Recognition and Specific Degradation of Bacteriophage T4 mRNAs

Corresponding author: Tetsuro Yonesaki, Department of Biology, Graduate School of Science, Osaka University, 1-16 Machikaneyama-cho, Toyonaka-shi, Osaka 560-0043, Japan., yonesaki{at}bio.sci.osaka-u.ac.jp (E-mail)

Communicating editor: G. R. SMITH

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Gene 61.5 of bacteriophage T4 has a unique role in gene expression. When this gene is mutated, mRNAs of many late genes are rapidly degraded, resulting in late-gene silencing. Here, we characterize an extragenic suppressor, ssf5, of a gene 61.5 mutation. ssf5 was found to be an amber mutation in motA, which encodes a transcription activator for T4 middle genes. When this gene is mutated, both degradation and specific cleavage of late-gene mRNA is induced after a delay, as exemplified by soc mRNA. Consequently, partial late-gene expression occurs. In an ssf5 genetic background, a gene 61.5 mutation exhibits a novel phenotype: in contrast to late-gene mRNA, middle-gene mRNA is stabilized and the expression of middle genes is prolonged. This is attributable to an activity of gene 61.5 specific for degradation of middle-gene mRNA. The degradation of middle-gene mRNA in the presence of a normal gene 61.5 appears in parallel with the degradation of late-gene mRNA in its absence. This observation suggests that the mRNA-degrading activity that silences late genes in cells infected with a gene 61.5 mutant is targeted to middle-gene mRNA when gene 61.5 is wild type. These results and the results obtained in the presence of a normal motA gene suggest that gene 61.5 protein functions to discriminate mRNAs for degradation in a stage-dependent manner.

BACTERIOPHAGE T4 has evolved a highly deterministic genetic program in which many scores of genes are expressed sequentially. T4 genes can be classified into three categories, early, middle, and late, according to when they are expressed. Gene 61.5 has an important role in T4 gene expression. When a 61.5 mutant infects Escherichia coli cells at low temperatures, mRNAs are destabilized and the expression of many late genes is blocked (KAI et al. 1996 ). Several observations argue that stabilization of late-gene mRNA in the presence of normal gene 61.5 requires a highly coupled mechanism involving multiple T4 genes (KAI et al. 1998 ). First, gene 61.5 is expressed immediately after infection, which implies that an intracellular process depending on the function of gp61.5 (the product of gene 61.5) continues to late stages. Second, other T4 early and middle genes are involved in stabilizing late-gene mRNAs. Third, extragenic suppressors of a gene 61.5 mutation can arise in at least three other T4 genes.

Three different classes of promoters, early, middle, and late, initiate T4 transcription (MOSIG and HALL 1994 ). Immediately after T4 infection, early genes are transcribed by the E. coli RNA polymerase from early promoters. Early-gene products alter the host RNA polymerase in two ways, both of which lead to the expression of middle genes: the RNA polymerase becomes able to read through a transcription terminator to transcribe a downstream region, and the enzyme is modified to recognize middle promoters. By the expression of middle genes, the host -factor of RNA polymerase is replaced by a T4-specific -factor, and the RNA polymerase ultimately recognizes late promoters to express late genes (MATHEWS 1994 ). Thus, the sequential expression of T4 genes is controlled primarily by differential transcription. However, T4 gene expression shifts from early to late within a period as short as 30 min (O'FARRELL and GOLD 1973 ; CHRISTENSEN and YOUNG 1984 ; KAI et al. 1996 ). Such rapid shifts of gene expression cannot be achieved solely by differential transcription from stage-specific promoters and presumably require the degradation of mRNAs at the following stage to repress their expression. The latter requirement is emphasized by the discovery of an endoribonuclease encoded by the regB gene. The regB protein cleaves early gene mRNAs, including regB mRNA but not middle-gene mRNAs (SANSON et al. 2000 ). This unique property of regB protein contributes to a rapid shift of gene expression from early to middle. By analogy, T4 might encode an activity for degrading middle mRNAs at the late stage.

In this article, we describe an extragenic suppressor, ssf5, of a gene 61.5 mutation. The ssf5 mutant turned out to carry an amber mutation in motA, which encodes a transcription activator for middle promoters (MATTSON et al. 1978 ; ADELMAN et al. 1998 ). In the ssf5 genetic background, both degradation and specific cleavage of late-gene mRNA was delayed by the ssf5 mutation and, consequently, partial late-gene expression occurred. Although partially stabilized by the ssf5 mutation, late-gene mRNA still degraded faster when gene 61.5 was mutated than when it was not. In contrast, middle-gene mRNA degraded more slowly at late stages when gene 61.5 was mutated than when it was not. This finding strongly suggests a new role for gene 61.5 in destabilization of middle-gene mRNAs as well as a role in stabilization of late-gene mRNAs.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Phage and bacterial strains:
The wild-type T4D and mutant phages amSF16 (gene 61.5^-), nd28 (denA^-), rIIPT8 (rII-denB), and GT7 were laboratory stocks. GT7 was used for making T4 deoxycytosine-containing (dC) DNA (WILSON et al. 1979 ). The ssf5 mutant was isolated and the double mutant amSF16 ssf5 was constructed previously (KAI et al. 1998 ). For the marker-rescue test (KAI et al. 1999 ), ssf5 nd28 rIIPT8 was constructed by recombination. We used the E. coli strains MH1 (hsdR^- sup⁰), BB (sup⁰), and CR63 (supD).

