A food-grade site-directed mutagenesis system for Streptococcus thermophilus LMG 18311.

AIMS: To develop a general method for site-directed mutagenesis in the dairy starter strain Streptococcus thermophilus LMG 18311 which does not depend on antibiotic-resistance genes or other selection markers for the identification of transformants. METHODS AND RESULTS: In a previous study, we demonstrated that Strep. thermophilus LMG 18311 can be made competent for natural genetic transformation by overexpression of the alternative sigma factor ComX. In the present study, we wanted to investigate whether the natural transformation mechanism of Strep. thermophilus LMG 18311 is efficient enough to make it feasible to perform site-directed mutagenesis in this strain without the use of a selection marker. Competent bacteria were mixed with a DNA fragment engineered to contain a nonsense and a frameshift mutation in the middle of the target gene (lacZ) and subsequently seeded on agar plates. By performing colony-lift hybridization using a digoxigenin-labelled oligonucleotide probe, we succeeded in identifying transformants containing the sought after mutation. CONCLUSIONS: By exploiting the natural transformability of Strep. thermophilus LMG 18311 and standard molecular methods, we have demonstrated that the genome of this bacterium can be altered at preselected sites without introduction of any foreign DNA. SIGNIFICANCE AND IMPACT OF THE STUDY: A food-grade site-directed mutagenesis system has been developed for Strep. thermophilus LMG 18311 that can be used by the dairy industry to construct starter strains with novel and/or improved properties.

c-Jun causes focus formation and anchorage-independent growth in culture but is non-tumorigenic.

The RCAS retroviral vector was used to express chicken and mouse cellular Jun proteins in chicken embryo fibroblasts. Both mouse and chicken proteins induced foci of transformed cells with low to moderate efficiency compared with viral Jun, but were as effective as the viral protein in promoting anchorage-independent growth. Viral Jun and a recombinant between viral and cellular Jun induced tumors in 1-day-old chicks; the cellular Jun proteins were uniformly non-tumorigenic.

Amino acid substitutions modulate the effect of Jun on transformation, transcriptional activation and DNA replication.

The retroviral oncogene v-jun and its cellular counterpart code for proteins that function as major components of the transcription factor complex AP-1. Jun proteins bind to the AP-1 consensus sequence as homodimers or heterodimers with members of the Fos protein family. This report compares the ability of viral and cellular Jun proteins (v-Jun and c-Jun) to activate transcription and to stimulate DNA synthesis. The effect of amino acid substitutions on cellular transformation is also described. In F9 cells c-Jun is a more effective transactivator than v-Jun, which carries two amino acid substitutions in the carboxy-terminal region that together down-regulate transactivation. The delta deletion, present in the amino-terminal region of v-Jun, does not affect transactivation in F9 cells; however, it does modulate the stimulation of DNA synthesis. When delta is deleted, the amino acid substitutions are without consequence on DNA synthesis. In the presence of delta the amino acid substitutions down-regulate DNA synthesis. Deletion of the Jun transactivation domain, which is required for cellular transformation, abolishes both transactivation and stimulation of DNA synthesis. We conclude that transformation, transactivation and stimulation of DNA synthesis all depend on the presence of the transactivation domain. The three functions are, however, not tightly correlated, and further work is needed to define the role of the biochemical activities of Jun in oncogenesis.

Efficient induction of fibrosarcomas by v-jun requires mutations in the DNA binding region and the transactivation domain.

v-jun is the transforming gene of ASV 17, a retrovirus isolated from a spontaneous chicken fibrosarcoma. There are three mutations in the viral Jun protein (v-Jun) as compared to its cellular progenitor c-Jun: a deletion in the transactivation domain (called delta) and two amino acid substitutions in and near the DNA binding region. The effect of each of these mutations on fibrosarcoma development is described. All three mutations contribute towards tumor formation, and their cumulative effect makes v-Jun more tumorigenic compared to Jun proteins that carry only one or two of the mutations. Viruses rescued from tumors induced by c-Jun carrying the two amino acid substitutions in the DNA binding region have increased transforming and tumorigenic potential. These increases are probably due to further mutations that result in the expression of a rearranged Jun protein. Taken together the results show that the evolution of the c-Jun oncoprotein to an efficient carcinogen requires mutations in the transactivation and DNA binding regions.

