Isolation, characterization and deletion mapping of amber mutations in the cll gene of phage lambda

Protection of bacteriophage-sensitive Escherichia coli by lysogens

SignificanceSome viruses that infect bacteria, temperate bacteriophages, can confer immunity to infection by the same virus. Here we report λ-immune bacteria could protect λ-sensitive bacteria from killing by phage λ in mixed culture. The protection depended on the extent to which the immune bacteria were able to adsorb the phage. Reconciling modeling with experiment led to identifying a decline in protection as bacteria stopped growing. Adsorption of λ was compromised by inhibition of bacterial energy metabolism, explaining the loss of protection as bacterial growth ceased.

Population Dynamics of Phage and Bacteria in Spatially Structured Habitats Using Phage λ and Escherichia coli

Bacteria living in physically structured habitats are exposed heterogeneously to both resources and different types of phages. While there have been numerous experimental approaches to examine spatially distributed bacteria exposed to phages, there is little theory to guide the design of these experiments, interpret their results, or expand the inferences drawn to a broader ecological and evolutionary context. Plaque formation provides a window into understanding phage-bacterium interactions in physically structured populations, including surfaces, semisolids, and biofilms. We develop models to address the plaque dynamics for a temperate phage and its virulent mutants. The models are compared with phage λ-Escherichia coli system to quantify their applicability. We found that temperate phages gave an increasing number of gradually smaller colonies as the distance increased from the plaque center. For low-lysogen frequency this resulted in plaques with most of the visible colonies at an intermediate distance between the center and periphery. Using spot inoculation, where phages in excess of bacteria were inoculated in a circular area, we measured the frequency and spatial distribution of lysogens. The spot morphology of cII-negative (cII(-)) and cIII(-) mutants of phage λ displays concentric rings of high-density lysogenic colonies. The simplest of these ring morphologies was reproduced by including multiplicity of infection (MOI) sensitivity in lysis-lysogeny decisions, but its failure to explain the occasional observation of multiple rings in cIII(-) mutant phages highlights unknown features of this phage. Our findings demonstrated advantages of temperate phages over virulent phages in exploiting limited resources in spatially distributed microbial populations.

IMPORTANCE

Phages are the most abundant organisms on earth, and yet little is known about how phages and bacterial hosts are influencing each other in density and evolution. Phages can be either virulent or temperate, a difference that is highlighted when a spatially structured bacterial population is infected. Phage λ is a temperate phage, with a capacity for dormancy that can be modified by single gene knockouts. The stochastic bias in the lysis-lysogeny decision's probability is reflected in plaque morphologies on bacterial lawns. We present a model for plaque morphology of both virulent and temperate phages, taking into account the underlying survival of bacterial microcolonies. It reproduces known plaque morphologies and speaks to advantages of temperate phages in a spatially structured environment.

Host responses influence on the induction of lambda prophage

Inactivation of bacteriophage lambda CI repressor leads almost exclusively to lytic development. Prophage induction can be initiated either by DNA damage or by heat treatment of a temperature-sensitive repressor. These two treatments also cause a concurrent activation of either the host SOS or heat-shock stress responses respectively. We studied the effects of these two methods of induction on the lytic pathway by monitoring the activation of different lambda promoters, and found that the lambda genetic network co-ordinates information from the host stress response networks. Our results show that the function of the CII transcriptional activator, which facilitates the lysogenic developmental pathway, is not observed following either method of induction. Mutations in the cro gene restore the CII function irrespective of the induction method. Deletion of the heat-shock protease gene ftsH can also restore CII function following heat induction but not following SOS induction. Our findings highlight the importance of the elimination of CII function during induction as a way to ensure an efficient lytic outcome. We also show that, despite the common inhibitory effect on CII function, there are significant differences in the heat- and SOS-induced pathways leading to the lytic cascade.

