Figure 1.
Wild and domesticated cotton have different architectures and flower at different times.
(A) TX701, a photoperiodic accession, has deeply lobed “okra” leaves and pronounced apical dominance. TX701 does not flower under long days and this plant is completely vegetative. (B) DP61, a day-neutral cultivar, has a bushy growth habit with normal leaves and flowers profusely. Arrows are pointing to floral buds (“squares”). Both plants were grown under long-day conditions (16/8 h day/night) in a greenhouse with supplemental lighting. Scale bars, 25 cm.
Figure 2.
Florigen promotes determinate leaf growth.
(A) Leaf shape along the fruiting branch (SU, sympodial units) of a dCLCrV::FT-infected TX701 plant transitions from highly lobed to lanceolate as the fruiting branch gets older. (B) dCLCrV-infected TX701 plants do not show the same transition. Note the leaf crumpling symptoms of dCLCrV-infected plants (A, B). (C) Changes in TX701 leaf correlate with changes in day length and reproductive vs. vegetative growth; these leaves are from the same plant that transitioned from long-day, to short day and back to long day growth conditions, with ∼10 weeks in each condition. See also Fig. S2 for plants grown exclusively in natural sunlight.
Figure 3.
dCLCrV::FT infection uncouples flowering from photoperiod.
(A) Arrows point to some of many squares on a representative dCLCrV::FT infected TX701 plant under long day conditions. Scale bar is 25 cm. (B) Close-up of sympodial growth on a fruiting branch of an FT-induced TX701 plant. (C) An open bloom from an FT-induced TX701 plant reveals characteristic dark red petal spots. (D) A boll on a DP61 plant resulting from cross-pollination with an FT-induced TX701 flower (See also Fig. S3). (E) RT-PCR demonstrates that FT expression is limited to plants inoculated with dCLCrV::FT; GAPDH expression serves as an internal control. Lanes are: 1 and 2, untransfected TX701; 3, no reverse transcriptase control; 4, dCLCrV::FT-infected plant shown in (C); 5, dCLCrV::FT-infected plant shown in (B); 6, dCLCrV::FT-infected plant shown in (A); 7 and 8, other dCLCrV::FT-infected TX701 plants that flowered (plants not shown); 9) dCLCrV::FT-bombarded TX701 plant which did not flower – note the absence of FT; 10) dCLCrV::FT plasmid template control; 11) no template control.
Figure 4.
FT promotes determinate growth and synchronizes flowering.
(A) dCLCrV::FT-infected DP61 plants exhibit a more compact growth habit. Shown are dCLCrV-infected (i.e., empty virus) and dCLCrV::FT-infected DP61 plants (left and right, respectively; both plants are 87 dpg and were grown at the same time in the same greenhouse). White circles highlight maturing bolls and arrows point to flowers before or in bloom. Note that the dCLCrV::FT-infected plant has only maturing bolls and no immature flowers. Scale bar is 25 cm. (B) dCLCrV::FT-infected plants demonstrate more determinate growth. Shown are the mean number of sympodial units along fruiting branches among untransfected plants (n = 4, black bar), dCLCrV-infected plants (n = 3, white bar), and three dCLCrV::FT-infected plants represented individually (green, blue and red bars) to show the range of variation; dCLCrV::FT-3 is the plant shown in (A). The severity of viral infection was scored as mild (+) or stronger (++) based on leaf crumpling. (C, D, E) Schematic representations of growth patterns observed among ∼90 d-old plants: (C) Uninfected and dCLCrV-infected DP61; (D) dCLCrV::FT infected DP61, (E) dCLCrV::FT infected TX701. Red circles represent maturing bolls; magenta circles represent immature or blooming flowers; green circles represent active buds reiterating sympodial growth; blue circles represent the monopodial bud of the main stem; and branches without a circle represent buds that have terminated without a flower or a fruit. Leaves are not represented, and the number of branches and internode lengths are not to scale. (F) Representative floral cluster (two floral buds inside a common bract whorl) terminating a fruiting branch on a dCLCrV::FT-infected TX701 plant. “SU branch” is the internode of the terminal sympodial unit; “petiole” is the petiole of the leaf subtending the floral cluster. No other vegetative growth is evident.
Figure 5.
F1 progeny of crosses with a VIF-treated parent have expected traits and are virus-free.
(A) The shape of main-stem leaves harvested from nodes 8–9 of TX701, DP61, and the F1, as labeled. TX701 leaves exhibit the okra leaf phenotype with five deep lobes and reduced lamina area. DP61 exhibits the normal leaf phenotype with three smaller lobes and a well-expanded lamina. The F1 from the DP61 × FT-induced TX701 cross exhibit an intermediate phenotype with leaves having five lobes of intermediate length. (B) Node of first fruiting branch is intermediate in the F1 population. Under non-inductive long days, the F1 produce floral buds (n = 46) in contrast to the TX701 parent which does not demonstrate reproductive growth by node 25 (n = 8 plants). Day-neutral DP61 produce floral buds earlier in development (n = 6 plants). (C) The maternal DP61 flowers (top) do not have the flower spots characteristic of the paternal TX701 pollen donor (see Fig. 3c); F1 progeny flowers (bottom) have the flower spot trait. (D) dCLCrV is not transmitted to the F1 progeny. From left to right, the viral Rep gene is not detected in uninfected DP61 or TX701 (lanes 1 and 2), nor in 9 different F1 progeny plants (lanes 3–11, collectively labeled F1), but is readily detected by PCR from a dCLCrV::FT-infected plant (lane 12, labeled dCLCrV::FT; this is the same plant shown in Fig. 3A and E, lane 6) and from a plasmid template (lane 13, labeled +). Detection of the endogenous magnesium chelatase subunit I (ChlI) serves as an internal control (plasmid control in the case of lane 13, +). No sequences were amplified in the absence of DNA (lane 14, labeled -).