Purification and partial characterization of an IgM-like serum immunoglobulin from Atlantic salmon (Salmo salar).

An IgM-like immunoglobulin was isolated from pooled sera collected from healthy Atlantic salmon. The immunoglobulin was purified by means of gel filtration followed by ion-exchange chromatography. It eluted from the ion-exchange column as two distinct peaks, but the two IgMs seems to be identical in their molecular natures except for net charge. The molecular weight of the unreduced (native) IgM was determined to be approximately 800 kilo Daltons (kD) when estimated by dodecyl sulphate polyacrylamide gel electrophoresis (SDS-PAGE). However, when the molecular weight of the native IgM was estimated by gel filtration, a molecular weight of 1000 kD was obtained. Furthermore, the molecular weights of the heavy and the light chains were estimated by SDS-PAGE analysis to be about 72 and 27 kD respectively. The amount of IgM was found to range from 80 to 130 mg/100 ml serum. Isoelectric focusing demonstrated that the major part of the IgM molecules focused between pH 5 and pH 6.

Stability of the nucleotide sequence of the phosphoprotein gene of measles virus during lytic infections.

Three clones with cDNA inserts encoding large portions of the measles virus phosphoprotein mRNA were characterized and compared with a previously published sequence of the Edmonston strain of measles virus. The two cloned viruses were separated by more than 100 passages. Only one out of 1477 nucleotides differed in the two sequences reflecting a very low mutation rate of the phosphoprotein gene during dilute lytic passages. The discovery that a third reading frame in the phosphoprotein gene may code for a novel peptide chain in addition to the P and C peptides may explain some of the high stability of the gene. The new reading frame was accessed by a translational shift caused by insertion of one extra G at a particular site in one of three otherwise identical cDNA sequences. A discrepancy was also found between the presumably high error rate of viral RNA polymerases and the stability of nucleotides in which mutations would not lead to amino acid substitutions. A few errors in the previously published sequence were discovered and the corrections are presented.

Two species of the phosphoprotein mRNA with differently edited reading frames discovered in measles virus infected Vero cells. Brief report.

One cDNA clone representing the phosphoprotein mRNA sequence of the Edmonston strain of measles virus contained 4 G nucleotides at a particular position. Two other clones contained 3 G nucleotides at the same position. Otherwise the nucleotide sequence were identical. The mRNA with 3 G nucleotides codes for the 70 kD phosphoprotein. The mRNA with 4 G nucleotides may code for a putative new peptide with 231 aminoterminal amino acids in common with the P protein whereas the 68 carboxyterminal amino acids are different from any amino acid sequence of the phosphoprotein. Thus the consequence of the insertion of one additional G is a translational shift to a shorter open reading frame. There are several indications that the observation is of biological significance.

Stimulation of plasmalemma adenosine triphosphatase from oat roots by inorganic and organic cations: concentration-dependence, selectivity, and sites.

K(+)-stimulated ATPase activity of a plasmalemma-enriched fraction from excised roots of oat was triphasic in the range 5 to 80 millimolar KCl. The phases obeyed Michaelis-Menten kinetics and were separated from each other by jumps or sharp breaks at about 10 and 20 millimolar. Stimulation by alkali cations was in the order K(+) > Rb(+) > Na(+) > Cs(+) > Li(+) or in a closely related sequence. The specificity reflected differences in V(max), not in affinity (K(m) (-1)). Stimulation by the organic cations ethanolamine and choline in the interval 11 to 80 millimolar appeared monophasic rather than biphasic. Substitution on the quaternary nitrogen of the amino alcohols decreased their effectiveness, as did extension and branching of the chain. Stimulation was maximal at about pH 7 both for K(+) and choline.The kinetics of K(+) stimulation are multiphasic, not cooperative, as was also found for uptake. The ATPase is also stimulated by organic cations, but the difference in kinetics indicates the existence of separate sites for stimulation and transition.

An unmodified heptadecapeptide pheromone induces competence for genetic transformation in Streptococcus pneumoniae.