Plasmids and marker-rescue test:
The 616-bp KpnI-PstI fragment, corresponding to T4 nucleotides 162884–163500 (GenBank accession no. NC_000866), was obtained by digestion of T4 dC DNA. pBOT401 was constructed by an insertion of the 616-bp fragment between the KpnI and PstI sites of pBluescript II KS+ (Stratagene, La Jolla, CA). The cloned fragment had one EcoRI site in its middle. pBOT402 or 403 was derived from pBOT401 by digestion with EcoRI plus PstI or with EcoRI plus KpnI, respectively, blunting with T4 DNA polymerase and self-ligation. The marker-rescue test was performed according to KAI et al. 1999 . The ssf5 nd28 rIIPT8 phage was propagated on MH1 cells harboring pBOT401, -402, or -403, and the progeny were examined for plaque size on BB cells. The efficiency of marker rescue calculated from the ratio of normal-size plaques to total plaques was 0.03 for pBOT401, 0.02 for pBOT403, and <0.001 for pBOT402.

Labeling and analysis of newly synthesized proteins:
MH1 cells were grown to a density of 5 x 10⁸ cells/ml in M9 minimal medium supplemented with 0.3% casamino acids, 1 µg/ml thiamine, and 20 µg/ml tryptophan and infected at 30° at a multiplicity of infection of 10 with T4 phage. [³⁵S]Methionine/cysteine (American Radiolabeled Chemicals, St. Louis, MO; >37 TBq/mmol) was added to a 0.1-ml culture at 3.7 MBq/ml at various times after infection to label newly synthesized proteins for 3 min. Under these conditions, 10⁵ cpm was usually incorporated into the trichloroacetic acid-insoluble fraction in the wild-type-infected cell culture. Labeled proteins from an equal portion of infected cells were analyzed by electrophoresis through an 8%, unless otherwise stated, polyacrylamide gel containing 0.1% SDS. After an autoradiograph was taken with a Bio-Image analyzer (Fuji BAS-2000II), the intensity of signal was quantified by the National Institutes of Health (NIH) image program.

Northern blot analysis:
RNAs were extracted as described previously (KAI et al. 1996 ), denatured by incubation at 65° for 5 min in a solution consisting of 0.1% SDS and 7% glycerol, and electrophoresed through a 1% agarose gel in TAE buffer or a 5% polyacrylamide gel containing 7 M urea in TBE buffer. The integrity and quantity of rRNAs were monitored by staining with ethidium bromide. RNAs were transferred to a nylon membrane and crosslinked by irradiation with UV light. The membrane was hybridized with a ³²P-labeled DNA probe at 45° overnight in 50% formamide, 0.25 M sodium phosphate (pH 7.0), 0.2 mM EDTA (pH 8.0), 0.25 M sodium chloride, 100 µg/ml sonified and heat-denatured herring-sperm DNA, and 3.5% SDS. The membrane was washed at 45° with 2x SSC containing 0.1% SDS and then washed at 45° with 1x SSC containing 0.1% SDS. A radioactive probe for each gene transcript was prepared by PCR using one primer 5' end-labeled with T4 kinase and [-³²P]ATP (Institute of Isotopes of the Hungarian Academy of Sciences, Hungary, 259 TBq/mmol) and another unlabeled primer; primers 15–20 nucleotides long were used to amplify T4 nucleotides (GenBank accession no. NC_000866) 167941–16881 for the rIIB gene; 15234–15647 for the soc gene; 114773–115153 for the uvsY gene; 106258–106518 for gene 23; 107361–107848 for gene 24; 31923–32260 for gene 45; and 34439–34924 for gene 46. Template DNA was previously prepared by PCR using the same primer set used for each probe synthesis with either plasmid or T4 DNA as a template and purified by polyacrylamide gel electrophoresis. The labeled probes were heat denatured immediately before use. After an autoradiograph had been taken with a Bio-Image analyzer, the signal intensity was quantified with the NIH image program.

Primer extension analysis:
The DNA primer, 5'-gttattaaccagttactttc, was complementary to the 3' terminal portion of soc mRNA. Primer extension was conducted with avian myeloblastosis virus reverse transcriptase as described previously (KAI et al. 1996 ). The reaction was terminated by addition of 0.1% xylene cyanol, 0.1% bromophenol blue, 10 mM EDTA (pH 8.0), and 95% (v/v) deionized formaldehyde. The reaction products were denatured by boiling for 2 min and analyzed by electrophoresis through a 4% polyacrylamide gel containing 7 M urea.

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

The phenotype and locus of the ssf5 mutation:
When a 61.5 mutant infects E. coli cells, the expression of many late genes is blocked, causing the growth defect of this mutant (KAI et al. 1996 ; refer to Fig 1). Previously we isolated the extragenic suppressor, ssf5, of the growth defect, which was mapped close to gene 52 (KAI et al. 1998 ). In this study, we characterized this mutation in more detail.

View larger version (118K): In this window In a new window Download PPT slide   Figure 1. Polyacrylamide gel electrophoresis of T4 proteins. MH1 cells were infected with wild type, 61.5- (amSF16), motA- (ssf5), or 61.5- motA-. Newly synthesized proteins were pulse labeled at various times indicated and electrophoresed through a 12.5% polyacrylamide gel as described in MATERIALS AND METHODS. Middle-gene products and motA protein are indicated by arrowheads and late-gene products by arrows.

The ssf5 mutant produced 30 progeny particles per infected cell (about one-third of the wild-type number) after a 10- to 20-min delay. These characteristics and the map position of ssf5 were reminiscent of the motA mutations. The calculated molecular mass of the motA protein is 23.6 kD (UZAN et al. 1990 ) and indeed a 24-kD protein (marked motA in Fig 1) synthesized 1 to 5 min after infection with the wild type was missing in the ssf5-infected cells. Furthermore, the ssf5 mutant also exhibited other phenotypes characteristic of motA mutants (PULITZER et al. 1985 ): a delay in late-gene expression (see below), a 10-min delay in the start of T4 DNA synthesis, and a reduction of transcripts initiated from middle promoters, such as those of the uvsY gene and gene 45, to 5–10% of the wild-type level (data not shown).

