mosaic virus (TGMV), a member of the geminivirus group, has a genome consisting of two DNA molecules designated the A and B components. Both are required for infectivity in healthy plants, although the former has been shown to replicate independently in transgenic plants containing tandem direct repeats of the A genome component. In the studies presented here, petunia plants transgenic for either both components (A×B hybrids) or the A component alone were examined for the presence of virus particles and encapsidated, single stranded viral DNA. The results of DNase protection experiments and direct observation of extracts from transgenic plants by electron microscopy indicate that single stranded TGMV DNA is in both cases packaged into paired particles identical to those obtained from virus-infected plants. DNase-treated virions isolated from A×B hybrid petunia are infectious when inoculated onto healthy Nicotiana benthamiana. Likewise, virions obtained from transgenic A petunia are infectious for plants transgenic for the B component.Our observations of TGMV replication in transgenic plants indicate that TGMV A DNA encodes all viral functions necessary for the replication and encapsidation of viral DNA. The possible role of the B component in TGMV replication is discussed.     

A geminivirus replication protein is a sequence-specific DNA binding protein.

The genome of the geminivirus tomato golden mosaic virus (TGMV) consists of two circular DNA molecules designated as components A and B. The A component encodes the only viral protein, AL1, that is required for viral replication. We showed that AL1 interacts specifically with TGMV A and B DNA by using an immunoprecipitation assay for AL1:DNA complex formation. In this assay, a monoclonal antibody against AL1 precipitated AL1:TGMV DNA complexes, whereas an unrelated antibody failed to precipitate the complexes. Competition assays with homologous and heterologous DNAs established the specificity of AL1:DNA binding. AL1 produced by transgenic tobacco plants and by baculovirus-infected insect cells exhibited similar DNA binding activity. The AL1 binding site maps to 52 bp on the left side of the common region, a 235-bp region that is highly conserved between the two TGMV genome components. The AL1:DNA binding site does not include the putative hairpin structure that is conserved in the common regions or the equivalent 5' intergenic regions of all geminiviruses. These studies demonstrate that a geminivirus replication protein is a sequence-specific DNA binding protein, and the studies have important implications for the role of this protein in virus replication.     

DNA sequences essential for replication of the B genome component of tomato golden mosaic virus.

The genome of the geminivirus tomato golden mosaic virus (TGMV) is divided between two DNA components, designated A and B, which differ in sequence except for a 230-nucleotide common region. The A genome component is known to encode viral functions necessary for viral DNA replication, while the B genome component specifies functions necessary for spread of the virus through the infected plant. To identify cis-acting sequences required for viral DNA replication, several mutants were constructed by the introduction of small insertions into TGMV B at selected sites within and just outside the common region. Other mutants had the common region inverted or deleted. All of the mutants were tested for their effects on infectivity and DNA replication in whole plants and leaf discs. Our results indicate that the common region in its correct orientation is required for infectivity and for replication of TGMV B. Furthermore, the conserved hairpin loop sequence located within the TGMV common region and found in all geminiviruses is necessary for DNA replication, and may be part of the viral replication origin.     

Selection for wild type size derivatives of tomato golden mosaic virus during systemic infection.

A chimeric tomato golden mosaic virus (TGMV) A component DNA, which results from replacement of the coding region of the viral coat protein gene (CP) with the larger bacterial beta-glucuronidase coding sequence (GUS), can replicate in agroinoculated leaf discs but is unstable in systemically infected plants (1). We have made similar replacements of the TGMV CP gene with the GUS coding sequence in both the sense and antisense orientations. Both derivatives replicated in leaf discs inoculated via Agrobacterium. However, systemic movement of the GUS substituted vectors was not detected in agroinoculated Nicotiana benthamiana plants. The only TGMV A derivatives detected in systemically infected leaves of inoculated plants were similar in size to the wild type viral component. Sequence analysis of derivatives from six independently inoculated plants revealed that they did not result from internal deletions of the larger replicons detected in leaf discs but, instead, were generated by fusion events occuring within the original T-DNA insert. These results indicate that systemic movement of TGMV in N. benthamiana plants provides a strong selective pressure favoring viral derivatives similar in size to the wild type virus components.     

