Cytogenomic analyses reveal the structural plasticity of the chloroplast genome in higher plants

Abstract

A DNA fiber-based fluorescence in situ hybridization (fiber-FISH) technique was developed to analyze the structure and organization of a large number of intact chloroplast DNA (cpDNA) molecules from Arabidopsis, tobacco, and pea. Using this cytogenomic approach, we determined that 25 to 45% of the cpDNA within developing leaf tissue consists of circular molecules. Both linear and circular DNA fibers with one to four copies of the chloroplast genome were present, with monomers being the predominant structure. Arabidopsis and tobacco chloroplasts contained previously unidentified multimers (>900 kb) consisting of six to 10 genome equivalents. We further discovered rearranged cpDNA molecules of incomplete genome equivalents, confirmed by both differential hybridizations and size estimations. The unique cpDNA organization and novel structures revealed in this study demonstrate that higher plant cpDNA is more structurally plastic than previous sequence and electrophoretic analyses have suggested. Additionally, we demonstrate how the fiber-FISH-based cytogenomic approach allows for powerful analysis of very rare events that cannot be detected by traditional techniques such as DNA gel blot hybridization or polymerase chain reaction.

Figure 1.

PFGE Analysis of Chloroplast DNAs from Tobacco and Pea. (A) Tobacco. (B) Pea. Ethidium bromide–stained pulsed-field gels with a λ marker in lanes 1 ([A] and [B]) and a low-range PFGE marker (New England Biolabs, Beverly, MA) in lane 2 (B). T and P indicate lanes containing embedded chloroplasts from tobacco and pea, respectively. Arrows designate the most prominent bands representing monomer (155), dimer (312), trimer (450), and tetramer (630) fragments in tobacco and monomer (130) and dimer (260) fragments in pea (in kilobases). PFGE conditions were 5 to 120 sec of pulse time at 4.5 V/cm, with a total run time of 46 hr.

Figure 2.

Fiber-FISH Images of Tobacco Chloroplast DNA. (A) A monomeric circular cpDNA molecule. The green signal represents the entire cpDNA genome, and the red signal represents the 12.2-kb IR. (B) A monomeric circular cpDNA molecule with ∼87 kb (24 μm) of the green signal representing the LSC and 68 kb (17 μm) of the red signal homologous to both the SSC and IR regions. (C) to (E) Three monomeric cpDNA molecules (green) together with hybridization signals (red) derived from the 12.5- (C), 4.6- (D), and 1.8-kb (E) cpDNA probes. (F) to (J) Schematic illustrations of the cpDNA images, corresponding to (A) to (E), respectively. Green and red lines represent the differential hybridization. .

Figure 3.

Structural Plasticity of cpDNA Molecules from Tobacco, Arabidopsis, and Pea. (A) A wide-field view (magnification ×400) showing three circular cpDNA molecules from tobacco. The two small circles are monomer size, and the larger molecule is trimer size (arrow). (B) A field of view (magnification ×630) showing tobacco chloroplast fibers hybridized with an IR probe (red FISH signal). (C) Selected monomeric and multimeric cpDNA molecules from tobacco. Red signals are from the 12.2-kb IR probe. (D) A complete linear hexameric tobacco molecule. The green signals represent LSC; the red signals show SSC and IR regions. (E) A circular tetrameric chloroplast molecule from Arabidopsis. The green signals were derived from P1 clones MAB17 and MCI3, and the red signals are from P1 clone MAH2 (Sato et al., 1999). (F) Two monomeric circular molecules from pea probed with a single-copy sequence possessing the origin of replication (red). .

Figure 4.

Size Variation of Circular Chloroplast DNA Molecules. Black bars represent tobacco and gray bars represent pea molecules. Sizes were converted to kilobases based on 1.0 kb equaling 0.26 ± 0.03 μm. Asterisks represent the monomeric, dimeric, trimeric, and tetrameric molecules within a 99% confidence interval around the mean.

Figure 5.

Atypical cpDNA Molecules. (A) A typical head-to-tail dimer (arrow) from tobacco showing the 12.2-kb IR (red signals) and two small (133 and 110 kb) circular molecules exhibiting only a single IR. (B) A typical head-to-tail dimer (arrow) from tobacco showing the IR (red signals). A smaller circular molecule (205 kb) exhibits three IRs. (C) Two Arabidopsis cpDNA molecules hybridized with three probes homologous with 40.4 kb of the LSC region (red signal). The arrow points to a typical monomeric structure, and the arrowhead points to an ∼190-kb molecule that has two hybridization signals. (D) An ∼660-kb cpDNA molecule from tobacco exhibiting five IRs (red signals). A significantly smaller (∼40 kb) DNA molecule shows no hybridization to the IR probe. (E) A putative D-loop-like molecule from Arabidopsis. The red signals are derived from P1 clone MAH2 that covers the IRs, SSC, and part of the LSC regions (Sato et al., 1999). (F) A schematic illustration of the image shown in (E). .