v . ¢ .x. .. . .2 "3,31 3333? m2, “3?: .3 ; p . m»; } _ ”runnia u... , n 3A,...hw. . 2...»? .. a: u .r .1 5am“: 2me ugh, .1 .K . I ‘ . ‘ 1 z . o 3 .3 x . .1 v1 . a ‘ . . ,. V1 nu 1.4. . . .. . .. 55.1.2.4”... Hum-m... 5...... , u... 5.... ‘5... : . . . . a i- v. . . , . , .43... ‘.u...fl..;u I . . .x . .4... u (I; . I.» 43”.»,qu . . .. {NHL 3.0. s. 1 .1 . .1152. :5. :‘c‘ . 51.1.31? . x ‘un. A .4 Y‘iéua .‘ I . , u p a v. .r . :5‘ 34Lulumwvfi. 54”;ch tn. :0}. .ommwwhurfi. . 3 . :32. ,, . in}. x v . ‘3 . 17‘ Ill.fl,%v«mnuvas¢¥‘.rysuw . .| desk L ‘ [VIA . .- rlfl‘ginun 43.! 38.53%}: L. i . y Yp‘ Wefvaff- .' ‘h. 4.)!!- .flh a»! . This is to certify that the thesis entitled ANTI-LATE RNA AND TRANSCRIPTIONAL CONTROL IN T4 BACTERIOPHAGE INFECTED ESCHERICHIA COLI presented by ROBERT J. FREDERICK has been accepted towards fulfillment of the requirements for PhD MICRO. PUB. HLTH. degree in Major professor Date June 16, 1976 0-7639 ABSTRACT ANTI-LATE RNA AND TRANSCRIPTIONAL CONTROL IN T4 BACTERIOPHAGE INFECTED ESCHERICHIA COLI BY Robert J. Frederick The regulation of transcription in T4 infected bacteria is both complex and diverse. I have examined some of the transcriptional events involved in the development of T4 using two separate approaches. The first takes advantage of the antibiotic rifampicin. The second involves the characterization of a unique RNA species called anti- late RNA. At a concentration of 100 ug/ml of rifampicin, only partial inhibition of 3H-uridine incorporation occurs in T4 infected E. coli K803. The residual transcription activity in the presence of the antibiotic, or rifampicin refractory synthesis, is dependent upon the time the antibiotic is added. It does not occur in E. coli B strains at the same rifampicin concentration or in the K12 strains at higher concentrations. The refractory synthesis is at least par- tially specific for certain RNA transcripts; that is, all of them are not reduced proportinately. In addition, some preliminary results suggest that the effect is also dependent upon the template being transcribed. Robert J. Frederick The second and major portion of this dissertation is devoted to the characterization of anti-late RNA in T4 infected E. coli. The anti-late transcripts are made early during infection but are com- plementary to late RNA. They constitute only 2% of the T4 specific RNA but are transcribed on over 80% of lfstrand (early strand) in the late region of the genome. Anti-late RNAs sediment in a broad peak on neutral sucrose gradients at about 20 to 22 S. They are made in the late region on the lfstrand. They are refractory to rifampicin added at 1 minute after infection. Furthermore, the examination of several T4 and E. coli mutants demonstrates that anti-late RNA synthesis is subject to the same control mechanisms as delayed early transcription. In cases where the delayed early expression is altered at the transcriptional level, the synthesis of anti-late RNA is also altered. In one instance where a translational control mechanism has been proposed, anti- 1ate transcripts were made normally. This suggests that quantitation of anti-late RNAs may serve as a criterion for distinguishing between translational and transcriptional regulation. In the discussion it is postulated that anti-late transcription is subject to strong polarity effects once initiation has occurred. When the polarity constraints are removed, synthesis occurs along the lfstrand in the late region making the complementary RNA. ANTI-LATE RNA AND TRANSCRIPTIONAL CONTROL IN T4 BACTERIOPHAGE INFECTED ESCHERICHIA COLI BY ‘ afi‘ \ Robert J: Frederick A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Microbiology and Public Health 1976 DEDICATION To Mark and Kathy, who unknowingly gave up many things for their father, and very especially to Barbara, who endured and persevered ab initio ad finem ii ACKNOWLEDGEMENTS I would like to thank Drs. H. Sadoff, R. Patterson, and E. McGroarty for serving on my committee. Special thanks to Drs. Patterson and Velicer for the use of their laboratories and equip- ment for some of this work. My deepest appreciation goes to Dr. Larry Snyder for the opportunity to work under his direction and showing me what enthusiasm for research is all about. I would also like to acknowledge the financial support from a departmental assistantship and National Institutes of Health Traineeship. iii INTRODUCTION. . . . . . LITERATURE REVIEW . . . TABLE OF CONTENTS The T4 Genome. . . . . . . . . . . . . . . . T4 and Transcriptional Regulation. . . . . . Shutoff of Host Transcription . . . T4 Induced Modifications of the Host RNA Polymerase. . . . . . . . . . . . . Regulation of Early or Prereplicative Transcription . . . . . . . . . . . Late Gene Expression. . . . . . . . . Anti-Late RNA . . . . . . . . . . . . REFERENCES. . . . . . . ARTICLE I O O O O O O O RIFAMPICIN REFRACTORY RNA SYNTHESIS IN T4 BACTERIO- PHAGE INFECTED ESCHERICHIA COLI K12 ARTICLE II. . . . . . . THE REGULATION OF ANTI-LATE RNA SYNTHESIS IN BACTERIOPHAGE T4: APPENDIX. . . . . . . . A DELAYED EARLY CONTROL iv Page 11 13 16 24 50 109 LIST OF TABLES Table ' Page ARTICLE I 1 Effect of rifampicin on the phage production by T4 glucosyltransferaseless mutants. . . . . . . . . . . . . 42 ARTICLE II 1 Bacteria and bacteriophage used. . . . . . . . . . . . . 55 2 In Vitro synthesis of anti-late on intact and sheared T4 DNA . . . . . . . . . . . . . . . . . . . . . 7O Figure 1 LIST OF FIGURES Topological model of prereplicative transcription units as proposed by O'Farrell and Gold (1973a). . . . ARTICLE I l Rifampicin refractory RNA synthesis in T4 infected E 0 C011 K803 . O O O O O O O O O O O O O O O O O O O O 2 The effect of rifampicin on phage production in E. coli K803 and BE- . . . . . . . . . . . . . . . . . 3 SDS polyacrylamide electrophoresis of T4 proteins made in the presence of rifampicin . . . . . . . . . . 4 The absence of rifampicin refractory RNA synthesis in the absence of late gene expression . . . . . . . . S Hybridization competition of T4 3H-RNA made in the presence of rifampicin . . . . . . . . . . . . . . . . + 6 The rate of 3H-thymidine incorporation in T4 and T4 HA5? (aBgt‘) infected E. coli K803 with and without rifampicin . . . . . . . . . . . . . . . . . . ARTICLE II 1 Sucrose density gradients of anti-late RNA . . . . . . 2 Degradation of unglucosylated T4 DNA in a restricting host, E. coli D675 . . . . . . . . . . . . . . . . . . 3 T4 early gene expression on degraded DNA in vivo . . . 4 Anti-late RNA made on degraded T4 DNA in vivo. . . . . 5 The effect of rifampicin on anti-late RNA synthesis. . 6 SDS polyacrylamide gel electrophoresis analysis of early gene products made on T4 am_N122 infected E. coli DG7S and DG75 RifR-Z. . . . . . . . . . . . . . . 7 The time of anti-late RNA synthesis in T4 infected E. coli xeoa and xeo3 RifR-Z . . . . . . . . . . . . . vi Page 30 33 35 38 41 44 61 65 67 69 74 77 79 Figure Page 8 Anti-late RNA synthesis on T4 infected E. coli K803 and xao3 RifR-Z. . . . . . . . . . . . . . . . . . . . . 82 9 Anti-late RNA synthesis in T4 tsGl infected E. coli BE at 41 C O O O O O O O O I O O O O I O C O O O O O O O 8 5 10 Anti-late RNA synthesis in T4 tsGl, am_E10 infected E. coil-1. BE at 43 C O O O O O O O O O O O O O O O O O O O 87 ll Anti—late RNA synthesis in T4 Acl9rev2 infected E. coli BE 0 O O C O O O O O C O O O O I O O O O O O O O 90 12 Anti-late RNA synthesis in T4 SP62, am_N122 infected E. coli BE . . . . . . . . . . . . . . . . . . . . . . . 92 APPENDIX 1 SDS polyacrylamide gel electropherogram of T4 proteins made in the presence of rifampicin at 1 minute after infection . . . . . . . . . . . . . . . . . 111 2 SDS polyacrylamide gel electropherogram of T4 tsGl gene products made at 43 C. . . . . . . . . . . . . 113 3 SDS polyacrylamide gel electropherogram of T4 Acl9rev2 gene products made at 23 C. . . . . . . . . . . 116 vii INTRODUCTION The transfer of genetic information from a deoxyribonucleic acid (DNA) genome into functional ribonucleic acid (RNA) and pro- tein molecules involves many and varied control processes. Inherent in these is the capacity to regulate growth, maintain physiological order, and preserve the temporal sequence of development. The problems to be examined and discussed in this dissertation are directed toward a better understanding of DNA dependent RNA synthesis (transcription) and its regulation using bacteriophage T4 infection of Escherichia coli as a model system. The study of T4 as a transcriptional model offers many advan- tages. It is a complex virus with a DNA coding capacity of 150 to 200 genes, approximately 60% of which have been identified. There- fore, a great deal is already known about the genetics of this phage. In addition, it has been well characterized biochemically. More important to this work, however, is T4's diversity of transcriptional regulation. The T4 phage maintain a strict transcription sequence that is regulated on at least two levels, the DNA template and the RNA polymerase. The complexity and diversity of T4 regulation, coupled with the ease of handling are why I chose to work with this system. i The first part of this dissertation describes a change in the sensitivity of the RNA polymerase to rifampicin after infection. 1 2 This antibiotic blocks initiation of RNA synthesis by binding to the 8 subunit of the enzyme. Evidence is presented that transcription is partially refractory to rifampicin at certain times after infec- tion. This can be demonstrated in K12 strains but not 8 strains of E. coli. It does not appear to be due to induced or intrinsic permeability differences, but is dependent upon late gene expression and the state of the template being transcribed. The second part of this dissertation concerns the characteriza- tion of anti-late RNA regulation in T4 infected bacteria. Anti-late RNA is a species of T4 transcripts which is made on the anti-sense strand of the late genes. It is made early during the infection on an extensive length of the late region but constitutes only 2% of the early T4 RNA. They are the size of late mRNA, are made from initiations within the late region and are controlled in the same way as delayed early mRNA transcription. The data are discussed in terms of T4 regulation of early gene expression and transcriptional control mechanisms in general. LITERATURE REVIEW The T4 Genome The T4 genome is a terminally redundant, 130 x 106 dalton double stranded DNA molecule. It is physically and genetically separable into four regions corresponding to early (2) and late (2) gene expression or alternatively pre- and postreplicative transcription units (Wood, 1974; Geiduschek et al., 1968). In general, each region has a distinct polarity; that is, transcription occurs only on one strand and, therefore, in one direction. Cuba at al. (1971) and Notani (1973) have shown that 98% or more of early mRNA is made on the DNA lfstrand and more than 80% of the late mRNA on the Efstrand. The T4 DNA contains hydroxymethylcytosine (HMC) instead of cytosine (Wyatt and Cohen, 1952). This unique base serves several functions for the virus. It has recently been determined, for example, that when cytosine replaces HMC in progeny DNA, the late genes are not transcribed (Wu and Geiduschek, 1975; Kutter et al., 1975). Szybalski et al. (1966) have shown that the HMC residues can exist in clusters. There is some evidence that these clusters have some regulatory role (Montgomery and Snyder, 1973; Snyder and Montgomery, 1974). Not only does T4 use hydroxymethylcytosine instead of cyto- sine, but this base is also glucosylated after replication. Lehman 3 4 and Pratt (1960) showed that all of the HMC residues are singly glucosylated, 70% with an a-linkage and 30% with a B-linkage. This pattern of glucosylation is characteristic of T4, the other T-even phage differing in the amount and type of glucosylation present (Mathews, 1971; Revel and Luria, 1970). It has been suggested that glucosylation has evolved as a protective mechanism against host restriction nucleases (Revel and Luria, 1970), thereby increasing the host range of the phage. T4 and Transcriptional Regulation Shutoff of Host Transcription Shortly