An AGAMOUS intron-driven cytotoxin leads to flowerless tobacco and produces no detrimental effects on vegetative growth of either tobacco or poplar

Abstract

Flowerless trait is highly desirable for poplar because it can prevent pollen- and seed-mediated transgene flow. We have isolated the second intron of PTAG2, an AGAMOUS (AG) orthologue from Populus trichocarpa. By fusing this intron sequence to a minimal 35S promoter sequence, we created two artificial promoters, fPTAG2I (forward orientation of the PTAG2 intron sequence) and rPTAG2I (reverse orientation of the PTAG2 intron sequence). In tobacco, expression of the β-glucuronidase gene (uidA) demonstrates that the fPTAG2I promoter is non-floral-specific, while the rPTAG2I promoter is active in floral buds but with no detectable vegetative activity. Under glasshouse conditions, transgenic tobacco plants expressing the Diphtheria toxin A (DT-A) gene driven by the rPTAG2I promoter produced three floral ablation phenotypes: flowerless, neuter (stamenless and carpel-less) and carpel-less. Further, the vegetative growth of these transgenic lines was similar to that of the wild-type plants. In field trials during 2014 and 2015, the flowerless transgenic tobacco stably maintained its flowerless phenotype, and also produced more shoot and root biomass when compared to wild-type plants. In poplar, the rPTAG2I::GUS gene exhibited no detectable activity in vegetative organs. Under field conditions over two growing seasons (2014 to the end of 2015), vegetative growth of the rPTAG2I::DT-A transgenic poplar plants was similar to that of the wild-type plants. Our results demonstrate that the rPTAG2I artificial promoter has no detectable activities in vegetative tissues and organs, and the rPTAG2I::DT-A gene may be useful for producing flowerless poplar that retains normal vegetative growth.

Keywords: AGAMOUS; Populus; Intron; flowerless; neuter (stamenless and carpel-less); transgene flow.

Figure 1

Histochemical staining of GUS activity in fPTAG2I::GUS and rPTAG2I:: GUS poplar and tobacco. (a, b) Shoot apices from 2‐month‐old transgenic poplar plants harbouring fPTAG2I::GUS (a) and rPTAG2I::GUS (b). (c, d) Longitudinal sections of shoot apices from five‐week‐old glasshouse‐grown transgenic tobacco plants harbouring fPTAG2I::GUS (c) and rPTAG2I::GUS (d). (e–g) Longitudinal sections of floral buds from three different 4‐month‐old fPTAG2I::GUS tobacco lines showed varying levels of fPTAG2I promoter activity. (h, i) Cross sections of leaf (h) and stem (i) from 4‐month‐old fPTAG2I::GUS tobacco exhibited GUS activity. (j–l) Longitudinal sections of floral buds from three different 4‐month‐old rPTAG2I::GUS tobacco lines showed varying levels of rPTAG2I promoter activity. (m, n) Cross sections of leaf (m) and stem (n) from 4‐month‐old rPTAG2I::GUS tobacco exhibited no GUS activity. se: sepal; pe: petal; st: stamen; ca: carpel.

Figure 2

Floral organ development of glasshouse‐grown rPTAG2I::DT‐A tobacco plants. (a–c) Four‐month‐old rPTAG2I::DT‐A transgenic plants (right) alongside wild type plants (left). Transgenic plants displayed normal vegetative growth and one of three floral ablation phenotypes: flowerless ( rPTAG2I::DT‐A‐Line‐8) (a), neuter (stamenless and carpel‐less, rPTAG2I::DT‐A‐Line‐27) (b), or carpel‐less ( rPTAG2I::DT‐A‐Line‐35) (c). (d) A closer look at the floral buds from rPTAG2I::DT‐A‐Line‐8. All floral buds were aborted before reaching stage 9 of floral development. (e) A longitudinal section of a floral bud from (d) showing petals, stamens and carpel were aborted. (f) A flower with no stamens or carpels ( rPTAG2I::DT‐A‐Line‐27). (g) A longitudinal section of floral buds from rPTAG2I::DT‐A‐Line‐27 having stamens and carpels aborted. (h) A longitudinal section of floral buds from rPTAG2I::DT‐A‐Line‐35 having carpels aborted. (i, j) A longitudinal section of a flower from rPTAG2I::DT‐A‐Line‐35 (j) having carpel aborted when compared to that from a wild type plant (i). se: sepal; pe: petal.

