Molecular Weights of DNA from Bacteriophages T5, T5st(0), BF23, and BF23st(4)

The DNA molecular weights were determined by calibrated electron microscopy. The results (in units of 10⁶) are: for T5, 77.4 ± 2.4; T5st(0), 72.4 ± 1.9; BF23, 76.7 ± 2.3; and BF23st(4), 71.4 ± 1.7.     

Irreversible binding to the receptor of bacteriophages T5 and BF23 does not occur with the tip of the tail.

Treatment of purified tails of bacteriophage T5 with 0.05% sodium dodecyl sulfate specifically removed pb2, a protein of 108,000 molecular weight (108K), from the tail. Although these tails were devoid of the single straight tail fiber, they still inhibited adsorption of T5 to Escherichia coli cells. Reconstitution of these tails with pb2 increased the efficiency of inhibition of T5 adsorption. Treatment of tails with 0.1% sodium dodecyl sulfate removed, in addition to pb2, a protein of 67K from phage T5 and one of 60K from phage BF23. These tails failed to inhibit phage adsorption, and no reconstitution was achieved. Reconstitution of T5 tails with pb2 from BF23, and of BF23 tails with pb2 from T5, did not alter the host receptor specificity of the tails. Binding of untreated T5 tails to small FhuA receptor particles revealed that binding occurred with the conical part of the tail and that pb2 was most likely released from the tail upon binding. From these results and from recent observations with T5-BF23 hybrid phages (K.J. Heller, Virology 139:11-21, 1984), we conclude that the receptor-binding proteins of T5 and BF23 are the 67K and 60K proteins, respectively, and that they are not located at the tip of the tail but rather at or near the site where the straight tail fiber is attached to the conical part of the tail.     

Cloning, sequencing, and recombinational analysis with bacteriophage BF23 of the bacteriophage T5 oad gene encoding the receptor-binding protein.

Binding of bacteriophage T5 to its receptor, the Escherichia coli FhuA protein, is mediated by tail protein pb5. In this article we confirm that pb5 is encoded by the T5 oad gene and describe the isolation, expression, and sequencing of this gene. In order to locate oad precisely, we analyzed recombinants between BF23, a T5-related phage with a different host range, and plasmid clones containing segments of the T5 chromosome. This analysis also showed that oad has little or no homology with hrs, the analogous BF23 gene. We were able to overproduce a protein that comigrates with pb5 after fusing a 2-kb segment containing oad to a phage T7 promoter. This segment contains an open reading frame that can encode a protein of the appropriate size. Its deduced amino acid sequence does not closely resemble that of any other protein in the database. The sequence upstream of the open reading frame shows typical characteristics of a promoter region with two overlapping, divergently orientated promoters.     

Arrangement on the Chromosome of the Known Pre-Early Genes of Bacteriophages T5 and BF23

Genetic crosses between mutants defective in the known pre-early genes of T5 and BF23 and the detection of putative N-terminal fragments have allowed the determination of the order of genes along the initially transferred 8% section of DNA of these phages.     

Terminally redundant deletion mutants of bacteriophage BF23.

Deletion mutants of bacteriophage BF23 were isolated and the positions of the deletions were determined. Two different deletable regions were detected: one in the same region as previously reported for bacteriophage T5, which is closely related to BF23; and the other within both terminal repetitions. The former deletable region lay between positions 0.31 and 0.36, which represented the fractional lengths of the BF23 ( + ) DNA as measured from its left end. The latter deletion was evenly divided between the two terminal repetitions. The deletion in the left terminal repetition lay between positions 0.044 and 0.078 and was repeated in the corresponding region of the right terminal repetition between positions 0.966 and 1.0. The size of the DNA transferred to host cells during the first step of DNA transfer by BF23 carrying deletions in the terminal repetitions of its DNA was less than the size of DNA transferred during the first step by wild-type BF23 by an amount equal to the size of the deletion in each terminal repetition. This finding suggests the existence of a specific mechanism for delineating the position at which the first step of DNA transfer is stopped.     

