Isolation of DNA from agarose gels using DEAE-paper. Application to restriction site mapping of adenovirus type 16 DNA

Influence of ionic constituents and electrical conductivity on the propagation of charged nanoscale objects in passivated gel electrophoresis

When determining the electric field E acting on charged objects in gel electrophoresis, the electrical conductivity of the buffer solution is often overlooked; E is typically calculated by dividing the applied voltage by a separation distance between electrodes. However, as a consequence of electrolytic reactions, which occur at the electrodes, gradients in the ionic content of the buffer solution and its conductivity can potentially develop over time, thereby impacting E and affecting propagation velocities of charged objects, v, directly. Here, we explore how the types and concentrations of ionic constituents of the buffer solution, which largely control its conductivity, when used in passivated gel electrophoresis (P-gelEP), can influence E, thereby altering v of charged nanospheres propagating through large-pore gels. We measure the conductivity of the buffer solution in the center of the gel region near propagating bands of nanospheres, and we show that predictions of E based on conductivity closely correlate with v. We also explore P-gelEP involving two different types of passivation agents: nonionic polyethylene glycol (PEG) and anionic sodium dodecyl sulfate (SDS). Our observations indicate that using a conductivity model to determine E from the local current density and the conductivity where spheres are propagating can lead to a better estimate than the standard approach of a voltage divided by a separation. Moreover, this conductivity model also provides a starting point for interpreting the complex behavior created by amphiphilic ionic passivation agents, such as SDS, on propagating nanospheres used in some P-gelEP experiments.

A chromosomally encoded type III secretion pathway in Yersinia enterocolitica is important in virulence

Numerous Gram-negative bacteria use a type III, or contact dependent, secretion system to deliver proteins into the cytosol of host cells. All of these systems identified to date have been shown to have a role in pathogenesis. We have identified 13 genes on the Yersinia enterocolitica chromosome that encode a type III secretion apparatus plus two associated putative regulatory genes. In order to determine the function of this chromosomally-encoded secretion apparatus, we created an in frame deletion of a gene that has homology to the hypothesized inner membrane pore, ysaV. The ysaV mutant strain failed to secrete eight proteins, called Ysps, normally secreted by the parental strain when grown at 28 degrees C in Luria-Bertani (LB) broth supplemented with 0.4 M NaCl. Disruption of the ysaV gene had no effect on motility or phospholipase activity, suggesting this chromosomally encoded type III secretion pathway is distinct from the flagella secretion pathway of Y. enterocolitica. Deletion of the ysaV gene in a virulence plasmid positive strain had no effect on in vitro secretion of Yops by the plasmid-encoded type III secretion apparatus. Secretion of the Ysps was unaffected by the presence or absence of the virulence plasmid, suggesting the chromosomally encoded and plasmid-encoded type III secretion pathways act independently. Y. enterocolitica thus has three type III secretion pathways that appear to act independently. The ysaV mutant strain was somewhat attenuated in virulence compared with the wild type in the mouse oral model of infection (an approximately 0.9 log difference in LD50). The ysaV mutant strain was nearly as virulent as the wild type when inoculated intraperitoneally in the mouse model. A ysaV probe hybridized to sequences in other Yersinia spp. and homologues were found in the incomplete Y. pestis genome sequence, indicating a possible role for this system throughout the genus.

Cloning and characterization of the Bg/II restriction-modification system reveals a possible evolutionary footprint

Bg/II, a type II restriction-modification (R-M) system from Bacillus globigii, recognizes the sequence 5'-AGATCT-3'. The system has been cloned into E. coli in multiple steps: first the methyltransferase (MTase) gene, bglIIM, was cloned from B. globigii RUB561, a variant containing an inactivated endonuclease (ENase) gene (bglIIR). Next the ENase protein (R.BglII) was purified to homogeneity from RUB562, a strain expressing the complete R-M system. Oligonucleotide probes specific for the 5' end of the gene were then synthesized and used to locate bglIIR, and the gene was isolated and cloned in a subsequent step. The nucleotide sequence of the system has been determined, and several interesting features have been found. The genes are tandemly arranged, with bglIIR preceding bglIIM. The amino acid sequence of M.BglII is compared to those of other known MTases. A third gene encoding a protein with sequence similarity to known C elements of other R-M systems is found upstream of bglIIR. This is the first instance of a C gene being associated with an R-M system where the R and M genes are collinear. In addition, open reading frames (ORFs) resembling genes involved with DNA mobility are found in close association with BglII. These may shed light on the evolution of the R-M system.

Carboxy-terminal sequence divergence and processing of the polyprotein antigen from Dirofilaria immitis

A polyprotein composed of multiple units arranged in direct tandem arrays has been identified in parasitic and free living nematodes. Analysis of previously cloned units from the Dirofilaria immitis polyprotein antigen (DiPA) indicated the units were nearly identical but here we demonstrate that they segregate into two related families. The consensus repeats, DiPA-CR1 and CR2, derived for each family are 80% identical. However, the repeats at the C-terminus of the polyprotein have diverged from DiPA-CR1 and CR2. This was shown by DNA sequence and Southern blot analysis of a 1.9 kb cDNA clone that encodes 4.4 C-terminal repeats (DiPA-TR1 through TR5). DiPA-TR3 through TR5 show 27-52% amino acid identity with the consensus repeats and 31-35% amino acid identity with one another. Metabolic labeling studies have shown that cleavage of DiPA generates a protein "ladder' from 14 to > 200 kDa. RRKR, a cleavage motif of subtilisin-like proprotein convertases, was identified as the natural cleavage site. In vitro digestion experiments with proteinase K suggest a structural model for DiPA consisting of protease resistant cores joined by protease sensitive linkers containing the RRKR site. This motif is absent between DiPA-TR3 and TR4 and has been altered to KR between DiPA-TR4 and TR5. An immunoblot of D. immitis extract probed with anti-DiPA-TR4/5 serum demonstrates the absence of cleavage at these sites. These divergent repeats provide an opportunity to investigate processing of the D. immitis polyprotein in vivo.