Next, the sequence from the wild-type motA gene was found to rescue the ssf5 mutation (see MATERIALS AND METHODS). In the corresponding region of ssf5 phage DNA, we detected one base substitution: G at nucleotide 313 of the motA coding region (UZAN et al. 1990 ) was changed to T, replacing Gln at codon 104 with an amber codon. All these results clearly show that ssf5 is a motA amber mutation. Hereafter we call ssf5, for simplification, motA^- or motA mutation.

The effect of motA mutation on late-gene expression in the gene 61.5 mutant:
Because it was isolated as a suppressor of the growth defect of a gene 61.5 mutant (KAI et al. 1998 ), the motA mutation should alleviate the silencing of late genes. To justify this notion, the rate of gene expression was examined by pulse labeling newly synthesized proteins with [³⁵S]methionine/cysteine for 3 min and analyzing the proteins using polyacrylamide gel electrophoresis (Fig 1).

In the wild-type-infected cells, different classes of proteins synthesized sequentially were discernible. Consistent with our previous work (KAI et al. 1996 ), late proteins were virtually undetectable in the gene 61.5 mutant-infected cells. The motA mutant synthesized early proteins normally except for a motA protein (3–6 min). Middle proteins such as gprIIA (membrane protein), gp39 (DNA topoisomerase subunit), gp43 (DNA polymerase), gp46 (putative exonuclease subunit for DNA recombination), and gpuvsX (DNA strand transfer protein) were synthesized in as timely a fashion as those in the wild-type-infected cells. The synthesis of late proteins such as gp7 (baseplate protein), gp10 (baseplate protein), gp23* (major head protein, processed from gp23), gp23, gp24 (vertex protein), and gp37 (tail fiber protein) was efficient but slightly delayed. More detailed analysis revealed that the delay in late protein synthesis was 5–8 min in comparison with that of wild type (data not shown). As expected, introducing a motA mutation allowed 61.5^- phage to synthesize the same late proteins. These results clearly confirmed the suppression effect of motA mutation on the silencing of late genes.

The burst size (mean number of progeny particles per infected cell) of 61.5^- motA^- was 3, significantly higher than 0.1 with 61.5^- phage but apparently lower than 30 with motA^- phage. Thus, the suppression of the gene 61.5 mutation by a motA mutation was rather weak. This observation prompted us to characterize the late gene expression of 61.5^- motA^- in more detail to look for a difference from that of motA^- (Fig 2A). Densitometric scanning of the radioactivity in each protein band (Fig 2B) revealed that the rates of synthesis of gp7, gp10, gp23, gp24, and gp37 were lower in cells infected with 61.5^- motA^- phage than in motA^--infected cells. Remarkably, the difference in rates of synthesis between 61.5^- motA^- and motA^- increased with time; rates in 61.5^- motA^- remained constant or declined slightly after 30 min, while increasing in motA^-.

View larger version (61K): In this window In a new window Download PPT slide   Figure 2. Gene expression of a gene 61.5 mutant in a motA- genetic background. (A) MH1 cells were infected with motA- or 61.5- motA- phage. Newly synthesized proteins were labeled and analyzed as described in MATERIALS AND METHODS. Middle-gene products are indicated by arrowheads and late-gene products by arrows. Gp43 forms a highly diffuse band in an 8% polyacrylamide gel (as seen here) for unknown reasons. The rate of synthesis of late-gene (B) or middle-gene products (C) at each time was measured by densitometry of each protein band and expressed in arbitrary units. Open and solid circles represent the rates of synthesis in motA--infected or 61.5- motA-infected cells, respectively. Because the gp23 band was close to other bands in A, the rate for this protein was derived from another experiment (not shown) in which the gp23 band was separated from others.

The fate of late-gene transcripts in the gene 61.5 mutant in a motA^- genetic background:
Because the lower proficiency of late-protein synthesis by 61.5^- motA^- than by motA^- might have resulted from a difference in mRNA abundance, we compared the quantities of some late-gene transcripts by Northern blotting. Three distinct species of full-length gene 23 transcripts can be produced by transcription initiated by late promoters located upstream of genes 21, 22, or 23 (YOUNG et al. 1981 ; KASSAVETIS and GEIDUSCHEK 1982 ). Consistent with our previous results (KAI et al. 1996 , KAI et al. 1998 ), none of these species was detectable when only gene 61.5 was mutated (data not shown). In a motA^- background, two distinct species of the full-length gene 23 transcript were detected regardless of the presence or absence of the gene 61.5 mutation (Fig 3A). The faster migrating species was monocistronic for gene 23 and the slower migrating one was polycistronic for genes 21–23. Detection of the species polycistronic for genes 22 and 23 would have been hampered by comigration with abundant 23S rRNA. The mono- and polycistronic transcripts increased with time in motA^--infected cells. In 61.5^- motA^--infected cells, on the other hand, the monocistronic transcript was detectable, although with reduced abundance. The polycistronic transcript was barely detectable, indicating that the abundance of these transcripts is lower in 61.5^- motA^--infected cells than in motA^--infected cells. We obtained similar results with the gene 24 monocistronic transcript (Fig 3B) and with the soc gene monocistronic transcript (Fig 3C). On the basis of these results, the reduced expression of late genes in 61.5^- motA^- relative to those in motA^- can be explained by partial recovery of the abundance of late-gene mRNA.