after infection, T4 causes the cessation of all E. coli macromolecular syntheses (for a review, see Mathews, 1971). Despite a great deal of study, the mechanisms involved have not been clearly elucidated. In one report, for example, two mechanisms were pro- posed (Nomura et al., 1966). The first did not require T4 protein synthesis and seemed to be active at the membrane level by means of a colicinogenic (El and K) type reaction. The second required a T4 gene product but just how this caused the arrest of E. coli mRNA synthesis was not determined. T4 Induced Modifications of the Host RNA Polymerase T4 bacteriophage have a very complex regulatory system for gene expression. Unlike other bacteriophage such as T7 and T3, T4 uses the host RNA polymerase throughout infection (Haselkorn et al., 1969; Mizuno and Nitta, 1969; di Mauro et al., 1969). The E. coli polymerase is a large molecule composed of four subunits designated 5 a, B, 8' and 6 in the ratio of 2:1:l:l (Burgess, 1971). During the course of infection, T4 causes many changes in the enzyme, some of which have been shown to be involved in transcriptional regulation. Snyder (1973) and Pitale and Jayaraman (1975) demonstrated a change in the sedimentation profile and template specificity of RNA polymerase after T4 infection. Schachner et a1. (1971) reported a change in the electrophoretic mobility of the 8 subunit. Goff and Weber (1970) and Seifert et a1. (1971) showed that adenosine and phosphorus are covalently linked to the alpha subunit after infec- tion. Goff (1974) later reported that this was accomplished by the enzymatic attachment of adenosinediphosphoribose residues. He also showed that the adenylation occurred by two distinct mechanisms, one requiring T4 gene expression (modification) and one which did not (alteration). Recently, Horvitz (1974b) presented evidence that alteration and modification are nonessential for phage production. Rohrer et al. (1975) have successfully isolated the alteration enzyme and characterized it as a 70,000 dalton NAD+:protein ADP- ribosyltransferase. It is carried in the phage heads and is injected with the DNA (Horvitz, 1974a; Rohrer et al., 1975). Not only are these structural modifications of cellular RNA polymerase induced, but phage specific activity directed to the enzyme also occurs. For example, both T4 induced sigma (Travers, 1970) and anti-sigma (Bogdanova et al., 1970; Khesin et al., 1972; Stevens, 1974) factors have been reported. Furthermore, several phage specific polypeptides have been shown to have an affinity for the polymerase (Rather, 1974). Some of these bind so tightly they can be isolated with the enzyme (Stevens, 1972; Horvitz, 1973). 6 Stevens (1972) tentatively identified two of these as the T4 gene 55 and gene 33 products. Subsequently, Horvitz (1973) proved that the latter was indeed the T4 gene 33 product. In his study of RNA polymerase binding proteins, Ratner (1974) used affinity chroma- tography to show that the gene 45 product bound to the RNA polymerase isolated from infected E. coli. The association of the gene 45 polypeptide was also implicated in vivo by Snyder and Montgomery (1974) and Coppo et al. (1975a,b). All of the above genes are required for late gene expression of T4 (Bolle et al., 1968b; Notani, 1973; Wu and Geiduschek, 1975). Very recently, Snyder (personal communication) has obtained evidence that another of the tightly bound polypeptides originally described by Stevens (1972) is responsible for the inability of the phage to make late mRNA on cytosine containing DNA. Regulation of Early or Prereplicative Transcription Early gene expression encompasses the transcriptional events occurring before the start of DNA replication. The early transcripts have been subdivided into three classes: immediate early (IE), delayed early (DE) and quasi-late. The nomenclature derives essen- tially from the time they appear after infection and the length of time they are synthesized. A diagram of representative transcrip— tional units as discussed by O'Farrell and Gold (1973) is shown in Figure 1. IE gene products can be detected by SDS polyacrylamide gel electrophoresis approximately one minute after infection, while DE gene products are not made until two or three minutes after infection '1' Delayed Immedi 4“ E ate '1' Dela ed 1 d —~ , W H was“ H P I Figure l. Topological model of prereplicative transcrip— tion units as proposed by O'Farrell and Gold (1973a). Abbrevi- ations: (PE) immediate early promoters; (TE) termination site; (P ) quasi-late promoters; A immediate early; B quasi-late; C gelayed early. {0' \ Alt n if": . rm DOV Um- NV} NH. uric 8 (O'Farrell and Gold, 1973a). Immediate early transcripts are made in the presence of chloramphenicol, but delayed early are not (Grasso and Buchanan, 1969; Salser et al., 1970; Lembach and Buchanan, 1970; Peterson et al., 1972). This led to the conclusion that delayed early transcription required a T4 protein. 0n the other hand, amino acid analogs did not have any effect on DE transcrip- tion (Black and Gold, 1971; Lembach and Buchanan, 1970). Furthermore, other methods to disrupt protein synthesis such as amino acid star- vation (Witmer et al., 1975; Baros and Witmer, 1975) or depletion of potassium in a cell which cannot concentrate or retain the ion did not inhibit DE transcription (Morse and Cohen, 1975). It was assumed, therefore, that chloramphenicol was affecting transcription by some other manner than inhibiting protein synthesis. In 1969, Morse demonstrated that this antibiotic exerted a polar effect on the transcription of the E£p_operon. If, then, the DE genes were read from IE promoters, chloramphenicol could have been causing an analogous polar effect on early T4 transcription. In fact, there has been considerable evidence, in vivo and in vitro, that many if not most prereplicative genes are transcribed as polycistronic messages (Brody et al., 1970; Stahl et al., 1970; Btilanesi et al., 1969, 1970; Black and Gold, 1971; Witmer, 1971; ()fFarrell and Gold, l973a,b; Hercules and Sauerbier, 1974). Milanesi et a1. (1970) , for example, showed that only IE transcripts were made in vitro on a sheared DNA template. That is, DE regions ooulxi be physically separated from their promoters. In in vivo and in VJ: tro experiments, O'Farrell and Gold (l973a,b) showed that DE 9 gene products could be made when transcription initiations were blocked with rifampicin at one minute after infection. Several authors (Schmidt et al., 1970; Brody and Geiduschek, 1970; Milanesi et al., 1970; Brody et al., 1970; Schachner et al., 1971) proposed that a termination site existed between the promoter proximal (IE) and promoter distal (DE) genes. According to this hypothesis a termination factor or intrinsic termination site would have to be somehow suppressed by an anti-terminator to allow read through to the DE genes. Roberts (1969) reported one such termina- tion factor, £22! in E. 0011 which prevented transcription of bacteriophage. There have been conflicting reports of the effect of £hg_on T4 transcription. Jayaraman (1972), Travers (1970) and Richardson (1970) all reported a positive rh2_effect (i.e., termina- tion) on the in vitro transcription products. O'Farrell and Gold (1973b), however, did not see any effect of the termination factor in a T4 transcription-translation system. This may have been due to the high salt concentration which is necessary for translation inhibiting the binding of £h9_(Richardson, 1970; Schafer and Zillig, 1973b). Recently, Ratner (1976) and Richardson et a1. (1975) have established that £13 is coded for by the §_1_1_A gene. Mutations to EBA have been shown to suppress the polarity effect of chloramphenicol on the £52 operon (Morse, 1970, 1971). However, chloramphenicol inhibition of delayed early T4 transcription was not affected in ggA hosts (Young, 1975; Baros and Witmer, 1975; Witmer et al., 1975). This suggests that Ehg_may not be involved in termination of immediate early transcripts in vivo. Of course, some other termination factor 10 (e.g., kappa, Schafer and Zillig, 1973) or intrinsic termination sites could be involved. In any event, if a termination step is part of the prereplicative regulatory events, then one might expect there also to be some anti-termination factor, perhaps analogous to the N_gene product of lambda bacteriophage (Roberts, 1970). There is no direct evidence for this type of mechanism in T4. An alternative to the termination factor model has been sug— gested by Black and Gold (1971) and O'Farrell and Gold (1973a4b). They propose a passive control due to the topography of the pre- replicative transcription units. The promoter proximal genes (IE) are transcribed first, followed by the promoter distal genes (DE) from pre-early promoters (PB). The time DE are synthesized thereby would depend upon the length of the IE transcripts, that is how far the polymerase had to move before reaching the DE regions. This hypothesis is also supported by Baros and Witmer (1975). The third class of prereplicative genes is the quasi-late. These are made from newly recognized promoters as early as two minutes after infection (O'Farrell and Gold, 1973a). This was concluded from experiments using rifampicin to block new RNA polymerase initiations. When the antibiotic was added at one minute after infection, the quasi—late gene products were either reduced in quantity or not made. Those made in reduced amounts were made from both P type promoters, i.e., promoter distal genes, E and from P (quasi-late promoters) located between the IE and DE Q regions. Those not made at all in the presence of rifampicin were made only from P promoters. Additional evidence that some early Q 11 genes have more than one promoter has been reported by Hercules and Sauerbier (1974). The study of prereplicative gene expression has been slowed somewhat by the lack of mutants affecting the early regulatory events. In 1974, Mattson et a1. isolated and characterized a T4 mutant which made the PE to PQ switch inefficiently at high tempera- tures. In addition, M. Nelson and L. Gold (unpublished observations) isolated an rIIB revertant that does not make the PE to PQ switch at all low temperature. Such mutants should help in the examination of prereplicative controls in T4. Late Gene Expression Our understanding of T4 late transcription has lagged behind that of early transcription. One reason for this has been the lack of a well defined in vitro system for making late transcripts. Only crude systems could be made to transcribe late regions (Snyder and Geiduschek, 1968; Maor and Shalitin, 1974). Despite a general lack of in vitro data, many interesting events involved in late transcrip- tion have been reported. Late gene transcription has been shown to require modifications in the host polymerase, processing of the DNA and simultaneous repdication, and HMC instead of cytosine in the template. RNA £x11ymerase modifications have already been discussed (see above). TTH3 coupling of replication to late transcription was shown by Riva et.£il. (1970a). Stopping DNA synthesis in temperature shift experi- ments also resulted in the cessation of late transcription. This was not an absolute requirement, however. In another report, Riva 12 et al. (1970b) demonstrated that the replication dependence could be bypassed by mutations in the T4 DNA ligase gene which resulted in nicked DNA. This led to the hypothesis that "competent" DNA was necessary for late transcription (Riva et al., 1970b). Subsequently, Wu and Geiduschek (1973, 1975) have shown that some late transcrip- tion does occur in the absence of replication but at a much reduced level. A more complete block of late transcription occurs when cytosine replaces HMC in the DNA. Kutter (1975) has shown that normal amounts of DNA are made in mutants which introduce cytosine into progeny DNA and in which certain T4 nucleases are not present. In these infections, less late protein synthesis occurs than when replication is inhibited. As mentioned earlier, there is new evidence that one of the T4 gene products which binds to the polymerase is responsible for the block to late transcription on cytosine containing DNA (Snyder, to be published). There is also some suggestion that glucosylation may have some role in late transcription. Riva et al. (1970a), for example, reported a more dramatic cessation of late gene expression after blocking replication when the progeny DNAs were unglucosylated. In addition, Montgomery and Snyder (1973) demonstrated that a block of T4 late transcription caused by a mutation to the host polymerase was suppressed by mutations in the T4 8 glucosyl transferase gene (Sgt). Interestingly, the Bgt- mutations did not suppress other defects in the T4 reproduction process in the host mutant (Snyder auui Montgomery, 1974; Montgomery, PhD Dissertation, M80, 1975). Betzi glucosylation occurs predominantly on adjacent HMC residues r .1 .l 197 r9"; ‘V. .. ?n., -.