Identification of the galE gene and a galE homolog and characterization of their roles in the biosynthesis of lipopolysaccharide in a serotype O:8 strain of Yersinia enterocolitica

A clone that complements mutations in Yersinia enterocolitica lipopolysaccharide (LPS) core biosynthesis was isolated, and the DNA sequence of the clone was determined. Three complete open reading frames and one partial open reading frame were located on the cloned DNA fragment. The first, partial, open reading frame had homology to the rfbK gene. The remaining reading frames had homology to galE, rol, and gsk. Analysis of the galE homolog indicates that although it can complement an Escherichia coli galE mutant, its primary function in Y. enterocolitica is not in the production of UDP galactose but, instead, some other nucleotide sugar required for LPS biosynthesis. This gene has been renamed lse, for LPS sugar epimerase. The rol homolog has been demonstrated to have a role in Y. enterocolitica serotype 0:8 O-polysaccharide antigen chain length determination. An additional galE homolog has been identified in Y. enterocolitica by homology to the E. coli gene. The product of this gene has UDP galactose 4-epimerase activity in both E. coli and Y. enterocolitica. This gene is linked to the other genes of the galactose utilization pathway, similar to what is seen in other members of the family Enterobacteriaceae. Although Y. enterocolitica 0:8 strains are reported to have galactose as a constituent of LPS, a strain containing a mutation in this galE gene does not exhibit any LPS defects.

Mutations affecting lipopolysaccharide enhance ail-mediated entry of Yersinia enterocolitica into mammalian cells

Two genes of Yersinia enterocolitica, inv and ail, have been identified as having a role in the bacterial adherence to and entry into mammalian cells in vitro. Expression of both genes is regulated by temperature. In stationary phase, ail gene expression is detectable only in bacteria at 37 degrees C, not at lower temperatures. An inv mutant derivative of Y. enterocolitica, which cannot enter mammalian cells when grown at 30 degrees C because of the lack of both inv and ail gene products, was mutagenized with the transposons mini-Tn10 and Tn5B50 to look for an increase in Ail-mediated cell entry. Sixteen mutants that could enter tissue culture cells after growth at 30 degrees C were selected. All of the mutants had increased cell surface Ail levels as detected by an Ail-specific monoclonal antibody. All of the ten Tn5B50 and one of the six mini-Tn10 mutants showed no increase in ail expression, but they had alterations in their lipopolysaccharide (LPS) such that no O side chains were detectable in bacteria grown at 30 degrees C. Thus, these mutants that are increased in their ability to enter cells appear to be so as a result of a change in the LPS on the surface resulting in increased levels of Ail protein able to interact with the mammalian cell surface. In the remaining mini-Tn10 mutants, LPS is normal, and the increase in cell surface Ail levels appears to be due to an increase in ail mRNA present in the cell. These mutants may therefore be affecting a repressor of ail gene expression.

Detection and characterization of cyanobacterial nifH genes

The DNA sequence of a 359-bp fragment of nifH was determined for the heterocystous strains Anabaena sp. strain CA (ATCC 33047), Nostoc muscorum UTEX 1933, a Nostoc sp., Gloeothece sp. strain ATCC 27152, Lyngbya lagerheimii UTEX 1930, and Plectonema boryanum IU 594. Results confirmed that the DNA sequence of the 359-bp segment is sufficiently variable to distinguish cyanobacterial nifH genes from other eubacterial and arachaeobacterial nifH genes, as well as to distinguish heterocystous from nonheterocystous nifH genes. Nonheterocystous cyanobacterial nifH sequences were greater than 70 and 82% identical on the DNA and amino acid levels, respectively, whereas corresponding values for heterocystous cyanobacterial nifH sequences were 84 and 91%. The amplified nifH fragments can be used as DNA probes to differentiate between species, although there was substantial cross-reactivity between the nifH amplification products of some strains. However, an oligonucleotide designed from a sequence conserved within the heterocystous cyanobacteria hybridized primarily with the amplification product from heterocystous strains. The use of oligonucleotides designed from amplified nifH sequences shows great promise for characterizing assemblages of diazotrophs.

Plasmodium falciparum genetic diversity can be characterised using the polymorphic merozoite surface antigen 2 (MSA-2) gene as a single locus marker

The genetic diversity of Solomon Island Plasmodium falciparum isolates was examined using MSA-2 as a single locus marker. Amplification of MSA-2 gene fragments showed size polymorphism and the presence of mixed infections. Sequence analysis indicated a global representation of MSA-2 alleles with representatives of 3D7/CAMP allelic subfamilies and the FCQ-27 allelic family being identified. A simplified method of characterisation, utilising PCR-RFLPs of MSA-2 gene fragments, was developed. The RFLPs allowed identification of allelic families and further distinction within the 3D7/CAMP family. The amplification of MSA-2 gene fragments from culture derived lines revealed a loss of diversity for a number of Solomon Island isolates. Genomic diversity was confirmed for Solomon Island lines, along with Papua New Guinean and Thai lines, by the generation of 7H8/6 fingerprints. All lines were distinct and band sharing frequencies and Wagner tree construction failed to identify any geographic clustering.

Comparison of properties of agarose for electrophoresis of DNA

Agarose as a medium for separation of DNA was first introduced in 1962 and since the early 1970s agarose submarine gel electrophoresis has been synonymous with separations of DNA molecules larger than 1 kilobase pair (kb). The large pore size, low electroendosmosis and strength of the matrix have advantages over other media such as polyacrylamide for many applications. The variety of grades of agarose, developed by chemical manipulation of the substitutions on the agarose polymer, provides a range of matrices for separation of DNA molecules from a few base pairs (bp) to over 5 megabase pairs (Mb) in length. The introduction of low-melting-temperature agarose has revolutionised the extraction and manipulation of chromosome-sized molecules. On the other hand, the demand for analysis of very small quantities of DNA will most likely lead to the increasing importance of capillary electrophoresis. Many theories have been propounded to explain the electrophoretic migration of DNA in agarose. The most popular of these has been reptation theory but none can account for all of the reported anomalies in migration. However, anomalous migration has been exploited to study DNA structure, topology and catenation. An example of the use of two-dimensional electrophoresis to demonstrate the complexity of DNA migration through agarose is given. Generally, for molecules smaller than 50 kb, electrophoretic separation is a function of length. By alternately electrophoresing DNA in two different directions, molecules as large as 5.7 Mb have been effectively separated, although with such large molecules DNA structure as well as size may determine migration. In the case of separations of chromosomes from the intestinal protozoan, Giardia duodenalis, for example, a discrepancy of 1 Mb in the size of one chromosome, with an apparent size of 0.7-2.0 Mb, depended on the boundary conditions of separation. Major challenges for the molecular biologist are separation of larger chromosomal sized molecules, greater number of samples and smaller formats. Towards this challenge computer-aided technology is a key component in the control of electrophoresis parameters and analysis.