View larger version (97K): In this window In a new window Download PPT slide   Figure 3. The effect of gene 61.5 on the abundance of individual transcripts. Total RNA of MH1 cells infected with motA- or 61.5- motA- phages was extracted at 10, 20, 30, or 40 min after infection. Ten micrograms of each total RNA was subjected to Northern analysis with a probe for the T4 genes indicated in A–G after agarose gel (A, B, and D–G) or polyacrylamide gel electrophoresis (C) as described in MATERIALS AND METHODS. Both sets of motA- and 61.5- motA- phage RNAs were analyzed on the same membrane. Arrowheads indicate the transcripts of each gene. The positions of 23S rRNA (2.9 kb) and 16S rRNA (1.5 kb) are indicated by solid circles and open circles, respectively, at the right margins. In F, RNAs were analyzed by dot-blot hybridization.

A mutation in gene 61.5 does not affect transcription, and reduced late-gene expression results from extensive degradation of mRNA (KAI et al. 1996 ). Therefore, the partial recovery of mRNA abundance suggested that, although the degradation of late-gene mRNA associated with gene 61.5 mutation was suppressed by the motA mutation, it was still faster in 61.5^- motA^--infected cells than in motA^--infected ones. This notion was confirmed by measuring the decay rate of the full-length gene 24 transcript (Fig 4A) and the full-length soc gene transcript (Fig 4B) after blocking transcription with rifampicin. The half-life of the gene 24 and soc transcripts was 11 min and 40 min, respectively, in motA^--infected cells and 6.8 min and 20 min, respectively, in 61.5^- motA^--infected cells. Thus, the decay rate of late mRNA in the motA^- genetic background was higher when gene 61.5 was mutated than when it was normal.

View larger version (38K): In this window In a new window Download PPT slide   Figure 4. The effect of gene 61.5 on the decay rates of specific transcripts. MH1 cells were grown and infected with motA- or 61.5- motA- phage as described in MATERIALS AND METHODS. Rifampicin (200 µg/ml) was added to the cultures at 20 min (A, C, and D) or 25 min (B) after infection to block further transcription. Total RNAs were isolated at the times indicated and 10 µg of each total RNA was analyzed by Northern blotting with a probe for the T4 genes indicated in A–D. The half-life of each transcript was calculated by measuring signal intensities and is shown below A–D. The arrowheads indicate the transcripts of each gene. The positions of 23S and 16S rRNAs are indicated by solid and open circles, respectively, at the right margins.

Two possibilities may account for the still significant but reduced rate of mRNA degradation in the 61.5^- mutant. One is that the activity degrading mRNA itself is partially impaired, and the other is that the activation of mRNA degradation is delayed, that is, that mRNA is not destabilized early at late stages of infection but is destabilized late. As shown in Fig 3A&NDASH;C, the abundance of late transcripts increased from 20 to 40 min in motA^--infected cells. In 61.5^- motA^--infected cells, it increased from 20 to 30 min, but decreased by 40 min. This observation favored the second possibility above.

To explore this possibility more directly, we measured the decay rate of the full-length soc transcript at two different times. Its half-life in 61.5^- motA^--infected cells was 22 min when rifampicin was added at 23 min after infection. However, it was only 4.9 min when measured after adding rifampicin at 40 min. As a control, we also measured transcript half-lives of wild-type, 61.5^-, and motA^- phages at the same times in parallel experiments. In all cases, the half-life was nearly constant regardless of when it was measured: 30 min for wild type, 3.0 min for 61.5^-, and 40 min for motA^- (Fig 5 and data not shown). Thus, the half-life measured at the early time in 61.5^- motA^--infected cells was as long as that in wild-type- or motA^--infected cells, while the half-life measured at the late time was as short as that in 61.5^--infected cells. Therefore, during the late stages of 61.5^- motA^- infection, the soc transcript was initially stable but later became unstable. This result strongly suggests that the activation of soc mRNA degradation is delayed in 61.5^- motA^--infected cells.

View larger version (57K): In this window In a new window Download PPT slide   Figure 5. The decay of soc transcripts at different times. MH1 cells were infected with wild-type, 61.5-, and 61.5- motA- phage and rifampicin was added to 200 µg/ml at 23 min (B and D) or 40 min (A and C) after infection. Total RNA was isolated at 5, 12, or 17 min after rifampicin addition, and 5 µg taken at each time was analyzed by polyacrylamide gel electrophoresis and Northern blotting. The half-life (t1/2) of the full-length soc transcript was determined by measuring the signal intensity at each time and is shown at the right margins.

Cleavage of soc transcripts:
To address whether the activity degrading mRNA in 61.5^- motA^--infected cells is the same as that in 61.5^--infected cells, we investigated the degradation intermediates of soc transcripts using the primer-extension method to detect cleavages specific to the gene 61.5 mutant. In Fig 6, two prominent bands correspond to the 5' end of the full-length soc transcript and the 5' end of transcripts from which 59 nucleotides at the 5' terminus of the full-length soc transcript were deleted. This truncated species is generated by processing early transcripts initiated from early promoters 1 kb and 1.5 kb upstream of soc (MACDONALD et al. 1984 ), and it remains stable (t_1/2 = 30 min) regardless of whether gene 61.5 is wild type or mutated (T. KAI and T. YONESAKI, unpublished results). In addition, when wild-type and 61.5^- lanes are compared, three 5' terminal sites, designated TC1, TC2, and TU, were detected only with transcripts from 61.5^--infected cells, corresponding to gene 61.5 mutant-specific cleavages.

View larger version (45K): In this window In a new window Download PPT slide   Figure 6. Primer-extension analysis of soc transcripts. MH1 cells were infected with wild-type, 61.5-, 61.5- motA-, or motA- phage. Total RNA was isolated at 20, 30, or 40 min after infection, and 10 µg taken at each time was used for primer-extension analysis as described in MATERIALS AND METHODS. The product derived from the full-length soc transcript or its 5'-truncated form (see text) is indicated by an arrow. See text for detail.