Phage lambda CIII: a protease inhibitor regulating the lysis-lysogeny decision

The ATP-dependent protease FtsH (HflB) complexed with HflKC participates in post-translational control of the lysis-lysogeny decision of bacteriophage lambda by rapid degradation of lambda CII. Both phage-encoded proteins, the CII transcription activator and the CIII polypeptide, are required for efficient lysogenic response. The conserved CIII is both an inhibitor and substrate of FtsH. Here we show that the protease inhibitor CIII is present as oligomeric amphipathic alpha helical structures and functions as a competitive inhibitor of FtsH by preventing binding of the CII substrate. We identified single alanine substitutions in CIII that abolish its activity. We characterize a dominant negative effect of a CIII mutant. Thus, we suggest that CIII oligomrization is required for its function. Real-time analysis of CII activity demonstrates that the effect of CIII is not seen in the absence of either FtsH or HflKC. When CIII is provided ectopically, CII activity increases linearly as a function of the multiplicity of infection, suggesting that CIII enhances CII stability and the lysogenic response. FtsH function is essential for cellular viability as it regulates the balance in the synthesis of phospholipids and lipopolysaccharides. Genetic experiments confirmed that the CIII bacteriostatic effects are due to inhibition of FtsH. Thus, the early presence of CIII following infection stimulates the lysogenic response, while its degradation at later times ensures the reactivation of FtsH allowing the growth of the established lysogenic cell.

Genetic analysis of bacteriophage lambda cIII gene: mRNA structural requirements for translation initiation

The bacteriophage lambda cIII gene product regulates the lysogenic pathway. The cIII gene is located in the leftward operon, which is transcribed from the pL promoter. We have previously shown (S. Altuvia and A. B. Oppenheim, J. Bacteriol. 167:415-419, 1986) that mutations that show elevated expression lie within the cIII coding sequence. We isolated mutants that show decreased CIII activity. All the mutations were found to cause a drastic reduction in the rate of initiation of cIII translation. Several mutations were found to be scattered within the first 40 nucleotides of the cIII coding region. Additional mutations affected the AUG initiation codon, the Shine-Dalgarno sequence, and the upstream RNaseIII processing site. Computer folding of the cIII mRNA suggested the presence of two alternative RNA structures. All the mutations within the coding region that reduce expression reduce the stability of one specific mRNA structure (structure B). Mutations that increase expression lie in the loops of this structure and may in fact stabilize it by interfering with the formation of the alternative structure (structure A). Thus, it appears that a specific mRNA secondary structure at the beginning of the cIII coding region is essential for efficient translation, suggesting that changes in mRNA structure regulate cIII expression.

Effect of photoreactivation on mutagenesis of lambda phage by ultraviolet light

There is disagreement in the literature as to whether the major mutagenic photoproduct induced in DNA by ultraviolet light is the cyclobutane dipyrimidine dimer, the most common product, or the [6-4] photoproduct, the next most frequent. In the experiments reported here, cyclobutane dimers were removed from irradiated lambda phage DNA by enzymatic photoreactivation, a process thought to affect no other photoproduct. Photoreactivation of lambda phage in host cells and of lambda DNA in solution reduced clear plaque mutants per plaque-forming unit by two-thirds, in host cells with a constant and near-maximal expression of the SOS functions required for mutagenesis. This result is interpreted to mean that removal of cyclobutane dimers in or near the mutated gene reduces mutation induced by ultraviolet light by two-thirds; therefore, cyclobutane dimers in the phage DNA are responsible for most observed mutations. DNA sequences of mutations in photoreactivated phage showed a smaller fraction of G.C to A.T transitions and a larger fraction of A.T to G.C transitions, compared to phage that were not photoreactivated. This suggests that cyclobutane dimers at TC and CC sites are particularly mutagenic.

Aberrant regulation of synthesis and degradation of viral proteins in coliphage lambda-infected UV-irradiated cells and in minicells

The patterns of bacteriophage lambda proteins synthesized in UV-irradiated Escherichia coli cells and in anucleate minicells are significantly different; both systems exhibit aberrations of regulation in lambda gene expression. In unirradiated cells or cells irradiated with low UV doses (less than 600 J/m2), regulation of lambda protein synthesis is controlled by the regulatory proteins CI, N, CII, CIII, Cro, and Q. As the UV dose increases, activation of transcription of the cI, rexA, and int genes by CII and CIII proteins fails to occur and early protein synthesis, normally inhibited by the action of Cro, continues. After high UV doses (greater than 2,000 J/m2), late lambda protein synthesis does not occur. Progression through the sequence of regulatory steps in lambda gene expression is slower in infected minicells. In minicells, there is no detectable cII- and cIII-dependent synthesis of CI, RexA, or Int proteins and inhibition of early protein synthesis by Cro activity is always incomplete. The synthesis of early b region proteins is not subject to control by CI, N, or Cro proteins, and evidence is presented suggesting that, in minicells, transcription of the early b region is initiated at a promoter(s) within the b region. Proteolytic cleavage of the regulatory proteins O and N and of the capsid proteins C, B, and Nu3 is much reduced in infected minicells. Exposure of minicells to very high UV doses before infection does not completely inhibit late lambda protein synthesis.

