High-level tuber expression and sucrose inducibility of a potato Sus4 sucrose synthase gene require 5' and 3' flanking sequences and the leader intron.

The 3.6 kb of 5' flanking sequence, leader intron, and 0.7 kb of 3' sequence from the potato sucrose synthase gene Sus4-16 are sufficient to direct high-level expression in developing tubers, in basal tissues of axillary buds and shoots, and in meristems and caps of roots, and to confer sucrose inducibility in leaves. By examining a series of deletion and substitution constructs in transgenic potato plants, we found that this pattern of expression requires 5' flanking sequences both upstream and downstream of position -1500 and that sequences between positions -1500 and -267 are essential for sucrose induction. Replacement of the native 3' sequence with the nopaline synthase 3' sequence resulted in the loss of sucrose inducibility and of expression in basal tissues of axillary buds. A general decrease in expression in other tissues was also observed. Removal of the 1612-bp leader intron also had a dramatic effect on both the pattern and level of expression.     

Secretion of biologically active human granulocyte colony-stimulating factor (G-CSF) in milk of transgenic mice

Two constructs were devised, containing the full-length gene of the human granulocyte colony-stimulating factor (G-CSF) fused with the 5' and 3' flanking promoter sequences of bovine alpha-S1-casein gene. Both constructs contained a 1518-bp fragment that included exons 18 and 19 and 320 bp of the 3' flanking region of bovine gene @CSN1S1, but differed in size of the 5' flanking sequences, which were of 721 bp, and exon 1 in construct pGCm1 and 2001 bp and exon 1 and intron 1 in construct pGCm2. With both constructs, transgenic mice were produced. The transgene expression was assessed using RT-PCR and immunochemically from the production of human G-CSF in milk of lactating females. Secretion of human G-CSF into the milk varied in a wide range, from 0.8 microg/ml to over 1 mg/ml, in mice with construct pGCml and was low (up to 60 microg/ml) or absent in mice with construct pGCm2. G-CSF glycosylation was incomplete in mice with transgene pGCml and complete in mice with pGCm2. G-CSF of transgenic mouse milk was shown to stimulate the formation and growth of granulocyte-containing colonies in human umbilical blood cell culture and be close or identical in physiological activity to the natural human G-CSF.     

Molecular Cloning and Sequence of Potato Granule-Bound Starch Synthase Gene

It is known that tuber-specific expressions of many genes exist in the process of tuber development from stolon in potato (Solanum tuberosum). Study on the regulation of those gene expression will share light on the mechanism of organ-specific gene expression. Potato GBSS (granule-bound starch synthase) gene, which is solely responsive for the pres- ence of amylose in potato tuber, expression is tuber-specific. The paper describes the construction of a genomic library of a Chinese potato cultivar "Dongnong 303" in which 20 clones were isolated using partial GBSS gene sequence ampified by PCR. 5428 bp DNA sequence of one clone (GBSS17-1) was determined, including 1823 bp 5' flanking region. 2964 bp structure gene, and 641 bp 3' flanking region. It is highly homologious with reported GBSS gene sequence. In addition, the 730 bp most upstream sequence of 5' flanking region which was not reported previously contained stem and loop structures. The present result may provide some important information for further study in the molecular mechanism of organ specific gene expression.     

A Leader Intron of a Soybean Elongation Factor 1A (eEF1A) Gene Interacts with Proximal Promoter Elements to Regulate Gene Expression in Synthetic Promoters

Introns, especially the first intron in the 5’ untranslated region (5’UTR), can significantly impact gene expression via intron-mediated enhancement (IME). In this study, we demonstrate the leader intron of a soybean elongation factor 1A (eEF1A) gene (GmScreamM8) was essential for the high activity of the native promoter. Furthermore, the interaction of the GmScreamM8 leader intron with regulatory element sequences from several soybean eEF1A promoters was studied using synthetic promoters, which consisted of element tetramers upstream of a core promoter used to regulate a green fluorescent protein (gfp) reporter gene. Element tetramers, placed upstream of a GmScreamM8 core promoter, showed very high activity using both transient expression in lima bean cotyledons and stable expression in soybean hairy roots, only if the native leader intron was included, suggesting an interaction between intronic sequences and promoter elements. Partial deletions of the leader intron showed that a 222 bp intronic sequence significantly contributed to very high levels of GFP expression. Generation of synthetic intron variants with a monomeric or trimeric repeat of the 222 bp intronic sequence, yielded almost two-fold higher expression compared to the original intron, while partial deletion of the 222 bp intronic repeated sequence significantly decreased gene expression, indicating that this intronic sequence was essential for the intron-element interaction enhancement.     