Methods for recovering nucleic acid fragments from agarose gels

Agarose gel electrophoresis is a powerful technique for the separation of nucleic acids on the basis of their size and conformation. The development of methods to recover size-fractionated nucleic acids molecules from agarose gels has greatly facilitated recombinant DNA technologies. Although several methods for recovering DNA and RNA molecules have been developed during the past fifteen years, none of them has been universally accepted. In this review we describe, discuss and evaluate the most common procedures with which we have had experience. Our evaluation is based on the criteria of yield, purity, speed, simplicity and low cost. We have considered three different approaches to the problem of recovering nucleic acids: chemical gel dissolution, physical gel disruption and physical extrusion from intact gels.

The ail gene of Yersinia enterocolitica has a role in the ability of the organism to survive serum killing

Two Yersinia enterocolitica genes, inv and ail, play a major role in the ability of this microorganism to enter cultured mammalian cells. ail-homologous sequences are present only in pathogenic species and strains of Yersinia. We previously demonstrated (D. E. Pierson and S. Falkow, Infect. Immun. 58:1059-1064, 1990) that four different nonpathogenic isolates of Y. enterocolitica are not able to invade tissue culture cells because they contain functionally inactive variants of the inv gene. When a functional version was introduced into these strains, they became invasive. In this study, we introduced a functional ail gene into the same strains and found that the ail gene was expressed but that these strains neither adhere to nor invade cultured animal cells. However, these recombinant strains became resistant to killing by human serum, whereas their parental strains were not. Using an ail mutant, we also demonstrate that the ail gene has a role in both invasion/adherence and serum resistance in a pathogenic isolate of Y. enterocolitica. These results support a role for Ail in the pathogenesis of Y. enterocolitica infection and disease.

Copy number and location of insertion sequences ISS1 and IS981 in lactococci and several other lactic acid bacteria

Genomic DNA from 49 lactococcal strains was screened by Southern hybridization for the presence and relative copy number of lactococcal insertion sequence ISS1: ISS1 was found in 47 of 49 strains giving 1 to 20 hybridizing bands per strain. Southern hybridizations of undigested plasmid DNA from 17 lactococcal strains probed with ISS1 and IS981 showed that ISS1 was present on plasmids in all 17 strains, whereas IS981 was present on plasmids in 14 of the 17 strains. Both insertion sequences were present primarily on larger plasmids (> 25 kb), and some plasmids contained copies of both insertion sequences. When probed with ISS1, Southern hybridizations of DNA isolated from Lactococcus lactis ssp. lactis ML3 frozen stock culture and from isolated colonies showed that the stock culture consisted of a mixture of cells having different ISS1-hybridizing bands, indicating that stock cultures may contain cells with varying locations of ISS1 sequences. The number of copies and their widespread distribution among lactococcal strains establish that insertion sequences will contribute significantly to genotypic and phenotypic events that may affect the industrial performance and stability of lactococcal strains.

Identification and analysis of all components of a gel retardation assay by combination with immunoblotting

A better method was developed for analysis and identification of protein and DNA components of gel-shift assays. The protein-DNA complexes, separated in polyacrylamide gels, were transferred onto stacked nitrocellulose and anion-exchange membranes. The proteins, bound to nitrocellulose, were identified by immunoblotting, while the DNA, which bound only to the anion-exchange membrane, was detected by autoradiography. The technique readily identified thyroid hormone receptors interacting with response elements representing inverted or direct repeats of the consensus half-site AGGTCA. In addition, specific antisera identified both the thyroid hormone and the retinoic acid receptors in heterodimeric complexes. Adding a third membrane and digoxigenin-labeled DNA probes allowed separate detection of [125I]T3 (labeled 3,5,3'-L-triiodothyronine), DNA, and protein from a single gel-shift reaction. The usefulness of this technique was also demonstrated by detecting the transcription factors P75gag-v-erbA and Jun in shifted complexes. Finally, proteins and DNA transferred to anion-exchange membranes can be eluted and subjected to further study. The combination of the gel shift and the immunoblot approaches (called "Shift-Western blotting") allows identification of the individual components of protein-DNA complexes containing multiple transcription factors, their cognate DNA elements and, when applicable, also the ligand.

The genomic organization and DNA sequence of the mouse vas deferens androgen-regulated protein gene

The gene for mouse vas deferens protein (MVDP) is expressed, under androgenic control, exclusively in the epithelial cells of the deferent duct. As a first step in correlating cell-specific and hormonal regulations with the structure of the gene, the complete sequence of the MVDP gene (11 kb) and 0.5 kb of the 5' flanking region have been determined. The size range for the 10 exons is 78 to 168 bp, whereas that of introns is 292 to 2833 bp. A major site of transcription is located on an A residue 46 nucleotides upstream from the A of the ATG initiation codon. A TATA (CATAA) box, a CAAT box, a GC-rich motif and a (5'-TGTTCT-3') element that closely resembles the consensus sequence of the androgen response elements are present in the 5' flanking region of the MVDP gene.

Cloning of a cuticular antigen that contains multiple tandem repeats from the filarial parasite Dirofilaria immitis

An unusual antigen composed of tandemly repeated protein units was cloned from the filarial parasite Dirofilaria immitis. The antigen was initially identified by screening a lambda gt11 cDNA library with serum from dogs immunized with irradiated D. immitis third-stage larvae. DNA sequence analysis of the cDNA clone, Di5, revealed a continuous open reading frame composed of two 399-base-pair repeats arranged in tandem. Southern blot analysis of genomic D. immitis DNA showed that the gene coding for Di5 is composed of a tandem array of 25-50 copies of this same 399-base-pair repeat. Antiserum raised against recombinant Di5 protein detected a protein "ladder," from about 14 to greater than 200 kDa with steps approximately 15 kDa apart, on immunoblots of D. immitis extract. Metabolic labeling of adult parasites with [35S]methionine showed that Di5 is synthesized as a large precursor that is subsequently cleaved to produce the ladder-like array. These results suggest that the characteristic ladder is created by proteolytic cleavage of the precursor at the same site in each monomer. The Di5 antigen was localized to the cuticle and hypodermis of adult D. immitis by immunoelectron microscopy. Both male and female parasites were found to release Di5 when cultured in vitro. DNA hybridization analysis demonstrated that Di5 is a member of a gene family present in many filarial parasites that infect both animal and human populations.

