De novo shoot organogenesis of Pinus eldarica Medw. in vitro : I. Reproducible regeneration from long-term callus cultures.

Seedling-derived explants of the Afghan pine, Pinus eldarica, were cultured in a triplicate experiment to produce callus that was serially subcultured for up to three years. Callus was removed at various times and induced to regenerate shoots by de novo organogenesis. The shoot regeneration process involved the identification of four discrete developmental steps, each requiring a separate cultural manipulation. In one case a regenerated shoot was induced to root following an auxin pulse treatment. Induction and limited development of buds in callus derived from mature-tree explants was also achieved. This is the first reproducible system for shoot regeneration from long-term callus cultures of a conifer.

De novo shoot organogenesis of Pinus eldarica Medw. in vitro : II. Macro- and micro-photographic evidence of de novo regeneration.

Semi-thin section microscopy was used to evaluate callus formation and subsequent shoot regeneration in the Afghan pine, Pinus eldarica, as correlated to macro-photography of the same processes. Evidence showed the development of unorganized callus required three to six months. Observations over the following two years of culture revealed that regeneration of buds involved induction of subsurface reorganization in the tissues. Buds emerged through the surface of the callus later during development. Evidence indicated regeneration was de novo in origin and proceeded by the mode of shoot organogenesis. A single shoot was adventitiously rooted with continuous vascular connection between shoot and root.

Two domains of the AL1 protein mediate geminivirus origin recognition.

The geminiviruses tomato golden mosaic virus (TGMV) and bean golden mosaic virus (BGMV) have bipartite genomes. Their A and B DNA components contain cis-acting sequences that function as origins of replication, while their A components encode the trans-acting replication proteins--AL1 and AL3. Earlier experiments demonstrated that virus-specific interactions between the cis- and trans-acting functions are required for TGMV and BGMV replication and that the AL1 proteins of the two viruses specifically bind their respective origins. In the current study, characterization of AL1 and AL3 proteins produced from plant expression cassettes in transient replication assays revealed that interaction between AL1 and the origin is responsible for virus-specific replication. The AL3 protein does not contribute to specificity but can be preferred by its cognate AL1 protein when replication is impaired. Analysis of chimeric proteins showed that two regions of AL1 act as specificity determinants during replication. The first domain is located between amino acids 1 and 116 and recognizes the AL1 origin binding site. The second region, which is between amino acids 121 and 209, is not dependent on the known AL1 DNA binding site. Analysis of wild type and chimeric proteins in transient transcription assays showed that AL1 also represses its own promoter in a virus-specific manner. Transcriptional specificity is conferred primarily by AL1 amino acids 1-93 with amino acids 121-209 making a smaller contribution. Together, these results demonstrated that the virus-specific interactions of AL1 during replication and transcription are complex, involving at least two discreet domains of the protein.

Geminivirus replication origins have a modular organization.

Tomato golden mosaic virus (TGMV) and bean golden mosaic virus (BGMV) are closely related geminiviruses with bipartite genomes. The A and B DNA components of each virus have cis-acting sequences necessary for replication, and their A components encode trans-acting factors are required for this process. We showed that virus-specific interactions between the cis- and trans-acting functions are required for TGMV and BGMV replication in tobacco protoplasts. We also demonstrated that, similar to the essential TGMV AL1 replication protein, BGMV AL1 binds specifically to its origin in vitro and that neither TGMV nor BGMV AL1 proteins bind to the heterologous origin. The in vitro AL1 binding specificities of the B components were exchanged by site-directed mutagenesis, but the resulting mutants were not replicated by either A component. These results showed that the high-affinity AL1 binding site is necessary but not sufficient for virus-specific origin activity in vivo. Geminivirus genomes also contain a stem-loop sequence that is required for origin function. A BGMV B mutant with the TGMV stem-loop sequence was replicated by BGMV A, indicating that BGMV AL1 does not discriminate between the two sequences. A BGMV B double mutant, with the TGMV AL1 binding site and stem-loop sequences, was not replicated by either A component, indicating that an additional element in the TGMV origin is required for productive interaction with TGMV AL1. These results suggested that geminivirus replication origins are composed of at least three functional modules: (1) a putative stem-loop structure that is required for replication but does not contribute to virus-specific recognition of the origin, (2) a specific high-affinity binding site for the AL1 protein, and (3) at least one additional element that contributes to specific origin recognition by viral trans-acting factors.

Multiple cis elements contribute to geminivirus origin function.

The genome of the geminivirus tomato golden mosaic virus (TGMV) consists of two circular DNA molecules which are dissimilar in sequence except for a highly conserved 200-bp common region that includes the origin for rolling circle replication. To better characterize the plus-strand origin, we analyzed the capacities of various TGMV common region sequences to support episomal replication in tobacco protoplasts when the viral replication proteins AL1 and AL3 were supplied in trans. These experiments demonstrated that the minimal origin is located in 89-bp common region fragment that includes the known AL1 binding motif and a hairpin structure containing the DNA cleavage site. Analyses of mutant origin sequences identified two additional cis elements--one that is required for origin activity and a second that greatly enhances replication. In contrast, a conserved partial copy of the AL1 binding site did not contribute to origin function. Mutational analysis of the functional AL1 binding site showed that both spacing and sequence of this motif are important for replication in vivo and AL1/DNA binding in vitro. Spacing changes between the AL1 binding site and hairpin also negatively impacted TGMV origin function in a position-dependent manner. Together, these results demonstrated that the organization of TGMV plus-strand origin is complex, involving multiple cis elements that are likely to interact with each other during initiation of replication.