The gene 61.5 mutant-specific cleavages were prominent at 20 min after infection with 61.5^-. These cleavages were at or below the detectable level at 20 min after infection with wild-type phage and at 40 min after infection with motA^-. These cleavages were also weak at 20 min after infection with 61.5^- motA^-. However, they were discernible with this double mutant at 30 min and became prominent at 40 min. Thus, cleavage in a gene 61.5 mutant-specific manner was detectable, but the induction of such activity occurred fairly late in 61.5^- motA^--infected cells.

The early termination of middle-gene expression by gene 61.5:
Comparing gene expression between 61.5^- motA^- and motA^- (Fig 2) made us aware of an additional difference—that the expression of middle genes was prolonged in 61.5^- motA^--infected cells. For example, synthesis of gp39, gp43, and gp46 was discernible until 40 min after infection with 61.5^- motA^-, although it almost ceased by 30 min with motA^- (Fig 2A). Densitometric measurements revealed that the rate of synthesis of these proteins declined abruptly after 20 min in motA^--infected cells, while it remained at nearly the maximal level until 30 min in 61.5^- motA^--infected cells (Fig 2C). This observation reveals a novel phenotype of gene 61.5 in the control of middle-gene expression, namely, that it has a role in the early termination of middle-gene expression.

The effect of gene 61.5 on the stability of middle-gene transcripts:
To determine whether gene 61.5 affects middle-gene mRNA, we compared the transcripts of genes 45 (DNA clamp protein) and 46. Both exhibited prolonged expression when gene 61.5 was mutated (Fig 2A and data not shown). In accordance with previous results (HSU and KARAM 1990 ), distinct species of transcripts of genes 45 and 46 were detected by Northern blotting. These species were designated E and M1-2 for gene 45 and E and M for gene 46 (Fig 3D and Fig E). According to HSU and KARAM 1990 , the E species corresponds to polycistronic transcripts encompassing many genes, including genes 45 and 46, and most of this species is transcribed from early promoters 7 kb upstream of gene 46. The other species correspond to transcripts from middle promoters closely upstream of each gene. At 10 min after infection, when middle-gene expression was increasing, all the E, M1, M2, and M species were abundant in both motA^-- and 61.5^- motA^--infected cells. At 20 min, when middle-gene expression was nearly maximal and late-gene expression became discernible, the quantity of each species was reduced in both motA^-- and 61.5^- motA^--infected cells. After 30 min, we could observe a critical difference in the quantity of these species between motA^- and 61.5^- motA^-. They were close to or below the detectable level in motA^--infected cells, while remaining detectable in 61.5^- motA^--infected cells through 40 min.

We also investigated other typical middle-gene transcripts, such as from rIIB (membrane protein) and uvsY (enhancer of uvsX protein). The rIIB protein is one of the easiest to identify by one-dimensional polyacrylamide gel electrophoresis. In fact, independent experiments revealed that the synthesis of rIIB protein abruptly decreased to less than the detectable level at 30 min after infection with motA^-, while only gradually decreasing and remaining still detectable until 40 min after infection with 61.5^- motA^- (refer to Fig 7). In spite of the ease of identifying rIIB protein, we detected not distinct but extensively smeared rIIB transcripts, presumably because of multiple transcription start sites and/or transcript processing. Therefore, we quantified the rIIB transcript by dot-blot hybridization. RNA hybridizable with an rIIB probe decreased rapidly at 30 min after infection with motA^-, while gradually decreasing through 40 min after infection with 61.5^- motA^- (Fig 3F). The result for uvsY transcripts is shown in Fig 3G. Distinct species designated L or M correspond to transcripts initiated from a 0.8-kb upstream late promoter or from a just-upstream middle promoter, respectively (GRUIDL et al. 1991 ). The amount of M species at 10 and 20 min after infection was similar in both motA^- and 61.5^- motA^- infections. It was decreased by 30 min and became hardly detectable by 40 min in motA^--infected cells. In contrast, it remained nearly constant from 30 through 40 min in 61.5^- motA^--infected cells. The L species in 61.5^- motA^--infected cells was less than in motA^--infected cells at all times, behaving like a late-gene transcript.

View larger version (110K): In this window In a new window Download PPT slide   Figure 7. The effect of gene 61.5 on the rate of functional decay of mRNAs. MH1 cells were infected with motA- or 61.5- motA- phage and rifampicin was added to 200 µg/ml at 25 min after infection. Newly synthesized proteins were pulse labeled for 3 min with [35S]methionine/cysteine at various times indicated and were analyzed by polyacrylamide gel electrophoresis as described in MATERIALS AND METHODS. Middle proteins are indicated by arrowheads and late proteins by arrows.

These results suggested that the prolonged expression of some middle genes in 61.5^- motA^--infected cells results from the persistence of mRNA. Transcription from middle promoters should cease after substitution of an RNA-polymerase -factor at late stages of T4 infection (STEVENS 1972 ; MATHEWS 1994 ). Furthermore, because the reduced expression of late genes described above should be caused by accelerated degradation of mRNA, it is possible that the persistence of middle-gene mRNA results from increased stability of mRNA. Indeed, the E and M transcripts of gene 46 had half-lives of 5.3 and 3.3 min, respectively, in motA^--infected cells. On the other hand, their half-lives were 11 and 10 min, respectively, in 61.5^- motA^--infected cells (Fig 4C). As judged by dot-blot analysis, the half-life of rIIB transcripts was 13 min for motA^- and 28 min for 61.5^- motA^- (data not shown). The half-life of the uvsY M transcript was 5.1 min in motA^--infected cells and 7.7 min in 61.5^- motA^--infected cells (Fig 4D). As described previously, the half-life of late-gene 24 and soc transcripts was about twofold longer in motA^--infected cells than in 61.5^- motA^--infected cells. Therefore, the effect of gene 61.5 on the stability of these middle-gene mRNAs was the opposite of that on mRNAs of the late genes 24 and soc.