Replication and temperature-sensitive maintenance functions of lactose plasmid pSK11L from Lactococcus lactis subsp. cremoris

The replication region of pSK11L, the lactose plasmid of Lactococcus lactis subsp. cremoris (L. cremoris) SK11, was isolated on a 14.8-kbp PvuII fragment by shotgun cloning into an Escherichia coli vector encoding erythromycin resistance and selection for erythromycin-resistant transformants of L. lactis subsp. lactis (L. lactis) LM0230. Deletion analysis and Tn5 mutagenesis of the resulting plasmid (pKMP1) further localized the replication region to a 2.3-kbp ScaI-SpeI fragment. DNA sequence analysis of this 2.3-kbp fragment revealed a 1,155-bp open reading frame encoding the putative replication protein, Rep. The replication origin was located upstream of rep and consisted of an 11-bp imperfect direct repeat and a 22-bp sequence tandemly repeated three and one-half times. The overall organization of the pSK11L replicon was remarkably similar to that of pCI305, suggesting that pSK11L does not replicate by the rolling-circle mechanism. Like pSK11L, pKMP1 was unstable in L. lactis LM0230. Deletion analysis allowed identification of several regions which appeared to contribute to the maintenance of pKMP1 in L. lactis LM0230. pKMP1 was significantly more stable in L. cremoris EB5 than in L. lactis LM0230 at all of the temperatures compared. This stability was lost by deletion of a 3.1-kbp PvuII-XbaI fragment which had no effect on stability in L. lactis LM0230. Other regions affecting stability in L. cremoris EB5 but not in L. lactis LM0230 were also identified. Stability assays conducted at various temperatures showed that pKMP1 maintenance was temperature sensitive in both L. lactis LM0230 and L. cremoris EB5, although the plasmid was more unstable in L. lactis LM0230. The region responsible for the temperature sensitivity phenotype in L. lactis LM0230 was tentatively localized to a 1.2-kbp ClaI-HindIII fragment which was distinct from the replication region of pSK11L. Our results suggest that the closely related L. lactis and L. cremoris subspecies behave differently regarding maintenance of plasmids.

Amplification, cloning, and sequencing of a nifH segment from aquatic microorganisms and natural communities

By use of the polymerase chain reaction and degenerate oligonucleotide primers for highly conserved regions of nifH, a segment of nifH DNA was amplified from several aquatic microorganisms, including an N2-fixing bacterium closely associated with the marine filamentous cyanobacterium Trichodesmium sp., a heterotrophic isolate from the root/rhizome of the seagrass Ruppia maritima, and the heterocystous freshwater cyanobacterium Anabaena oscillarioides. nifH segments were amplified directly from DNA extracted from the rhizosphere of roots of the seagrass Halodule wrightii. The nifH fragments were then cloned and sequenced. The DNA and deduced amino acid sequences were compared with known sequences, revealing distinct differences between taxonomic groups. This technique was shown to be useful for (i) the detection of N2-fixing microorganisms and (ii) rapidly obtaining the DNA sequence of the nifH gene, which provides information about general taxonomic groups of N2-fixing microorganisms.

A simple, effective method for the construction of subtracted cDNA libraries

A simple method is described for the construction of subtracted cDNA libraries. The technique was used to create a human pancreatic tumor cDNA library that was screened using either hybridization with cDNA probes or antibodies. cDNA from a well-differentiated tumor cell line (CD-11) was subtracted against RNA from an undifferentiated tumor cell line (Panc-1). The subtracted cDNA was purified from RNA-cDNA hybrids by oligo-dA cellulose affinity chromatography. Single-stranded subtracted cDNA was used as a template for random primed second-strand synthesis using the Klenow's fragment of DNA polymerase. After ligation with Eco R1 adapters, cDNA was inserted into lambda gt11. A library of 140,000 primary pfu was obtained that contained 92% recombinants. A small portion of this library (40,000 pfu) was subjected to probe screening with a mucin cDNA probe known to be differentially expressed by CD-11 cells. The ratio of mucin cDNA clones to actin cDNA clones was increased by greater than 300-fold in the subtracted cDNA library compared to a standard cDNA library from the same cell line. The absolute number of mucin cDNA clones per 40,000 pfu was also increased 32-fold in the subtracted library. Pancreatic tumor mucin cDNAs were also identified in the subtracted library by antibody screening. The subtraction procedure yielded a 50-fold enrichment in differentially expressed cDNA detected by antibodies, compared to a nonsubtracted library from the same cell line.

Mutational hotspot in the p53 gene in human hepatocellular carcinomas

Human hepatocellular carcinomas (HCC) from patients in Qidong, an area of high incidence in China, in which both hepatitis B virus and aflatoxin B1 are risk factors, were analysed for mutations in p53, a putative tumour-suppressor gene. Eight of the 16 HCC had a point mutation at the third base position of codon 249. The G----T transversion in seven HCC DNA samples and the G----C transversion in the other HCC are consistent with mutations caused by aflatoxin B1 in mutagenesis experiments. No mutations were found in exons 5,6,8 or the remainder of exon 7. These results contrast with p53 mutations previously reported in carcinomas and sarcomas of human lung, colon, oesophagus and breast; these are primarily scattered over four of the five evolutionarily conserved domains, which include codon 249 (refs 4-9). We suggest that the mutant p53 protein may be responsible for a selective clonal expansion of hepatocytes during carcinogenesis.

Analysis of the lacZ sequences from two Streptococcus thermophilus strains: comparison with the Escherichia coli and Lactobacillus bulgaricus beta-galactosidase sequences

The lacZ gene from Streptococcus thermophilus A054, a commercial yogurt strain, was cloned on a 7.2 kb PstI fragment in Escherichia coli and compared with the previously cloned lacZ gene from S. thermophilus ATCC 19258. Using the dideoxy chain termination method, the DNA sequences of both lacZ structural genes were determined and found to be 3071 bp in length. When the two sequences were more closely analysed, 21 nucleotide differences were detected, of which only nine resulted in amino acid changes in the proteins, the remainder occurring in wobble positions of the respective codons. Only three bases separated the termination codon for the lacS gene from the initiation codon for lacZ, suggesting that the lactose utilization genes are organized as an operon. The amino acid sequence of the beta-galactosidase, derived from the DNA sequence, corresponds to a protein with a molecular mass of 116860 Da. Comparison of the S. thermophilus amino acid sequences with those from Lactobacillus bulgaricus, E. coli and Klebsiella pneumoniae showed 48, 35 and 32.5% identity respectively. Although little sequence homology was observed at the DNA level, many regions conserved in the amino acid sequence were identified when the beta-galactosidase proteins from S. thermophilus, E. coli and L. bulgaricus were compared.

