The genus Cuscuta (dodder) is composed of parasitic plants, some species of …

• Abstract
• Introduction
• Materials and methods
• Results
• Discussion
• References

Biology Articles » Botany » Plant Taxonomy » Plastid genome structure and loss of photosynthetic ability in the parasitic genus Cuscuta » Results

Results

- Plastid genome structure and loss of photosynthetic ability in the parasitic genus Cuscuta

A molecular phylogeny for 15 species of Cuscuta
To provide maximum phylogenetic resolution, a molecular phylogeny of 15 Cuscuta species was constructed using the internal transcribed spacer between 18S and 26S nuclear rDNA, the intergenic spacer between the chloroplastic trnL 3' exon and trnF, the intron of the chloroplast trnL gene, and a partial coding sequence of the chloroplast rbcL gene. These combined nuclear and chloroplastic regions represented 2758 positions, of which 44% were phylogenetically informative and allowed a strongly supported molecular phylogeny with high boot strap values for each node to be produced (Fig. 1A). The only exception was a weakly supported clade proposed to include C. monogyna, C. reflexa, C. lupuliformis, otherwise, bootstrap values for each clade ranged from 97% to 100%. This analysis allowed phylogenetic relationships amongst these dodders to be proposed (Fig. 1B), and indicated that the species of Cuscuta included in this analysis can be partitioned into four major clades. The most basal clade containing C. lupuliformis, C. reflexa, and C. monogyna has undergone the fewest speciation events (Fig. 1B) and the group containing C. australis, C. epithymum, C. indecora, and C. epilinum has been subject to the greatest extent of evolutionary change (Fig. 1B). This sequence data has all been deposited in Genbank. 

Inferred variation in amino acids within the large subunit of Rubsico
The sequence information collected for each rbcL coding region was used to infer the amino acid sequence of part of the large subunit of Rubisco for each species. This allowed conservation of the large subunit of Rubisco within the 15 species of Cuscuta to be assessed. The first amino acid presented in this study corresponds to the 10th amino acid of the tobacco large subunit of Rubisco (Shinozaki and Suguira, 1982) and the last the 454th amino acid. The amino acid sequence of the Rubisco large subunit in each Cuscuta species was, in general, highly conserved: alterations to amino acid sequence of the active site of Rubisco were not detected (data not shown). The number of alterations in amino acids from the 15 species ranged between 3 and 33 within the region of the rbcL gene that was sequenced. Interestingly, rbcL in the 15 Cuscuta species is becoming more AT-rich than rbcL of tobacco and Convolvulus: AT content was, on average, 58% in the Cuscuta species compared with 55% in Convolvulus arvensis.

Estimated size of plastid genomes and gene content within the dodders
Fourteen BamHI fragments from the tobacco chloroplast genome (Suguira et al., 1986) were used to probe total cellular DNA from five species of Cuscuta (Fig. 2). Three species (C. australis, C. epilinum, and C. epithymum) were used from the most rapidly evolving clade, one from the most basal clade (C. reflexa) and C. pentagona from an intermediate clade. Because C. arvensis is more closely related to the dodders than is tobacco, C. arvensis was also used as a control. This DNA blot analysis allowed the size of sections of the plastid genome in C. arvensis and five species of Cuscuta to be estimated (Table 1) and indicated that there were differences in plastid genome structure in each species (Fig. 2). By combining all of this information, the total plastome size, and the relative sizes of the LSC, SSC, and IRs were estimated (Fig. 3). All visible bands on the blots were included in this analysis, and because some of the weaker bands may represent plastid DNA that has transferred to the nucleus, this approach estimates the largest likely size of each plastid genome. The plastome of C. arvensis appears to be larger than that of tobacco (186 kbp compared with 156 kbp), but all five species of Cuscuta have smaller plastid genomes than both C. arvensis and tobacco. C. reflexa appears to possess the largest plastid genome (65% the size of C. arvensis and estimated to be 122 kbp), and C. australis the smallest (43% the size of C. arvensis and estimated to be 80 kbp). Plastid genome size varied significantly (from 43% to 57% that of C. arvensis) even within the three species (C. australis, C. epilinum, and C. epithymum) from the fastest evolving clade (Fig. 3).