Competence for genetic transformation in Streptococcus pneumoniae has been known for three decades to arise in growing cultures at a critical cell density, in response to a secreted protease-sensitive signal. We show that strain CP1200 produces a 17-residue peptide that induces cells of the species to develop competence. The sequence of the peptide was found to be H-Glu-Met-Arg-Leu-Ser-Lys-Phe-Phe-Arg-Asp-Phe-Ile-Leu-Gln-Arg- Lys-Lys-OH. A synthetic peptide of the same sequence was shown to be biologically active in small quantities and to extend the range of conditions suitable for development of competence. Cognate codons in the pneumococcal chromosome indicate that the peptide is made ribosomally. As the gene encodes a prepeptide containing the Gly-Gly consensus processing site found in peptide bacteriocins, the peptide is likely to be exported by a specialized ATP-binding cassette transport protein as is characteristic of these bacteriocins. The hypothesis is presented that this transport protein is encoded by comA, previously shown to be required for elaboration of the pneumococcal competence activator.

The leader peptide of colicin V shares consensus sequences with leader peptides that are common among peptide bacteriocins produced by gram-positive bacteria.

Colicin V is a ribosomally synthesized antimicrobial peptide produced by Escherichia coli. Four recently characterized genes, arranged in two convergent operons on the plasmid pCoIV-K30, are required for colicin V synthesis, export and immunity. We report the purification and N-terminal amino acid sequencing of the colicin V protein. Our results demonstrate that the colicin V primary translation product, which consists of 103 amino acids, is proteolytically processed. A leader peptide, consisting of 15 amino acid residues, is removed from the N-terminus during maturation of colicin V. This leader peptide is not related to the N-terminal signal sequences which direct proteins across the cytoplasmic membrane via the Sec pathway. The molecular mass of colicin V, obtained by mass spectrometry analysis, showed that the peptide consists of only unmodified amino acids. The deduced amino acid sequence of the leader peptide was highly homologous to the N-terminal extensions found in non-lantibiotic, peptide bacteriocins produced by Gram-positive bacteria. These findings strongly indicate that colicin V belongs to a family of small peptide bacteriocins that have been found previously only among the Gram-positive lactic acid bacteria.

Identification of DNA binding sites for ComE, a key regulator of natural competence in Streptococcus pneumoniae.

Natural competence in Streptococcus pneumoniae is regulated by a quorum-sensing mechanism consisting of a competence-stimulating peptide (CSP), its dedicated secretion apparatus (ComAB), its histidine kinase receptor (ComD) and a response regulator (ComE). In this report, we show that ComE is a DNA-binding protein that acts autocatalytically by binding to a region in its own promoter. Two additional ComE binding sites were identified in the pneumococcal genome, one in the promoter region of comAB and the other upstream of an ABC transporter of unknown function. A comparison of the ComE-binding sequences with the sequence motif previously found to be involved in the co-ordinated expression of the late genes revealed that they are unrelated. These findings indicate that ComE activates transcription of the late genes indirectly, i.e. via one or more intermediate factors.

A family of bacteriocin ABC transporters carry out proteolytic processing of their substrates concomitant with export.

Lantibiotic and non-lantibiotic bacteriocins are synthesized as precursor peptides containing N-terminal extensions (leader peptides) which are cleaved off during maturation. Most non-lantibiotics and also some lantibiotics have leader peptides of the so-called double-glycine type. These leader peptides share consensus sequences and also a common processing site with two conserved glycine residues in positions -1 and -2. The double-glycine-type leader peptides are unrelated to the N-terminal signal sequences which direct proteins across the cytoplasmic membrane via the sec pathway. Their processing sites are also different from typical signal peptidase cleavage sites, suggesting that a different processing enzyme is involved. Peptide bacteriocins are exported across the cytoplasmic membrane by a dedicated ATP-binding cassette (ABC) transporter. Here we show that the ABC transporter is the maturation protease and that its proteolytic domain resides in the N-terminal part of the protein. This result demonstrates that the ABC transporter has a dual function: (i) removal of the leader peptide from its substrate, and (ii) translocation of its substrate across the cytoplasmic membrane. This represents a novel strategy for secretion of bacterial proteins.

Characterization of the locus responsible for the bacteriocin production in Lactobacillus plantarum C11.