Conservation of the gene for outer membrane protein OprF in the family Pseudomonadaceae: sequence of the Pseudomonas syringae oprF gene

The conservation of the oprF gene for the major outer membrane protein OprF was determined by restriction mapping and Southern blot hybridization with the Pseudomonas aeruginosa oprF gene as a probe. The restriction map was highly conserved among 16 of the 17 serotype strains and 42 clinical isolates of P. aeruginosa. Only the serotype 12 isolate and one clinical isolate showed small differences in restriction pattern. Southern probing of PstI chromosomal digests of 14 species from the family Pseudomonadaceae revealed that only the nine members of rRNA homology group I hybridized with the oprF gene. To reveal the actual extent of homology, the oprF gene and its product were characterized in Pseudomonas syringae. Nine strains of P. syringae from seven different pathovars hybridized with the P. aeruginosa gene to produce five different but related restriction maps. All produced an OprF protein in their outer membranes with the same apparent molecular weight as that of P.aeruginosa OprF. In each case the protein reacted with monoclonal antibody MA4-10 and was similarly heat and 2-mercaptoethanol modifiable. The purified OprF protein of the type strain P. syringae pv. syringae ATCC 19310 reconstituted small channels in lipid bilayer membranes. The oprF gene from this latter strain was cloned and sequenced. Despite the low level of DNA hybridization between P. aeruginosa and P. syringae DNA, the OprF gene was highly conserved between the species with 72% DNA sequence identity and 68% amino acid sequence identity overall. The carboxy terminus-encoding region of P. syringae oprF showed 85 and 33% identity, respectively, with the same regions of the P. aeruginosa oprF and Escherichia coli ompA genes.

Structure of the snake short-chain neurotoxin, erabutoxin c, precursor gene

A 2.4-kb genomic clone, which encodes the precursor of a snake neurotoxin, erabutoxin c, was isolated from the liver of the sea snake, Laticauda semifasciata. The erabutoxin c gene is about 1.2 kb long and consists of three exons and two introns. The first intron is found at the position corresponding to the signal peptide between amino acid residues 18(Leu) and 19(Gly). The second intron is located at the position corresponding to the central loop of the mature toxin, between amino acid residues 33(Arg) and 34(Gly). A TATA box consensus sequence, characteristic of promoter regions, is identified 29-33 nucleotides upstream from the transcription-initiation site identified by nuclease S1 analysis. The erabutoxin c cDNA nucleotide sequence, deduced from the present work, is compared with the cDNA sequences of erabutoxins a and b reported previously. Replacements are found at three positions, two of which correspond to the amino acid replacements found among the toxins, while the other is silent.

The level of a transcript required for production of a Streptomyces coelicolor antibiotic is conditionally dependent on a tRNA gene

In Streptomyces coelicolor A3(2), bldA mutants are conditionally defective in aerial mycelium formation and fail to synthesize all four antibiotics produced by bldA+ strains. Previous studies showed that bldA specifies the tRNA for the rarely used leucine codon UUA. Here we describe experiments examining the abundance in a bldA mutant of a transcript involved in antibiotic production. With use of a bacteriophage-based integrative vector, a promotorless xylE reporter gene was inserted into a previously undescribed gene for an early step in biosynthesis of the red antibiotic undecylprodigiosin, located in the red gene cluster. With this transcriptional fusion present at unit copy number in the chromosome, xylE expression in a bldA+ strain was maximal late in growth in a liquid production medium and was virtually absent in a bldA mutant. On plates of a different medium, the bldA mutant was able to produce undecylprodigiosin and to express the red::xylE fusion, but both abilities were repressed by increasing the concentration of phosphate in the medium. These experiments showed that the undecylprodigiosin deficiency of bldA mutants cannot be accounted for by the presence of TTA codons in the red structural genes, but rather that bldA influences red gene mRNA abundance. In low-phosphate conditions, an alternative regulatory pathway can lead to red gene expression.

Growth-dependent modulation of type I collagen production and mRNA levels in cultured human skin fibroblasts

Five human skin fibroblast lines were studied for type I collagen production and type I procollagen mRNA levels through the different growth phases. The cells were plated at low density and followed for 11 days at daily intervals through the stages of rapid growth and visual confluency until the cultures reached stationary growth phase. Each day one culture flask was labeled with [3H]proline for 24 h, and analyzed for production of radiolabeled type I collagen into culture medium. The cell layers were counted and subjected to isolation of cytoplasmic RNA and determination of type I procollagen mRNA levels. The results revealed an approx. 2-fold increase in procollagen production and mRNA levels when the cells reached visual confluency. Thereafter the synthesis rates and mRNA levels remained relatively constant, although a decreasing tendency of both parameters was observed upon further culturing. The results confirm that determination of cell density is important when cell cultures are used for measurement of collagen synthesis or mRNA levels. For determination of pro alpha 2(I) collagen mRNA an 1193 bp cDNA clone was constructed using RNA extracted from human fetal calvaria. Sequencing of the clone revealed some nucleotide and amino acid differences between the previously published sequences. This suggests the presence of more individual variation in procollagen coding sequences than expected.

Antibodies against the large subunit of the EBV-encoded ribonucleotide reductase in patients with nasopharyngeal carcinoma

The open reading frame corresponding to BORF2 and encoding the large subunit of the Epstein-Barr virus (EBV) ribonucleotide reductase has been inserted into the prokaryotic expression vector pUC19. A 90-kDa protein was produced when the intact plasmid was used as a template for in vitro DNA-directed protein synthesis. Using templates generated by restriction digests within the BORF2 open reading frame, in the same system, truncated polypeptides resulted confirming the identity of the 90-kDa protein. The protein was then produced in a heterologous expression system and used in protein immunoblotting to screen for antibodies in sera from nasopharyngeal carcinoma (NPC), Burkitt's lymphoma (BL) or control subjects. Twenty out of 33 NPC sera were positive for antibodies against the large subunit, 13 of these were positive for both IgG and IgA, whilst 7 were positive for IgG only. Out of 15 BL sera and 10 control sera, none were positive. These results are similar to those observed for other EBV-encoded enzymes, including the DNase which had been used as an early marker for the development of NPC. The results presented here indicate that antibodies against the large subunit of ribonucleotide reductase could serve as an additional marker for NPC.