   
Similar regions of the LSC, SSC, and IRs appear to have been lost from the plastome in the different Cuscuta species. For example, the size of the LSC from these five Cuscuta species was either reduced by 32% or 49%. In addition, the size of the IRs was reduced by 20%, 50%, or 69% compared with C. arvensis, and the SSC was reduced to 5 kbp in C. reflexa and C. epilinum, and 14 kbp in C. pentagona, C. epithymum, and C. australis (Fig. 3). The limited variation in the size of each region of the plastid genomes from Cuscuta does not appear to relate to phylogenetic position determined from sequencing. For example, C. epilinum, C. epithymum, and C. australis are all positioned within the same clade, and yet they possess all the sizes of the LSC, SSC, and IRs that were detected.

The gene content of the Cuscuta plastid genomes was investigated by generating probes for 101 genes from the tobacco plastid genome, and immobilizing these on nylon membranes prior to challenging them with total cellular DNA isolated from C. reflexa. Hybridization to all 101 probes was apparent, indicating the presence of homologous DNA fragments in C. reflexa to each gene from tobacco (Fig. 4) although there was variation in the extent to which hybridization occurred. For example, there was very clear hybridization between atpF, psaA, rpl23, rbcL, and accD from tobacco and total DNA from C. reflexa, whereas there was low hybridization to rps16, psbK, psbF, and psaC (Fig. 4). However, in all cases, there was more hybridization between the tobacco plastid genes and C. reflexa DNA than there was to the gfp gene which was used as a negative control (Fig. 4).

 
To investigate whether genes that are normally located on plastid DNA have transferred to, and are transcribed in the nucleus of these Cuscuta species, RNA was extracted and cDNA made using poly dT primers in the presence of radiolabelled dATP. This radiolabelled cDNA from C. arvensis (as a control), C. reflexa, and C. australis was incubated with dot blots and showed that for the non-parasitic plant C. arvensis the only hybridization was with the LhcB gene (located in the nucleus in all plants so far studied). However, with C. reflexa there was hybridization to eight genes and with C. australis there was hybridization to nine genes from the tobacco plastid genome. Because this cDNA was made using poly dT primers that will hybridize to the polyA tail of RNAs made in the nucleus, these results may provide evidence that regions of plastid DNA have transferred from the plastid to the nucleus, and are actively transcribed there.

Chlorophyll content and photosynthetic parameters in dodders
Chlorophyll content and photosynthesis were investigated for each Cuscuta species to determine whether photosynthetic competence correlated with phylogenetic position. Chlorophyll content of the dodders varied along the length of the filament, but was maximal at the growing tip (data not shown), and so the most apical 5 cm was used for this analysis. Between species, there were clear variations in chlorophyll content (Fig. 5A). All of the species had significantly less chlorophyll than stems of Convolvulus arvensis (400 µg chl g^–1 FW), and C. reflexa, C. lupuliformis, C. monogyna, and C. gronovii contained the most chlorophyll. However, when related to phylogenetic position as determined from the molecular phylogeny, there was no clear evidence that species within the more slowly evolving clades possessed higher levels of chlorophyll than those in clades evolving more quickly (Fig. 5A).

 
Pulse-modulated chlorophyll fluorescence was used to investigate the photosynthetic competence of these dodders. F_v/F_m varied significantly within the 15 species, but again this did not correlate clearly with phylogenetic position (Fig. 5B). Infrared gas analysis was also used to investigate the effect of increasing light intensity on CO₂ exchange rates in each of the Cuscuta species. The rate of CO₂ fixation never exceeded the rate of CO₂ evolution from respiration, and so the compensation point was never achieved. In the light, the rate of respiration was reduced by between 0.9 and 3.7 µmol CO₂ m^–2 s^–1 indicating that there were clear differences in photosynthetic ability between species (Fig. 5C). This variation in the apparent ability to incorporate CO₂ did not correlate clearly with phylogenetic position (Fig. 5C).