The effect of gene 61.5 on the functional decay of mRNA:
The functional decay rate is estimated from the ability of mRNA to direct the synthesis of protein after transcription has been blocked by rifampicin. Because this method can provide information simultaneously about the stability of various mRNAs, we used it to test the universality of the opposite effects of gene 61.5 on the stability of middle-gene and late-gene mRNAs. Rifampicin was added to the culture at 25 min after infection with motA^- or 61.5^- motA^- and rates of synthesis of individual proteins were measured by pulse labeling with [³⁵S]methionine/cysteine and analyzing by SDS-polyacrylamide gel electrophoresis (Fig 7). The decline in the rate of protein synthesis was apparently steeper for all the middle proteins, indicated on the right side of Fig 7, when gene 61.5 was normal than when it was inactivated. In contrast, the opposite tendency was observed for all the late proteins. Table 1 summarizes the functional half-lives of mRNA calculated from the results shown in Fig 7. The half-life of middle-gene mRNA in the presence of wild-type 61.5 was 1.2- to 4.0-fold shorter than in its absence, whereas that of late-gene mRNA was 1.5- to 2.6-fold longer in the presence than in the absence of wild-type 61.5.

We also measured the functional half-life of middle-gene mRNAs in the presence of wild-type motA gene. As shown in Fig 1, middle-gene expression of wild type and 61.5^- phage was similar at 9 min after infection. It was increased in wild-type-infected cells at 18 min. By contrast, it was apparently reduced by 18 min in 61.5^--infected cells. This reduction would be attributable to gene 61.5 mutant-specific middle-gene mRNA degradation between middle and late stages (see DISCUSSION) and usually occurs between 10 and 16 min. Thereafter, middle-gene expression was gradually decreased. To measure the functional half-life of middle-gene mRNA at late stages, rifampicin was added to the culture at 16 min after infection with wild type or 61.5^-, and rates of synthesis of individual proteins were measured (Fig 8). Table 1 also summarizes the functional half-lives of mRNA calculated from the results shown in Fig 8. The half-life of middle-gene mRNA in the presence of wild-type 61.5 was 2.2-- to 3.0-fold shorter than in its absence.

View larger version (105K): In this window In a new window Download PPT slide   Figure 8. The rate of functional decay of mRNAs in the presence of a normal motA gene. MH1 cells were infected with wild-type or 61.5- phage and rifampicin was added to 200 µg/ml at 16 min after infection. Newly synthesized proteins were analyzed in the same manner as in Fig 7, except that threefold more [35S]methionine/cysteine was used to label proteins in 61.5--infected cells.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

The ssf5 mutation was isolated through its ability to suppress the growth defect of a gene 61.5 mutation (KAI et al. 1998 ). Here, ssf5 was found to be an amber mutation in motA. Middle genes are transcribed by two different mechanisms, transcription from middle promoters and readthrough transcription from upstream early promoters. Because motA mutants cannot activate transcription from middle promoters, transcription of middle genes is reduced (MATTSON et al. 1978 ; PULITZER et al. 1985 ; HINTON 1991 ). Such an effect may reduce the expression of some middle genes and secondarily delay the expression of late genes (MATTSON et al. 1978 ; PULITZER et al. 1985 ), a characteristic of motA mutants including ssf5.

A mutation in gene 61.5 strongly destabilizes late-gene mRNA, and the resulting mRNA scarcity silences many late genes (KAI et al. 1996 ). Because the motA mutation increased the synthesis of late proteins in cells infected with the gene 61.5 mutant, the stability of late-gene mRNA should be increased. The gene 61.5 mutant-specific degradation of soc transcripts is associated with mutant-specific cleavages (T. KAI and T. YONESAKI, unpublished results). The mutant-specific cleavages at TC1, TC2, and TU were also induced in 61.5^- motA^--infected cells (Fig 6). Cleavage activities at late stages were initially low and gradually became high as phage development proceeded. Therefore, it is likely that the motA mutation affects the timing of activation of gene 61.5 mutant-specific cleavages. This effect can explain why the soc transcript was initially stable at late stages and thereafter became unstable (Fig 5). Extrapolating such soc-mRNA effects to other late mRNAs may explain why the abundance of late-gene mRNA in the gene 61.5 mutant partially recovered (Fig 3) and why the synthesis of late proteins and the abundance of late transcripts ultimately ceased to increase or actually declined during late stages (Fig 2 and Fig 3). From these considerations, we conclude that suppression by the motA mutation is attributable primarily to a delay in the activation of late-gene mRNA degradation in cells infected with the gene 61.5 mutant. Because of the coincidence with a delay in the expression of late genes, the above conclusion suggests that T4 genes controlled directly or indirectly by motA activate the mechanism of late-gene mRNA degradation. Characterization of other suppressors of the gene 61.5 mutation (KAI et al. 1998 ) will help to identify such genes.

An intriguing outcome from this study is the discovery of a novel phenotype of gene 61.5 in the motA^- genetic background. As shown by measurements of chemical and functional decay rates, middle-gene mRNA was more slowly degraded at late stages of infection when gene 61.5 was mutated than when it was wild type (Fig 4 and Fig 7, Table 1). Consequently, the expression of middle genes was prolonged in cells infected with a gene 61.5 mutant (Fig 2). These effects of the gene 61.5 mutation on middle genes contrast with those on late genes (rapid degradation of late-gene mRNA and reduced expression of late genes). These phenotypes strongly suggest that gene 61.5 has opposing roles in mRNA metabolism at late stages of infection, roles of both destabilization of middle-gene mRNA and stabilization of late-gene mRNA. This unique property may be attributable solely to gene 61.5 mutation but not to a combination with motA mutation, because gene 61.5 mutation also stabilized middle-gene mRNAs at late stages of infection in the presence of the wild-type motA gene (Fig 8 and Table 1).