‘.q ‘! ‘1- «“EE (A Vb". l3 (diWaard et al., 1967). Since HMC can occur in clusters (Szybalski et al., 1966) and these would be primarily B glucosylated, it was suggested that these regions may have some significance in the regulation of late transcription (Montgomery and Snyder, 1973). As mentioned before, the glucosylated state of the DNA pro- tects it from host nucleases (Revel and Luria, 1970). This is observed as a 100,000-fold difference in plating efficiencies between the wild type T4 and unglucosylated T4 (Revel and Luria, 1970; Cox and Conway, 1973). If the glucosylated state did evolve under what appears to be strong selective pressure, it seems reasonable to think that there are phage specific mechanisms to facilitate the transcription of glucosylated DNA. This may be indicated in the decreased burst size of unglucosylated phage when grown on a non-restricting host (Georgopoulos and Revel, 1971) and the data of Montgomery and Snyder (1973) mentioned above. Anti-Late RNA The accepted view that transcription is strictly asymmetric has recently come under some question. There is now increasing evidence that symmetrical RNA is made in a number of biological systems. These include the animal viruses: SV40 (Aloni, 1972, 1973), vaccinia (Colby and Duesberg, 1969), Adenovirus 2 (Zimmer and Raskas, 1976), herpes simplex (Zeev and Beker, 1975), and polyoma (Aloni and Locker, 1973); HeLa cell mitochondria (Aloni and Attardi, 1971, 1972; Young and Attardi, 1975; Murphy et a1. , 1975); and the bacterial viruses A (Ehdvre and Szybalski, 1969), the Pseudomonas phage, gh-l (James 3011);, PhD Dissertation, M80, 1976), and T4 (Jurale et al., 1970; Geidtlschek and Grau, 1970; Notani, 1973). n I ‘V I W P r u. v. Er AG “‘4 ‘ ml 5. he; #L ..,5 Cam 14 In 1970, Jurale et al. reported the isolation of double stranded RNA fragments from T4 infected Escherichia coli. These were shown to be phage specific molecules with molecular weights estimated at 8.5 x 104 to 8 x 105. The authors presumed these arose from the overlapping transcription of opposite DNA strands; a pro— posal that was previously suggested by B¢vre and Szybalski (1969) for the b2 region of lambda phage. Geiduschek and Grau also published data in 1970 that demon- strated symmetrical RNA synthesis by the T4 phage. Complementary RNA was assayed by using in vitro annealing conditions and digesting noncomplementary single strand molecules with RNases. Using this technique, they determined that this RNA was made early and hybridized with late mRNA. They called this complementary species antimessenger RNA and postulated that it was made by termination errors resulting in "read through" from the early to late regions on the lystrand of the DNA. In related work, Notani (1973) characterized T4 antimessenger RNA further. He found that antimessage transcripts were made between two and six minutes after infection. They are not made in the presence of chloramphenicol. The decay rates of this RNA were comparable to normal mRNA. Antimessenger RNA constituted approxi- rnately 2% of the early RNA being transcribed at any one time, but was made from about 81% of the late region. Young (1975) later cxnrfirmed this estimate by showing that up to 90% of the lfstrand (was transcribed early during infection. From his results, Notani (1973) argued against a "read through" type synthesis of anti- messenger RNA. If one allowed that T4 transcription proceeds at 15 the rate of one gene per half minute (Bremer and Yuan, 1968), and remembering that late genes of T4 are clustered, synthesis of the complementary RNA species in the amounts observed would require about 14 minutes by a defective termination mechanism. In view of this, Notani postulated that these anti-sense transcripts may be the result of mistakes in initiation or a change in initiation specificity of RNA synthesis. However they are made, it is possible that antimessenger RNA may be functional. Wu et al. (1973) have suggested some possibili- ties. These included a role in regulating protein synthesis, replication or initiation of RNA synthesis. There is, however, no clue to any functional role for these RNAs at the present time. REFERENCES Aloni, Aloni, Aloni, Aloni, Aloni, Baros , Black, REFERENCES Y. 1972. Extensive symmetrical transcription of simian virus 40 DNA in virus-yielding cells. Proc. Nat. Acad. Sci., USA 62: 2404—2409. Y. 1973. Poly A and symmetrical transcription of SV40 DNA. Nat. New Biol. 243: 2-5. Y., and Attardi, G. 1971. Symmetrical in vivo transcription of mitochondrial DNA in HeLa cells. Proc. Nat. Acad. Sci., USA 66: 1757-1761. Y., and Attardi, G. 1972. Expression of the mitochondrial genome in HeLa cells: XI. Isolation and characterization of the transcription complexes of mitochondrial DNA. J. Mol. Biol. 29; 363-373. Y., and Locker, H. 1973. Symmetrical in viva transcription of Polyoma DNA and the separation of self-complementary viral and cell RNA. Virology E23 495-505. A., and Witmer, H. J. 1975. Effect of chloramphenicol and starvation for an essential amino acid on the synthesis and decay of T4 bacteriophage-specific messengers transcribed from early and quasi-late promoters. Arch. Biochem. Biophys. 162: 415-427. L. W., and Gold, L. M. 1971. Pre-replicative development of the bacteriophage T4: RNA and protein synthesis in vivo and in vitro. J. Mol. Biol. 66: 365-388. Bogdanova, E. S., Zagraff, I. H., Yu, N., Bass, I. A., and Shemyakin, Bolle, Bolle, M. F. 1970. Free subunits of RNA polymerase in normal and phage infected cells of E. coli. Molec. Biol. 2; 435-444. A., Epstein, R. H., Salser, W., and Geiduschek, E. P. 1968a. Transcription during bacteriophage T4 development: Synthesis and relative stability of early and late RNA. J. Mol. Biol. ‘61: 325-348. A., Epstein, R. H., Salser, W., and Geiduschek, E. P. 1968b. Transcription during bacteriophage T4 development: Require- ments for late messenger synthesis. J. Mol. Biol. 3;: 339-362. 16 3m Brew: Bur; Cox dew c134 l7 B¢vre, K., and Szybalski, W. 1969. Patterns of convergent and overlapping transcription within the b2 region of coli phage A. Virology 66: 614-626. Bremer, H., and Yuan, D. 1968. Chain growth rate of messenger RNA in Escherichia coli infected with bacteriophage T4. J. Mol. Brody, E. N., Diggelman, H., and Geiduschek, E. P. 1970. Transcrip- tion of the bacteriophage T4 template. Obligate synthesis of T4 prereplicative RNA in vitro. Biochem. 2; 1289-1299. Brody, E. H., and Geiduschek, E. P. 1970. Transcription of the bacteriophage T4 template. Detailed comparison of in vitro and in vivo transcripts. Biochem. 2; 1300-1309. Burgess, R. R. 1971. RNA polymerase. Ann. Rev. Biochem. 39: 711- 740. Colby, C., and Duesberg, P. H. 1969. Double stranded RNA in vaccinia virus infected cells. Nature 222: 940-944. Coppo, A., Manzi, A., Pulitzer, J. F., and Takahashi, H. 1975a. Host mutant (tabD)-induced inhibition of bacteriophage T4 late transcription. 1. Isolation and phenotypic characteri- zation of the mutants. J. Mol. Biol. 26; 579-600. Coppo, A., Manzi, A., Pulitzer, J. F., and Takahashi, H. 1975b. Host mutant (tabD)-induced inhibition of bacteriophage T4 late transcription. II. Genetic characterization of mutants. J. Mol. Biol. 26; 601-624. Cox, G. S., and Conway, T. W. 1973. Template properties of glucose deficient T-even bacteriophage DNA. J. Virol. lg: 1279-1287. deWaard, A., Ubbink, T. E. C. M., and Beukman, W. 1967. On the specificity of bacteriophage induced hydroxymethylcytosine glucosyl transferases. Eur. J. Biochem. 33 303-308. diMauro, E., Snyder, L., Marino, P., Lamberti, A., Coppo, A., and Tocchini-Valentini, G. P. 1969. Rifampicin sensitivity of the components of DNA-dependent RNA polymerase. Nature 282: 533-537. Geiduschek, E. P., and Grau, O. 1970. In Lepetit C011. Biol. Med. I, ed. by L. Silvestri, p. 190-203. North Holland Publ. Co., Amsterdam. Geiduschek, E. P., Brody, E. N., and Wilson, D. L. 1968. Some aspects of RNA transcription. Mblecular Associations in Biology. Academic Press, New York, ed. by B. Fullman. 18 Georgopoulos, C. P., and Revel, H. R. 1971. Studies with glucosyl transferase mutants of T-even bacteriophages. Virology 33; 271-285. Goff, C. G. 1974. Chemical structure of a modification of the Escherichia coli ribonucleic acid polymerase polypeptides induced by T4 infection. J. Biol. Chem. 249: 6181-6191. Goff, C. G., and Weber, K. 1970. A T4 induced RNA polymerase a subunit modification. Symp. Quant. Biol. 26: 101-108. Grasso, R. J., and Buchanan, J. M. 1969. Synthesis of early RNA in bacteriophage T4 infected Escherichia coli B. Nature 224: 882-888. Guha, A., Szybalski, W., Salser, W., Bolle, H., Geiduschek, E. P., and Pulitzer, J. F. 1971. Controls and polarity of transcrip- tion during bacteriophage T4 development. J. Mol. Biol. 62; 329-349. Haselkorn, R., Vogel, M., and Brown, R. D. 1969. Conservation of the rifamycin sensitivity during T4 development. Nature 221: 836-838. Hercules, K., and Sauerbier, W. 1974. Two modes of in vivo tran- scription for genes 43 and 45 of phage T4. J. Virol. 23; 341-348. Horvitz, R. H. 1973. Polypeptide bound to the host RNA polymerase is specified by T4 control gene 33. Nat. New Biol. 244: 137-140. Horvitz, R. H. 1974a. Control by bacteriophage T4 of the two sequential phosphorylations of the alpha subunit of Escherichia coli RNA polymerase. J. Mol. Biol. 22: 727-728. Horvitz, R. H. 1974b. Bacteriophage T4 mutants deficient in altera- tion and modification of the alpha subunit of Escherichia coli RNA polymerase. J. Mol. Biol. 29; 739-750. Jayaraman, R. 1972. Transcription of bacteriophage T4 DNA by Escherichia coli RNA polymerase in vitro: Identification of some immediate-early and delayed-early genes. J. Mol. Biol. 19; 253-263. Jurale, C., Kates, J. R., and Colby, C. 1970. Isolation of double- stranded RNA from T4 phage infected cells. Nature (London) 226: 1027-1029. Khesin, R. B., Bogdanova, E. S., Goldfarb, A. D., and Zograff, Yu N. 1972. Competition for the DNA template between RNA polymerase molecules from normal and phage infected E. coli. Molec. Genetics 222: 299-314. hit a: Mi 19 Kutter, E., Beug, A., Sluss, R., Jensen, L., and Bradley, D. 1975. The production of undegraded cytosine-containing DNA by bacteriophage T4 in the absence of dCTPase and endonucleases II and IV, and its effects on T4-directed protein synthesis. J. Mol. Biol. 22; 591-607. Lehman, I. R., and Pratt, E. A. 1960. On the structure of the glucosylated hydroxymethyl cytosine nucleotides of coli phages T2, T4 and T6. J. Biol. Chem. 235: 3254-3259. Lembach, K. J., and Buchanan, J. M. 1970. The relationship of protein synthesis to early transcriptive events in bacterio- phage T4 infected Escherichia coli B. J. Biol. Chem. 246: 1575-1587. Maor, G., and Shalitin, C. 1974. Competence of membrane-bound T4rII DNA for in vitro "late" mRNA transcription. Virology 62: 500-511. Mathews, C. 1971. Bacteriophage Biochemistry. Van Nostrand Reinhold Co., New York. Mattson, T., Richardson, J., and Goodin, D. 1974. Mutant of bacteriophage T4D affecting expression of many early genes. Nature 250: 48-50. Milanesi, G., Brody, E. N., and Geiduschek, E. P. 1969. Sequence of the in vitro transcription of T4 DNA. Nature (London) 221: 1014-1016. Milanesi, G., Brody, E. N., Grau, 0., and Geiduschek, E. P. 1970. Transcriptions of the bacteriophage T4 template in vitro: Separation of "delayed early" from "immediate early" transcrip- tion. Proc. Nat. Acad. Sci., USA 66: 181-188. Mizuno, S., and Nitta, K. 1969. Effect of streptovaricin on RNA synthesis in phage T4 infected Escherichia coli. Biochem. Biophys. Res. Commun. 