Figure 3

Relative expression levels of the DT‐A gene in the 0.7‐mm floral buds (floral stage 6) of three representative rPTAG2I::DT‐A tobacco plants, showing that the severity of the floral phenotype correlated with the level of the DT‐A gene expression. WT: wild‐type buds; Line 8: flowerless buds from rPTAG2I::DT‐A‐Line 8; Line 27: neuter buds from rPTAG2I::DT‐A‐Line 27; Line 35: carpel‐less buds from rPTAG2I::DT‐A‐Line 35. Expression levels of the tobacco elongation factor 1α gene in each biological replicate were used as an internal reference (Schmidt and Delaney, 2010). Data represent means from three independent biological replicates. Bars show standard errors.

Figure 4

Performance of flowerless transgenic tobacco plants under field conditions. (a) Flowers developed from a 3‐month‐old wild‐type plant. (b) No flowers observed in a 3‐month‐old flowerless plant ( rPTAG2I::DT‐A‐Line 8) because all floral buds were aborted before floral stage 9. (c) Abundant seeds/seed pods produced from a 4‐month‐old wild‐type plant. (d) No seeds/seed pods produced but more branch shoots developed from a 4‐month‐old flowerless plant ( rPTAG2I::DT‐A‐Line 8). (e) A better root system (in right) observed in a 4‐month‐old flowerless plant ( rPTAG2I::DT‐A‐Line 8) when compared to that of a wild‐type plant (in left).

Figure 5

PCR confirmation of stable incorporation of the rPTAG2I::DT‐A gene into the genome of representative poplar plant lines used for the field evaluation. PCRs were performed as described in the Experimental Procedures section with primer sequences for the DT‐A gene within the T‐DNA region and for the tetR gene within the backbone of the Ti‐plasmid, using genomic DNA isolated from representative putative transgenic poplar plants as templates. The Lane M: molecular weight marker. Lanes 1 and 2: the rPTAG2I::DT‐A Ti‐plasmid as template with the tetR primers (Lane 1) and DT‐A primers (Lane 2). Lane 3: Wild‐type poplar plant DNA as template with the DT‐A primers. Lanes 4–5, 6–7, 8–9, 10–11, 12–13 and 14–15 are for PCR products using genomic DNA isolated from putative rPTAG2I::DT‐A transgenic poplar lines 2, 16, 29, 36, 45 and 57, respectively, with the tetR primers for even numbers and the DT‐A primers for odd numbers. The presence of the DT‐A gene and the absence of the tetR gene in the putative transgenic poplar lines demonstrate that these lines should be transgenic. On the other hand, the presence of both the DT‐A gene and the tetR gene indicates that the genomic DNA from that putative transgenic poplar plant is contaminated with the Ti‐plasmid DNA, and thus, the presence of the DT‐A gene does not necessarily support that the plant is transgenic.

Figure 6

Representatives of rPTAG2I::DT‐A transgenic poplar plants grown under field conditions. (a–d) No morphological differences were observed between transgenic (c, d) and wild‐type poplar plants (a, b) after 1‐month growth in the field.

Figure 7

Performance of rPTAG2I::DT‐A transgenic poplar plants under glasshouse and field conditions. (a–d) No significant differences in height or biomass index (estimated using height × diameter²) between WT and transgenic plants were observed after 2‐month growth in glasshouse for both 2014 and 2015 transformations (a, b), or after two growing seasons in field for 2014 transformation (c, d) according to Student's t‐test with the pooled variance at P = 0.05. Brackets represent 95 % confidence intervals. Blue bars show data of wild‐type poplar plants and red bars show data of rPTAG2I::DT‐A transgenic poplar plants.

Figure 8

Gene constructs used for tobacco and poplar transformation. LB: left border sequence of T‐DNA. RB: right border sequence of T‐DNA. tNOS : nopaline synthase terminator. NptII : neomycin phosphotransferase gene. pNOS : nopaline synthase gene promoter sequence. 35S: cauliflower mosaic virus 35S gene promoter sequence. GUS : the coding sequence for the β‐glucuronidase gene. fPTAG2I: the forward orientation of the second intron of P. trichocarpa AGAMOUS (AG) 2 gene. Min35S: 60 basepairs of the 35S gene leader and promoter sequence that has no promoter activity. rPTAG2I: the reverse orientation of the second intron sequence of the P. trichocarpa AGAMOUS ( AG ) 2 gene. DT‐A: the coding sequence for the Diphtheria toxin A ( DT‐A) gene, which codes for a ribosome inactivating protein.