BIOSYNTHESIS OF /3-D-GALACTOSIDASE CONTROL LED BY PHAGE-CARRIED GENES. II. THE BEHAVIOR OF PHAGE- TRANSDUCED z+ GENES TOWARD REGULATORY MECHANISMS* BY H. R. REVEL AND S. E. LURIA MASSACHUSETTS INBTITUTE OF TECHNOLOGY Cummunicnted October 23,1961 The preceding paper' described the kinetics of induced B-D-galactosidase produc- tion in z- bacteria following transduction of the z+ gene by the defective phage Pldl. It was concluded that, in the presence of an inducer, the newly entered z+ genes begin to function shortly after having gained access to the recipient cells and continue to function at a constant rate, even though they do not multiply and are not integrated into the recipient cell chromosome. The present paper is con- cerned with the function of the i locus and with the functional interplay between the specific repressors for galactosidase and for phage Pl on the expression of newly entered z+ genes. The results reveal some novel features of the control of gene function in bacterial cells. Materials and methods were the same as in the preceding paper.' Results.--Synthesis of galactosidase after transduction in the absence of inducer: Figures 1 and 2 show the results of infection of E. coli i-z- or i+z- with high fre- Minutes After Dilution FIGS. I and Z.-Formation of galactosidase following transduction in the presence or absence of an inducer. Bacteria, infected in TGA medium for 15 min, were diluted at time 0 in medium with or without lo--" A4 TMG. Pldlu, i+.z'- and PI& i-z-+ were used a~ donor lysatea. 1968 VOL. 47, 1961 BIOCHEMISTRY: REV&L AND LURIA 1969 quency transducing ( = HFT) lysates containing Pldl particles that carry either i-z+ or i+z+ genes in the absence of an external inducer; results obtained in the presence of inducer' are included for purposes of comparison. The results can be summarized as follows : 1. When no i+ gene is present, galactosidase synthesis proceeds.as it would in the presence of an inducer. 2. When the if gene is present in the phage only, synthesis of the enzyme begins as usual, then slows down and ceases at about 60 min. This finding is similar to that obtained in mating experiments2 and, as in the case of mating, can be ascribed to a delayed expression of the i+ gene in producing the specific repressor for the Zac genes. 3. When the recipient is i+z- and the transducing phage carries i-z+, there is a constitutive synthesis of galactosidase at about 20 per cent the rate observed in the presence of inducer. This synthesis continues at a linear rate for at least 3 hours. 4. When both the recipient bacterium and the transducing phage carry the if gene, there is an early constitutive enzyme production ceasing at about 60 min. If the amounts of enzyme produced in cases 2 and 4 are normalized to the corre- sponding rates of synthesis by these two systems in the presence of inducer, the value for case 4 is 20 to 25 per cent that for case 2. The results of cases 3 and 4 are unexpected. According to current views, based mainly on the study of heterogenotes, in which the exogenote is part of a male chro- mosome or of a fertility factor,3 one copy of the i+ gene should be enough to saturate with repressor all the Zac operons present in a cell. Instead, we observe that a z+ gene in the phage is only incompletely repressed after entering an i+ recipient, although it soon becomes completely repressed by the activity of an if gene carried in the phage itself. Similar results were obtained with other i+ bacterial recipients, such as S. dys- enteriae strain Sh, and with another type of HFT lysate containing a Pldl i-z+ phage of different origin and with different properties. Before proceeding to further analysis of these results, we should point out that the experiment analogous to case 3, namely, the introduction of an i-z+ set of genes into an i+z- recipient, could not have been done in mating experiments,2 because the enzyme in the donor cells would preclude detection of enzyme made by the zygotes (barring studies with labeled protein). Likewise, in a mating analogous to case 4, that is, the introduction of an i+z+ gene into an i+z- recipient, it would be difficult to detect the small amounts of enzyme made by the zygotes over the back- ground level of enzyme in the uninduced i+z+ parent cells. In phage transduction, of course, there are no limitations