Isolation of five rice nonendosperm tissue-expressed promoters and evaluation of their activities in transgenic rice

Abstract

Using promoters expressed in nonendosperm tissues to activate target genes in specific plant tissues or organs with very limited expression in the endosperm is an attractive approach in crop transgenic engineering. In this article, five putative nonendosperm tissue-expressed promoters were cloned from the rice genome and designated P_OsNETE1 , P_OsNETE2 , P_OsNETE3 , P_OsNETE4 and P_OsNETE5 . By qualitatively and quantitatively examining GUSplus reporter gene expression in transgenic rice plants, P_OsNETE1 -P_OsNETE5 were all found to be active in the roots, leaves, stems, sheaths and panicles but not in the endosperm of plants at different developmental stages. In addition, P_OsNETE2 , P_OsNETE4 and P_OsNETE5 were also inactive in rice embryos. Among these promoters, P_OsNETE4 and P_OsNETE5 exhibited higher activities in all of the tested tissues, and their activities in stems, leaves, roots and sheaths were higher than or comparable to those of the rice Actin1 promoter. We also progressively monitored the activities of P_OsNETE1 -P_OsNETE5 in two generations of single-copy lines and found that these promoters were stably expressed between generations. Transgenic rice was produced using P_OsNETE4 and P_OsNETE5 to drive a modified Bt gene, mCry1Ab. Bt protein expressed in the tested plants ranged from 1769.4 to 4428.8 ng/g fresh leaves, whereas Bt protein was barely detected in the endosperm. Overall, our study identified five novel nonendosperm tissue-expressed promoters that might be suitable for rice genetic engineering and might reduce potential social concern regarding the safety of GMO crops.

Keywords: Bacillus thuringiensis; GUSplus; nonendosperm tissue-expressed promoters; transgenic rice.

Figure 1

Expression profile of five nonendosperm tissue‐expressed genes. The temporal and spatial expression patterns of five genes were examined by semi‐quantitative RT–PCR using 28‐cycle amplification in different tissues, including the roots (1) and leaves (2) of 10 DAG rice seedlings; the roots (3), leaves (4), stems (5) and sheaths (6) of 60 DAG plants; the panicles of 75 DAG plants (7) and the endosperm of maturing seeds collected at 21 DAF (8). OsACTIN1 were used to normalize the samples.

Figure 2

Validation of the tissue specificity of the putative nonendosperm tissue‐expressed promoters in transgenic rice. Histochemical staining was conducted in a series of tissues from transgenic plants including the roots, leaves, stems and sheaths of 60 DAG rice plants and the panicles of 75 DAG plants (a), anthesis and seeds at different developmental stages (b). The seeds were cut into median longitudinal sections. In transgenic plants carrying the construct of the tissue‐expressed promoters, strong blue staining was observed in various nonendosperm tissues after only 2 to 12 h of incubation, while no GUS staining was detected in endosperms after 24 h of X‐gluc incubation. Scale bars represent 2 mm.

Figure 3

qRT–PCR analysis of the reporter gene expression in six single‐copy transgenic lines harbouring promoter:: GUSplus. Total RNA was prepared from various tissues including the panicles of 75 DAG plants and the endosperm of maturing seeds (21 DAF). The relative levels of GUSplus were measured by qRT–PCR, and OsACTIN1 was used as the internal control gene. The levels of promoter‐driven reporter gene transcripts were calculated relative to those of GUSplus in the P _{O s ACT} ::GUSplus transgenic lines. The average was calculated from three biological replicates. ±SD is indicated.

Figure 4

Quantitative analysis of promoter activities in T₃ lines harbouring promoter:: GUSplus. GUS activity was measured in the roots and leaves of 10 DAG seedlings; the roots, leaves, stems and sheaths of 60 DAG plants; the panicles of 75 DAG plants and the endosperm of mature seeds. Six independent single‐copy lines from each construct were analysed. GUS activity, given as nmol 4‐methylumbelliferone (MU) per minute per milligram protein, was averaged from three biological replicates. ±SD is indicated.

Figure 5

mCry1Ab expression pattern of fresh leaves from three T₂ homozygous transgenic rice lines. a to c, mCry1Ab transcript levels were determined by qRT–PCR in the leaves at the tillering stage (a) and filling stage (b) or in the endosperm of maturing seeds 21 DAF (c) of transgenic plants containing P _{O s NETE 4} ::mCry1Ab, P _{O s NETE 5} ::mCry1Ab and P _{O s ACT} ::mCry1Ab. The levels of promoter‐driven mCry1Ab transcripts were calculated relative to those of mCry1Ab in the P _{O s ACT} ::mCry1Ab transgenic lines. d to f, mCry1Ab protein levels were measured by ELISA in the leaves at the tillering stage (d) and filling stage (e) or in the endosperm of mature seeds (f). For each construct, three single‐copy transformants were used. Data are the mean ± SD of three independent biological replicates.