Lactobacillus plantarum C11 secretes a small cationic peptide, plantaricin A, that serves as induction signal for bacteriocin production as well as transcription of plnABCD. The plnABCD operon encodes the plantaricin A precursor (PlnA) itself and determinants (PlnBCD) for a signal transducing pathway. By Northern (RNA) and sequencing analyses, four new plantaricin A-induced operons were identified. All were highly activated in concert with plnABCD upon bacteriocin induction. Two of these operons (termed plnEFI and plnJKLR) each encompass a gene pair (plnEF and plnJK, respectively) encoding two small cationic bacteriocin-like peptides with double-glycine-type leaders. The open reading frames (ORFs) encoding the bacteriocin-like peptides are followed by ORFs (plnI and -L, respectively) encoding cationic hydrophobic proteins resembling bacteriocin immunity proteins. On the third operon (termed plnMNOP), a similar bacteriocin-like ORF (plnN) and a putative immunity ORF (either plnM or -P) were identified as well. These findings suggest that two bacteriocins of two-peptide type (mature PlnEF and PlnJK) and a bacteriocin of one-peptide type (mature PlnN) could be responsible for the observed bacteriocin activity. The last operon (termed plnGHSTUV) contains two ORFs (plnGH) apparently encoding an ABC transporter and its accessory protein, respectively, known to be involved in processing and export of peptides with precursor double-glycine-type leaders. Promoter structure was established. A conserved regulatory-like box encompassing two direct repeats was identified in the promoter regions of all five plantaricin A-induced operons. These repeats may serve as regulatory elements for gene expression.

Regulation of competence for genetic transformation in Streptococcus pneumoniae by an auto-induced peptide pheromone and a two-component regulatory system.

The regulation of competence for genetic transformation in Streptococcus pneumoniae depends on a quorum-sensing system, but the only molecular elements of the system whose specific role have been identified are an extracellular peptide signal and an ABC-transporter required for its export. Here we show that transcription of comC, the gene encoding a predicted 41-residue precursor peptide that is thought to be processed and secreted as the 17-residue mature competence activator, increased approximately 40-fold above its basal level of expression in response to exogenous synthetic activator, consistent with earlier experiments indicating that the activator acts autocatalytically. We also describe two new genes, comD and comE, that encode members of histidine protein kinase and response-regulator families and are linked to comC. Disruption of comE abolished both response to synthetic activator peptide and endogenous competence induction.

A bacteriocin-like peptide induces bacteriocin synthesis in Lactobacillus plantarum C11.

In this study, we show that bacteriocin production in Lactobacillus plantarum C11 is an inducible process triggered by a secreted protein factor produced by the bacteriocin producer itself. The induction factor was identified to be plantaricin A, a bacteriocin-like peptide whose gene (plnA) is located in the same operon as a two-component regulatory system (plnBCD). When L. plantarum C11 cultures were depleted for plantaricin A, either by growing individual colonies on agar plates or by starting a new culture with a highly diluted inoculum, no bacteriocin was produced during the following growth. When chemically synthesized plantaricin A or purified bacterially produced plantaricin A was added to non-producing cultures, bacteriocin production was induced. Only 1 ng ml-1 plantaricin A is sufficient to induce the bacteriocin production in non-producing L. plantarum C11, and bacteriocin activity appears in the growth medium approximately 150 min after induction. Northern analyses, using a plnA-specific probe, demonstrated that plantaricin A is able to induce its own synthesis by transcription of the plnABCD operon, and this is observed approximately 15 min after adding plantaricin A. Furthermore, heterologous expression of the plnABCD operon in a Lactobacillus sake strain showed that the conditioned growth medium contained the active induction factor. Neither synthetic nor expressed plantaricin A from the heterologous system possesses any bacteriocin activity, suggesting that plantaricin A is primarily an induction factor and not a bacteriocin as claimed earlier.

Natural competence in the genus Streptococcus: evidence that streptococci can change pherotype by interspecies recombinational exchanges.

To map the incidence of natural competence in the genus Streptococcus, we used PCR to screen a number of streptococcal strains for the presence of the recently identified competence regulation operon, containing the comC, -D, and -E genes. This approach established that the operon is present in strains belonging to the S. mitis and S. anginosus groups, but it was not detected in the other strains examined. Competence is induced in S. pneumoniae and S. gordonii by strain-specific peptide pheromones, competence-stimulating peptides (CSPs). With its unique primary structure, each CSP represents a separate pheromone type (pherotype), which is recognized by the signalling domain of the downstream histidine kinase, ComD. Thus, all bacteria induced to competence by a particular CSP belong to the same pherotype. In this study, we identified a number of new pherotypes by sequencing the genes encoding the CSP and its receptor from different streptococcal species. We found that in several cases, these genes have a mosaic structure which must have arisen as the result of recombination between two distinct allelic variants. The observed mosaic blocks encompass the region encoding the CSP and the CSP-binding domain of the histidine kinase. Consequently, the recombination events have led to switches in pherotype for the strains involved. This suggests a novel mechanism for the adaptation of naturally competent streptococci to new environmental conditions.