Mutations of bacteriophage lambda that define independent but overlapping RNA processing and transcription termination sites

Bacteriophage lambda int gene expression is regulated differentially from transcripts originated at the pL and pI promoters. Transcripts initiated at pI terminate at the site tI and express int gene product efficiently. Polymerases starting at pL do not terminate at tI, due to the antiterminating activity of lambda N protein. The pL transcripts are unable to express Int protein efficiently because sib, a control site overlapping tI in the unterminated RNA, is processed by host RNase III. We have isolated lambda sib- mutants by their inability to inhibit int expression from pL transcripts. sib mutations were genetically mapped to the left of the lambda attachment site, and do not structurally alter this site for recombination. Several sib mutations do alter the nucleotide sequence of the overlapping sib and tI sites. The lambda sib- mutants tested prevent RNA processing but do not affect transcription termination in vivo.

Base substitution mutations induced in the cI gene of lambda phage by neocarzinostatin chromophore: correlation with depyrimidination hotspots at the sequence AGC

Treatment of intact lambda phage with the nonprotein chromophore of neocarzinostatin resulted in efficient phage inactivation and generation of clear-plaque mutants. Both effects required a preincubation at low pH to allow diffusion of chromophore into the phage head. Chromophore activation was then effected by addition of a sulfhydryl cofactor, followed by a shift to neutral pH. Sequence analysis of mutations mapped to the DNA-binding region of the cI gene revealed that nearly all were single base substitutions. Significant numbers of all possible base changes were found, with A:T to G:C transitions being the most frequent events. Of 11 G:C to A:T transitions, 7 were found at C residues in the trinucleotide sequence AGC, which has previously been shown to be a hotspot for chromophore-induced depyrimidination. This result, as well as the SOS dependence of mutagenesis and the overall distribution of various types of base substitutions, is consistent with the hypothesis that apurinic/apyrimidinic sites are important mutagenic lesions.

Pyrimidine dimers are not the principal pre-mutagenic lesions induced in lambda phage DNA by ultraviolet light

Experiments were performed to examine the role of cyclobutyl pyrimidine dimers in the process of mutagenesis by ultraviolet (u.v.) light. Lambda phage DNA was irradiated with u.v. and then incubated with an Escherichia coli photoreactivating enzyme, which monomerizes cyclobutyl pyrimidine dimers upon exposure to visible light. The photoreactivated DNA was packaged into lambda phage particles, which were used to infect E. coli uvr- host cells that had been induced for SOS functions by ultraviolet irradiation. Photoreactivation removed most toxic lesions from irradiated phage, but did not change the frequency of induction of mutations to the clear-plaque phenotype. This implies that cyclobutyl pyrimidine dimers can be lethal, but usually do not serve as sites of mutations in the phage. The DNA sequences of mutants derived from photoreactivated DNA showed that almost two-thirds (16/28) were transitions, the same fraction found for u.v. mutagenesis without photoreactivation. These results show that in this system, the lesion inducing transitions (the major type of u.v.-induced mutation) is not the cyclobutyl pyrimidine dimer; a strong candidate for a mutagenic lesion is the Pyr(6-4)Pyo photoproduct. On the other hand, photoreactivation of SOS-induced host cells before infection with u.v.-irradiated phage reduced mutagenesis substantially. In this case, photoreversal of cyclobutyl dimers serves to reduce expression of the SOS functions that are required in the process of targeted u.v. mutagenesis.

Recombinational hotspot activity of Chi-like sequences

Chi sites, consisting of the nucleotide octamer 5' G-C-T-G-G-T-G-G 3', stimulate coliphage lambda recombination mediated by the Escherichia coli RecBC recombination pathway. In a sensitive genetic assay using phage lambda crosses, three of four Chi-like sequences tested, namely 5' A-C-T-G-G-T-G-G 3', 5' G-T-T-G-G-T-G-G 3' and 5' G-C-T-A-G-T-G-G 3', had about 6%, 11% and 38% of full Chi activity, respectively. We conclude that certain Chi-like sequences manifest a spectrum of recombinational hotspot activities and may account for RecBC-mediated generalized recombination of lambda lacking Chi sites.