26; 127-130. Montgomery, D. L., and Snyder, L. R. 1973. A negative effect of B-glucosylation on T4 growth in certain RNA polymerase mutants of Escherichia coli: Genetic evidence implicating pyrimidine rich sequences of DNA in transcription. Virology 62; 349-358. bhorse, D. F. 1970. "Delayed-early" mRNA for the tryptophan operon. An effect of chloramphenicol. Symp. Quant. Biol. 26; 495-496. amorse, D. E. 1971. Polarity induced by chloramphenicol and relief by ggA. J. Mol. Biol. 66; 113-118. .Morse:, J. W., and Cohen, P. S. 1975. Synthesis of functional bacteriophage T4-de1ayed early mRNA in the absence of protein synthesis. J. Virol. 26; 330-339. tier. 5:: .. Us ‘- 4. FE 1‘! 20 Murphy, W. I., Attardi, 8., Tu, C., and Attardi, G. 1975. Evidence for complete symmetrical transcription in vivo of mitochondrial DNA in HeLa cells. J. Mol. Biol. 22: 809-814. Nomura, M., Witten, C., Mantei, N., and Echols, H. 1966. Inhibi- tion of host nucleic acid synthesis by bacteriophage T4 effect of chloramphenicol at various multiplicities of infection. J. Mol. Biol. 21: 273-278. Notani, G. W. 1973. Regulation of bacteriophage T4 gene expression. J. Mol. Biol. 12: 231-249. O'Farrell, P. 2., and Gold, L. M. l973a. Bacteriophage T4 gene expression: Evidence for two classes of prereplicative cistrons. J. Biol. Chem. 248: 5502-5511. O'Farrell, P. 2., and Gold, L. M. 1973b. Transcription and transla- tion of prereplicative bacteriophage T4 genes in vitro. J. Biol. Chem. 248: 5512-5519. Peterson, R. F., Cohen, P. 8., and Ennis, H. L. 1972. Properties of phage T4 messenger RNA synthesized in the absence of protein synthesis. Virology 46: 201-206. Ratner, D. 1974. The interaction of bacterial and phage proteins immobilized Escherichia coli RNA polymerase. J. Mol. Biol. 66: 373-383. Ratner, D. 1976. Evidence that mutations in the 66A polarity suppressing gene directly affect termination factor rho. Nature 259: 151-153. Revel, H. R., and Luria, S. E. 1970. DNA glucosylation in T-even phage: Genetic determination and role in phage host inter- action. Ann. Rev. Genetics 4: 177-192. Richardson, J. P. 1970. Rho factor function in T4 RNA transcrip- tion. Symp. Quant. Biol. 26: 127-133. Richardson, J. P., Grimley, C., and Lowery, C. 1975. Transcription termination factor rho activity is altered in Escherichia coli with 66A gene mutations. Proc. Nat. Acad. Sci., USA 12: 1725-1728. Iliva, 8., Cascino, A., and Geiduschek, E. P. 1970a. Coupling of late transcription to viral replication in bacteriophage T4 development. J. Mol. Biol. 64: 85-102. Riwna, S., Cascino, A., and Geiduschek, E. P. 1970b. Uncoupling of late transcription from DNA replication in bacteriophage T4 development. J. Mol. Biol. 64: 103-119. 21 Roberts, J. W. 1969. Termination factor for RNA synthesis. Nature 224: 1168-1174. Roberts, J. W. 1970. The p factor: Termination and antitermina- tion in lambda. Symp. Quant. Biol. 66: 121-126. Rohrer, H., Zillig, W., and Mailhammer, R. 1975. ADP-Ribosylation of DNA-dependent RNA polymerase of Escherichia coli by an NAD+:protein ADP-ribosyltransferase from bacteriophage T4. Eur. J. Biochem. 62: 227-238. Salser, W., Bolle, A., and Epstein, R. 1970. Transcription during bacteriophage T4 development: A demonstration that distinct subclasses of the "early" RNA appear at different times and that some are "turned off" at late times. J. Mol. Biol. 42: 271-295. Schachner, M., Seifert, W., and Zillig, W. 1971. A correlation of changes in host and T4 bacteriophage specific RNA synthesis with changes of DNA dependent RNA polymerase in Escherichia coli infected with bacteriophage T4. Eur. J. Biochem. 22: 520-528. Schafer, R., and Zillig, W. 1973. K (kappa), a novel factor for the arrest of transcription in vitro by DNA dependent RNA polymerase from Escherichia coli at specific sites of natural templates. Eur. J. Biochem. 66: 201-206. Schmidt, D. A., Mazaitis, A. J., Kasai, T., and Bautz, E. R. F. 1970. Involvement of a phage T4 6 factor and an anti- terminator protein in the transcription of early T4 genes in vivo. Nature 226: 1012-1016. Seifert, W., Rabussay, D., and Zillig, W. 1971. On the chemical nature of alteration and modification of DNA dependent RNA polymerase of E. coli after T4 infection. FEBS Letters 46: 175-179. Snyder, L. R., and Montgomery, D. L. 1974. Inhibition of T4 growth by an RNA polymerase mutation of Escherichia coli: Physio- logical and genetic analysis of the effect during phage development. Virology 62: 184-196. Snyder, L. 1973. Change in RNA polymerase associated with the shut off of host transcription by T4. Nat. New Biol. 243: 131-134. Snyder, L., and Geiduschek, E. P. 1968. In vitro synthesis of T4 late messenger RNA. Proc. Nat. Acad. Sci., USA 62: 459-466. Stahl, F. W., Crasemann, J. M., Yegian, C. D., Stahl, M. M., and Nakata, A. 1970. Cotranscribed cistrons in bacteriophage T4. Genetics 64: 157-170. 0 ‘5‘ ('1‘ all 22 Stevens, A. 1972. New small polypeptides associated with DNA dependent RNA polymerase of Escherichia coli after infection with bacteriophage T4. Proc. Nat. Acad. Sci., USA 62: 603- 607. Stevens, A. 1973. An inhibitor of host sigma-stimulated core enzyme activity that purifies with DNA dependent RNA polymerase of E. coli following T phage infection. Biochem. Biophys. Res. Commun. 64: 48 -493. Szybalski, W., Kubinski, H., and Sheldrick, P. 1966. Pyrimidine clusters on the transcribing strand of DNA and their possible role in initiation of RNA synthesis. Symp. Quant. Biol. 64: 123-127. Travers, A. A. 1970. Positive control of transcription by a bac- teriophage sigma factor. Nature 225: 1009-1012. Witmer, H. J. 1971. In vitro transcription of T4 deoxyribonucleic acid by Escherichia coli ribonucleic acid polymerase: Sequential transcription of immediate early and delayed early cistrons in the absence of release factor, rho. J. Biol. Chem. 142: 5220-5227. Witmer, H. J., Baros, A., and Forbes, J. 1975. Effect of chloram- phenicol and starvation for an essential amino acid on poly- peptide and polyribonucleotide synthesis in Escherichia coli infected with bacteriophage T4. Arch. Biochem. Biophys. 462: 406-414. Wood, W. B. 1974. Bacteriophage T4, 12_Handbook of Genetics, ed. R. C. King, Vol. 1, pp. 327-331. Plenum Press, New York. Wu, R., and Geiduschek, E. P. 1975. The role of replication pro- teins in the regulation of bacteriophage T4 transcription. I. Gene 45 and hydroxymethyl-C-containing DNA. J. Mol. Biol. 26: 513-538. Wu, R., Geiduschek, E. P., Rabussy, D., and Cascino, H. 1973. Regulation of transcription in bacteriophage T4-infected E. coli - A brief review and some recent results, 22_Virus Research, eds. C. F. Fox and W. S. Robinson, pp. 181-204. Academic Press, New York. hbfiitt, G. R., and Cohen, S. S. 1952. A new pyrimidine base from bacteriophage nucleic acids. Nature 170: 1072-1073. Young, E. T. 1975. Analysis of T4 chloramphenicol RNA by DNA: RNA hybridization and by cell free protein synthesis, and the effect of E. coli polarity suppressing alleles on its synthesis. J. Mol. Biol. 26: 393-424. Young Zeev 23 Young, P. G., and Attardi, G. 1975. Characterization of double stranded RNA from HeLa cell mitochondria. Biochem. Biophys. Res. Commun. 66: 1201-1207. Zeev, A. B., and Beker, Y. 1975. Symmetrical transcription of herpes simplex virus DNA in infected BSC-l cells. Nature 254: 719-727. Zillig, W., Zechel, K., Rabussay, D., Schachner, M., Sethi, V. 8., Palm, P., Heil, A., and Seifert, W. 1910. On the role of different subunits of DNA dependent RNA polymerase from E. coli in the transcription process. Symp. Quant. Biol. 66: 47-58. Zimmer, S. G., and Raskas, H. J. 1976. Synthesis of complementary viral transcripts early in productive infection with adeno- virus 2. Virology 22: 118-126. ARTICLE I RIFAMPICIN REFRACTORY RNA SYNTHESIS IN T4 BACTERIOPHAGE INFECTED ESCHERICHIA COLI K12 ROBERT J. FREDERICK AND IOREN R. SNYDER Manuscript to be submitted for publication 24 tray 9‘. £19 25 ABSTRACT Rifampicin is a potent and specific inhibitor of RNA synthesis. In this report we describe a rifampicin refractory RNA synthesis in T4 infected E. coli K12 strains. This occurs at concentrations of the antibiotic that completely inhibit host transcription. It is dependent upon the time rifampicin is added, occurring only after 5 minutes at 37 C. In addition, we present evidence that only some transcription is refractory and that the effect is dependent upon the template being transcribed. INTRODUCTION T4 bacteriophage use the E. coli RNA polymerase during infec- tion. This was affirmed by the conservation of the phenotype of polymerase mutations, with respect to its rifampicin or streptoly- digin sensitivity, throughout the latent period (Haselkorn et al., 1969; di Mauro et al., 1969; Mizuno and Nitta, 1969). All of these studies were performed using E. coli B as the host. In attempting to repeat these results using a K12 strain of E. coli (K803), we found that at a concentration of 100 ug/ml, rifampicin depressed the rate of 3H-uridine incorporation 70 to 80% of that of an infected untreated control. The same concentration, however, resulted in a greater than 97% inhibition in the uninfected bacteria. In this investigation, we have partially characterized the effect of rifampicin on T4 infected E. coli K803. We present evidence thatzsuggests a differential sensitivity to the antibiotic at intermediate concentrations. In addition, we postulate that it is 26 dependent upon the state of the enzyme as well as the template being transcribed. MATERIALS AND METHODS Bacteria and bacteriophage: Escherichia coli K803 (erK' rgl-) and NF58 (EEsu-) have been described (Montgomery and Snyder, 1973). Escherichia coli BE was obtained from L. Gold. T4+ is a laboratory wild type strain. All other T4 strains have been previously referenced (Montgomery and Snyder, 1973; Snyder and Montgomery, 1974). All of the phage were purified on 3 m1 CsCl step gradients at 35,000 rpm in a Spinco SW50 Rotor. The phage band was collected and dialyzed against sequential changes of 2, l, and 0.5 M NaCl and finally M9 buffer. Media and chemicals: M98 and tryptone are the same as described previously (Snyder and Montgomery, 1974). M9AA is M9 medium supplemented with 20 ug/ml each of alanine, aspartate, cysteine, glutamate, glycine, histidine, isoleucine, lysine, ornithine, phenylalanine, proline, serine, threonine, tryptophan, tyrosine, valine, arginine and methionine. 4C-leucine and rifampicin were purchased from Schwarz-Mann Company. Stock solutions of the antibiotic were prepared by dissolving it first in methanol (10 mg/0.1 m1) and then diluting with continuous mixing in water or M9S to give a final concentration of 1 mg/ml. Uridine-S-BH and thymidine-methyl-3H were purchased from New England Nuclear. 27 Preparation of cultures: The bacteria for each experiment were transferred to fresh medium (1:50) from overnight cultures stored at 4 C. Absorbance at 625 nm was measured on a Spectronic 20 spec- trometer. An A625 of 0.4 corresponded to 4 x 108 bacteria/m1. The bacteria were infected at a multiplicity of infection (MOI) of 5. Bacterial survivors were determined at 2 or 3 minutes after infection by diluting them with 0.85% saline and plating them on tryptone agar. In all experiments reported here, survivors were less than 2%. Phage production was measured as previously described (Snyder and Montgomery, 1974). Measuring the rate of RNA and DNA synthesis: The procedure for measuring the rate of 3H-thymidine incorporation was previously described (Snyder and Montgomery, 1974). The rate of RNA synthesis was measured in the same fashion using 1 uCi and 1 ug/ml of 5-3H- uridine. Trichloroacetic acid (TCA) precipitates were collected on Gelman cellulose triacetate filters, dried and counted in a Packard ambient temperature scintillation counter using a toluene base fluor. Alkali stability of 3H-uridine labeled TCA precipitates: TCA ;precipitates of cultures labeled in the presence of rifampicin were resuspended in 0.5 M KOH. One half was reprecipitated immediately .and the other stored overnight at 37 C. After this incubation, the sample was precipitated with 10 volumes of 5% TCA. The precipitates were collected and counted as described above. 