Stability and expression of bacterial genes in replicating geminivirus vectors in plants.

Bacterial beta-glucuronidase (gus) and neomycin phosphotransferase (neo) genes were introduced into coat protein replacement vectors based on DNA A of tomato golden mosaic virus (TGMV). Recombinant gus and neo vectors up to 1.1 kbp larger than DNA A were shown to replicate stably in transgenic plants containing partial dimers (master copies) of the vectors integrated into their chromosomal DNA in the absence of DNA B. Beta-glucuronidase and neomycin phosphotransferase activities in independently transformed plants were proportional to the copy number of the double-stranded forms of the vector. Deletion analysis has shown that an essential part of the TGMV coat protein promoter, including a TATA box, lies within 76 nt upstream of the initiation codon of the gene. An increase in expression of a neo gene was obtained by replacing this 76 nt sequence by an 800 nt sequence containing a cauliflower mosaic virus 35S RNA promoter with no effect on the ability of the vector to replicate or on its stability in transgenic plants. Systemic infection of plants by agroinoculation with TGMV vectors larger than DNA A in the presence of DNA B resulted in deletions in the vector DNA in some, but not all, plants. Possible reasons for vector instability in systemically infected plants, and vector stability in transgenic plants containing master copies of the vector, are discussed.     

Delimitation of essential genes of cassava latent virus DNA 2

Insertion and deletion mutagenesis of both extended open reading frames (ORFs) of cassava latent virus DNA 2 destroys infectivity. Infectivity is restored by coinoculating constructs that contain single mutations within different ORFs. Although frequent intermolecular recombination produces dominant parental-type virus, mutants can be retained within the virus population indicating that they are competent for replication and suggesting that rescue can occur by complementation of trans acting gene products. By cloning specific fragments into DNA 1 coat protein deletion vectors we have delimited the DNA 2 coding regions and provide substantive evidence that both are essential for virus infection. Although a DNA 2 component is unique to whitefly-transmitted geminiviruses, the results demonstrate that neither coding region is involved solely in insect transmission. The requirement for a bipartite genome for whitefly-transmitted geminiviruses is discussed.     

Expression of Potato Virus Y Coat Protein Gene in Transgenic Potato

Potato virus Y (PVY) N coat protein (CP) coding sequence was cloned into a plant expression vector pMON316 under the CaMV 35S promoter. Leaf discs of potato (Solanum tuberosum) were used to Agrobacterium-mediated gene transfer. A large number of regenerated putative transgenic plants were obtained based on kanamycin resistance. Using total DNA purified from transgenic plants as templates and two oligonucleotides synthesized from 5' and 3' of the PVY coat protein gene as primers, the authors carried out polymerase chain reaction (PCR) to check the presence of this gene and obtained a 0. 8 kb specific DNA fragment after 35 cycles of amplification. Southern blot indicated that the PCR product was indeed PVY CP gene which had been integrated into the potato genome. Enzyme-linked immunosorbent assay (ELISA) of our transgenic plants showed that CP gene was expressed in at least some transgenic potato plants.     

The Nucleotide Sequence and Genome Structure of Mung Bean Yellow Mosaic Geminivirus

Complete nucleotide sequences of the infectious cloned DNA components (DNA 1 and DNA 2) of mung bean yellow mosaic virus (MYMV) were determined. MYMV DNA 1 and DNA 2 consists of 2,723 and 2,675 nucleotides respectively. DNA 1 and DNA 2 have little sequence similarity except for a region of approximately 200 bases which is almost identical in the two molecules. Analysis of open reading frames revealed nine potential coding regions for proteins of mol. wt. > 10,000, six in DNA 1 and three in DNA 2. The nucleotide sequence of MYMV DNA was compared with that of bean golden mosaic virus (BGMV), tomato golden mosaic virus (TGMV) and African cassava mosaic virus (ACMV). The 200-base region common to the two DNAs of each virus had little sequence similarity, except for a highly conserved 33-36 base sequence potentially capable of forming a stable hairpin structure. The potential coding regions in the MYMV DNAs had counterparts in the BGMV, TGMV and ACMV, suggesting an overall similarity in genome organization, except for absence of 1L3 in MYMV DNA 1. The most highly conserved ORFs, MYMV 1R1, BGMV 1R1, TGMV 1R1 and ACMV 1R1, are the putative genes for the coat proteins of MYMV, BGMV, TGMV and ACMV, respectively. MYMV 1L1 has also a high degree of sequence similarity with BGMV 1L1, TGMV 1L1 and ACMV 1L1.     