Nonpathogenic isolates of Yersinia enterocolitica do not contain functional inv-homologous sequences

Previous studies have demonstrated a correlation between the ability of isolates of Yersinia enterocolitica to cause disease and to invade tissue culture cells in vitro. Two genes, inv and ail, isolated from a pathogenic strain of Y. enterocolitica have each been shown to confer this invasive phenotype upon Escherichia coli. Eighty pathogenic, invasive isolates studied by Miller et al. (Infect. Immun. 57:121-131, 1989) contained sequences homologous to both of these genes. Thirty-five nonpathogenic, noninvasive isolates similarly studied had no ail homology but carried inv-homologous sequences. We investigated inv-homologous sequences from four nonpathogenic isolates. Recombinant clones of these inv-homologous sequences did not confer the invasive phenotype upon E. coli. No RNA transcripts capable of encoding a full-length Inv protein were detected in the four noninvasive Yersinia strains. When the inv gene from a pathogenic isolate was introduced into two of these strains, the resulting transformants invaded tissue culture cells in vitro. The inv gene was transcribed in a pathogenic Yersinia isolate grown at 30 degrees C but not at all in these cells grown at 37 degrees C. The production of RNA transcripts homologous to inv in transformants was not regulated by temperature to the same degree as was seen for pathogenic isolates. We conclude that the inv gene in nonpathogenic strains of Y. enterocolitica is nonfunctional. Y. enterocolitica isolates epidemiologically linked to disease contain both a functional inv gene and a functional ail gene. Environmental isolates not associated with disease have a nonfunctional inv gene and no ail gene.

Pattern of transcription of the genome of equine infectious anemia virus

The pattern of expression of the equine infectious anemia virus (EIAV) genome in a persistently infected canine cell line was determined. Five EIAV-specific transcripts (8.2, 5.0, 4.0, 2, and 1.8 kilobases [kb]) were detected by using subgenomic restriction enzyme fragments of EIAV DNA and EIAV-specific oligonucleotides as probes. The 8.2-kb mRNA could be shown to represent viral genomic RNA, whereas the smaller transcripts were generated by splicing events. Evidence was obtained that indicated that each subgenomic RNA species shared a common 5'-splice donor. The 5.0-kb mRNA was found to be expressed at relatively low levels, was difficult to detect consistently, and appeared to be generated by a single splicing event which linked the 5' exon to the 3' region of pol. The 4.0-kb transcript was concluded to be the env mRNA on the basis of its hybridization pattern with the various probes and its abundance. The 2-kb species was found to be multiply spliced and was encoded by sequences derived from orf2 but was not detected by probes representing 3'-env/3'-orf sequences. The 1.8-kb species was shown to consist of sequences representing orf1, part of orf2, and the 3'-orf/env and may represent the message for the EIAV trans-activator gene.

Expression of the Cellulomonas fimi cellulase genes cex and cenA from the divergent tet promoters of transposon Tn10

A cartridge was constructed which contained the divergent tet promoters of transposon Tn10 between an exoglucanase gene (cex) and an endoglucanase gene (cenA) of Cellulomonas fimi. When carried in a broad-host-range vector, the cartridge gave expression of cex and cenA in Escherichia coli, Rhodobacter capsulatus and Klebsiella pneumoniae.

Wound-inducible nuclear protein binds DNA fragments that regulate a proteinase inhibitor II gene from potato

Deletion analysis from the 3' to the 5' end of the promoter region of the wound-inducible potato proteinase inhibitor IIK gene has identified a 421-base sequence at -136 to -557 that is necessary for expression. Utilizing DNA band-shift assays, a 10-base sequence within the 421-base region was found to bind a nuclear protein from wounded tomato leaves. This 10-base sequence is adjacent to an 8-base consensus sequence at -147 to -155 that is present in the promoter region of several elicitor-inducible genes from various other plants. The evidence suggests that a complex set of cis- and trans-acting elements within the -136 to -165 region of the potato IIK gene may be involved with the signaling mechanisms that regulate the inducibility of this gene in response to pest and pathogen attacks.

Identification of a Locus Upstream from the Hydrogenase Structural Genes That Is Involved in Hydrogenase Expression in Bradyrhizobium japonicum

A locus involved in the expression of the uptake hydrogenase system of Bradyrhizobium japonicum was identified adjacent to genes encoding the hydrogenase subunits. A cloned fragment of DNA was used to complement to autotrophy a Hup − putative regulatory mutant of B. japonicum . The mutant strain lacked hydrogenase activity and synthesized low levels of the large subunit of hydrogenase as determined by Western gels. Tn 5 -induced mutagenesis located the region within the fragment which was necessary for complementation of the mutant phenotype. The locus identified is adjacent to that encoding the small subunit of hydrogenase; its right border is <0.5 kilobase upstream from the hydrogenase transcriptional start site, and its left border is between 1 and 2.5 kilobases from that start site. However, the locus is outside the region previously shown to contain hup -related genes of B. japonicum . Thus, the localization of this gene describes a previously unidentified hup -related gene on a region of DNA not previously shown to contain hup -specific DNA.

Cloning and sequencing of immunoglobulin variable-region genes using degenerate oligodeoxyribonucleotides and polymerase chain reaction

A procedure is described for using the polymerase chain reaction (PCR) to amplify and clone the cDNA from mouse immunoglobulin (Ig) variable (V) regions. This method uses a set of universal 5'-oligodeoxyribonucleotide primers that are degenerate and allow for the amplification of Ig V-region sequences from gamma and mu heavy chains and from kappa light chains. Selective first-strand cDNA synthesis is performed using Ig constant region primers and then a PCR is achieved by using the appropriate universal 5'-primer. The universal Ig heavy-chain primer was used to amplify the V-region cDNA from gamma and mu isotypes and the universal light-chain primer was used to amplify three separate kappa light V-region sequences. This procedure was used to obtain Ig V-region gene sequences from hybridomas secreting IgG1/kappa, IgG2b/kappa and IgM/kappa isotypes.

Use of degenerate oligonucleotides for amplification of the nifH gene from the marine cyanobacterium Trichodesmium thiebautii

Trichodesmium spp. are marine filamentous, nonheterocystous, nitrogen-fixing cyanobacteria which are an important component of marine ecosystems. This organism has never been maintained in axenic culture, and there has remained some doubt as to the identity of the organism responsible for nitrogen fixation in Trichodesmium aggregates. By using degenerate oligonucleotide primers, it has been possible to amplify, clone, and sequence a segment of the nifH gene from a natural assemblage of Trichodesmium thiebautii. Examination of the DNA and presumed amino acid sequence shows that the gene is most closely related to that of Anabaena spp. and therefore is most likely a cyanobacterial nifH gene. The use of degenerate oligonucleotides, in concert with the polymerase chain reaction, can be a powerful tool for the cloning and sequencing of a variety of genes from microorganisms in the environment.

