Barrie E. Davidson

Analysis of the DNA sequence, gene expression, origin of replication and modular structure of the Lactococcus lactis lytic bacteriophage sk1

Bacteriophage sk1 is a small isometric‐headed lytic phage belonging to the 936 species. It infects Lactococcus lactis, a commonly used dairy starter organism. Nucleotide sequence data analysis indicated that the sk1 genome is 28 451 nucleotides long and contains 54 open reading frames (ORFs) of 30 or more codons, interspersed with three large intergenic regions. The nucleotide sequence of several of the sk1 ORFs demonstrated significant levels of identity to genes (many encoding proteins of unknown function) in other lactococcal phages of both small isometric‐headed and prolate‐headed morphotype. Based on this identity and predicted peptide structures, sk1 genes for the terminase, major structural protein and DNA polymerase have been putatively identified. Genes encoding holin and lysin were also identified, subcloned into an Escherichia coli expression vector, and their function demonstrated in vivo. The sk1 origin of replication was located by identifying sk1 DNA fragments able to support the maintenance in L. lactis of a plasmid lacking a functional Gram‐positive ori. The minimal fragment conferring replication origin function contained a number of direct repeats and 179 codons of ORF47. Although no similarity between phage sk1 and coliphage λ at the nucleotide or amino acid sequence level was observed, an alignment of the sk1 late region ORFs with the λ structural and packaging genes revealed a striking correspondence in both ORF length and isoelectric point of the ORF product. It is proposed that this correspondence is indicative of a strong conservation in gene order within these otherwise unrelated isometric‐headed phages that can be used to predict the functions of the sk1 gene products.

Nucleotide sequence and transcription of the phenylalanine and tyrosine operons of Escherichia coli K12

A 4509 base-pair DNA fragment containing the phenylalanine and tyrosine operons of Escherichia coli K12 has been sequenced, and the pattern of transcription of these operons examined by S1 mapping, primer extension and galK fusion analyses. The phe operon consists of promoter, operator, leader region containing the phe attenuator and the pheA gene encoding chorismate mutase/prephenate dehydratase. The tyr operon consists of promoter, operator, a short leader region without an attenuator, and two structural genes aroF and tyrA encoding the tyrosine-sensitive isoenzyme of 3-deoxy-D-arabinoheptulosonate-7-phosphate (DAHP) synthetase and chorismate mutase/prephenate dehydrogenase, respectively. A bidirectional transcription terminator occurs between the two operons. The predicted amino acid sequences of chorismate mutase/prephenate dehydrogenase and chorismate mutase/prephenate dehydratase are homologous at their N termini, while the tyrosine-sensitive isoenzyme of DAHP synthetase is closely homologous to the phenylalanine-sensitive isoenzyme encoded by aroG.

TyrR protein of Escherichia coli and its role as repressor and activator

The TyrR protein regulates the expression of eight transcriptional units that comprise the TyrR regulon. In all but one case, regulation is by repression, while in two cases activation of expression can occur. Notwithstanding the fact that the TyrR protein contains an ATP‐binding domain and a helix‐turn‐helix DNA‐binding domain which are structurally homologous to domains of similar functions in proteins such as NifA, NtrC, DctD and XylR, it differs from them in a number of respects. It is not a part of a two‐protein component system and it lacks the amino‐terminal domain that is present on NtrC and DctD. It activates transcription from‘Eσ70’promoters but not from‘Eσ54’promoters. ATP binding seems to be essential for tyrosine‐mediated repression but not for activation. In addition, the activity of the TyrR protein is modulated by the binding of one or more of the aromatic amino acids. The consensus sequence for TyrR‐binding sites in DNA, referred to as TyrR boxes, is TGTAAAN6TTTACA. Tyrosine‐mediated repression occurs at operators containing a pair of adjacent boxes. These have unequal affinities for the TyrR protein. The box that overlaps the RNA polymerase binding site is only bound by TyrR in the presence of both ATP and tyrosine, and binding appears to involve co‐operativity between two TyrR protein dimers.

Genomic organization of lactic acid bacteria.