Expression of Multiple Virus-derived Resistance Determinants in Transgenic Plants Does Not Lead to Additive Resistance Properties

Plants can be protected against infection by potyviruses by expressing different portions of potyviral genomes as transgenes. This strategy has proven effective with several potyvirus genes, including the Nla, Nlb, and coat protein coding regions. Given the effectiveness of separate potyvirus coding regions as determinants of resistance, we tested the hypothesis that combinations of potyvirus coding regions would provide additively greater protection of plants against potyviruses. For this, we compared transgenic plant lines that expressed either the coat protein (CP) or the Nla+Nlb+coat protein (NNC) coding regions from tobacco vein mottling virus (TVMV). We found that plants that carry the NNC gene combination were invariably less resistant to TVMV than were lines that contain a CP gene alone. Additionally, we found that NNC lines displayed virtually no resistance to tobacco etch virus (TEV), in contrast to the CP lines. We conclude that combining more than one virus-derived resistance determinant in a single construct is detrimental to the production of virus-resistant plants.     

苜蓿花叶病毒外壳蛋白基因对红三叶的遗传转化及转基因植株的抗病性分析

以红三叶(Trifolium pratense L．)子叶为转化体,用农杆菌介导法将外源的苜蓿花叶病毒外壳蛋白基因AMV4转入红三叶,经过筛选、分化和再生,得到了具有卡拉霉素抗性的转基因植株。对这些植株进行PCR、Southern印迹杂交和Northern杂交分析,结果表明,外源目的基因已经整合到红三叶基因组中并且得到了表达。对Northern分析呈阳性的植株进行了抗病性检测,结果表明,表达苜蓿花叶病毒外壳蛋白基因的植株病症减轻,发病率、病情指数及病毒积累量都明显低于对照,有的甚至不表现症状,达到了免疫的程度。     

Isolation of pea matrix attachment region and study on its function in transgenic tobaccos

A DNA fragment containing consensus sequence of matrix attachment region (MAR) has been isolated from pea genome. Compared with original DNA sequence, one 115 bp-long repeat sequence is deleted in the obtained DNA sequence. DNA fragments located upstream and downstream of repeat DNA sequence respectively share 84% and 93% homology to the corresponding original sequence, and contain A-box or T-box and TATAA sequence, which is characteristics short sequence of MARs. To test the function of the DNA sequence, the plant expression vectors in which β-glucuronidase gene (GUS, uidA) was used as reporter gene were constructed and transferred into tobaccos via Agrobacterium-mediated transformation procedure. Quantitative GUS assay showed that the average level of uidA expression was increased twofold for the presence of MAR, and the highest level of GUS activity of transgenic plants could be increased six times. The results cited above suggest that the isolated DNA sequence contains consensus sequence of MARs and     

Isolation of pea matrix attachment region and study on its function in transgenic tobaccos

A DNA fragment containing consensus sequence of matrix attachment region (MAR) has been isolated from pea genome. Compared with original DNA sequence, one 115 bp-long repeat sequence is deleted in the obtained DNA sequence. DNA fragments located upstream and downstream of repeat DNA sequence respectively share 84% and 93% homology to the corresponding original sequence, and contain A-box or T-box and TATAA sequence, which is characteristics short sequence of MARs. To test the function of the DNA sequence, the plant expression vectors in which β-glucuronidase gene (GUS, uidA) was used as reporter gene were constructed and transferred into tobaccosvia Agrobacterium- mediated transformation procedure. Quantitative GUS assay showed that the average level of uidA expression was increased twofold for the presence of MAR, and the highest level of GUS activity of transgenic plants could be increased six times. The results cited above suggest that the isolated DNA sequence contains consensus sequence of MARs and has capability to increase expression level of gene in transgenic plants.