The Glycine max sucrose binding protein (GmSBP2) promoter directs phloem-specific expression of reporter genes in transgenic tobacco. Here, we identified cis-regulatory domains (CRD) that contribute with positive and negative regulation for the tissue-specific pattern of the GmSPB2 promoter. Negative regulatory elements in the distal CRD-A (-2000 to -700) sequences suppressed expression from the GmSBP2 promoter in tissues other than seed tissues and vascular tissues of vegetative organs. Deletion of this region relieved repression resulting in a constitutive promoter highly active in all tissues analyzed. Further deletions from the strong constitutive -700GmSBP2 promoter delimited several intercalating enhancer-like and repressing domains that function in a context-dependent manner. Histochemical examination revealed that the CRD-C (-445 to -367) harbors both negative and positive elements. This region abolished promoter expression in roots and in all tissues of stems except for the inner phloem. In contrast, it restores root meristem expression when fused to the -132pSBP2-GUS construct, which contains root meristem expression-repressing determinants mapped to the 44-bp CRD-G (-136 to -92). Thus, the GmSBP2 promoter is functionally organized into a proximal region with the combinatorial modular configuration of plant promoters and a distal domain, which restricts gene expression to the vascular tissues in vegetative organs.

The Glycine max sucrose binding protein (GmSBP2) promoter directs vascular tissue-specific expression of reporter genes in transgenic tobacco. Here we showed that an SBP2-GFP fusion protein under the control of the GmSBP2 promoter accumulates in the vascular tissues of vegetative organs, which is consistent with the proposed involvement of SBP in sucrose transport-dependent physiological processes. Through gain-of-function experiments we confirmed that the tissue-specific determinants of the SBP2 promoter reside in the distal cis-regulatory domain A, CRD-A (position -2000 to -700) that is organized into a modular configuration to suppress promoter activity in tissues other than vascular tissues. The four analyzed CRD-A sub-modules, designates Frag II (-1785/-1508), Frag III (-1507/-1237), Frag IV (-1236/-971) and Frag V (-970/-700), act independently to alter the constitutive pattern of -92pSBP2-mediated GUS expression in different organs. Frag V fused to -92pSBP2-GUS restored the tissue-specific pattern of the full-length promoter in the shoot apex, but not in other organs. Likewise, Frag IV confined GUS expression to the vascular bundle of leaves, whereas Frag II mediated vascular specific expression in roots. Strong stem expression-repressing elements were located at positions -1485 to -1212, as Frag III limited GUS expression to the inner phloem. We have also mapped a procambium silencer to the consensus sequence CAGTTnCaAccACATTcCT which is located in both distal and proximal upstream modules. Fusion of either repressing element-containing module to the constitutive -92pSBP2 promoter suppresses GUS expression in the elongation zone of roots. Together our results demonstrate the unusual aspect of distal sequences negatively controlling tissue-specificity of a plant promoter.