How does a single gene oppositely affect the stabilities of different classes of mRNA? When rates of protein synthesis were compared in the presence and absence of a normal gene 61.5 (Fig 2), there was no significant difference until 20 min after infection, when middle-gene expression was maximized. The destabilization of middle-gene mRNA by gene 61.5 became evident only thereafter. The difference in the abundance of middle-gene mRNA in the presence and absence of a normal gene 61.5 appeared from 20 to 30 min (Fig 3). The abundance of middle transcripts such as those of genes 45, 46, rIIB, and uvsY was similar at 10 and 20 min, regardless of whether gene 61.5 was normal or mutated. However, these transcripts abruptly disappeared at 30 min when gene 61.5 was normal, suggesting that the gene 61.5-specific degradation of middle-gene mRNA was activated between 20 and 30 min. These kinetics appear to parallel the kinetics of destabilization of late-gene mRNA by a gene 61.5 mutation. From these observations, we conclude that gene 61.5 has a role in targeting mRNAs for degradation. In this hypothesis, the mRNA degradation process is activated when infection proceeds into late stages, and when gene 61.5 is mutated, the process targets late-gene mRNAs. Gene 61.5 plays a role in targeting the process toward otherwise untargeted middle-gene mRNAs. As a consequence, late-gene mRNAs escape degradation. In other words, late-gene mRNA degradation specific to a gene 61.5 mutant was originally aimed at middle-gene mRNAs, when their function is completed, by the function of gp61.5.

Stabilization of late-gene mRNA and destabilization of middle-gene mRNA by gene 61.5 must require an intracellular process starting at early stages during which gp61.5 is synthesized (KAI et al. 1998 ). What, then, is the role of gene 61.5 before the late stages? The observation that middle-gene expression of a gene 61.5 mutant underwent a sudden drop after it had started normally (Fig 1) is quite suggestive of a role for this gene. In this connection, we reported a still-unexplained phenotype of the gene 61.5 mutation: uvsY mRNA and other middle-gene mRNAs were rapidly destabilized by a gene 61.5 mutation in the absence of late-gene expression (KAI et al. 1996 ). These observations suggest that gene 61.5 has a role in stabilizing middle-gene mRNA prior to late-gene expression. In support of this idea, we note that middle transcripts of genes uvsY and 45 start degrading earlier in cells infected with the gene 61.5 mutant than in cells infected with the wild type (KAI et al. 1996 ). On the basis of these facts, the role of gene 61.5 in stabilizing mRNA can now be extended to middle-gene mRNA at stages between middle and late. Middle-gene mRNA is stabilized before late stages and is selectively degraded at late stages when gene 61.5 is normal. In contrast, when gene 61.5 is mutated, middle-gene mRNA suffers from untimely degradation between middle and late stages, and at late stages late mRNA is erroneously degraded. These considerations have led us to hypothesize that gp61.5 has a function in specifying mRNAs for degradation in a stage-dependent manner and therefore we propose to call gene 61.5 the dmd (discrimination of messages for degradation) gene.

In T4-infected cells, mRNAs can be classified into four groups: host mRNA and T4 early-gene, middle-gene, and late-gene mRNAs. Each group, except late-gene mRNA, would be sequestered at suitable stages to help a quick shift of gene expression. At the early stage, host gene expression is rapidly shut off. This shut-off may be caused by degradation of host mRNA. At the middle stage, early-gene mRNA must be eliminated. T4 endoribonuclease regB can cleave both host mRNA and early-gene mRNA in the middle of the 5'-GGAG located in the ribosome binding site (UZAN et al. 1988 ), and this activity promotes specific degradation of certain host and T4 early mRNAs but not of middle and late mRNAs (SANSON et al. 2000 ). At the late stage, middle-gene expression is rapidly abolished by sequestering middle mRNA by the dmd-mediated degradation pathway. These mechanisms undoubtedly contribute to rapid shifts of gene expression and efficient multiplication of T4 phage.

	ACKNOWLEDGMENTS

We cordially thank Dr. John W. Drake at the National Institute of Environmental Health Sciences for critical reading and invaluable help with the manuscript. We thank all the staff of the Radioisotope Research Center at Toyonaka, Osaka University, where all our experiments using radioisotope were carried out, for facilitation of our research.

Manuscript received November 2, 2000; Accepted for publication January 29, 2001.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

ADELMAN, K., E. N. BRODY, and M. BUCKLE, 1998  Stimulation of bacteriophage T4 middle transcription by the T4 proteins, MotA and AsiA, occurs at two distinct steps in the transcription cycle. Proc. Natl. Acad. Sci. USA 95:15247-15252[Abstract/Free Full Text].

CHRISTENSEN, A. C., and E. T. YOUNG, 1984 Characterization of T4 transcripts, pp. 184–188 in Bacteriophage T4, edited by C. K. MATHEWS, E. KUTTER, G. MOSIG and P. B. BERGET. American Society for Microbiology, Washington, DC.

GRUIDL, M. E., T. C. CHEN, S. GARGANO, A. STORLAZZI, and A. CASCINO et al., 1991  Two bacteriophage T4 base plate genes (25 and 26) and the DNA repair gene uvsY belong to spatially and temporally overlapping transcription units. Virology 184:359-369[Medline].

HINTON, M. D., 1991  Transcription from a bacteriophage T4 middle promoter using T4 MotA protein and phage-modified RNA polymerase. J. Biol. Chem. 266:18034-18044[Abstract/Free Full Text].