28 SDS-slab gel electrophoresis: To label T4 proteins, the bacteria were grown in M9AA. Samples (2 ml) of the infected cells were transferred to test tubes containing 2 uCi/ml of 14C-leucine (312 mCi/mMole). Samples were prepared and electrophoresed on a 12.5% polyacrylamide gel according to Studier (1973). The gel was dried under vacuum and developed on Kodak no screen X-ray film. Hybridization competition assays: 3H-RNA.was labeled, extracted and assayed as described by Bolle et al. (1968a). All competitions were done at DNA excess. The T4 DNA was extracted from CsCl puri- fied virus according to the procedure of Mandell and Hershey (1960). RESULTS Rifampicin refractory3H-uridine incorporation: When E. coli K803 is treated with 100 ug/ml of rifampicin, the rate of 3H- uridine incorporation decreases to less than 4% of untreated bac- teria within 10 minutes. However, when these same bacteria are infected with T4+ and subsequently treated with the same concentra- tion of antibiotic, uridine incorporation drops to between 20 and 40% of an untreated, infected control (Figure 1). We shall refer to this persistent incorporation in the presence of the antibiotic as rifampicin refractory activity. As seen in Figure l, the refractory synthesis is dependent upon the time of addition of rifampicin. In addition, it is concen- tration dependent. Since doubling the amount of antibiotic added at 5 minutes resulted in a 90% decrease of uridine incorporation (data not shown), this tends to exclude the possibility of some 29 Figure 1. Rifampicin refractory RNA synthesis in T4 infected E. coli K803. The rate of 3H-uridine incorporation was measured for 0.1 ml samples of a K803 culture infected at time 0. Rifampicin at a final concentration of 100 ug/ml was added at 2 min (0), 5 min (Cl) , and 10 min (0) after infection. Results are reported as the percent of uridine incorporated in an infected, untreated control during the same 2 min interval. lrnfirwrrarnnrrfirorl l lrir'h'np L43 50 A 0 OJ 0 H‘ Uridine Incorporation ES (% of untreated control) N O O 30 P l J l l l l O 5 l0 IS 20 25 30 35 4 Time after infection (min) Figure 1 45 31 mechanism for uridine incorporation other than the host polymerase which may be resistant to rifampicin. To insure that we were measuring RNA synthesis, we determined the alkali sensitivity of acid-insoluble counts (see Methods). The alkali treated samples lost 96 to 97% of the TCA precipitable radioactivity. Since previous studies done with B strains of E. coli had not shown any rifampicin refractory synthesis (Haselkorn et al., 1969; di Mauro et al., 1969), we tried to duplicate our results in T4 infected E. coli BE. In this case, RNA synthesis was inhibited to the same extent (approximately 90%) independent of the time of addition of rifampicin (data not shown). The difference between K12 and B strains is also reflected in the phage production and synthesis of T4 gene products when rifampicin was added at 5 minutes after infection at 37 C (Figures 2 and 3). 0n BE' phage production was less than 1% of that in the untreated control. Similar data were also reported by Rosenthal and Reid (1973). On K803, however, phage production reached 25% of the control. Similarly, late gene product synthesis in rifampicin treated BE is greatly reduced within 15 minutes after the addition of the antibiotic (Figure 3). In addition, the continued synthesis of early gene products is also indicated (see arrows). This effect of rifampicin was also seen by O'Farrell and Gold (1973). The con- tinued early gene expression was attributed to the lack of new mRNA transcripts which could successfully compete with old transcripts for translation. In contrast to the rifampicin effect on BE' on K803 all of the ruxrmally occurring late gene products were made. The overall amounts 32 Figure 2. The effect of rifampicin on phage production in E. coli K803 and BE- The bacteria were grown and infected in M9AA medium at 37 C. Rifampicin, 100 ug/ml, was added at 5 minutes after infection. (Cl) K803 (control), (0) 83 (control), (I) K803 + rifMpicin, (0) BE + rifampicin. Log No. Bacteriophage/Cell 33 _ 4.. K803 - -¢ :3 BE - I, 1+ 4K803 +Rif. ABE ' +Rif. I0 20 30 40 50 6 0 Time after infection (min) Figure 2 34 Figure 3. SDS polyacrylamide electrophoresis of T4 proteins made in the presence of rifampicin. Samples (2 ml) from the infections shown in Figure 2 were labeled with 14C—leucine (2 uCi, 312 mCi/mMole) from 20 to 25 minutes after infection. Samples were treated identically and equal volumes (0.02 ml) containing 2.6 to 6.2 x 103 TCA precipitable CPM were analyzed. 35 K 803 BE Ce! pIPlII— H an m».— Lu. o... Rifom picin Figure 3 36 appeared to be less, but the gel pattern was almost identical to the untreated control. This implied that late mRNA transcripts were being initiated in the presence of rifampicin since late synthesis had not yet begun at 5 minutes after infection. Rifampicin refractory RNA synthesis does not occur in the absence of late gene expression: The implications from the time dependence and differences in late gene expression prompted a test to see if the rifampicin refractory synthesis could occur in the absence of late gene expression. To do this, an Emsu- K12 strain (NF58) was infected with various T4 amber mutants which do not make late gene products on nonsuppressing hosts and the effect of rifampicin on uridine incorporation followed. A comparison of wild type and gmN8l (gene 41) is shown in Figure 4. Rifampicin refractoy synthesis did not occur in T4+ infected NF58 until 10 minutes after infection. This was expected since this strain grows slightly less well than K803 and the latent periods are protracted accordingly. The decrease in refractory synthesis at 30 to 35 minutes for the 15 minute curve was due to lysis of the culture. In the gmNBl infection, 3H-uridine incorporation was inhibited to the same extent independent of when the antibiotic was added to the culture. These results were repeated with nglO (gene 45), gleZZ (gene 42) and the double mutant gmBL292,N134 (genes 55 and 33). In all cases, when late gene expression was prevented, there was no rifampicin refractory activity seen. 37 Figure 4. The absence of rifampicin refractory RNA synthesis in the absence of late gene expression. The experiments were performed as described in the legend of Figure l. The times indicate when rifampicin was added. A. T4+ infected nrss (§p_su-). B. T4 §m_N81 (gene 41) infected NF58. 38 50 ' ' ' ' ' I I T I 40- lOmin. 30' l5min. 20b 5“ ' g 2 4 . 25 IE '()_ nnut O (93-0 V 2:78 0 1 1 1 J J 1 1 l l E 8 V V I 1 T I I 1 .5}: B :2 5 c: '33 40- 0 ”It 30" lOmin. 20L IO” 0 l l I l l l l o 5 |0l5 202 3035404550 Time after infection (min) Figure 4 39 Hybridization competition of T4 RNA made on K803 in the presence of rifampicin: An analysis of the rifampicin refractory tran- scripts was done using hybridization competition techniques. The RNA was labeled in 2 minute pulses at 5, 10 and 15 minutes after the addition of the antibiotic. The specific activities of the RNA preparations from the treated cultures was 44.5, 26.1, and 27.8% of the equivalent untreated 10-12, 15-17 and 20-22 minute RNAs, respectively. Each RNA sample was competed with 5 and 20 minute RNA from T4+ infected K803. The results are shown in Figure 5. In all cases, 20 minute RNA competed the RNA made in the presence or absence of the antibiotic equally well. On the other hand, 5 minute RNA competed the transcripts made in the presence of rifampicin less well than those made in the absence of the antibiotic. Thus, at least some RNA species may be differen- tially inhibited by rifampicin. An effect of DNA glucogylation on rifampicin refractory synthesis: It has been shown that the glucosylation of T4 DNA may play a role in late gene expression (Montgomery and Snyder, 1973). In Table l are results from several experiments showing the effect of rifampicin on the production of phage by T4 with different glucosylation patterns. All experiments were done on K803 (rgl-) to avoid restriction of unglucosylated phage. The Bgt- phage do not have B-glucosyl transferase and as a consequence the DNA is only a- glucosylated (approximately 80% of the hydroxymethylcytosine residues) (Revel and Luria, 1970). Although phage production may be more sensitive at early times (2 minutes) it appears equally as 40 Figure 5. Hybridization competition of T4 3H-RNA made in the presence of rifampicin. Escherichia coli K803 was grown in M93 and infected with T4+ at an MOI of 5. RNA was labeled with 3H-uridine (S uCi and 5 ug/ml) at 10 to 12 min (A), 15 to 17 min (B), and 20 to 22 min (C) after infection in the presence (0 ,I) or absence (0,0) of 100 ug/ml of rifampicin added at 5 min after infection. Competition was done in the presence of 10 ug/ml T4 DNA, with 5 (O, O) or 20 (C), I) minute unlabeled RNA from T4+ infected K803. Hybridization efficiencies were 28 :_4%. One hundred percent hybridization equaled: A, 9651 and 4290; B, 5589 and 2028; and C, 3982 and 2654 CPM for the untreated and treated samples, respectively. 41 m madman 2.: ~ 0:: 42m. 00:000.:3 63 ”6 OJ ”.0 a; 03 “.0 DJ “.0 linfl . A... Hu/o uHo“. I l I l I J luooaod ) i. unvnut + 3|an" 8 42 Table 1. Effect of rifampicin on the phage production by T4 glucosyltransferaseless mutants Time of Rifampicin Phage/Cell Addition (min) T4+ Bgt‘ agt' aBgt' --- 130 110 101 33.5 2 12.5 0.2 --— _-- 5 27 15 3.8 0.8 10 120 100 25 3.4 15 --- -—— 30 8.5 refractory at later times. The agt- (a-glucosyltransferaseless) phage have all their hydroxymethylcytosine residues fully B- glucosylated. The aBgt- mutants are fully unglucosylated (Revel and Luria, 1970). Both the agt- and aBgt- phage are sensitive to rifampicin even at 15 minutes after infection. The interpretation of these results is complicated by several facts regarding the synthesis of unglucosylated or partially glu- cosylated phage. For example, even on the nonrestricting hosts, phage production is depressed to varying degrees but is most severe in aBgt- pHage infection (Georgopoulos and Revel, 1971). In addi- tion, DNA synthesis is affected in agt- phage infection. In experi- ments not shown here, rifampicin inhibited the rate of 3H-thymidine incorporation when added after DNA synthesis had begun. This would explain the effect shown on phage production. For the aBgt- mutant, however, DNA synthesis is the same as wild type in the presence or absence of rifampicin added at 5 minutes after infection (Figure 6). 43 Figure 6. The rate of 3H-thymidine incorporation in T4+ and T4 HA57 (oBgt‘) infected E. coli K803 with and without rifampicin. Samples (10 m1) of the bacteria grown in M98 were infected with the appropriate phage at an MOI of 5. Rifampicin was added at 5 minutes after infection (arrow). ((3) T4+; (e) T4+ + rifampicin; (D) T4 HA57; (I) T4 HAS? + rifampicin. 