of the sort. The mechanism of cmtitutive gahctosidase sph&s after transduction: Several trivial explanations of the constitutive synthesis observed in case 3 were tested and excluded by simple experiments. Thus, heterogeneity in the recipient Z- populations with respect to i+ * i- mutation was excluded by using as recipients several single colony isolates from an i+z- strain; all gave results like those of Figure 2. The possibility that the Pldl phages that were supposedly i-0+2+ (PldE type 13-4) had instead a lac region with the genetic composition i+o%+ or i-o%+ (dom- inant constitutive) was excluded, among other reasons, because the heterogenote strains with i+z- in their chromosome and a Pldl 13-4 prophage behaved as fully 1970 BIOCHEMISTRY: REVEL AND LURIA PROC. N. A. S. or almost fully repressed (see below), whereas heterogenotes i+z- carrying Pldl i-o%+ behaved as typical oC strains (constitutive synthesis f&7 per cent of the in- duced leve13p4). Another interpretation, that the i+ gene in the i+z- recipient cells may not be fully functional because the adjacent z portion of the lac region is inactive, was excluded by the observation that constitutive synthesis, as in case 3, was also ob- served when phage with i-z+ was used to infect i+z+ bacteria in the absence of in- ducer, as shown in Figure 3. This experiment also indicated that the escape of a --I 0 i+r-racipicnl; TMG Y i+ z-rocipienli no TMG s i+z+rocipirnta no TMG Fw. X-Formation of galactosidase fol- lowing transduction of B. coli i+z- or a'+~+ by phage Pldl i-z+. The values for the i+z+ recipient were corrected by subtracting the small amounts of fl-galactosidaae formed by a control culture without phage. phage-carried gene z + from repression is probably not shared by a Z+ gene in the recipient, because the rate of constitutive synthesis was not higher than in the case of a z- recipient. In addition, it was easy to show that escape from repression was not due to phage infection per se since infection of i+z+ bacteria with normal phage Pl does not lead to constitutive galactosidase synthesis; nor does mixed infection of i+z- cells with l'ldl i-z+ and normal Pl increase the level of constitutive synthesis. The escape phenomenon is related to the entry of z+ genes as part of a phage and not simply to their Zecation within a phage element. In fact, the study of several strains of different genetic constitutions, listed in Table 1, showed that once a stable (or quasi-stable) lysogenic relation is established, repression of the z+ gene in phage Pldl by an i+ gene in the host chromosome, that is, in "trans" position, is quite effective, although not always as complete as repression by an i+ gene in "cis" position.s Also, an i+ gene in an established prophage represses a y + gene in the host chromosome and vice versa, as shown by permease measurement on strains i-z-y+ (Pldl i+z+g-) and i+z+y- (PM i-z-y+).4 The repression of z+ or y+ VOL. 47, 1961 BIOCHEMISTRY: REVEL AND LURIA TABLE 1 EFPEcTIVENEfiS OF I~Tw~SION RY I+IN VARKKJ~ 14ACfkhAINS Strain ------Genotype----- ---Enzyme units/O. D. units- Chromosome Pldl -%fG +&SC W4032-W-5 :h:-:2-w-1 Sh 31-20 3.050-13-4-l 3.050-13-4-2 Sh 125-13-4 i+z+ 1.2 1700 i+z- i+z+ i+z+ 1.5 1.0 1800 900 i+z- i+z+ 1.6 2500 i+z- i-z+ i+z- i-.2+ i+z- i-z+ 1971 Ratio (b)/(a) 1100 1 1400 1250 900 1600 600 2: The bacterial strains were grown in TGA medium in the presence or absence of 10-a M TMG. Ali- quota were removed at intervals for measurement of optical density and of galaotosidase. genes by an i+ gene located in trans is analogous, of course, to the repression ob- served in similar situations between genes carried in two chromosomal fragments in mating and in F-duction heterogenotes2v 3 It seemed possible that the z+ gene in a newly entered phage escaped repression because of an insufficient cytoplasmic level of the product of the internal i+ gene. If so, the Z+ gene might be brought under control by addition of glucose, which antagonizes external inducers, supposedly by increasing production of internal repressors6 The results shown in Table 2 indicate, however, that the escape syn- TABLE 2 EFFECTOF GLUCOSE: ON ENZYME SYNTHESIS Carbon source in nrowth medium Enzyme synthesis after transduction i a +PldZ + i +z - cells (1) (?p$&) T%l (3) Induced (+TMG) units per ml per hour (2-l) Enzyme synthesis 3.000 "i.