HSU, T. and J. D. KARAM, 1990  Transcriptional mapping of a DNA replication gene cluster in bacteriophage T4. J. Biol. Chem. 265:5305-5316.

KAI, T., H. E. SELICK, and T. YONESAKI, 1996  Destabilization of bacteriophage T4 mRNAs by a mutation of gene 61.5. Genetics 144:7-14[Abstract].

KAI, T., H. UENO, and T. YONESAKI, 1998  Involvement of other bacteriophage T4 genes in the blockade of protein synthesis and mRNA destabilization by a mutation of gene 61.5. Virology 248:148-155[Medline].

KAI, T., H. UENO, Y. OTSUKA, W. MORIMOTO, and T. YONESAKI, 1999  Gene 61.3 of bacteriophage T4 is the spackle gene. Virology 260:254-259[Medline].

KASSAVETIS, G. A. and E. P. GEIDUSCHEK, 1982  Defining a bacteriophage T4 late promoter: bacteriophage T4 gene 55 protein suffices for directing late promoter recognition. Proc. Natl. Acad. Sci. USA 81:5101-5105.

MACDONALD, P. M., E. KUTTER, and G. MOSIG, 1984  Regulation of a bacteriophage T4 late gene, soc, which maps in an early region. Genetics 106:17-27[Abstract/Free Full Text].

MATHEWS, C. K., 1994 An overview of the T4 developmental program, pp. 1–8 in Molecular Biology of Bacteriophage T4, edited by J. D. KARAM, J. W. DRAKE, K. N. KREUZER, G. MOSIG, D. H. HALL, F. A. EISERLING, L. W. BLACK, E. K. SPICER, E. KUTTER, K. CARLSON and E. S. MILLER. American Society for Microbiology, Washington, DC.

MATTSON, T. G., V. HOUWE, and H. R. EPSTEIN, 1978  Isolation and characterization of conditional lethal mutations in the motA gene of bacteriophage T4. J. Mol. Biol. 126:551-570[Medline].

MOSIG, G., and D. H. HALL, 1994 Gene expression: A paradigm of integrated circuits, pp. 127–131 in Molecular Biology of Bacteriophage T4, edited by J. D. KARAM, J. W. DRAKE, K. N. KREUZER, G. MOSIG, D. H. HALL, F. A. EISERLING, L. W. BLACK, E. K. SPICER, E. KUTTER, K. CARLSON and E. S. MILLER. American Society for Microbiology, Washington, DC.

O'FARRELL, Z. P. and M. L. GOLD, 1973  The identification of prereplicative bacteriophage T4 protein. J. Biol. Chem. 248:5499-5505[Abstract/Free Full Text].

PULITZER, J. F., M. COLOMBO, and M. CIARAMELLA, 1985  New control elements of bacteriophage T4 pre-replicative transcription. J. Mol. Biol. 182:249-263[Medline].

SANSON, B., R. M. HU, E. TROITSKAYA, N. MATHY, and M. UZAN, 2000  Endoribonuclease RegB from bacteriophage T4 is necessary for the degradation of early but not middle or late mRNAs. J. Mol. Biol. 297:1063-1074[Medline].

STEVENS, A., 1972  New small polypeptides associated with DNA-dependent RNA polymerase of Escherichia coli after infection with bacteriophage T4. Proc. Natl. Acad. Sci. USA 69:603-607[Abstract/Free Full Text].

UZAN, M., R. FAVRE, and E. BRODY, 1988  A nuclease that cuts specifically in the ribosome binding site of some T4 mRNAs. Proc. Natl. Acad. Sci. USA 85:8895-8899[Abstract/Free Full Text].

UZAN, M., E. BRODY, and R. FAVRE, 1990  Nucleotide sequence and control of transcription of the bacteriophage T4 motA regulatory gene. Mol. Microbiol. 4:1487-1496[Medline].

WILSON, G. G., K. K. Y. YOUNG, and G. J. EDLIN, 1979  High-frequency generalized transduction by bacteriophage T4. Nature 280:80-82[Medline].

YOUNG, E. T., R. C. MENARD, and J. HARADA, 1981  Monocistronic and polycistronic bacteriophage T4 gene 23 message. J. Virol. 40:790-799[Abstract/Free Full Text].

This article has been cited by other articles:

				H. Yamanishi and T. Yonesaki RNA Cleavage Linked With Ribosomal Action Genetics, October 1, 2005; 171(2): 419 - 425. [Abstract] [Full Text] [PDF]

				Y. Otsuka and T. Yonesaki A Novel Endoribonuclease, RNase LS, in Escherichia coli Genetics, January 1, 2005; 169(1): 13 - 20. [Abstract] [Full Text] [PDF]

				L. Piesiniene, L. Truncaite, A. Zajanckauskaite, and R. Nivinskas The sequences and activities of RegB endoribonucleases of T4-related bacteriophages Nucleic Acids Res., October 14, 2004; 32(18): 5582 - 5595. [Abstract] [Full Text] [PDF]

				E. S. Miller, E. Kutter, G. Mosig, F. Arisaka, T. Kunisawa, and W. Ruger Bacteriophage T4 Genome Microbiol. Mol. Biol. Rev., March 1, 2003; 67(1): 86 - 156. [Abstract] [Full Text] [PDF]

				Y. Otsuka, H. Ueno, and T. Yonesaki Escherichia coli Endoribonucleases Involved in Cleavage of Bacteriophage T4 mRNAs J. Bacteriol., February 1, 2003; 185(3): 983 - 990. [Abstract] [Full Text] [PDF]

				T. Kai and T. Yonesaki Multiple Mechanisms for Degradation of Bacteriophage T4 soc mRNA Genetics, January 1, 2002; 160(1): 5 - 12. [Abstract] [Full Text] [PDF]