44 z: 0| <- V E“ o T '0: l Ha-Thymidine. Incorporation (CPM x l0 2’/ 2 min.) 6 0.5 » a A HAsmnifl T; a Bit . r; 0 5 I0 I5 20 25 .Time (min) middle 2' pulse Figure 6 30 45 Furthermore, uridine incorporation appears more sensitive to rifampicin at 10 or 15 minutes after infection than at 5 minutes after infection (data not shown). Therefore, despite the low phage yields, we suggest that the rifampicin refractory activity is affected by the state of the DNA template. DISCUSSION We have described a rifampicin refractory RNA synthesis in T4 infected E. coli K12 strains. It occurs at an antibiotic concen- tration of 100 ug/ml, an amount capable of inhibiting RNA synthesis in uninfected bacteria by greater than 97% in 10 minutes. The refractory synthesis is dependent upon the time of addition of the antibiotic. When added before 5 minutes at 37 C to T4 infected E. coli K803, inhibition of 3H-uridine incorporation is greater than 90%. After 5 minutes, inhibition is only reduced 70 to 80% and this refractory synthesis is maintained until the cells lyse. 0n SDS gels, it appears that all the resolvable late gene products are made, while hybridization competition data suggest there is a differential effect on at least some early transcripts. There are at least two possibilities that come to mind to explain the apparent decrease in sensitivity to rifampicin at late times during infection: a) a T4 induced change in permeability of the cell and b) a change(s) in the RNA polymerase itself. We sug- gest that our data would argue against a change in the permeability of the cell to rifampicin late in infection. The kinetics of inhibi- tion, for example, are not what would be expected from a decreased (peremeability. The initial rate of inhibition immediately after 46 rifampicin was added was the same whether it was added at 2, 5 or 10 minutes after infection. In addition, the rate of uridine incorporation plateaus. That is, for an interval of 10 to 15 minutes, the rate of RNA synthesis is maintained at a level of about 25% of that of the untreated control (see Figure 1). A second observation that cannot be explained by a permeability change is the differential effect on certain RNAs shown in the hybridization-competition data. The antibiotic must be entering the cell and reacting with the polymerase. We might postulate, therefore, that there is a change in the K12 polymerase that makes it less sensitive to rifampicin at late times in the T4 infection. This could be a phage induced change in the enzyme or a change in the state of the enzyme. Many phage induced changes in the host RNA polymerase have been demonstrated. These include the association of T4 coded polypeptides with the enzyme (Stevens, 1972; Horvitz, 1973; Ratner, 1974a,b). These polypeptides include the gene 33, 55 and 45 products, all of which are known to be required for late gene expression (Epstein et al., 1963; Bolle et al., 1968b; Notani, 1973). Since rifampicin inhibits by binding to the 8 subunit of the enzyme (Hell and Zillig, 1970), such changes in the enzyme might make it less sensitive to the antibiotic. A change in the state of the enzyme might be in the formation of (initiation) complexes no longer accessible to the antibiotic. This is suggested by the fact that the refractory synthesis is not seen when late gene expression is blocked or when unglucosylated DNA is used as a template. Furthermore, we do not have any evidence that new initiations are occurring in the presence 47 of rifampicin; that is, transcription may be from pre-initiated polymerase. The rifampicin refractory state could also be a combination of the events suggested above. This will have to await further analysis. The findings of Riva et al. (1972) and Romero et a1. (1973) may have some bearing on the effect noted here. They were able to demonstrate a differential sensitivity of transcription of cer- tain classes of genes to rifampicin. At sublethal concentrations which had no effect on bacterial syntheses or extrachromosomal DNA replication, the transcription of the latter was prevented. It was also possible to select rifampicin resistant mutants that selectively did not transcribe extrachromosomal genes, yet replica- tion continued and the plasmids or episomes were maintained. A similar effect was noted by Snyder (1972) for a rifampicin resistant mutant of E. coli (RifR-Z) which would not grow T4. The major block to phage growth was the inability to transcribe late genes. Interestingly, when the RifR-Z mutation was transduced to E. coli B, the inhibition of T4 growth was lost (Snyder, personal obser- vation). We are not sure if this observation and the data pre- sented here are related except that they both reflect differences between K12 and B strains. In conclusion, we have described a partial inhibition of transcription by rifampicin in T4 infected E. coli K12 strains. The refractory synthesis is dependent upon the time of addition and the concentration of the antibiotic used. We suggest that this potentially may provide a useful means for the examination of late gene expression, both at the molecular and genetic level. In 48 addition, these results portend possible misinterpretations where rifampicin is used exclusively to define temporal transcription sequences. REFERENCES Bolle, A., Epstein, R. H., Salser, W., and Geiduschek, E. P. 1968a. Transcription during bacteriophage T4 development: Synthesis and relative stability of early and late RNA. J. Mol. Biol. .22: 325-348. Bolle, A., Epstein, R. H., Salser, W., and Geiduschek, E. P. 1968b. Transcription during bacteriophage T4 development: Requirements for late messenger synthesis. J. Mol. Biol. 66: 339-362. de Mauro, E., Snyder, L., Marino, P., Lamberti, A., Copp, A., and Tocchini-Valentini, G. P. 1969. Rifampicin sensitivity of the components of DNA-dependent RNA polymerase. Nature 222: 533-537. Epstein, R. H., Bolle, A., Steinberg, C. M., Kellenberger, E., Boy de la Tour, E., Chevally, R., Edgar, R. S., Susman, M., Denhardt, G. H., and Lielausis, A. 1963. Physiological studies of conditional lethal mutants of bacteriophage T4D. Symp. Quant. Biol. 22: 375-396. Georgopoulos, C. P., and Revel, H. R. 1971. Studies with glucosyl transferase mutants of T-even bacteriophages. Virology 44: Haselkorn, R., Vogel, M., and Brown, R. D. 1969. Conservation of the rifamycin sensitivity during T4 development. Nature 221: Heil, A., and Zillig, W. 1970. Reconstitution of bacterial DNA dependent RNA polymerase from isolated subunits as a tool for the elucidation of the role of the subunits in transcription. FEBS Lett. 42: 165-168. Horvitz, R. H. 1973. Polypeptide bound to the host RNA polymerase is specified by T4 control gene 33. Nat. New Biol. 244: 137- 140. Mandell, J. D., and Hershey, A. D. 1960. A fractionating column for analysis of nucleic acids. Anal. Biochem. 2: 66-77. Mizuno, S., and Nitta, K. 1969. Effect of streptovaricin on RNA synthesis in phage T4 infected Escherichia coli. Biochem. Biphys. Res. Commun. 66: 127-130. 49 Montgomery, D. L., and Snyder, L. R. 1973. A negative effect of B-glucosylation on T4 growth in certain RNA polymerase mutants of Escherichia coli: Genetic evidence implicating pyrimidine rich sequences of DNA in transcription. Virology 66: 349-358. Notani, G. W. 1973. Regulation of bacteriophage T4 gene expression. J. Mol. Biol. 16: 231—249. O'Farrell, P. 2., and Gold, L. M. l973a. Bacteriophage T4 gene expression: Evidence for two classes of prereplicative cistrons. J. Biol. Chem. 248: 5502-5511. Ratner, D. 1974. The interaction of bacterial and phage proteins immobilized Escherichia coli RNA polymerase. J. Mol. Biol. Ratner, D. 1974. Bacteriophage T4 transcriptional control gene 55 codes for a protein bound to Escherichia coli RNA polymerase. Revel, H. R., and Luria, S. E. 1970. DNA glucosylation in T-even phage: Genetic determination and role in phage host inter- action. Ann. Rev. Genetics 4: 177-192. Riva, S., Fietta, A. M., Silvestri, L. G., and Romero, E. 1972. Effect of rifampicin on expression of some episomal genes in E. coli. Nat. New Biol. 235: 78-80. Romero, E., Riva, S., Berti, M., Fietta, A., and Silvestri, L. 1973. Pleiotropic effects of a rifampicin resistant mutation in E. coli. Nat. New Biol. 246: 225-228. Rosenthal, D., and Reid, P. 1973. Rifampicin resistant DNA synthesis in phage T4 infected Escherichia coli. Biochem. Biophys. Res. Snyder, L. R., and Montgomery, D. L. 1974. Inhibition of T4 growth by an RNA polymerase mutation of Escherichia coli: Physio- logical and genetic analysis of the effects during phage development. Virology 62: 184-196. Snyder, L. 1972. An RNA polymerase mutant of Escherichia coli defective in the T4 viral transcription program. Virology 50: 396-403. '— Stevens, A. 1972. New small polypeptides associated with DNA dependent RNA polymerase of Escherichia coli after infection with bacteriophage T4. Proc. Nat. Acad. Sci., USA 62: 603-607. Studier, F. W. 1973. Analysis of bacteriophage T7 early RNAs and proteins on slab gels. J. Mol. Biol. 22: 237-248. ARTICLE II THE REGULATION OF ANTI-LATE RNA SYNTHESIS IN BACTERIOPHAGE T4: A DELAYED EARLY CONTROL ROBERT J. FREDERICK AND LOREN R. SNYDER Manuscript to be submitted for publication 50 51 SUMMARY T4 bacteriophage make mutually complementary RNA transcripts. The anti-sense RNA is made in the late region but at early times after infection. Hence it is called anti-late RNA. We have studied some of the physical characteristics and possible regulatory mechan- isms involved in the synthesis of these unique early RNA species. Anti-late RNAs sediment on 5 to 20% sucrose gradients with an average sedimentation coefficient of 20-22 8. They are comparable in size to late mRNA. We provide evidence that anti-late transcripts are initiated in the late region and they are made in vivo on degraded DNA templates. Their synthesis is refractory to rifampicin at one minute after infection. Anti-late RNA production is examined under several different conditions of altered early gene expression. In all circumstances where the delayed early gene transcription is altered, anti-late synthesis is also altered. From the data presented here, we postulate that anti-late RNA synthesis is controlled by the same mechanism regulating delayed early genes. The data are discussed with respect to the current understanding of T4 gene expression. INTRODUCTION Until recently, the transcription of double stranded DNA was thought to be strictly an asymmetric process. That is, for any particular region of the molecule, RNA is made in only one direction from one of the strands (see, for example, Geiduschek et al., 1968, and Davidson, 1972). There are, however, increasing numbers of 52 reports of symmetrical transcription in many different biological systems. These include the animal viruses SV40 (Aloni, 1972, 1973), Vaccinia (Colby and Duesberg, 1969), Adenovirus 2 (Zimmer and Raskas, 1976), herepes simplex (Zeev and Beker, 1975), and polyoma (Aloni and Locker, 1973); HeLa cell mitochondria (Aloni and Attardi, 1971, 1972; Young and Attardi, 1975; Murphy et al., 1975); and the bacterial viruses 1 (B¢vre and Szybalski, 1969), gh-l (Jolly, PhD Dissertation, Michigan State University, 1976) and T4 (Jurale et al., 1970; Geiduschek and Grau, 1970; Notani, 1973). Symmetrical RNA synthesis by T4 was evident from the isolation of virus specific double stranded RNA from infected bacteria (Jurale et al., 1970) and the self-annealing properties of RNAs from infected cells (Jurale et al., 1970; Geiduschek and Grau, 1970). In their characterization, Geiduschek and Grau (1970) determined that the symmetric RNA was made early during infection and was com- plementary to late mRNA. Notani (1973) later demonstrated that this anti-late RNA was made for a short interval, approximately 2 to 6 minutes after infection at 30 C, and was synthesized on 81% of the late region from the early (4) strand. The origin of anti- late RNA, at first attributed to overlapping transcription regions of opposite polarities (Jurale et al., 1970) or mistakes in termina- tion recognition which allowed read through into the late regions from early promoters (Geiduschek and Grau, 1970), was later inter- preted by Notani (1973) to be from mistakes in initiations. The reasoning was that, due to the clustering of late genes, it would take substantially longer than the four or five minutes of synthesis for the RNA polymerase to traverse the extensive late region. 53 Several observations, however, suggest that anti-late RNA may be made in response to or as a consequence of some other mechanism than simply initiation errors. This RNA is not made when chloram- phenicol is added at the time of infection. Anti-late RNA is made temporally, starting at 2 minutes and ending at 6 minutes after infection. It is specific for the late regions; i.e., no comple- mentary early RNA is synthesized, at least at early times (Geiduschek and Grau, 1970; Notani, 1973). Consequently, we decided to investigate the synthesis of anti-late RNA to see if it is regulated, if so what controls are involved, and what relation- ship, if any, it may have with other classes of early RNAs. In discussing the various pre-replicative mRNA species, we shall use the terms and abbreviations as defined by O'Farrell and Gold (1973a). Immediate early (IE) are made as promoter proximal cistrons from pre-early promoters (PE). Delayed early (DE) are made as promoter distal cistrons from pre-early promoters and/or from other promoters recognized later in the course of infection (P ). Quasi-late are made from P Q be made late in infection. Q type promoters and continue to In this paper we present evidence substantiating the conclusions of Notani (1973) that T4 anti-late RNA is made by initiations on the DNA lystrand (early strand) in the late region. In addition, we show that anti-late transcription is subject to a delayed early type of regulation. The data are discussed in terms of prereplicative transcription and its regulation. 