Association of the lactococcin A immunity factor with the cell membrane: purification and characterization of the immunity factor.

The physicochemical characteristics of the lactococcin A immunity protein, as deduced from its gene sequence, were used to devise a procedure for its purification. The protein was purified from cell extracts by cation-exchange and reverse-phase chromatography. As judged from the amino acid composition and amino acid sequencing, the immunity protein is not post-translationally processed by cleavage at its N- or C-terminus. Consequently, the absorption coefficient at 280 nm, the isoelectric point, and the molecular mass of the immunity protein may be calculated to be, respectively, 8.2 x 10(3) M-1 cm-1, 10.2 and 11,163 Da from the amino acid sequence predicted from the nucleotide sequence. The immunity protein is a major cell protein component--one cell may contain (to an order of magnitude) 10(5) molecules--and it is in part associated with the cell membrane, as judged by immunoblot analysis of membrane vesicle-associated proteins. Exposing lactococcin-A-sensitive cells to an excess of the immunity protein did not affect the lactococcin-A-induced killing of the cells, indicating that the immunity protein does not protect cells by simply binding to lactococcin A, nor to externally exposed domains on the cell surface. Exposing immune-positive cells to antiserum against the immune protein did not sensitize the cells to lactococcin A, suggesting that the immunity protein in fact does not act extracellularly.

The gene encoding plantaricin A, a bacteriocin from Lactobacillus plantarum C11, is located on the same transcription unit as an agr-like regulatory system.

Purification and amino acid sequencing of plantaricin A, a bacteriocin from Lactobacillus plantarum C11, revealed that maximum bacteriocin activity is associated with the complementary action of two almost-identical peptides, alpha and beta (J. Nissen-Meyer, A. G. Larsen, K. Sletten, M. Daeschel, and I. F. Nes, J. Gen. Microbiol. 139:1973-1978, 1993). A 5-kb chromosomal HindIII restriction fragment containing the structural gene of plantaricin A was cloned and sequenced. Only one gene encoding plantaricin A was found. The gene, termed plnA, encodes a 48-amino-acid precursor peptide, of which the 22 and 23 C-terminal amino acids correspond to the purified peptides. Northern (RNA) blot analysis demonstrated that a probe complementary to the coding strand of the plantaricin A gene hybridized to a 3.3-kb mRNA transcript. Further analysis of the 3.3-kb transcript demonstrated that it contains three additional open reading frames (plnB, plnC and plnD) downstream of plnA. The DNA sequences of plnB, plnC, and plnD revealed that their products closely resemble members of bacterial two-component signal transduction systems. The strongest homology was found to the accessory gene regulatory (agr) system, which controls expression of exoproteins during post-exponential growth in Staphylococcus aureus. The finding that plnABCD are transcribed from a common promoter suggests that the biological role played by the bacteriocin is somehow related to the regulatory function of the two-component system located on the same operon.

Mapping of neutralization epitopes on infectious pancreatic necrosis viruses.

We have characterized and mapped variable and conserved neutralization epitopes of serogroup A strains of aquatic birnaviruses. Epitope mapping using monoclonal antibodies (MAbs) and Escherichia coli-expressed deletion fragments of VP2 of the N1 strain of infectious pancreatic necrosis virus (IPNV) demonstrated that two variable epitopes, H8 and B9, depend on the variable region between amino acid 204-330. A conserved neutralization epitope, F2, was shown to depend on the same region as epitopes H8 and B9 but was additionally dependent on amino acids between 153-203. The neutralization epitopes H8, B9 and F2 were also shown to overlap by a competitive binding assay. One conserved neutralization epitope, AS-1, was not exposed on any of the recombinant VP2 deletion fragments and was therefore not possible to map. However, the MAbs AS1 and F2 were partly competitive indicating that these epitopes are overlapping. All neutralization epitopes were independent of a conserved non-neutralization epitope, E4. Our results demonstrate that the central third of VP2 contains several partly overlapping neutralization epitopes, both variable and conserved among serogroup A strains of IPNV.