Study of the transfer RNAs coded by T2, T4, and T6 bacteriophages.

T2, T4, and T6 bacteriophage tRNAs coding for arginine, leucine, proline, isoleucine, and glycine were isolated under conditions of short term and long term infection of Escherichia coli B cells. The corresponding phage tRNA species were examined for sequence homology by RNA-DNA hybridization analysis and by their relative behavior on reversed phase chromatography. The results indicate that all three T-even phages code for similar tRNA species; however, some tRNA species are homologous, others are not, and not all of the same tRNA species are coded by each bacteriophage. Reversed phase chromatography showed the presence of isoacceptor tRNAs for each phage aminoacyl-tRNA species. Pulse-chase experiments for [32P]tRNAGly suggest that the multiple isoacceptor species observed derive from the intracellular modification of a single tRNAGly gene product.

T2, T4, and T6 bacteriophage tRNAs coding for arginine, leucine, proline, isoleucine, and glycine were isolated under conditions of short term and long term infection of Escherichia coli B cells. The corresponding phage tRNA species were examined for sequence homology by RNA. DNA hybridization analysis and by their relative behavior on reversed phase chromatography.
The results indicate that all three T-even phages code for similar tRNA species; however, some tRNA species are homologous, others are not, and not all of the same tRNA species are coded by each bacteriophage. Reversed phase chromatography showed the presence of isoacceptor tRNAs for each phage aminoacyl-tRNA species. Pulse-chase experiments for [32PltRNAG*Y suggest that the multiple isoacceptor species observed derive from the intracellular modification of a single tRNAGLS gene product.
Based on criteria of nucleotide composition, DNA sequence homology, serological cross-reactivity, morphology, and genetic recombination and complementation, T2, T4, and T6 bacteriophages are considered to be closely related. The discovery of tRNAs coded by the T4 genome (1,2) made it likely that T2 and T6 phages code for similar tRNA molecules. Weiss et al. (1) had originally shown that %-labeled T4 tRNAs hybridized equally well to T2 and T4 DNAs. Waters and Novelli (3) and Tillack and Smith (4) independently suggested the presence of phage-specific tRNAs in T2-infected cells. Kim and Davidson (5) carried out a study on the sequence relationship of T2, T4, and T6 DNAs by using the heteroduplex method. They concluded that all of the T-even phages contained tRNA genes, some of which were homologous.
In the present study, we infected Escherichia coli B cells separately with the three different T-even phages and then attempted to examine the RNA extracts for phage-specific tRNA species to compare their sequence homology by RNA. DNA hybridization analysis and to study their relative behavior on reversed phase chromatography.
The results presented here indicate that all three T-even phages code for several different tRNA species; some are homologous, others are not, and not all of the same tRNA species are coded by the genome of each bacteriophage.
Isoacceptor tRNA species were observed by reversed phase chromatography for most of the phage tRNAs examined. In the case of tRNAG'J, 32Pi pulsechase experiments suggest that the multiple isoacceptor species derive from the intracellular modification of a single tRNAG'" gene product. A preliminary report on a part of this work has been presented (6). MATERIALS AND METHODS Growth and Purification ofPhages -Phages T2 and T4 were grown in TYN medium (1 liter contained 10 g of Difco tryptone, 2 g of yeast extract, 6 g of NaCll withE.scherichia coli B as host. T6 was grown in TYN medium, except that it contained 5 g of yeast extract and 10 g of NaCl, per liter, and E. coli B strain RH288 (obtained from Dr. A.

Markowitz,
The University of Chicago) was used as host. Cells were infected at a density of 7 x lOa per ml, with a multiplicity of infection of 0.2 for T2 and T4, and of 2 for T6. Prior to infection, tryptophan was added to a concentration of 100 pg/ml. Infection was allowed to proceed for 3 h at 37", and lysis was aided by the addition of CHCl,. Phages were purified from crude lysates by differential centrifugation, and then by banding in CsCl. The specific activity and the conditions for annealing were the same as in A, except that the charged tRNA preparation contained 30 x IO6 cpmiml.
The specific activity and the conditions for annealing were the same as in A, except that the charged tRNA preparation contained 54 x lo6 cpm/ml. The hybridized values shown were corrected by subtraction of radioactivity found in washout control filters which was less than 100 cpm. were the same as those described for Fig. 1 were the same as for Fig. 1. B, hybridization with crude T4 [3H]prolyl-tRNA.
The specific activity and conditions for annealing were the same as in A, except that the charged tRNA preparation contained 13.2 x lo6 cpm/ml. infected cells, charged with VHlproline, showed no hybridization to T6 DNA, suggesting the absence of this tRNA gene in the T6 chromosome. Table I summarizes the results presented in Figs. 1 to 5 and shows the presence (+) or absence (-) of tRNA sequence homology for the five ammo acid tRNA species coded by each T-even phage. The absence of a detectable phage tRNA species is indicated by the letters ND.