Current knowledge of the genomes of the lactic acid bacteria, Lactococcus lactis and Streptococcus thermophilus, and members of the genera Lactobacillus, Leuconostoc, Pediococcus and Carnobacterium is reviewed. The genomes contain a chromosome within the size range of 1.8 to 3.4 Mbp. Plasmids are common in Lactococcus lactis (most strains carry 4–7 different plasmids), some of the lactobacilli and pediococci, but they are not frequently present in S. thermophilus, Lactobacillus delbrueckii subsp. bulgaricus or the intestinal lactobacilli. Five IS elements have been found in L. lactis and most strains carry multiple copies of at least two of them; some strains also carry a 68-kbp conjugative transposon. IS elements have been found in the genera Lactobacillus and Leuconostoc, but not in S. thermophilus. Prophages are also a normal component of the L. lactis genome and lysogeny is common in the lactobacilli, however it appears to be rare in S. thermophilus. Physical and genetic maps for two L. lactis subsp. lactis strains, two L. lactis subsp. cremoris strains and S. thermophilus A054 have been constructed and each reveals the presence of six rrn operons clustered in less than 40% of the chromosome. The L. lactis subsp. cremoris MG1363 map contains 115 genetic loci and the S. thermophilus map has 35. The maps indicate significant plasticity in the L. lactis subsp. cremoris chromosome in the form of a number of inversions and translocations. The cause(s) of these rearrangements is (are) not known. A number of potentially powerful genetic tools designed to analyse the L. lactis genome have been constructed in recent years. These tools enable gene inactivation, gene replacement and gene recovery experiments to be readily carried out with this organism, and potentially with other lactic acid bacteria and Gram-positive bacteria. Integration vectors based on temperate phage attB sites and the random insertion of IS elements have also been developed for L. lactis and the intestinal lactobacilli. In addition, a L. lactis sex factor that mobilizes the chromosome in a manner reminiscent to that seen with Escherichia coli Hfr strains has been discovered and characterized. With the availability of this new technology, research into the genome of the lactic acid bacteria is poised to undertake a period of extremely rapid information accrual.

Conservation of gene arrangement and an unusual organization of rRNA genes in the linear chromosomes of the Lyme disease spirochaetes Borrelia burgdorferi, B. garinii and B. afzelii

Physical maps of the chromosomes of the Lyme disease spirochaetes Borrelia garinii and Borrelia afzelii have been elucidated for the enzymes CspI, SgrAI, I-CeuI, SmaI, EagI, BssHII, MluI and ApaI by two-dimensional pulsed-field gel electrophoresis techniques. The maps contain 42 sites for B. garinii and 32 for B. afzelii. The mapping studies showed that the two chromosomes are linear DNA molecules of 953 and 948 kbp, respectively. A comparison of the physical maps of B. garinii and B. afzelii and the published map of the other Lyme disease spirochaete, Borrelia burgdorferi [Davidson, B.E., MacDougall, J. & Saint Girons, I. (1992) J Bacteriol 174, 3766-3774] revealed that the three chromosomes have few endonuclease sites in common, apart from a cluster in rrl (encoding 23S rRNA) and rrs (encoding 16S rRNA). Cloned borrelial genes were used as specific hybridization probes to construct genetic maps, using the physical maps as a basis. The resulting maps contain 41 genetic loci for B. burgdorferi, 39 for B. garinii, and 33 for B. afzelii. In contrast to the physical maps, the three genetic maps are closely related, with no detectable differences in gene order along the entire length of the chromosome. It is concluded that the chromosomes of these three borrelial species have undergone no major rearrangements, deletions or insertions during their evolution from a common ancestor. Detailed mapping of the region of the B. garinii and B. afzelii chromosomes that encodes rRNA revealed that each chromosome contains one copy of rrs separated by 5 kbp from two copies each of rrl and rrf (encoding 5S rRNA). (ABSTRACT TRUNCATED AT 250 WORDS)

Construction of plasmid vectors for the detection of streptococcal promoters

Plasmid vectors have been constructed for detecting DNA fragments that exhibit promoter activity in Streptococcus sanguis. The plasmids are able to replicate in both S. sanguis and Escherichia coli, and contain an erythromycin resistance marker which is expressed in both hosts. Selection for promoter activity is dependent upon the insertion of appropriate DNA fragments upstream from a promoterless chloramphenicol acetyl transferase gene (cat) from Staphylococcus aureus. To facilitate this insertion, a pair of vectors, pMU1327 and pMU1328, were constructed with the polylinker from M13mp 18 in either orientation. The to transcriptional terminator of phage lambda is present downstream from cat. Translation stop codons in all reading frames are located between the polylinker and the initiation codon of cat. These plasmids have been used to isolate DNA fragments from S. sanguis, S. lactis and S. cremoris that exhibit promoter activity in S. sanguis.

Inhibition of Listeria monocytogenes by piscicolin 126 in milk and Camembert cheese manufactured with a thermophilic starter

The effect of bacteriocin, piscicolin 126, on the growth of Listeria monocytogenes and cheese starter bacteria was investigated in milk and in Camembert cheese manufactured from milk challenged with 102 cfu ml−1L. monocytogenes. In milk incubated at 30°C, piscicolin 126 added in the range of 512–2048 AU ml−1 effectively inhibited growth of L. monocytogenes for more than 20 d when challenged with approximately 102 cfu ml−1L. monocytogenes. At higher challenge levels (104 and 106 cfu ml−1), piscicolin 126 reduced the viable count of L. monocytogenes by 4–5 log units immediately after addition of the bacteriocin ; however, growth of Listeria occurred within 24 h. The minimum inhibitory concentration (MIC) of piscicolin 126 against lactic acid cheese starter bacteria was generally greater than 204 800 AU ml−1, and the viable count and acid production of these starter cultures in milk were not affected by the addition of 2048 AU ml−1 piscicolin 126. Camembert cheeses made from milk challenged with L. monocytogenes and with added piscicolin 126 showed a viable count of L. monocytogenes 3–4 log units lower than those without piscicolin 126. Inactivation of piscicolin 126 by proteolytic enzymes from cheese starter bacteria and mould together with the emergence of piscicolin 126‐resistant isolates was responsible for the recovery of L. monocytogenes in the cheeses during ripening.