$ i+s+ Cpd&ti~ Induced (+TMG) units per 1 unit O.D. keg' I % Inhibition by glucose 0.85 0.5 0.7 3.7 0.2 2.85 1270 516 1050 110 40% . 93% 59% 89% thesis, like the constitutive synthesis in i-z+ bacteria, is reduced only to about one- half by glucose, whereas the TMG-induced synthesis is reduced lo- to 20-fold. Thus, if the escape phenomenon is due to an insufficient amount of repressor, this insufficiency is not remedied by glucose; it may be an insufficiency of aporepressor only. An insufficient amount of repressor might be a reflection of early multiplica- tion of the i+ gene copies introduced with phage. If so, when these gene copies become diluted in the course of cell division, one might expect escape synthesis of enzyme t,o cease. For at least 3 hours this was not observed. Another explanation of the escape phenomenon would be an early interaction between the newly entered genes and the host genes, leading to inactivation of the host gene i+ in some of the bacteria, presumably by recombination. Apart from the fact that most recipient cells probably have several nuclei, hence several i+ copies, we have obtained no evidence to suggest such an early recombination be- tween exogenote and endogenote. If integration of newly entered genes could occur almost immediately, some of the Zac + transductants may be expected to start to multiply quite early; instead, the number of Zac + transductant colonies remains constant for hours. In addition, several experiments were done in which HFT 1972 BIOCHEMISTRY: REVEL AND LlJRIA PROC. N. A. 5. lysates containing Pldl phage with the gfxles o+z- were used to infect either another (J+Z-- or WI ~J"Z+ rC?cipkIlt. .In these cases, recombination within tho o-z region is required to give a functional z + locus, and the transduction frequencies are cor- respondingly lower than with o+z+ donors. If a substantial part of the recombina- tional events could occur very early after infection, some galactosidase activity should be found; none was observed (beyond the very low levels corresponding to the uninfected levels of the recipient cells in the cases of two leaky mutant strains, E. coli 3.320 and 3.OUO). In summary, we conclude that the escape pheqomenon is not due to failure or displacement of the i+ gene of the host, but reflects a peculiar behavior of a lac region that has newly entered a bacterial cell as part of a phage. Transduction to Pl-lysogenic recipients: Another peculiarity of the behavior of phage-transduced z+ genes toward repression is revealed by a comparison of Pl- sensitive and Pl-lysogenic recipient strains. The course of enzyme production in the presence of an inducer is qualitatively similar to both cases; but the rate of enzyme production is much lower with the lysogenic recipients (see ref. 1, Fig. 5). Table 3 gives values for several recipients and for a variety of phage TABLE 3 SYNTHESIS OF GALACTOSIDASE AVER TRANSDUCTION TO Pl-SENSITIVE AND PI-LYSOQENIC RECIPIENTS - Donor phsge -- Pldlw i +z + Pldl i+c+ Pldl i+oCz+ R;ci&rb l?nxvmr. mit~/tnl/lll Ena me. iy lh~~WlP. Ratio unita ml/hr Ratio nrrit~/IId/h1 Rnt,io E. coli i +z- 5.2 4.5 0.6 8 6.6 2.3 E. colii+z-(Pl) 0.65 0.67 0.26 E. coli i-z- 2.7 1.1 67 Ii.6 E. coli i-z-(PI ) 0.04 0.2 Shigells i +z- 8.4 4.2 1.3 3/, 3.8 7.2 Shigellai+z-(Pl) 0.25 1.1 0.18 The veluen for enryme correspond to the linear phase of synthesis in tht! presence of 10-s M TMG. The values in italics BTB the ratios between the values for Pl-sensitive snd for Pl-lysogenic recipient ntrainn. The ratas of Pl adsorption to lysogenic and sensitive recipimta 8~8 identiaal. types. The ratios between rates of synthesis in lysogenic vs sensitive cells range from 1: 2.3 to 1: 67. It is noteworthy that the repression by the Pl prophage is more effective with those types of Pldl phages, such as Pldlw, that are known to be them- selves less effective in phage-immunity functions and less able to compete with a normal Pl prophage.' Discussion.