Microarray analysis of selenium-depleted and selenium-supplemented mice.

Nutritional selenium deficiency is associated with Keshan disease in humans and white muscle disease in ruminant livestock. In this study, mice were fed a selenium-deficient diet for three generations. Female mice from the third depleted generation of these mice were given water containing either no added selenium or 0.1 or 1.0 ppm selenium as sodium selenate; DNA microarrays were used to compare gene expression in the muscle from mice fed the selenium diets to that from mice remaining on the depleted diet. The most prominent expression increases were observed with Ptger2 (a prostaglandin E receptor), Tcrb-V13 (a T-cell receptor beta), Tcf-7 (a T-cell transcription factor), and Lck (lymphocyte protein tyrosine kinase), and the major consistent decrease was Vav2, an oncogene in mice consuming the selenium containing diets.

Deletion analysis of the rodent selenoprotein W promoter.

To identify regulatory elements in the rat selenoprotein W (SeW) promoter, 2090, 1265, 741, and 404 base pair truncations of genomic DNA lying immediately upstream of the SeW coding sequence were cloned into a luciferase reporter vector (pGL3-Basic from Promega, Madison, WI, USA). 3656 and 406 base pair mouse SeW promoter constructs were also compared. SeW promoter activity was assayed in two rat cell lines: L8 muscle cells and C6 brain cells. The SeW promoter was 2-7 times more active (p<0.01) than SV40 promoter. Promoter activity of constructs of the SeW promoter ranging from 200 base pairs to 51 base pairs gradually decreased to zero in brain cells, but fell precipitously to zero in muscle cells. Some truncations stimulated promoter activity, suggesting the full-length promoter may contain binding sites for factors that suppress SeW expression.

Selenoprotein W accumulates primarily in primate skeletal muscle, heart, brain and tongue.

The human selenoprotein W coding region with the selenocysteine codon (TGA) changed to a cysteine codon (TGT) was fused to six histidine codons (at its 3' end), cloned into a prokaryotic expression vector (pTrc99a), and the corresponding mutated selenoprotein W was expressed in bacteria. The protein was purified by Ni-NTA agarose column and reverse phase HPLC. Polyclonal antibodies raised against this protein were used in Western blots to determine tissue distribution of selenoprotein W from rhesus monkeys fed a commercial chow. Selenoprotein W was found in several tissues with highest amounts in skeletal muscle and heart (muscle 6 fold greater than liver) and lowest levels in liver, but selenium concentrations were highest in kidneys (10 fold greater than muscle) and lowest in skeletal muscle. Northern blots using a human selenoprotein W cDNA probe indicated that mRNA levels were highest in monkey skeletal muscle and heart (2-2.5 fold greater than in liver), which is similar to the pattern found with a human multiple tissue Northern blot. However, as in the monkey, selenium concentrations were highest in human kidney and lowest in skeletal muscle and heart. Thus, selenoprotein W protein levels correlated with selenoprotein W mRNA levels but not with tissue selenium concentrations.

Selenoprotein W cDNAs from five species of animals.

The nucleotide sequences of the open reading frames of cDNAs for selenoprotein W from skeletal muscle of rat, mouse, sheep, rhesus monkey and human are reported. Theoretical translation of the coding sequences indicated highly similar proteins of 88 (mouse and rat) or 87 (human, monkey and sheep) amino acids. In 73 of 88 positions the specified amino acids are identical for all five proteins. TGA encoding selenocysteine is the 13th codon of all the cDNAs. The mouse, rat and sheep open reading frames terminate with TGA but the human and rhesus monkey coding regions terminate with TAA. The encoded amino acid sequences are identical for the rat and mouse proteins, and for the human and monkey proteins. The similarity of the cDNAs continues in the 3' noncoding regions through the putative selenocysteine insertion sequence (SECIS) elements which are required for correct interpretation of the selenocysteine codon. The region between the SECIS elements and the polyadenylation signals showed much lower similarity. The cloned rat gene for selenoprotein W is 5000 bases long, with the 663 bases of the cDNA in six exons. The transcription start site was identified by nuclease protection assay to be 16 bases upstream of the longest cDNA clone. A canonical TATA box occurs 150 bases upstream, but the assay did not indicate the presence of longer mRNAs.

T-DNA border sequences required for crown gall tumorigenesis.