UV-mutagenesis at a cloned target sequence: converted suppressor mutation is insensitive to mutation frequency decline regardless of the gene orientation

Premutational lesions produced by ultraviolet radiation in the Gln2 tRNA genes of E. coli B/r show differing sensitivities to a mutation avoidance phenomenon known as mutation frequency decline (MFD). A mutation event that changes the wild-type gene to an amber (UAG) suppressor is normally sensitive to MFD. Mutation of this amber suppressor to an ochre (UAA) suppressor is not sensitive to MFD. These two mutation events occur in the same anticodon region of the DNA. The dissimilarity of MFD sensitivity between these two mutations may result because the respective premutational photoproducts for the two are located in opposite strands of duplex DNA. To examine the effect of strand position of the premutational lesions on MFD, recombinant lambda phage were constructed that contained the amber suppressor as a mutation target in the two possible orientations. Comparison of MFD in bacterial lysogens containing either of the two types of recombinant prophage indicated that reversing the orientation of the target sequence relative to adjacent bacterial DNA had no effect on MFD. Since rotational inversion of the target sequence did not alter the sensitivity to MFD of mutation occurring at the cloned target gene, the antimutation process inherent to MFD can not be attributed to an asymmetrical interaction between the template strands and the DNA-replication complex.

Non-targeted mutagenesis of unirradiated lambda phage in Escherichia coli host cells irradiated with ultraviolet light

Non-targeted mutagenesis of lambda phage by ultraviolet light is the increase over background mutagenesis when non-irradiated phage are grown in irradiated Escherichia coli host cells. Such mutagenesis is caused by different processes from targeted mutagenesis, in which mutations in irradiated phage are correlated with photoproducts in the phage DNA. Non-irradiated phage grown in heavily irradiated uvr+ host cells showed non-targeted mutations, which were 3/4 frameshifts, whereas targeted mutations were 2/3 transitions. For non-targeted mutagenesis in heavily irradiated host cells, there were one to two mutant phage per mutant burst. From this and the pathways of lambda DNA synthesis, it can be argued that non-targeted mutagenesis involves a loss of fidelity in semiconservative DNA replication. A series of experiments with various mutant host cells showed a major pathway of non-targeted mutagenesis by ultraviolet light, which acts in addition to "SOS induction" (where cleavage of the LexA repressor by RecA protease leads to din gene induction): (1) the induction of mutants has the same dependence on irradiation for wild-type and for umuC host cells; (2) a strain in which the SOS pathway is constitutively induced requires irradiation to the same level as wild-type cells in order to fully activate non-targeted mutagenesis; (3) non-targeted mutagenesis occurs to some extent in irradiated recA recB cells. In cells with very low levels of PolI, the induction of non-targeted mutagenesis by ultraviolet light is enhanced. We propose that the major pathway for non-targeted mutagenesis in irradiated host cells involves binding of the enzyme DNA polymerase I to damaged genomic DNA, and that the low polymerase activity leads to frameshift mutations during semiconservative DNA replication. The data suggest that this process will play a much smaller role in ultraviolet mutagenesis of the bacterial genome than it does in the mutagenesis of lambda phage.

Changes in DNA base sequence induced by targeted mutagenesis of lambda phage by ultraviolet light

In targeted mutagenesis of lambda phage by ultraviolet light, the mutations are caused by radiation-induced lesions in the phage DNA. Of 62 mutations in the lambda cI gene that were sequenced, 41 (63%) of the targeted mutations were transitions, with similar numbers of C X G to T X A and T X A to C X G base changes. The remaining 21 mutations were about equally divided among eight transversions, seven frameshifts (5 additions and 2 deletions), and six double events with either two nearby base changes or a base change and a nearby frameshift. Of the 62 mutations, 60 could be associated with -Pyr-Pyr- sequences in the DNA, sites of likely photoproducts. For more information on this point, lambda phage were irradiated with 313 nm light in the presence of acetophenone, for which the major photoproduct is reported to be the thymine-thymine cyclobutyl dimer, with no measurable Pyr(6-4)Pyo photoproducts. Of 22 mutations sequenced, 19 were transversions and only one was a transition, permitting the conclusion that thymine-thymine cyclobutyl dimers are not the primary cause of ultraviolet light-induced transitions. A consideration of all the data strongly suggests that Pyr(6-4)Pyo photoproducts are mutagenic lesions.

