BRIEF DESCRIPTION OF THE DRAWINGS
The present invention is illustrated by way of example in the following drawings wherein:
FIG. 1 shows COX5c genes schemes (A, upper panel) and (B, lower panel), which contain an intron within the 5′-non-coding region.
(A) Scheme of the six sequenced COX5c genes available in data banks: three genes from Arabidopsis thaliana (AtCOX5c) and one each from Ipomoea batatas (IbCOX5c), Oryza sativa (OsCOX5c), and Lotus corniculatus (LcCOX5c). All genes contain a single intron (white boxes) located in the 5′-non-coding region. The OsCOX5c intron is not drawn to scale due to its length (2.4 kbp). Non-coding and coding regions of exons are shown in light and dark grey, respectively.
(B) Scheme of the different constructs used to analyse COX5c regions required for expression. Different regions of the COX5c-1 and COX5c-2 genes, containing 5′non coding region (black boxes), non coding regions of exons 1 and 2 located upstream of the start codon (light grey boxes), and the leader intron (white box) were fused to the gus coding region and introduced into plants. In a similar way, COX5c-2 untranslated leader sequences were fused to the COX5b-1 promoter (striped boxes) in both orientations.
FIG. 2 shows histochemical localization of GUS activity in Arabidopsis plants transformed with constructs carrying different portions of the COX5c-2 gene fused to gus. Plants transformed with fragments containing non-transcribed upstream regions plus either the complete transcribed 5′-non-coding region (A-J), only exon 1 sequences (K-O), or the 5′-non-coding region without the intron (P-T) were analysed. Two-(A), 3-(B, K, P), 15-(D, F), and 20-d-old plants (C, E) are shown. Leaves (G, L, Q, R), flowers (H, I, M, N, S), anthers (T), and siliques (J, O) from adult plants are also shown.
FIG. 3 shows bar-graphic illustrating that removal of the COX5c intron produces a pronounced decrease in expression of the reporter gene. GUS-specific activity of extracts from plants transformed with construct pBI5c2 (5c2+I) or the respective constructs in which either the intron plus a portion of exon 2 (5c2-I/E2) or only the intron (5c2-I) were removed was measured using the fluorogenic substrate MUG and protein extracts prepared from rosettes of 20-d-old plants. GUS activity was also measured in extracts from non-transformed plants (wt) or from plants carrying the promoterless gus gene (pBI101) or the gus gene under the control of the CaMV 35S promoter (pBI121). Error bars represent SE of three independent measurements with one representative line from each construct. Similar results were obtained with different independent lines. The inset shows an enlargement of the bars corresponding to plants that display low GUS activity values.
FIG. 4 shows histochemical localization of GUS activity in Arabidopsis plants transformed with the COX5b-1 promoter fused to the COX5c-2 leader intron. (A, B) Three- and 15-d-old plants transformed with a 609 bp COX5b-1 promoter fragment. (C-F) Similar plants transformed with the same promoter fragment fused to the COX5c-2 leader intron in the sense (D, E) or antisense (C, F) orientation. (G-L) Leaves from adult plants carrying the promoter alone (G, J) or with the intron in the sense (H, K) or the antisense (I, L) orientation. (M-P) Flowers and siliques from plants transformed with the COX5b-1 promoter fragment without (M, N) or with the COX5c-2 intron (O, P).
FIG. 5 shows a graph illustrating that the COX5c-2 leader intron increases expression from the unrelated COX5b-1 promoter. Specific GUS activity in extracts from plants transformed with a 609 bp COX5b-1 promoter fragment fused to gus (5b) or with a similar construct in which the COX5c-2 leader intron was inserted between the promoter and the gus coding region either in the sense (5b+Is) or antisense (5b+Ias) orientation was measured using the fluorogenic substrate MUG. GUS activity was also measured in extracts from non-transformed plants (wt) or from plants carrying the promoterless gus gene (pBI101) or the gus gene under the control of the CaMV 35S promoter (pBI121). Error bars represent SE of three independent measurements with one representative line from each construct. Similar results were obtained with different independent lines.
FIG. 6 shows a northern blot analysis of gus steady-state transcript levels using total RNA from plants transformed with constructs containing different portions of the COX5c-2 gene fused to gus. Total RNA (20 μg) from plants transformed with fragments containing COX5c-2 non-transcribed upstream regions and either the complete transcribed 5′-non-coding region (lane 1), only exon 1 sequences (lane 2), or the 5′-non-coding region without the intron (lane 3). Lanes 4, 5, and 6 contain RNA from plants transformed with a 609 bp COX5b-1 promoter fragment or the same fragment fused to the COX5c-2 leader intron in the antisense or sense orientation, respectively. Plants transformed with the promoterless gus gene (lane 7), non-transformed plants (lane 8), and plants transformed with the gus gene under the control of the CaMV 35S promoter (lane 9) were also analysed.
FIG. 7 shows DNA sequences isolated in the present invention:
SEQ No 1: COX5c-1 promoter;
SEQ No 2: COX5c-1 exon 1;
SEQ No 3 COX5c-1 intron;
SEQ No 4: COX5c-1 exon 2
FIG. 8 shows DNA sequences isolated in the present invention:
SEQ No 5: COX5c-2 promoter;
SEQ No 6: COX5c-2 exon 1;
SEQ No 7: COX5c-2 intron;
SEQ No 8: COX5c-2 exon 2
FIG. 9 shows DNA sequences isolated in the present invention:
SEQ No 9: COX5c-3 promoter;
SEQ No 10: COX5c-3 exon 2;
SEQ No 11: COX5c-3 intron;
SEQ No 12: COX5c-3 exon 2