Physical map of bacteriophage BF23 DNA: terminal redundancy and localization of single-chain interruptions.

The DNA of bacteriophage BF23 possesses two structural features, localized single-chain interruptions and a large terminal repetition, previously described for T5, a closely related virus. As is the case for T5, single-chain interruptions occur with variable frequencies at a small number of fixed sites within one strand of the double-stranded BF23 genome. The sites where interruptions occur with the highest frequencies were napped by an electrophoretic analysis of the single-stranded fragments produced by denaturation of BF23 DNA. The positions of these fragments were determined by degrading BF23 DNA to various extents with lambda exonuclease and observing the relative order with which they were (i) degraded or (ii) released intact from the undenatured duplex. The exact locations of the interruptions were determined from analysis of analogous duplex fragments produced by degrading exonuclease III-treated BF23 DNA with a single-strand-specific endonuclease. BF23 has five principal sites (located at 7.9, 18.7, 32.4, 65.8, and 99.6% from the left end of the DNA) where interruptions occur in most molecules. The principal interruptions in T5 DNA occur at similar positions. The locations of eight secondary interruptions in BF23 DNA were also determined. In general, BF23 DNA has fewer secondary interruptions than t5 dna, although there is at least one location where an interruption occurs with a greater frequency in BF23. The presence of a terminal repetition in BF23 DNA was demonstrated by annealing ligase-repaired molecules that had been partially digested with lambda exonuclease. If the complementary sequences at both ends of the DNA were exposed by exonuclease treatment, the duplex segment that resulted from annealing could be released by digestion with a single-strand-specific endonuclease. This segment was analyzed by agarose gel electrophoresis and found to represent 8.4% of BF23 DNA.     

Early Abortive Lysis by Phage BF23 in Escherichia coli K-12 Carrying the Colicin Ib Factor

Growth of phage BF23 was restricted in Escherichia coli K-12 strains carrying a colicin I factor (ColIb); most infected cells lysed early without producing progeny phages. Either addition of chloramphenicol before phage infection or ultraviolet irradiation of phage prevented early abortive lysis, an indication that certain phage functions are required for this phenomenon. Very little or no phage-induced lysozyme was synthesized in the infected ColI(+) cells. This result suggests that early abortive lysis was not due to the lysozyme action. A small fraction (0.05) of BF23-infected ColI(+) cells showed normal phage growth. This "escaped growth" may reflect the physiological state of the host bacteria rather than the heterogeneity of the infecting phage. Host-controlled modification was not observed. A phage mutant, BF23hI, able to grow on ColI(+) cells, was isolated and was characterized to be recessive to the wild-type BF23 in its ability to undergo early abortive lysis. Among the T series phages, T5 induced early abortive lysis, and growth of T5 was restricted upon infection to ColI(+) cells. These results and the other observations, including the occurrence of phenotypic mixing between BF23 and T5, suggest that these two phages are related to each other even though the receptor sites for BF23 and T5 are apparently different.     

Arrangement of the single-stranded fragments in E. coli bacteriophage BF23 DNA

Bacteriophage BF23st(0) DNA was denatured with alkali and fractionated by agarose gel electrophoresis. Seven single-stranded fragments (designated Fragments I--VII) were identified as the major constituents of the phage DNA. The presence of several minor fragments which represent minor populations of the phage genome was also observed. The largest fragment (Fragment I) represents the intact strand of phage DNA, whereas the other fragments form the complementary strand. Thus, BF23st(0) DNA carries single-strand interruptions in only one strand. The arrangement of the major fragments in the nicked strand was determined by use of gamma-exonuclease and agarose gel electrophoresis. From the mode of action of this nuclease, and from the kinetics of release or disappearance of the fragments, the polarity of the fragments in BF23st(0) DNA was specified. In addition, the presence of two types of major phage populations differing in their composition of the fragments was demonstrated. One type has an additional nick (yielding Fragment IV and Fragment V) in a specific fragment (Fragment II) of other type.     