54 METHODS AND MATERIALS (a) Phage and Bacteria. The bacteria and phage used are described in Table l. (b) E2922: M93 and tryptone media were the same as used previously (Snyder and Montgomery, 1974). M9AA is M9 (Bolle et al., 1968) substituted with 20 ug/ml of the following amino acids: alanine, aspartate, cysteine, glutamate, glycine, histidine, iso- leucine, lysine, ornithine, phenylalanine, proline, serine, threonine, tryptophan, tyrosine, valine, arginine, and methionine. (c) Chemicals. 3H-S-Uridine and 32P04 were purchased from Amersham/Searle Co., and 3H-thymidine from New England Nuclear Co. 3H-Uridine triphosphate (3H-UTP), 14C-leucine, and rifampicin came from Schwartz-Mann. Sucrose (grade 1), ribonuclease A (bovine pancreatic, SX recrystallized), deoxyribonuclease (DN-EP), egg white lysozyme, and calf thymus DNA were purchased from Sigma Chemical Co. Ribonucleotide triphosphates were from P and L Biochemicals. Ribo- nuclease T1 (B grade) was purchased from Calbiochem. (d) Preparation of RNA Polymerase. RNA polymerase was isolated and purified through the DEAE cellulose step and assayed according to Burgess (1969). (e) Nucleic Acid Extractions. RNA was extracted using the hot phenol method of Bolle et al. (1968). Each preparation was addi- tionally treated with DNase (20 ug/ml, 37 C, 20 minutes) and re-extracted twice with hot phenol. Concentrations were determined spectrophotometrically (Bolle et al., 1968). 55 Table l. Bacteria and bacteriophage used Relevant Strain Characteristics Source or Reference Bacteria: E. coli B BE 22.5“- L. Gold K12 HR112 SE su-, rgl- H. Revel + - K803 22.2“ II'EKEK' (Wood, 1966) rgl K803 K803,rifr Pl transductant of RifR-Z K803 (Montgomery and Snyder, 1973) DG75 §m_su- (Worcel and Burgi, 1972) DG75 §m_su-, rifr Pl transductant of 1:1sz DG75 Bacteriophage: T 4 SP62 reg A- J. Wiberg am N122 gene 42 (Epstein et al., 1963) SP62, reg A-, gene (constructed in EE.N122 42 our laboratory) 2E E10 gene 45 (Epstein et al., 1963) Ac19rev2 cold sensitive M. Nelson and L. on BE Gold tsGl mot- (Mattson, 1974) gp_E10, gene 45, mot- (constructed in tsGl our laboratory) HA57 cBgt- J. Wiberg gm_E10, gene 45, aBgt- (constructed in HA5? our laboratory) 56 DNA was extracted from CsCl-purified virus according to Mandell and Hershey (1960). Sheared DNA was prepared as described by Brody et a1. (1970). The size of the sheared DNA was determined by ultracentrifugation on a S to 20% sucrose gradient in 0.1 5 NaCl, 0.05 §_EDTA. The gradients were centrifuged at 35,000 rpm for 2 hours at 20 C in a Beckman SW 50.1 rotor. The sheared DNA sedimented with a broad peak at approximately 16 S as determined from the location of H-labeled E. coli ribosomal RNA on the same 3H gradient. The molecular weight was calculated from the formula of Eigner and Doty (1965) as discussed by Kutter and Wiberg (1969) using this S value or the distance intact T4 DNA sedimented on a companion gradient and assuming a molecular weight of 130 x 106 daltons. These gave values of 3.16 and 3.04 x 106 daltons, respectively. (f) Radioisotopic Labeling. RNA was labeled using appropriate pulses of 3H-uridine as indicated in Results. The bacteria were growtn to an absorbance of 0.4 at 625 nm measured on a Spectronic 20 spectrometer and infected at a multiplicity of infection (MOI) of 5 unless otherwise indicated. 3H-S-Uridine was added to give a final concentration of SO uCi and l ug/ml. The incorporation was stopped by pouring the cell culture over an equal volume of ice and the RNA extracted as indicated above. T7 viruses were labeled by adding 3H-thymidine (10 uCi and 10 ug/ml) to log phase E. coli at a concentration of 4 x 108/ml. Fifteen minutes later, T7 phage were added at an MOI of 0.2. The culture lysed spontaneously in approximately 45 minutes. The 57 labeled viruses were purified on a CsCl step gradient. The final suspension contained 6 x 104 CPM/ml. T4 §m_E10, HA57 was labeled with 32F according to the procedure of Hattman (1964) except the bacteria were grown in tryptone broth. The CsCl purified viruses had a specific activity of 1.7 x 10-5 CPM/plaque forming unit. Proteins for SDS polyacrylamide electrophoresis were labeled as described in the appropriate legends. The bacteria for each experiment were grown overnight in M9AA and transferred to fresh medium at a 50-fold dilution. Label incorporation was stopped by adding an equal volume of ice to the labeling tubes or pouring the contents of each tube over ice with enough L-leucine to give 100 ug/ml. The samples were centrifuged at 3-5000 x g for 10 minutes. The pellets were resuspended in 0.05 ml sterile water and mixed with an equal volume of 2X concentrated lysing buffer (Studier, 1973). All samples were boiled for 2 minutes before running on 12.5% SDS gels. (9) SDS Slab Gel Electrophoresis. The procedure was the same as described by Studier (1973). The gels were run at 10 MA per gel for 2 to 3 hours until the marker dye approached the end of the gel. Kodak No Screen X—ray film was exposed to the dried gels for 3 days to 3 weeks. The identification of T4 protein bands in the gels was done using appropriate amber mutants and from published data (O'Farrell and Gold, l973a,b). 58 (h) Measurement of Unglucosylated T4 DNA Degradation in a Restrictive Host. A modification of the procedure described by Kutter and Wiberg (1968) was used. Fifteen milliliters of bacteria grown in M98 to 4 x 108/ml were infected with 32P labeled T4 am_ 210, HA5? (genes 45, asgt‘) at an MOI of 0.5. At 1, 2, and 4 minutes after infection, 5 m1 aliquots were taken, poured over an equal volume of ice, and centrifuged at 8000 rpm for 10 minutes in a Sorvall refrigerated centrifuge. The pellet was washed once in 5 m1 of cold M9 buffer to remove unadsorbed phage and centrifuged as above. The resulting pellet was resuspended in 0.2 m1 of 0.01 M_ Tris HCl, pH 7.9, 0.01 M EDTA and 300 ug/ml egg white lysozyme. Labeled T7 phage (0.02 ml) were added and the tubes heated to 65 C for 3 minutes. SDS was added to 0.1% and the incubation continued for 10 minutes. A fraction (0.1 ml) of the lysed samples was put into a scintillation vial, diluted in 0.9 ml water and counted in 5 ml of Aquafluor Scintillation cocktail (New England Nuclear). An equal volume was layered on a 2.8 m1, 5 to 20% sucrose gradient (0.01 M_Tris, pH 7.9 and 0.5 M_NaC1) over a 0.2 ml shelf of 60% sucrose. The gradients were centrifuged at 38,000 rpm in a Beckman SW 50.1 rotor for 3.5 hours at 20 C. Fractions were collected from the bottom of each tube directly into scintillation vials and treated as above. Samples were counted on a Packard ambient tempera- ture scintillation counter. (i) Anti-Late RNA Assay. The procedure was the same as pre- viously described (Geiduschek and Grau, 1970). Several different 17 to 20 minute pulse labeled RNA samples were used during the course 59 of this investigation. The amount of labeled RNA protected from RNAses varied from 8 to 32% depending upon the preparation used. In all cases, the results were consistent and reproducible with all of the assays reported being done at least twice. Pancreatic and T1 RNase were used in concentrations of 10 and l ug/ml, respectively. RESULTS (a) Size of Anti-Late RNA in vivo. An estimation of the in vivo size of the anti-late transcripts was made with the thought that the size distribution would give some indication about possible regula- tory mechanisms and/or functions of the anti-late RNA. For example, relatively small fragments or a very heterogeneous distribution would be consistent with the suggestion of initiation errors in the late region (Notani, 1973). The size distribution of anti-late RNA at various times after infection is shown in Figure l. A zero time control is included to demonstrate the background of the assay since uninfected E. coli does not make symmetric RNA (Geiduschek and Grau, 1970; Notani, 1973). Portions of each sample were annealed with labeled 2-4 minute RNA before the gradients were run. Since only background levels of protection were obtained, we concluded there was no T4 DNA in the samples and therefore only complementary RNA is being measured. On the 5 to 20% sucrose gradients, anti—late RNA sediments as a heterodisperse population in a broad peak. In the 5 minute sample, the bulk of the anti—late RNA (more than 68% of the assayed activity) sediments with a sedimentation coefficient greater than 60 Figure 1. Sucrose density gradients of anti-late RNA. Escherichia coli BE was grown to 4 x 108/ml in M93 at 30 C. The bacteria were infected with T4+ at an MOI of S. Aliquots (125 ml) were taken at 0, 2, 5, and 8 minutes after infection. The RNA was extracted as described in Materials and Methods. The gradients were prepared and centrifuged as described by Ricard and Salser (1975). Each RNA sample was disaggregated before layering onto a 37.5 ml 5 to 20% sucrose gradient. The samples were centrifuged for 17 hours at 27,000 rpm and 4 C. The gradients had 0.62, 0.68, 1.10 and 1.50 mg of the 0, 2, 5 and 8 minute RNAs, respectively. After centrifugation, 1.2 ml samples were collected from the top using an ISCO Fraction Collector and uv monitor. The fractions were precipitated and the RNAs pelleted as previously described (Ricard and Salser, 1975). The drained pellets were dissolved in 0.1 ml of sterile water. Then 0.3 m1 of a solution containing 1 ug of 3H (17 to 20 minute) RNA and sufficient 10X SSC to give a final concentra- tion of 2X was added to each tube. They were then incubated at 70 C for 3 hours and treated with T1 and pancreatic RNases as described by Geiduschek and Grau (1970). Control tubes without unlabeled RNA and :_RNases had 231 (average of 3 tubes) and 6399 (average of 3 tubes) TCA precipitable CPM. The results are reported as CPM/tube after background subtraction (-o-) and relative absorbance units at 254 nm (---). 61 €25 9:29; 855.32 a. a 4 a O O a 0. d d I 1 ow u 5 “ 2 . .HHHHHHnIu nu "" 2 “-¥u I '4' ........ . m . . . a m m . I 2 5 m m 3 5 ‘1' 2 u ........ r 2 ' “H'IUU'I o . III-I.|“\\ M v “Niall-III 5 O I!" . | . a I . o .m _\\\\ .m . .V |. t o “ v Q fill! 5 AO‘\ . 5 w b b h ilr b "I L b b P ’I. | . nw aw nu «w nu nu so so nu so no 9 7 6 4 3 I on 7. 