Expression of biologically active human chorionic gonadotropin and its subunits by recombinant vaccinia virus

Vaccinia virus (VV) expression vector was used to clone the genes for coding alpha and beta subunits of human chorionic gonadotropin (hCG). Recombinant viruses VSL3 and VSS1 containing these genes were selected as blue coloured plaques on the basis of co-expression of Escherichia coli beta-galactosidase in the infected cells. CV-1 cells when infected with VSL3 or VSS1 secreted 2.4 and 1.8 micrograms of alpha and beta hCG subunits, respectively, per 3 x 10(6) cells after 24 h of infection. The subunit proteins expressed individually had immunoreactivity with monoclonal and polyclonal antibodies specific to hCG. The subunit hormonal peptides associated with each other during co-infection to form the complete hCG dimer, which was biologically active as evident from the induction of steroidogenesis in a mouse Leydig cell system.

Molecular cloning of a gene for indole-3-acetamide hydrolase from Bradyrhizobium japonicum

A pLAFR1 cosmid genomic library of wild-type Bradyrhizobium japonicum J1063 was constructed. A cosmid clone designated pBjJ4, containing a 26-kilobase (kb) DNA insert, was identified as being able to confer the ability to convert alpha-naphthaleneacetamide acid on B. japonicum J1B7 Rifr, which cannot perform this conversion. The gene coding for the enzyme that converts alpha-naphthaleneacetamide to alpha-naphthaleneacetic acid was localized in the 3.5-kb region of pBjJ4 by recloning in plasmid pSUP202. The gene coding for the enzyme was also mapped by Tn5 insertion mutagenesis to a region of ca. 2.3 kb. When the gene was placed behind the lacZ promoter and used to transform Escherichia coli, a high level of expression of indole-3-acetamide hydrolase activity was found. Since there have been no reports of this activity in E. coli, we have thus confirmed that the gene cloned here is a structural gene for indole-3-acetamide hydrolase and have designated it as the bam (Bradyrhizobium amidehydrolase) gene. Southern hybridization with the central region of the bam gene indicated that a high degree of similarity exists among the bam gene, the iaaH gene from Pseudomonas savastonoi, and the tms-2 gene from Agrobacterium tumefaciens. The result suggests that there is a common origin for the gene that encodes the enzyme that catalyzes the biosynthesis of indoleacetic acid.

Isolation of two alleles of the b locus of Ustilago maydis

The basidiomycete Ustilago maydis is the causative agent of the disease corn smut. To be pathogenic, strains of U. maydis must be heterozygous for a locus called "b," which appears to control both pathogenicity and sexual development. Two alleles of the b locus of U. maydis were isolated by complementation and hybridization. The clones have the specificities of b1 and b2 alleles as demonstrated by their effects on the colony morphology and pathogenicity of haploid and diploid strains upon transformation. For example, nonpathogenic haploid and diploid strains of U. maydis carrying b2 alleles became pathogenic when transformed with the cloned b1 allele. Furthermore, an a2 b2 haploid strain could be transformed to an a2 b1 genotype by gene replacement using a DNA fragment containing a b1 allele and a selectable marker. The isolation of b alleles represents an important step toward understanding the control of dicaryon formation, dicaryon maintenance, and pathogenicity in U. maydis.

Nucleic acid hybridization for detection of cell culture-amplified adenovirus

A number of recombinant plasmids containing genomic segments of adenovirus were constructed. Seven cloned probes, as well as total adenovirus type 2 (Ad2) and Ad16 genomic DNA, were tested by a nucleic acid hybridization technique for sensitivity and specificity in detecting adenoviruses in infected cells. Adenovirus DNA was spotted onto a nitrocellulose filter and hybridized with 32P-labeled DNA probes. The probes, total Ad2 genomic DNA, and plasmid pAd2-H (containing the hexon gene from Ad2 DNA) all detected 10 reference serotypes of five genomic subgroups (A through E) with similar sensitivities. However, plasmid pAd2-H required less preparation time than did total Ad2 DNA. Probes pAd2-F (containing the fiber gene from Ad2) and pAd16-BD (containing the BamHI D fragment from Ad16) hybridized only with reference serotypes from the homologous subgroups (C and B, respectively). Of 101 patient isolates amplified in cells, pAd2-H detected 100% of all isolates from both the homologous and the heterologous subgroups. The detection rates for pAd2-F were 100% (subgroup C) and 3.6% (subgroups A, B, and D), and those for pAd16-BD were 100% (subgroup B) and 9.4% (subgroups A, C, and D). A commercial biotinylated product (Pathogene II) was also included in this study for comparison.

DNA probe technology for rapid detection of Haemophilus influenzae in clinical specimens

In a previous study, we reported that a 5-kilobase Haemophilus influenzae DNA fragment involved in penicillin-binding protein expression could be used as a probe for specific detection of H. influenzae strains (F. Malouin and L. E. Bryan, Mol. Cell. Probes 1:221-232, 1987). Here, we report the ability of this probe to detect H. influenzae in clinical specimens. In a bacterial dot experiment, there was strong hybridization of the 32P-labeled probe to nonencapsulated and serotype a through f H. influenzae strains. The detection of H. influenzae in body fluids was then evaluated by using pooled human serum, urine, cerebrospinal fluid, and sputum as dilution media for H. influenzae, Haemophilus aegyptius, Haemophilus parainfluenzae, and Escherichia coli cells. At 65 degrees C, the probe hybridized to H. influenzae and H. aegyptius (greater than or equal to 10(5) cells) in all fluids. There was no hybridization with the E. coli negative control, and H. parainfluenzae hybridized when greater than or equal to 10(7) cells were used. Experiments performed at 73 and 80 degrees C permitted elimination of H. parainfluenzae hybridization. The detection of H. influenzae in 232 sputa from patients with respiratory tract infections was very specific (96 to 97%) and sensitive (74 to 100%) when the total time of the procedure was sufficient (6 to 24 h) and when the experiments were performed at 80 degrees C. In addition, the probe detected three of three and four of four H. influenzae-infected cerebrospinal fluids and blood cultures, respectively, and did not react with pneumococcus- or streptococcus-infected cerebrospinal fluids. Finally, by using a small-scale procedure, the probe rapidly detected H. influenzae in cerebrospinal fluid and sputum specimens (4 and 8 h, respectively). These results imply prompt diagnosis of H. influenzae infections caused by nonencapsulated and serotype a through f strains.