Similar 23-base-pair (bp) direct repeats occur at the ends of two adjacent but noncontiguous T-DNAs, TL and TR (left and right T-DNA), in the tumor-inducing plasmid pTiA6NC. Thus, three border repeats lie right and one lies left of TL, which carries the genes needed for tumor maintenance. To determine whether T-DNA transfer and integration (subsequently called T-DNA transmission) require sequences in addition to the 23-bp border repeat, we constructed a deletion removing the three potential TL right borders (the TL right border and both TR borders). Since this deletion severely attenuated virulence, we reintroduced restriction fragments containing the TL right border repeat at a new location to the right of TL and tested their ability to restore virulence. Fragments that carried the border repeat flanked by at least 67 bp of wild-type Ti plasmid sequences on the left and 1035 bp on the right restored virulence completely. Smaller fragments restored virulence significantly but not fully, even though the border repeat remained intact. Therefore, T-region sequences flanking the border repeat in the fully active fragments stimulated T-DNA integration. Fragments that restored virulence fully when inserted in the wild-type orientation stimulated virulence only slightly in the opposite orientation. Thus, the right border sequence promotes T-DNA transfer and integration best in one direction.

Molecular analysis of the recF gene of Escherichia coli.

We analyzed the nucleotide sequence of a 1.325-kilobase region of wild-type Escherichia coli containing a functional recF gene and six Tn3 mutations that inactivate recF. The analysis shows a potentially translatable reading frame of 1071 nucleotides, which is interrupted by all six insertions. A protein of 40.5 kilodaltons would result from translation of the open reading frame, and a radioactive band of protein of an apparent molecular weight of approximately 40 kilodaltons was seen by the maxicell method using a recF+ plasmid. Putative truncated peptides were seen when two recF::Tn3 mutant plasmids were used. Differential expression of dnaN and recF from a common promoter was noted. recF332::Tn3 was transferred to the chromosome where, in hemizygous condition, it produced UV sensitivity indistinguishable from that produced by two presumed recF point mutations.

Cloning and deletion mapping of the recF dnaN region of the Escherichia coli chromosome.

By cloning a 3.6-kb EcoRI fragment of the Escherichia coli chromosome with pBR322 we located more precisely recF relative to dnaN. By deletion mapping we localized functional recF to a 1.65-kb region of the cloned fragment and allowed rough mapping of the C terminus of dnaN. Cloned recF+, separated from functional flanking genes dnaN and gyrB, complemented chromosomal recF mutations presumably by coding for a cytodiffusible product. The protein encoded by dnaN was observed as a band on a polyacrylamide gel from minicells. Identification of a recF protein was not made.

Multiple mutations in the T region of the Agrobacterium tumefaciens tumor-inducing plasmid.

Three genetic loci affecting tumor morphology lie within pTiA6NC T-DNA: tms, tmr, and tml. Using deletions and multiple transposon insertions, we constructed tumor-inducing (Ti) plasmids representing every possible double and triple mutant combination. tms tmr and tms tmr tml mutants did not incite tumors on most plants and produced a very weak response on a few other hosts but tms tml and tmr tml mutants were virulent. Thus, either tms+ or tmr+ alone can promote significant tumor growth but tml+ by itself is not sufficient. On hosts where tms mutants induce tumors accompanied by shoot proliferation, addition of a tml mutation reduces or eliminates shoot proliferation, suggesting that tml+ promotes shoot development. The small calli incited by tms tmr and tms tmr tml mutants contain agropine, an indication that these plant cells incorporate T-DNA in the absence of substantial tumor growth.

Crown gall disease and prospects for genetic manipulation of plants.

Agrobacterium tumefaciens incites crown gall tumors when bacterial DNA integrates into plant nuclear DNA. Plant cells can express these integrated bacterial genes. Following insertion of desired genes into bacterial DNA using recombinant DNA techniques, this system permits introduction of these new genes into plant DNA. We discuss the potential for genetic manipulation of plants using Agrobacterium tumefaciens and the related organism Agrobacterium rhizogenes.

Genetic and physical mapping of recF in Escherichia coli K-12.

Two factor transductional crosses place recF at approximately 82 min on the E. coli chromosome; recF is highly cotransducible with dnaA and gyrB (cou). Transductional analysis with a series of lambda tna specialized transducing phages carrying chromosomal DNA from the tnaA region place recF between dnaA and gyrB. This analysis also indicates that a gene lying in the same region and producing an easily detectable protein (estimated MW of 45 kD) is dnaN and not recF.