Transfection of Escherichia coli spheroplasts. VI. Transfection of nonpermissive spheroplasts by T5 and BF23 bacteriophage DNA carrying amber mutations in DNA transfer genes.

DNA was extracted from T5 and BF23 phage carrying amber mutations in genes A2, A1, or D9 and tested for its ability to transfect su minus spheroplasts. DNA from T5 am231, defective in gene A2, transfects Escherichia coli su minus recB minus spheroplasts with an efficiency of 16% of that of wild-type T5 DNA, whereas DNA from T5 am16d or BF23 am57, both defective in gene A1 or its equivalent, transfects E. coli su minus recB minus spheroplasts with an efficiency of 1.4% of that of wild-type T5 DNA, provided E. coli su+ bacteria is used as the indicator in all cases. More than 95% of the progeny from the am231, am16d, and am57 DNA that transfects su minus recB minus spheroplasts is still amber mutant. From these efficiencies of transfection we conclude that the product of gene A2 functions mainly in the mechanism of transfer of phage DNA to intact host cells, and that this function is not essential for transfection of spheroplasts. We also conclude that gene A1 controls functions in addition to DNA transfer, in agreement with previous studies which show that mutations in gene A1 have a pleiotropic effect. Apparently, the absence of these additional functions controlled by gene A1 leads to a high frequency of abortive infection. DNA from amber mutants defective in either gene A1 or A2 does not appreciably transfect su minus rec+ spheroplasts, indicating that the products of these two genes may both be needed to protect T5 DNA from the very active rec BC nuclease in spheroplasts.     

Physical map of the dispensable region of the genome of E. coli bacteriophage BF23

Using 13 deletion mutants of bacteriophage BF23, physical as well as genetic structures of that portion of the genome which is dispensable for phage growth were investigated. The dispensable region covers at least 15% of the genome of wild type BF23, extending from about 0.2 to 0.35 map unit. Restriction endonuclease (EcoRI and HindIII) cleavage sites and the sites of single-strand interruptions in this dispensable region were localized. It was found that the dispensable region contains an interruption site, which is missing in the mutant BF23st(0) used by Okada and Shimura (1980). Wild-type phage DNA is heterogeneous in the presence or absence of specific single-strand interruptions in this or in a neighboring region of the genome.     

Transport of vitamin B12 in Escherichia coli: common receptor system for vitamin B12 and bacteriophage BF23 on the outer membrane of the cell envelope.

We showed previously that the outer membrane of the Escherichia coli cell envelope normally contains about 200 to 250 B12 receptors, and that these receptors function both in B12 transport and as receptors for the E colicins. This paper shows that this receptor system is also shared with bacteriophage BF23. A strong positive correlation was observed between the number of B12 receptors per cell and the rate of adsorption of BF23. Cells from mutant strains that lacked B12 receptors did not adsorb BF23 particles. The rate of adsorption of BF23 to cells of a merodiploid strain (RK4151), with about 550 B12 receptors per cell, was approximately double that to cells of a normal, haploid strain. The adsorption of BF23 to hole cells, cell envelopes, outer membrane particles, and solubilized outer membranes was inhibited by vitamin B12, with 50% inhibition at B12 concentrations in the range of 0.5 to 2.0 nM. These values are close to the observed KS for B12 binding to the B12 receptors. Vitamin B12 concentrations as high as 100 nM did not inhibit adsorption of bacteriophages T5, T6, and lambdacI to cells of sensitive strains of E. coli. Bacteriophage BF23 inhibited B12 transport by whole cells and was shown to be a competitive inhibitor of B12 binding to isolated cell envelope particles. The B12/BF23 receptors from E. coli strains KBT069 (btuB69) and RK4104 (btuB69) were fully active, but the number per cell was reduced to an average value of about 0.5.     