Chromatography of Short Term Infected Phage tRNAs-The crude tRNA recovered from 13-to 16-min phage-infected cells was used for the preparation of phage-specific tRNAs as described under "Materials and Methods." The phage-specific tRNAs were charged with a single 3H-aminoacid and subjected to chromatography in the RPC-5 system of Pearson et al. (101. Fig. 6 shows the chromatography profile for T2, T4, and T6 not with T6 DNA. However, tRNA preparations from T6-tRNAs charged with [3H]leucine. The profile of Escherichia coli [Wlleucyl-tRNA is shown only in the upper panel of Fig.  6, but it was included in each chromatographic run. Multiple, and similar, isoacceptor leucyl-tRNA species were observed for T2 and T6 tRNAs, different from the tRNALeU isoacceptors of E. coli. When T4 tRNAs were charged with [3H]leucine, they repeatedly gave poorly defined chromatographic profiles. Chromatography of the three T-even phage-specific tRNAs charged with [3H]glycine showed very similar profiles (Fig. 7). Two isoacceptor phage glycyl-tRNA species were seen in each case, the major species eluting in a position almost identical with the major glycyl-tRNA species of E. coli. Contamination of the phage tRNA preparations with E. coli tRNAs is unlikely, since the chromatographic profiles of phage tRNAs charged with other ammo acids (Figs. 6, 8, and 10) did not show any significant coincidence with the charged E. coli tRNA markers.
Multiple tRNA*'g isoacceptors appear for all three T-even phage tRNAs when these preparations are subjected to RPC-5 chromatography (Fig. 8). Although the profiles for T4 and T6 arginyl-tRNA were nearly identical, the profile for T2 arginyl-tRNA was distinctly different. In the latter case, higher salt concentrations were required for the elution of T2 tRNA*'g isoacceptors.
No profiles are given for I6 tRNA"" or T2 tRNA"e since these phage tRNA species were not detected. The elution profiles for T2 and T4 prolyl-tRNAs were qualitatively similar, showing two tRNA"O isoacceptor species; the relative amounts of the T2 and T4 isoacceptors were different, however. For isoleucyl-tRNA, T4 tRNA showed two distinct RPC-5 peaks, but T6 tRNA gave only one.
Chromatography of Long Term Infected Phuge tRNAs-Phage-specific tRNAs were also prepared from cells infected with T-even phages for 60 min in the presence of chloramphenicol. For most of the "long term" phage tRNA preparations, RPC-5 profiles similar to those for the "short term" phage tRNA species were observed. However, as shown in Fig. 11, the chromatographic profile of tRNAGIY, isolated from 60-min infected cells, was significantly different from that for short term infected tRNAGIY (Fig. 7) in that an extra isoacceptor peak appeared which eluted at lower salt concentrations; in addition, there were quantitative changes in the same isoacceptors observed previously. It is possible that the altered 60min infected tRNAGLY profile might be attributable to some nonspecific effect of chloramphenicol (14), but similarly altered tRNAG'Y profiles were found for phage tRNAs prepared from 60 min-T4am61-infected cells, defective in T4 lysozyme production, in the absence of chloramphenicol.
The appearance of multiple isoacceptor tRNA species might result from the enzymatic modification of a single tRNA gene transcript; this possibility was explored for T4 tRNAG1y by pulse-chase experiments with inorganic "Pi.