Tissue-specific regulation of a chicken delta-crystallin gene in mouse cells: involvement of the 5' end region.

A cloned delta-crystallin gene of the chicken is preferentially expressed in lens cells after introduction into various mouse tissues. The level of expression in the lens epithelium is 20 times higher than in fibroblasts. Taking advantage of this system, we attempted to define regulatory regions of the delta-crystallin gene using a variety of deletion and substitution mutants. The results indicate that tissue-specific regulation of the delta-crystallin gene is mediated by the 5' end region of the gene; sequences upstream from -93 are not required for expression and sequences downstream from +58 are not involved in tissue specificity. The high expression in lens cells requires 5' flanking sequences of 80-bp long from the cap site, whereas the low expression in fibroblasts requires an additional 12 bp upstream sequence. Expression of both types is lost in a mutant with only 51 bp of the 5' flanking sequence. Thus, fine deletion analysis demonstrated that expression in lens cells and expression in fibroblasts are distinct not only in level but in regulation.     

A novel selection system for recombinational and mutational events within an intron of a eucaryotic gene.

In order to identify a poison sequence that might be useful in studying illegitimate recombination of mammalian cell chromosomes, several DNA segments were tested for their ability to interfere with gene expression when placed in an intron. A tRNA gene and its flanking sequences (267 bp) were shown to inhibit SV40 plaque formation 100-fold, when inserted into the intron in the T-antigen gene. Similarly, when the same DNA segment was placed in the second intron of the adenosine phosphoribosyl transferase (APRT) gene from CHO cells, it inhibited transformation of APRT-CHO cells 500-fold. These two tests indicated that the 267-bp DNA segment contained a poison sequence. The poison sequence did not affect replication since the replication of poisoned SV40 genomes was complemented by viable SV40 genomes and poisoned APRT genes were stably integrated into cell chromosomes. Cleavage of the poison sequence in the SV40 T-antigen intron by restriction enzymes indicated that the tRNA structural sequences and the 5' flanking sequences were not required for inhibition of SV40 plaque formation. Sequence analysis of viable mutant SV40, which arose after transfection of poisoned genomes, localized the poison sequence to a 35 bp segment immediately 3' of the tRNA structural sequences.     

Expression Enhancement of a Rice Polyubiquitin Gene Promoter

An 808 bp promoter from a rice polyubiquitin gene, rubi3, has been isolated. The rubi3 gene contained an open reading frame of 1140 bp encoding a pentameric polyubiquitin arranged as five tandem, head-to-tail repeats of 76 aa. The 1140 bp 5′ UTR intron of the gene enhanced its promoter activity in transient expression assays by 20-fold. Translational fusion of the GUS reporter gene to the coding sequence of the ubiquitin monomer enhanced GUS enzyme activity in transient expression assays by 4.3-fold over the construct containing the original rubi3 promoter (including the 5′ UTR intron) construct. The enhancing effect residing in the ubiquitin monomer coding sequence has been narrowed down to the first 9 nt coding for the first three amino acid residues of the ubiquitin protein. Mutagenesis at the third nucleotide of this 9 nt sequence still maintains the enhancing effect, but leads to translation of the native GUS protein rather than a fusion protein. The resultant 5′ regulatory sequence, consisting of the rubi3 promoter, 5′ UTR exon and intron, and the mutated first 9 nt coding sequence, has an activity nearly 90-fold greater than the rubi3 promoter only (without the 5′ UTR intron), and 2.2-fold greater than the maize Ubi1 gene promoter (including its 5′ UTR intron). The newly created expression vector is expected to enhance transgene expression in monocot plants. Considering the high conservation of the polyubiquitin gene structure in higher plants, the observed enhancement in gene expression may apply to 5′ regulatory sequences of other plant polyubiquitin genes.