Relationships among genes and gene products of bacteriophage BF23

Twenty-five gene products of bacteriophage BF23 were identified by sodium dodecyl sulfate-polyacrylamide gel electrophoresis, and their functions were studied in relation to type I and II genes classified by means of genetic complementation tests. All the type I mutants were defective in the synthesis of a tail protein, L3. In addition, 4 type I gene products, L5 (gp21), L7 (gp20), L8 (gp29), and L9 (gp25), were identified as constituents of tails (gp21 denotes that a protein is a product of gene 21). Three type IIb mutants in genes 10, 14, and 19 diminished substantially the production of late proteins, including tail and head proteins, and the two other type IIb mutants in genes 1 and 2 were defective in the synthesis of both early and late proteins. Of 14 type IIa mutants, at least 6 were defective in phage DNA synthesis and 2 were defective in the synthesis of head proteins. The defect in the head donor activities of type IIa mutants in extract complementation tests was due to the failure of the formation of mature heads containing DNA. The above results support directly the results of the genetic characterization of BF23 genes.     

Physical map of bacteriophage BF23 DNA: restriction enzyme analysis.

Cleavage maps of bacteriophage BF23 DNA have been constructed for the restriction endonucleases SalI (3 fragments), BamHI (5 fragments), EcoRI, (8 fragments), BalI (13 fragments), and HpaI (49 fragments, 32 of which have been ordered). The maps were determined by (i) analysis of deletion mutants, (ii) digestion with two endonucleases, (iii) digestion of isolated fragments with a second enzyme, (iv) analysis of partial digests, and (v) digestion after treatment with lambda exonuclease.     

Genes affecting coliphage BF23 and E colicin sensitivity in Salmonella typhimurium.

Rough strains of Salmonella typhimurium were sensitive to coliphage BF23. Spontaneous mutants resistant to BF23 (bfe) were isolated, and the trait was mapped using phage P1. The bfe gene in S. typhimurium was located between argF (66% co-transducible) and rif (61% co-transducible). The BF23-sensitive S. typhimurium strains were not sensitive to the E colicins. Cells of these rough strains absorbed colicin, as measured by loss of E2 or E3 killing units from colicin solutions and by specific adsorption of 125I-colicin E2 to bfe+ cells. Sensitivity to colicins E1, E2, and E3 was observed in a S. typhimurium strain carrying the F'8 gal+ episome. This episome complemented the tolB mutation of Escherichia coli. We conclude that the bfe+ protein satisfies requirements for adsorption of both phage BF23 and the E colicins. In addition, expression of a gene from E. coli, possibly tolB, is necessary for efficient E colicin killing of S. typhimurium.     

Cloning, sequencing, and expression of bacteriophage BF23 late genes 24 and 25 encoding tail proteins.

Two bacteriophage BF23 late genes, genes 24 and 25, were isolated on a 7.4-kb PstI fragment from the phage DNA, and their nucleotide sequences were determined. Gene 24 encodes a minor tail protein with the expected M(r) of 34,309, and gene 25 located 4 bp upstream of gene 24 encodes a major tail protein with the expected M(r) of 50,329. When total cellular RNA isolated from either phage-infected cells or cells bearing the cloned genes was analyzed by the primer extension method using the primers specific to either gene 25 or gene 24, we identified a possible late gene promoter, designated P25, in the 5'-flanking region of gene 25. This promoter was similar in structure to Escherichia coli promoters for sigma 70. Studies of the translational gene 25- and gene 24-lacZ fusions in the cloned gene system revealed that the promoter P25 was responsible for the expression of both genes 25 and 24 even in the absence of the regulatory genes which were absolutely required for late gene expression in the normal phage-infected cells. These results indicate that the two genes constitute an operon under the control of P25 and that the regulatory gene products of BF23 do not participate directly in specifying the late gene promoter.     

Abortive infection by bacteriophage BF23 due to the colicin Ib factor. II. Involvement of pre-early proteins

The mechanism by which the growth of phage BF23 is arrested in cells carrying colicinogenic factor Ib involves certain phage-specific pre-early proteins. BF23 induces the extensive formation of proteins lc and ld, but very little formation of protein le, whereas BF23h^? (a mutant that is not arrested in cells carrying colicinogenic factor Ib) induces very small amounts of proteins lc and ld, but extensive amounts of protein le. Proteins lc and ld may be oligomers of protein le, and it is the presence of these putative oligomers that is necessary for the arrest of phage growth.     