Detection of bacteriocins of lactic acid bacteria isolated from foods and comparison with pediocin and nisin

A total of 663 533 colonies from 72 dairy and meat sources showed a detection rate of 0·2% for bacteriocin producers using direct plating techniques. A further 83 000 colonies from 40 fish and vegetable sources showed a detection rate of 3·4% for bacteriocin producers using selective enrichment procedures. A collection of seven purified isolates showing a different host spectrum of bacteriocin activity and with the ability to produce bacteriocins in broth culture were compared with nisin and pediocin (with respect to their inhibitory activity, determined by the critical dilution method), against various indicator bacteria in agar and broth. The sensitivity of Listeria species to various bacteriocins was influenced by the agar and broth test systems used. A Lactobacillus curvatus strain was found to be the most suitable indicator for quantitating antimicrobial effects of all the bacteriocins investigated in both agar and broth test systems. The bacteriocin‐producing isolates were characterized by biochemical reactions and DNA restriction enzyme profiles and taxonomic identification revealed species of Lactobacillus, Carnobacterium and Lactococcus assigned on the basis of 16S rDNA sequences.

Nucleotide sequence of feline panleukopenia virus: comparison with canine parvovirus identifies host-specific differences.

The nucleotide sequence of feline panleukopenia virus (FPV) strain 193 was determined and compared with the sequence of canine parvovirus (CPV) strain N and partial sequences of FPV strain Carl and CPV strain b. Base differences were identified at 115 positions in these 5.1 kb genomes and predicted amino acid differences occurred at 40 positions. The two overlapping capsid protein genes contained almost twice as many base differences as the single non-structural protein gene (49 compared to 26) and about the same ratio was calculated for predicted amino acid differences (27 compared to 13). The 27 variant amino acids in the capsid proteins were clustered at three sites in the primary sequence, whereas 10 of the 13 variant amino acids in the non-structural protein occurred in the 130 C-terminal amino acids. The two FPV strains differed consistently from the two CPV strains at 31 bases: 12 base changes in the capsid protein genes resulted in six amino acid changes, six base changes in the non-structural protein gene resulted in three amino acid changes, and 13 base changes occurred in the non-coding sequence.

Current research on the genetics of lactic acid production in lactic acid bacteria

Abstract Lactic acid derived from lactose is a major by-product of energy production in lactic acid bacteria. The uptake of lactose by these organisms is mediated either by the lactose phosphoenolpyruvate-phosphotransferase system (lactose PEP-PTS), or by a lactose-proton symport system. The disaccharide is then converted to lactate with the concomitant production of ATP. In Lactococcus lactis the genes encoding the lactose PEP-PTS, phospho-β-galactosidase and the tagatose 6-phosphate pathway enzymes are plasmid encoded, while other genes required for lactate synthesis, including those of the Embden-Meyerhof pathway, are on the chromosome. We have compiled a current list of genes required for lactate synthesis in the lactic acid bacteria that have been cloned and characterized and discuss the present status of genetic research in this area. The analyses of the L. lactis lac operon have yielded one of the most detailed pictures of genetic regulation in this bacterium. The operon has been fully sequenced, the regulatory protein LacR which represses lac operon transcription has been purified and its properties determined, and the operon promoters and operators have been identified. Investigations of chromosomally encoded L. lactis genes have resulted in the identification and characterization of pfk, pyk, ldh, tpi and gap , which encode phosphofructokinase, pyruvate kinase, l -(+)-lactate dehydrogenase, triosephosphate isomerase and glyceraldehyde-3-phosphate dehydrogenase, respectively. All of these enzymes (except triosephosphate isomerase) are known from previous studies to be important in metabolite level regulation of the pathway, pfk, pyk and ldh are organized into a tricistronic operon (the las operon), while tpi and gap are in monocistronic units. The las operon is so far unique to L. lactis . A number of investigators have studied the effect of gene dosage on glycolytic flux in lactic acid bacteria and their results are reviewed. We have introduced multiple copies of pfk, pyk, Idh and the las operon into L. lactis and report the effect of the increase in gene dosage on enzyme levels and the rate of lactic acid production.