Nucleotide sequence of the gene for the delta 5-3-ketosteroid isomerase of Pseudomonas testosteroni

The structural gene for the delta 5-3-ketosteroid isomerase of Pseudomonas testosteroni has been sequenced by the dideoxy method. The sequence obtained confirms the amino acid (aa) sequence of Benson et al. [J. Biol. Chem. 246 (1971) 7514-7525] at all but 5 aa residues of the 125-aa polypeptide. Amino acid residues 22, 24, 33, and 38, reported to be asparagines by Benson et al., are found to be encoded by aspartic acid codons. Amino acid residue 77, reported to be a glutamine by Benson et al., is encoded by a glutamic acid codon. The identification of aa 38 as aspartic acid, coupled with its presence in the active site, as indicated by previous affinity and photoaffinity-labeling studies and confirmed independently by x-ray crystallographic studies, strengthens the hypothesis that Asp-38 is the aa responsible for the 4 beta to 6 beta proton transfer which is part of the enzymatic reaction.

Two nuclear location signals in the influenza virus NS1 nonstructural protein

The NS1 protein of influenza A virus has been shown to enter and accumulate in the nuclei of virus-infected cells independently of any other influenza viral protein. Therefore, the NS1 protein contains within its polypeptide sequence the information that codes for its nuclear localization. To define the nuclear signal of the NS1 protein, a series of recombinant simian virus 40 vectors that express deletion mutants or fusion proteins was constructed. Analysis of the proteins expressed resulted in identification of two regions of the NS1 protein which affect its cellular location. Nuclear localization signal 1 (NLS1) contains the stretch of basic amino acids Asp-Arg-Leu-Arg-Arg (codons 34 to 38). This sequence is conserved in all NS1 proteins of influenza A viruses, as well as in that of influenza B viruses. NLS2 is defined within the region between amino acids 203 and 237. This domain is present in the NS1 proteins of most influenza A virus strains. NLS1 and NLS2 contain basic amino acids and are similar to previously defined nuclear signal sequences of other proteins.

The predicted DBL oncogene product defines a distinct class of transforming proteins

The DBL transforming gene was originally identified by transfection of NIH 3T3 cells with DNA from a human B-cell lymphoma. This gene was found to have arisen as a result of recombination of the 3' portion of the DBL protooncogene coding sequences with an unrelated segment of human DNA. It encodes a cytoplasmic protein that is equally distributed between cytosol and crude membrane fractions. To further characterize this transforming gene, a biologically active cDNA clone of the DBL transforming gene mRNA was isolated. Analysis of the sequence of the DBL oncogene cDNA revealed a long open reading frame that encodes a hybrid protein whose first 50 amino acids (at least) derive from a complete exon of a different locus. No significant homology with known oncogenes or any known protein sequences was demonstrated. The computer analysis of the predicted DBL protein indicated it is highly hydrophilic with no hydrophobic domains characteristic of a membrane-spanning region or signal peptide. Thus, the DBL oncoprotein is distinct among known transforming gene products.

DNA probe technology for detection of Haemophilus influenzae

A 5 kb Haemophilus influenzae DNA fragment involved in penicillin-binding proteins expression was used as a probe for specific detection of H. influenzae strains. The 32p-labeled probe specificity was assessed by hybridization to bacterial dots and 75 strains were tested. All H. influenzae (18) and H. aegyptius (1) strains reacted very strongly with the probe. The H. influenzae serotypes tested (a, b, and non-typable strains) did not differ in their hybridization. Some hybridization was also found with the 12 other Haemophilus species tested as well as other Pasteurellaceae such as Actinobacillus lignieresii and Pasteurella multocida. Two other less related species (Klebsiella ozaenae and Providencia stuartii) also showed low hybridization. The probe detected as low as 10(5)-10(6) H. influenzae cells and 0.1 microgram of DNA in a dot sensitivity test. Hybridization to electroblotted, digested DNA from different species which reacted in the bacterial dot test revealed strong hybridization to H. influenzae and H. aegyptius only. This DNA probe should prove useful for H. influenzae and possibly H. aegyptius detection due to its high specificity and sensitivity under stringent hybridization conditions.

Quantitative electroelution of oligonucleotides and large DNA fragments from gels and purification by electrodialysis

We have designed a new device [Biotrap (Elutrap in the U.S.A. and Canada), available from Schleicher & Schuell] for electroelution, -concentration , and -dialysis of DNA and other charged macromolecules above 5 kDa. In an electric field, the DNA migrates in an open channel out of the gel slice through a microporous membrane, BT2, into the trap section, where it is retained by a very dense, non-adsorbant, and inert membrane BT1. Specifically designed for use in an electric field, the matrix of this new membrane is much denser than that of dialysis membranes. In contrast to dialysis membranes, BT1 will not adsorb large DNA fragments nor allow passage of small DNA fragments when subjected to an electric field. In the absence of an electric field, BT1 and BT2 effectively seal the trap, maintaining the final elution volume of the purified sample. The trap can contain from 200-600 microliter and is collected from above with a pipet. In the experiments described here, 85-95% of oligonucleotides (14-mer) and large (150 kb) DNA fragments were recovered, independent of fragment length. The purity of the eluted DNA was demonstrated by restriction enzyme digestion, nick-translation, primer extension, end-labeling with polynucleotide kinase, and ligation. Electrodialysis was successfully used for the complete removal of common contaminants inhibiting the polynucleotide kinase reaction and for the removal of CsCl from DNA samples.

A yeast operator overlaps an upstream activation site

The product of the BAR1 gene of Saccharomyces cerevisiae is synthesized only in the a cell type and inactivates alpha-factor, the mating pheromone made by alpha cells. The MAT alpha 2 protein represses the transcription of a-cell-specific genes, including BAR1, in alpha and a/alpha diploid cells. Transcription of BAR1 in a cells in stimulated upon exposure to alpha-factor. Deletion analysis of the 5' noncoding region of the BAR1 gene revealed that the major upstream activation site (UAS) overlaps the 31 bp operator sequence required for MAT alpha 2 repression. This result has implications for the negative control of transcription in yeast. The deletion analysis also indicated that the sequence TGAAACA mediates alpha-factor stimulation.