Acute exercise reduces the response to colon distension in T(5) spinal rats

Individuals with spinal cord injuries above thoracic level 6 (T(6)) experience life-threatening bouts of hypertension, termed autonomic dysreflexia (AD). AD is mediated by peripheral alpha-adrenergic receptor supersensitivity as well as a reorganization of spinal pathways controlling sympathetic preganglionic neurons. A single bout of dynamic exercise may be a safe therapeutic approach to reduce the severity of AD because mild-to-moderate dynamic exercise reduces postexercise alpha-adrenergic receptor responsiveness, lowers postexercise sympathetic nerve activity, and reduces the postexercise response to stress. Therefore, this study was designed to test the hypothesis that mild-to-moderate dynamic exercise attenuates the postexercise response to colon distension (mechanism to elicit AD). To test this hypothesis, six male Wistar rats (406 +/- 23 g), 5 wk post-T(5) spinal cord transection, were instrumented with an arterial catheter. After recovery, the response to graded colon distension (10, 30, 50, and 80 mmHg, in random order) was determined before and after a single bout of mild-to-moderate dynamic exercise (9-12 m/min, 0% grade for 40 min). After exercise, the pressor response to graded colon distension was significantly attenuated (preexercise change: 2 +/- 1, 9 +/- 1, 14 +/- 1, and 24 +/- 2 vs. postexercise change: 2 +/- 1, 2 +/- 1, 9 +/- 1, and 12 +/- 3 mmHg). Thus acute exercise is a safe, therapeutic approach to reduce the severity of AD in paraplegic subjects.     

Growth of coliphage BF23 on rough strains of Salmonella typhimurium: the bfe locus

Summary Coliphage BF23 develops in Salmonella typhimurium rough strains. The phage is neither restricted nor modified by S. typhimurium. The growth patterns of the phage were slightly different in S. typhimurium than in Escherichia coli, although phage propagated on S. typhimurium is identical to the phage propagated in E. coli by several criteria used. Mutants of S. typhimurium resistant to BF23 were isolated and found to map (by P22-and Plmediated transduction) in the same position as bfe mutants of E. coli. The order of genes was: metB-argC-bfe-rif-purD-metA.Phage BF23 does not form plaques on smooth S. typhimurium strains, since the phage fails to adsorb irreversibly to smooth cells. Nevertheless, on solid agar, the phage prevents growth of many (but not all) smooth strains. Moreover, UV-and alkali-inactivated phage BF23, although unable to form plaques on sensitive hosts, retains the ability to prevent growth of the host on solid medium. This ability is sensitive to protease and resistant to DNAse and RNase. Heat treatment of the phage causes rapid loss of the cell-growth-preventing-ability whereas the ability to form plaques is lost much more slowly. These results lead to a proposal that phage BF23 virions carry a colicin-like factor that kills sensitive cells.     

The chromosome of bacteriophage T5 : II. Arrangement of the single-stranded DNA fragments in the T5 and T5st(O) chromosomes

Denatured bacteriophage T5 DNA contains a large number of single-stranded DNA fragments which have been separated by agarose gel electrophoresis and classified as “major” or “minor” species on the basis of their relative abundances (Hayward & Smith, 1972). For further study of these fragments we have centrifuged denatured T5 DNA in CsCl density-gradients in the presence of poly(G). Gel electrophoretic analysis of fractions from these gradients shows that the 37.0 and 13.9 million major fragments of T5⁺ DNA and the 35.3 and 17.2 million of T5st(O) DNA are found in the high buoyant density regions. The other fragments vary in the extent of their interactions with poly(G) and a minor fragment, which has anomalous electrophoretic properties, exhibits the strongest poly(G) interaction.