-Two main findings emerge from these results: the partial escape of newly transduced z+ genes from repression by the i+ genes of the recipient cells, and the inhibition by a Pl prophage of the expression of a .z+ gene newly entered as part of a phage. The escape phenomenon indicates that repression by i+, which is fully effective on a z+ gene in cis position, is not always so in trans. Repression appears to work effectively in trans between genes in the chromosome and in an F' factor$ it is al- most as effective, although often incomplete, between chromosome and Pldl pro- phage.6 Yet, internally produced repressor is only about 80 per cent effective in repressing z+ genes tha.t have recently entered with a phage. Individual-cell ex- VOL. 47, 1961 BIOCHEMISTRY: REVEL AND LURIA 1973 periments with fluorogenic substrate' will be needed to decide whether the escape is due to full enzyme production in 20 per cent of the cells or to enzyme production at 20 per cent the maximal rate in all the i+z- cells that have received an i-z+ phage . It remains to be seen whether the escape phenomenon is peculiar to lac+ genes that have entered with a phage or occurs with any "newcomer" genes. Experi- ments with F' elements carrying i-z+ genes are in progress. The lower level of galactosidase synthesis after transduction to recipient bacteria that are Pl-lysogenic might reflect either a suppression of multiplication of the newly-entered phage element or an effect of the prophage, probably through the immunity repressor, on the function of the lac+ genes. Arguments against ex- tensive multiplication of the recently entered Pldl phage have been given.' Also, a phase of accelerated enzyme synthesis is still observed with PI-immune recipients (see ref. 1, Fig. 5) indicating that phage multiplication is probably not the explana- tion of the accelerated phase. It seems more likely that the effect of lysogeny on galactosidase production reflects a submission of the z+ gene in Pldl to the restrain- ing effect of the immunity repressor: the genes of the lac operon, being part of a (defective) Pl phage, become part of a "superoperon" controlled by the Pl-im- munity repressor. It is known that the functions of most phage genes are repressed by immunity. 3 The Zac genes in the phage are apparently subject to this control; more or less so, depending on the complement of Pl genes with which they are as- sociated in different Pldl types. In fact, specific phage genes are responsible, not only for production of the immunity repressor, but also for sensitivity to it (im- munity-operator genes3). It is remarkable that immunity should repress the function of z+ genes in a newly entered PldE phage and not in an established Pldl prophage of the same type (see Table 1). Clearly, with regard both to repression by the i+ gene and to the effects of immunity, the recently entered genes behave differently from genes in stabilized components of the cellular genome. The differences may reflect either differences in intracellular topography or differences in status of the genetic elements with re- gard to multiplication. This question may be discussed more profitably after con- sidering the experiments on derepression to be reported in the following paper.6 summary.-Introduction of z+ genes into i+z- bacteria by transduction, in the absence of an external inducer, results in a constitutive synthesis of @-D-galactosi- dase at about one-fifth the maximum induced rate. This "escape phenomenon" is attributed to a partial failure of phage-associated z+ genes to respond to the chro- mosomally-controlled repressor when they first enter the recipient cell, but not when they are part of an established prophage. In addition, the z+ gene in a newly en- tered phage element, but not in an established prophage, is subject to partial repres- sion by the specific phage-immunity repressor. The z+ gene in the phage, however, is fully subject to repression by an i+ gene in the sa.me phage element. * Aided by grants from the National Science Foundation, G-8808, and the National Institute of Allergy and Infectious Diseases, National Institutes of Health, E3038 (Cl ). 1 Revel, H. R., S. E. Luria, and B. Rotman, these PROCEEDINQS, 47,1956 (1961). * Pardee, A. B., F. Jacob, and J. Monod, J. Mol. Biol., 1,165 (1959). 3 Jacob, F., and J. Monod, ibid., 3,318 (1961). 1974 BIOCHEMISTRY: REVEL, LURIA, AND YOUNG PROC. N. A. S. ' Franklin, N. C., and S. E. Luria, Virology, in press (1961) and unpublished results. 5 Revel, H. R., S. E. Luria, and N. L. Young, these PROCEEDINQS, 47, 1974 (1961). 6 Neidhardt, F. C., and B. Magasanik, J. Bacterial., 73,253 (1957). `I Luria, S. E., J. N. Adam, and R. C. Tmg, Virology, 12,348 (1960).