Cotton leaf curl disease (CLCuD) is a major constraint to cotton production in Pakistan. Infectious clones of the monopartite begomovirus cotton leaf curl virus (CLCuV), associated with diseased cotton, are unable to induce typical symptoms in host plants. We have identified and isolated a single-stranded DNA molecule approximately 1350 nucleotides in length which, when coinoculated with the begomovirus to cotton, induces symptoms typical of CLCuD, including vein swelling, vein darkening, leaf curling, and enations. This molecule (termed DNA beta) requires the begomovirus for replication and encapsidation. The CLCuV/DNA 1/DNA beta complex, together with a similar complex previously identified in Ageratum conyzoides, represent members of an entirely new type of infectious, disease-causing agents. The implications of this finding to our understanding of the evolution of new disease-causing agents are discussed.

DNA beta molecules are symptom-modulating, single-stranded DNA satellites associated with monopartite begomoviruses (family Geminiviridae). Such molecules have thus far been shown to be associated with Ageratum yellow vein virus from Singapore and Cotton leaf curl Multan virus from Pakistan. Here, 26 additional DNA beta molecules, associated with diverse plant species obtained from different geographical locations, were cloned and sequenced. These molecules were shown to be widespread in the Old World, where monopartite begomoviruses are known to occur. Analysis of the sequences revealed a highly conserved organization for DNA beta molecules consisting of a single conserved open reading frame, an adenine-rich region, and a region of high sequence conservation [the satellite conserved region (SCR)]. The SCR contains a potential hairpin structure with the loop sequence TAA/GTATTAC; similar to the origins of replication of geminiviruses and nanoviruses. Two major groups of DNA beta satellites were resolved by phylogenetic analyses. One group originated from hosts within the Malvaceae and the second from a more diverse group of plants within the Solanaceae and Compositae. Within the two clusters, DNA beta molecules showed relatedness based both on host and geographic origin. These findings strongly support coadaptation of DNA beta molecules with their respective helper begomoviruses.

Despite rapidly decreasing costs and innovative technologies, sequencing of angiosperm genomes is not yet undertaken lightly. Generating larger amounts of sequence data more quickly does not address the difficulties of sequencing and assembling complex genomes de novo. The cotton (Gossypium spp.) genomes represent a challenging case. To this end, a coalition of cotton genome scientists has developed a strategy for sequencing the cotton genomes, which will vastly expand opportunities for cotton research and improvement worldwide.

DNA 1 components are satellite-like, single-stranded DNA molecules associated with begomoviruses (family Geminiviridae) that require the satellite molecule DNA beta to induce authentic disease symptoms in some hosts. They have been shown to be present in the begomovirus-DNA beta complexes causing cotton leaf curl disease (CLCuD) and okra leaf curl disease (OLCD) in Pakistan as well as Ageratum yellow vein disease (AYVD) in Singapore. We have cloned and sequenced a further 17 DNA 1 molecules from a diverse range of plant species and geographical origins. The analysis shows that DNA 1 components are associated with the majority of begomovirus-DNA beta complexes, being absent from only two of the complexes examined, both of which have their origins in Far East Asia. The sequences showed a high level of conservation as well as a common organization consisting of a single open reading frame (ORF) in the virion sense, a region of sequence rich in adenine and a predicted hairpin structure. In phylogenetic analyses, there was some evidence of grouping of DNA 1 molecules according to geographic origin, but less evidence for grouping according to host plant origin. The possible origin and function of DNA 1 components are discussed in light of these findings.