4. am 1.. Nb. x28 Enema «82¢ 42m 23-5.4 Fraction No. Figure l 62 16 S and an average value of approximately 20 S. In two separate determinations using independently extracted 5 minute RNA, the same results were found. At 8 minutes, more than 74% of the assayed activity sediments faster than the 16 S ribosomal fraction and has an average S value of 22 S. We did not detect any extremely long transcripts which could be indicative of "read through" on the lfstrand. In addition, there was very little anti-late RNA at the top of the gradient, although it is possible that the sensitivity of this assay may be limiting with very small RNA pieces in low concentrations. In general, the size distribution is similar to that observed for late mRNA (Richard and Salser, 1975). (b) §ynthesis of Anti—Late RNA on Less than Unit Length DNA. T4 pre-early transcription can be separated from delayed early transcription in vitro on sheared DNA templates (Milanesi et al., 1969, 1970). Considering the tandem nature of delayed early and immediate early cistrons, it would certainly seem reasonable that if anti-late RNA were made by a "run through" from early regions, its synthesis would be limited on a sheared template. In view of this, we developed an in vivo system containing degraded T4 DNA and examined the production of anti-late RNA. The procedure takes advantage of the E. coli restriction system for unglucosylated phage DNA. T4 phage normally have fully glucosyl- ated hydroxymethylcytosine residues; 70% with an alpha and 30% with a beta glucosidic bond (Lehmann and Pratt, 1965). Mutants lacking one or both of the respective glucosyltransferase enzymes can be isolated (Georgopoulos and Revel, 1971). If both enzymes are 63 inactive, such as in T4 HA57, the phage DNA will be unglucosylated and subject to restriction in rgl? bacteria. On E. coli DG75 (gut, Eglf), the unglucosylated DNA is degraded to pieces smaller than 2 x 106 daltons within 2 minutes after infection (Figure 2). The molecular weight was determined by the formula of Eigner and Doty (1965) using 25 x 106 daltons as the molecular weight of T7 DNA. At 4 minutes after infection, nearly 80% of the injected DNA was recovered at the top of the gradient. This is a more extensive and dramatic degradation than noted previously for unglucosylated DNA (Hattmann, 1969; Georgopoulos and Revel, 1971; Dharmalingam and Goldberg, 1976a,b). In these studies, however, degradation was measured as the loss of TCA insoluble radioactivity and there was no correction for unadsorbed viruses. In addition, the phage were grown on £:m:_bacteria. For our study, the phage were grown on K803 Egg; and therefore would also be subject to this restriction on DG75 £;m;. Early proteins made under conditions where the T4 DNA is degraded are shown in Figure 3. On DG75, only one band (P42) is seen in any significant amounts in a 2 to 5 minute pulse. At later times, other T4 polypeptides are made but in very much reduced amounts. RNA extracted from DG75 (gut, rgl?) and HR112 (s27, rgl?) 5 minutes after infection protected equivalent amounts of 17 to 20 minute 3H-labeled T4 RNA (Figure 4). Pulse labeling experiments indicate that 0.4 to 1.7% of the RNA made early is complementary to late RNA. This is approximately 50 to 100% of control values for the unglucosylated phage in the non-restrictive host (data not 64 Figure 2. Degradation of unglucosylated T4 DNA in a restricting host, E. coli DG75. 32P-labeled DNA was monitored on 5 to 20% sucrose radients as described in Materials and Methods. The total 2P CPM/gradient were from 101 to 584 (after background subtraction), representing recoveries of 64 to 100%. The percentages indicated are the percent of the total CPM in the top 6 or 7 fractions. Sedimentation was from right to left. 3H-T7 DNA was run on each gradient as a marker. 65 HR IIZ DG 75 «Iv 385an 2a .28 .\. 0000 0000 0 00 $4420864ZOBW420 5 2 O 2 w m 3"“! on”- 5 .MWMWh-n —mr 0.110 I I 2_ an . an 5 .m 6 2 8 - 9 3 5 O 2 w m 00000000,. . I 432... 432m0w A Ollllo V .IO-XZQUInI A I c 18.85an «mg .28 .\. F raction No. Figure 2 66 Figure 3. T4 early gene expression on degraded DNA in vivo. Escherichia coli HR112 and DG75 were grown in M9AA at 30 C. The bacteria (4 x 108/m1) were infected with T4 am_E10, HA57 at an MOI of 10. T4 proteins were labeled as described in Materials and Methods at the times indicated. HR112 (a,c,e); DG75 (b,d,f). ‘ h n ; a b 2311)) a, it WM .5 cd 67 Figure 3 IO‘I3 mln. H “ h — M N . A'" w ‘n—v ‘III — ”W‘- ‘IIIII' "“" III-D ‘" II. VL 4-h- GI... III-I e f _p43 —erA —p 32 —p42 _.pIPI|1 68 Figure 4. Anti-late RNA made on degraded T4 DNA in vivo. Escherichia coli HR112 and DG75 were grown in 100 m1 of M95 at 30 C to 4 x 108/m1. Each culture was infected with T4 am_ElO, HA5? at an MOI of 5. At 5 minutes after infection, they were poured over an equal volume of ice and the RNA was extracted. The assay was done with 0.5 ug of 3H (17 to 20 minute) RNA per tube. Total CPM added = 4039; background (without unlabeled RNA) was 137 CPM. (O) HR112 (333 su", rgl"); (O) DG75 (§m_su‘, rgl+). 69 m9 b - p b _ — - 87.65432 8.8280 42m 9.04 .6 o\o 500 200 300 400 #9 RNA / ml IOO Figure 4 70 included). Since unglucosylated DNA is degraded to small fragments less than 2.5% of the size of the intact genome by the time anti- late synthesis begins (2 minutes at 30 C), it seems unlikely that it is made by a "run through" mechanism from early promoters. We also attempted to synthesize anti-late RNA on sheared DNA templates in vitro. The results from an experiment using intact and sheared DNA for the in vitro synthesis of anti-late RNA are shown in Table 2. Escherichia coli B RNA polymerase holoenzyme was used at DNA excess to decrease the number of nonspecific initiations Table 2. In vitro synthesis of anti-late on intact and sheared T4 DNA 3H-Complementary RNA made in vitroa In vivo RNA Intact DNA Sheared DNA (250 ug/ml) RNase CPM % CPM % l --— --- 3018 ——- 983 --- 2 --- + 178 —-- 140 --- 3 5 min + 273 3.3b 197 6.8 4 20 min + 364 6.5 191 6.0 aIn vitro RNA was synthesized in the presence of 3H-UTP as described by Milanesi et a1. (1970) and extracted as described by Brody et a1. (1970), including the DNase step. The assay procedure was that of Geiduschek and Grau (1970). bPercentage of added label (1) after background (2) subtraction. 71 (Milanesi et al., 1970). As previosuly demonstrated (Geiduschek and Grau, 1970), anti-late RNA is made in vitro on the intact DNA template as noted by the greater protection with 20 minute in Vivo RNA. Differences between the published amounts synthesized and those reported here are probably due in part to the fact that rifampicin was not used here and therefore reinitiations could occur. Since this should result in more copies of early mRNA, the concentration of anti-late RNA may be diluted, especially if the majority of the anti-late transcripts in vitro are made by a "read through" as suggested by the data of Geiduschek and Grau (1970). If this is true, then we would not expect as much anti-late RNA synthesis on the sheared DNA. Annealing with 5 minute RNA is a measure of the symmetry of transcription (Brody et al., 1970). As expected, the symmetry increases on the sheared template (3.3 vs. 6.8%). Since, however, there is no significant difference in the amount of labeled RNA protected by the 5 or 20 minute in vivo RNA, we conclude that anti- late RNA either is not made on the sheared template or is made in such low amounts that it is not detectable in our assay. Although the background levels are high, this experiment was repeated with different polymerase and DNA preparations with similar results. Despite reports demonstrating the fidelity of in vitro T4 transcription (Brody et al., 1970; Brody and Geiduschek, 1970; Milanesi et al., 1969, 1970; O'Farrell and Gold, 1973b). it may be very different than in vivo transcription. For example, Cohen et al. (1974) reported substantial differences in the time of synthesis of deoxynucleotide kinase. Synthesized immediately in vitro, the 72 message is not synthesized until 2 minutes after infection in vivo. If our hypothesis is correct, anti-late RNA synthesis may also reflect differences in in vivo and in vitro transcription reactions, since in vitro it is probably made to some extent by a "runthrough" mechanism. (C) Anti-Late RNA is Made in the Presence of Rifampicin. When rifampicin is added to the infected bacteria 1 minute after infec- tion, the synthesis of delayed early gene products occurs but the gene products are made in lower amounts and some, those made exclu- sively from P promoters, are not made at all (O'Farrell and Gold, Q 1973a). Anti-late RNA is made under these conditions but in reduced amounts (Figure 5). We interpret the lower slope of the treated sample to mean that less of particular areas (analogous to promoter proximal regions) are being made per unit time in the treated culture. This is a reasonable assumption knowing that rifampicin blocks the initiation but not the elongation of transcription (wehrli and Staehelin, 1971). It also implies that initiations can occur in the late region on the lfstrand. The fact that a satura- tion concentration of RNA has not been reached at 500 ug/ml indi- cates that there is a considerable portion of the late region being transcribed (Notani, 1973). (d) Effect of a Host RNA Polymerase Mutant on Anti-Late RNA Synthesis. RifR-2 is a mutation to rifampicin resistance which also inhibits the growth of T4 bacteriophage (Snyder, 1973). Although the major block to phage synthesis is on late transcription, many other defects occur during infection. These include delays in 73 Figure 5. The effect of rifampicin on anti-late RNA synthesis. Escherichia coli BE was grown in M95 at 30 C and infected with T4+ at an MOI of 5. At 1 minute after infection, one half of the culture (125 ml) was transferred to a flask containing sufficient rifampicin to give a final concentration of 100 ug/ml. At 5 minutes the cultures were poured over ice and the RNA extracted. The assay tubes contained 0.2 ug 3H (17 to 20 minute) RNA. Total CPM added = 2963; background (without added unlabeled RNA) was 46 CPM. ((3) RNA from untreated culture; (0) RNA from rifampicin treated culture. % of Late RNA Protected 25 20 IS IO 74 -Rif +Rif 0 ICC 200 300 400 500 #9 RNA/ml Figure 5 75 a) replication, b) shutoff of host macromolecular synthesis and c) early gene expression (Snyder, 1972; Snyder and Montgomery, 1974; Figure 6). The autoradiograms in Figure 6 are of labeled proteins from T4 am_N122_(gene 42) infected DG75 and DG75 RifR-Z. DNA negative (DO) phage were used because of the defect in replication on RifR-Z which may itself affect early gene expression. All phage specific proteins appear with a lag ranging from 1 to 2 minutes for some to 4 minutes for others. The gene 43 product may be a special case since there is a 7 to 8 minute delay before it is synthesized in RifR-Z. We have no explanation for this at this time; however, gene 43 has been shown to have autoregulatory control (Russel, 1973). The overall lag in early gene expression is probably not simply due to the lower rate of amino acid incor- poration since the delays are too long and different gene products appear to be affected to varying extents independent of their molecular weight. In any event, if the delay in appearance of the respective polypeptides is due to an altered early transcription pattern, this may also be reflected in the synthesis of anti-late RNA. Complementary RNA is made in RifR-Z 2 to 3 minutes after it is made in the parental bacteria (Figure 7). The percentages reported here are low since anti-late RNA makes up only 1 to 2% of the phage RNA (see also Notani, 1973). However, repeated assays (3) from separate experiments gave the same results. Additional controls were done with 500 ug/ml of 5 minute RNA. These were all equivalent to background controls without added unlabeled RNA. In the earliest 76 Figure 6. SDS polyacrylamide gel electrophoresis analysis of early gene products made on T4 am_N122 infected E. coli DG75 and DG75 RifR-z. The procedure is described in Materials and Methods. The proteins were labeled for 1 minute intervals with l4C-leucine. The rate of incorporation of label was the same throughout the interval examined for the respective cul- tures, although the rate in DG75 RifR-Z was approximately 76% of that in DG75 (data not shown). 77 gene products —p43 _pr ‘P 32 jams Lrh- 0-1 2-3 4-5 6-7 8-9 0111213 1445 1617 ‘8-1‘5 mm 1-2 3-4 5-6 7-8 9-10 11-12 G-14 6‘5 1748 —p43 —pr IIA —P46 —p32 —prllB \wx .. 7" —p45 ”-Q-o’ ' “- (mun-04 2-3 4-5 6-7 8-9 10-11 12-13 14-15 1517 18-19 fected 1-2 3-4 5-6 7-8 9-10 11-12 1344 15-16 17-18 Time After Infection (min) Figure 6 78 Figure 7. The time of anti-late RNA synthesis in T4 infected E. coli K803 and K803 RifR-Z. The bacteria were grown to 4 x 108/m1 at 27 C and infected at an MOI of 5. At the times indicated, 10 m1 samples were labeled with 10 uCi and l ug/ml of 3H-uridine and stopped by pouring over ice. The“ RNA was extracted as described in Materials and Methods. Annealing reactions were run with 20 minute RNA as described by Geiduschek and Grau (1970). Each 3H labeled RNA was assayed at a concentration of 5 ug/ml. Total CPM added and background (without unlabeled RNA) were: for K803 (CD). 2 to 4 min, 8641 and 31 CPM; 5 to 7 min, 39,121 and 74 CPM; 8 to 10 min, 39,296 and 216 CPM; 11 to 13 min, 40,304 and 140 CPM; for K803 RifR—Z (0), 2 to 4 min, 21,945 and 107 CPM; 5 to 7 min, 20,146 and 67 CPM; 8 to 10 min, 17,561 and 95 CPM; 11 to 13 min, 18,019 and 113 CPM. 79 h whamam _s\