E. coli. cells infected with T4am61 were exposed to 32Pi between 6 and 12 min after infection (in the absence of chloramphenicol); a portion of these cells was removed and subjected to phenol extraction. To the remaining cells, an excess of cold inorganic phosphate and rifampicin was added, and they were allowed to incubate for a total infection time of 60 min. The "chased" and "nonchased" infected cells were proc-essed in the same way for the isolation of [32P]tRNAc1r (see under "Materials and Methods"). The ["P]tRNAG*J profile (Fig. 12A) for the 12-min infected cells (nonchased) was almost identical with that found for 12-min infected T4 [3H]glycyl-tRNA (Fig. 7). Fig. 12R shows that, after an extended chase period, the major radioactive tRNAGiy peak of Fig. 12A is reduced, whereas the minor peak of Fig. 12A is considerably enhanced. At the same time, an additional radioactive peak appears which elutes earlier than the other two isoacceptors, at slightly lower salt concentrations. In the absence of rifampicm (Fig. 120, cells infected with the mutant phage T4am61 and continuously exposed to 32Pi for 60 min produce multiple tRNAo" isoacceptor species on RPC-5, coincident with the nonchased and chased isoacceptors shown in Fig. 12, A and B . DISCUSSION These studies demonstrate that all three T-even phages carry genes coding for tRNAs, and that this information is expressed following phage infection ofEscherichia coli. Of the five amino acid tRNA species examined (leucine, glycine, arginine, proline, and isoleucine), T4 RNA extracts contained all five species, whereas T2 and T6 extracts were missing tRNA*ie and tRNAbo respectively. For similar amino acid tRNA species expressed by each of the three T-even phages, a significant degree of sequence homology was detected, except in the case of T2 tRNAAm which showed no homology with either T4 or T6 tRNAA". This difference between the tRNA**g species of the T-even phages was also reflected by RPC-5 chromatography, in which the profile of T2 arginyl-tRNA was  Fig. 6 were used for charging with 3H-(E.C.) B tRNAs (described for Fig. 6) were used for charging with and W-labeled glycine, respectively. The procedure for chromato-3H-and W-labeled arginine, respectively. The procedure for chrographic analysis was the same as for Fig. 6, except that a 0.4 M to matographic analysis was the same as for Fig. 6, except that a 0 applied radioactivity was greater than 85%.
distinctly different from those obtained with T4 and T6 arginyl-tRNAs; the latter two profiles resembled each other rather closely. The quantitative differences observed for the hybridization of the various charged tRNA species to the different phage DNAs is not altogether clear. In some cases, saturation of Teven DNA with homologous tRNA species was not achieved, hence, interpretation of these results is difficult. For those reactions in which near saturation or saturation was achieved, hybridization to T2 DNA appeared to be significantly higher than to T4 and T6 DNA (e.g. leucyl-and prolyl-tRNAs).
These results might indicate the presence of multiple tRNA gene copies in the T2 genome; however, the hybridization results with other tRNA species (e.g. glycyl-tRNA] showed relatively small differences among the three phage DNAs. It is possible, therefore, that the variations in the levels of hybridization observed for each DNA were due to some intrinsic factor in the annealing procedure itself.
For the three T-even phage RNA extracts, multiple tRNA isoacceptors were detected on RPC-5 for the five different amino acid tRNA species. In some instances, coincident profiles for phage and E. coli tRNAs were observed on RPC-5 chromatography.
One would expect that, if the phage-specific tRNAs were contaminated with host tRNAs, all of the phage 13H]aminoacyl-tRNA species examined on RPC-5 would also have shown 3H-radioactivity coincident with that of E. coli l'4C]aminoacyl-tRNAs used as markers; this was not the case. B tRNAs described for Fig. 6 were used for charging with 3Hand 'V-labeled proline, respectively. The procedure for chromatographic analysis was the same as for Fig. 6 10. Reversed phase chromatography of short term infected T-even phage isoleucyl-tRNAs.
The phage-specific and Escherichiu coli (EC.) B tRNAs described for Fig. 6 were used for charging with 3H-and "C-labeled isoleucine.
The procedure for chromatographic analysis was the same as for Fig. 6 conditions. 3ZP-labeled tRNA was isolated from cells infected with T4am61, which had been exposed to 32P1 under various conditions.
The labeled tRNAolY species were isolated as described under "Materials and Methods" and subjected to chromatography as outlined in Fig. 7. Escherichia coli (E.C.) B 13Hlglycyl-tRNA served as a marker for each of the above runs. A, [3zPltRNAG1' isolated from T4am61-infected cells pulsed with 52P, from 6 to 12 min of the infection period. B, [32PltRNAG13 isolated from T4amGl-infected cells, pulsed with 32Pi as in A, and then chased in the presence of a large excess of cold inorganic phosphate and rifampicin for a total infection time of 60 min. C , [32P]tRNAo'S isolated from cells infected with T4am61 phage and exposed to 32Pi from 6 to 60 min of infection, in the absence of rifampicin. E. co2i B tRNA charged with 13HIglycine was included in each chromatography but is shown only in A. (9,16,17), the presence or absence of multiple tRNA isoacceptor gene loci in T-even phage DNAs has not been firmly resolved.
The similarity in the phage tRNA species expressed by the three T-even phages supports the information obtained previously which indicates the close genetic relationship between these three bacteriophages.
Nevertheless, some differences in phage tRNA synthesis and sequence homology have been found; this emphasizes that the degree of genetic relatedness for different biological systems is often a function of the particular biochemical parameter examined. Except for highly selected bacterial strains (18), bacteriophage tRNAs have not been shown to be required for phage growth. Our findings are consistent with the view that variations in genetic information between highly related organisms are most likely to be found in those genes which are not essential for growth and survival.