For bipartite begomoviruses (family Geminiviridae) trans-replication of the DNA B component by the DNA A-encoded replication-associated protein (Rep) is achieved by virtue of a shared sequence, the "common region", which contains repeated motifs (iterons) which are sequence-specific Rep binding sites and form part of the origin of replication. Recently cotton leaf curl disease (CLCuD), a major constraint to cotton production on the Indian subcontinent, has been shown to be caused by a monopartite begomovirus ( Cotton leaf curl Multan virus [CLCuMV]) and a novel single-stranded DNA satellite molecule termed CLCuD DNA beta. The satellite molecule is trans-replicated by CLCuMV but does not possess the iteron sequences of this virus. We have investigated the ability of CLCuD DNA beta to interact with three further clones of monopartite begomoviruses, isolated from cotton, that have distinct Rep binding specificities. All three cloned viruses were capable of trans-replicating the satellite molecule and inducing CLCuD symptoms in cotton, indicating that the interaction between begomovirus and DNA beta is relaxed in comparison to the interaction between DNA A and DNA B components. Field surveys across all the cotton growing regions of Pakistan indicate that dual and multiple infections are the norm for CLCuD with no evidence of synergism. Despite the diversity of begomoviruses associated with CLCuD, only a single class of DNA beta has been detected, suggesting that this satellite has the capacity to be recruited by unrelated begomoviruses.

Recent reports have suggested that cotton leaf curl virus (CLCuV), a geminivirus of the genus Begomovirus, may be responsible for cotton leaf curl disease in Pakistan. However, the causal agent of the disease remains unclear as CLCuV genomic components resembling begomovirus DNA A are unable to induce typical disease symptoms when reintroduced into plants. All attempts to isolate a genomic component equivalent to begomovirus DNA B have been unsuccessful. Here, we describe the isolation and characterisation of a novel circular single-stranded (ss) DNA associated with naturally infected cotton plants. In addition to a component resembling DNA A, purified geminate particles contain a smaller unrelated ssDNA that we refer to as DNA 1. DNA 1 was cloned from double-stranded replicative form of the viral DNA isolated from infected cotton plants. Blot hybridisation using probes specific for either CLCuV DNA or DNA 1 was used to demonstrate that both DNAs co-infect naturally infected cotton plants from different geographical locations. DNA 1 was detected in viruliferous Bemisia tabaci and in tobacco plants infected under laboratory conditions using B. tabaci, indicating that it is transmitted by whiteflies. Sequence analysis showed that DNA 1 is approximately half the size of CLCuV DNA but shares no homology, indicating that it is not a defective geminivirus component. DNA 1 has some homology to a genomic component of members of Nanoviridae, a family of DNA viruses that are normally transmitted by aphids or planthoppers. DNA 1 encodes a homologue of the nanovirus replication-associated protein (Rep) and has the capacity to autonomously replicate in tobacco. The data suggest that a nanovirus-like DNA has become whitefly-transmissible as a result of its association with a geminivirus and that cotton leaf curl disease may result from a mutually dependent relationship that has developed between members of two distinct DNA virus families that share a similar replication strategy.

Random amplified polymorphic DNA (RAPD) analysis was used to evaluate the genetic diversity of elite commercial cotton varieties. Twenty two varieties belonging to Gossypium hirsutum L. and one to G. arboreum L. were analyzed with 50 random decamer primers using the polymerase chain reaction (PCR). Forty nine primers detected polymorphism in all 23 cotton varieties, while one produced monomorphic amplification profiles. A total of 349 bands were amplified, 89.1% of which were polymorphic. Cluster analysis by the unweighted pair group method of arithmetic means (UPGMA) showed that 17 varieties can be placed in two groups with a similarity ranging from 81.51% to 93.41%. G. hirsutum L. varieties S-12, V3 and MNH-93 showed a similarity of 78.12, 74.46 and 69.56% respectively with rest of the varieties. One variety, CIM-1100, showed 57.02% similarity and was quite distinct. The diploid cotton G. arboreum L. var. Ravi was also very distinct from rest of its tetraploid counterparts and showed only 55.7% similarity. The analysis revealed that the intervarietal genetic relationships of several varieties is related to their center of origin. As expected, most of the varieties have a narrow genetic base. The results obtained can be used for the selection of possible parents to generate a mapping population. The results also reveal the genetic relationship of elite commercial cotton varieties with some standard "Coker" varieties and the diploid G. arboreum L. var. Ravi (old world cotton).

No abstract