Genetic analysis of a streptomycin-resistant Escherichia coli mutant temperature-sensitive for nonsense suppression

Continuing the genetic and biochemical characterization of the streptomycin-resistant Escherichia coli mutant LD1, we confirmed that LD1 is temperature-sensitive for suppression of nonsense codons, and that this phenotype of the mutant and its streptomycin-resistance are genetically linked and are probably caused by a single mutation, strA(LD1). We also isolated a spontaneous revertant, called LD1-R, which partially relieves the restriction of nonsense suppression caused by the strA(LD1) mutation. LD1-R is derived by an additional mutation (revA) which is closely linked to strA(LD1). We further demonstrate that the weak suppression of a lacZUGA mutation in a suppressor-free strain, which probably takes place by normal tRNA1rp, can be detected by the use of the chromagenic substance x-gal (5-Bromo-4-chloro-3-indolyl-beta-D-Galactopyranoside).

Posttranscriptional control of bacteriophage lambda gene expression from a site distal to the gene

The bacteriophage lambda int gene product, integrase, recombines the phage DNA with the host DNA at specific sites on each to accomplish lysogeny. The int gene is transcribed from two promoters, PL and PI, each regulated positively by lambda proteins. The expression of integrase is also controlled from a site, sib, in the b region of the phage genome. This is a unique regulatory site because it is located distal to the structural gene in relation to the promoters. The expression of int from the PL promoter is inhibited when sib is present. This effect appears to be specific for PL because sib does not cause inhibition of PI-dependent int synthesis. lambda mutants that contain alterations in the site have been isolated. Sequence analyses of the mutations reveal single base changes, spanning 37 base pairs (bp) in the b region, some 240 bp beyond the int gene. Another mutant, hef13, which has a phenotype similar to that of sib, introduces a nucleotide change within the same 37-bp region. The sib and hef mutations cluster within a region of dyad symmetry. Regulation of int synthesis by sib occurs after transcription of the int gene. There is no difference in the rate of PL-promoted int mRNA synthesis in either sib+ or sib- phage infections, yet int mRNA is less stable in the sib+ infection. Because RNase III host mutants are defective in sib regulation, processing of the PL mRNA at sib by this endoribonuclease may cause int mRNA decay and decrease int synthesis.

Fusion of the promoter region of rRNA operon rrnB to lac Z gene

A Lambda phage was constructed in which the structural gene for beta galactosidase is fused to a DNA segment carrying the ribosomal promoter rrnB of E. coli. In this hybrid operon beta galactosidase synthesis in vitro is repressed by ppGpp. Repression of beta galactosidase synthesis by cAMP is reported.

The cII-independent expression of the phage lambda int gene in RNase III-defective E. coli

This study compares the rates of lambda protein synthesis after infection of rnc- cells, which are defective in ribonuclease III (RNase III), with the analogous rates in an isogenic rnc+ host. Temporal differences in gene expression are reflected in a delay in turn-off of lambda early proteins as well as in the delayed appearance of late phage functions in rnc- host cells. Moreover, in the two hosts there is a striking difference in the regulation of gene int expression, which in wild-type cells requires the product of the lambda cII (and cIII )genes, whereas Int synthesis occurs in the absence of cII in RNase III-defective cells. These results suggest that RNase III may be a negative regulator of Int synthesis. The expression of int is also shown to be cII- and cIII-independent in rnc+ cells infected with b2-deleted phages, thus confirming previous studies on the negative regulation of int by the b2-region. Possible mechanisms of these two inhibitory effects on int expression are considered and the significance of int regulation in the control of site-specific recombination is discussed.

Inceptor and origin of DNA replication in lambdoid coliphages. I. The lambda DNA minimal replication system

In a pBR313-lambda dv hybrid plasmid system, stepwise deletion and serial cloning procedures have led to a functional dissection of the DNA replication region of lambdoid bacteriophages lambda, 434 and 21. A simple system for initiation of DNA replication has been detected within lambdoid replicator DNAs, which is active in the absence of several normal replication elements, including the origin of replication (ori) and product of gene O. This "minimal" (or "mini") initiation system depends on the p0 or substitute leftward promoter in conjunction with the newly discovered "inceptor" (ice) element, which is located within the cII gene. Even the fragments containing ori are unable to initiate replication in these hybrid plasmids as long as fragments containing ice are missing. The base sequence of ice resembles transcriptional terminators and it appears to control both termination of primer RNA and inception of daughter strand DNA synthesis. Initiation in the p0-ice mini system of lambda or 21 phages requires the gene P product. Hwever, mini replication of 434 DNA hybrid plasmids required neither O nor P proteins, although there are only two single-base changes in the 434 inceptor sequence. The mini system is repressed by the elements of the maximal lambda replication system, as described in the accompanying publication.

The requirement of nonsense suppression for the development of several phages

A spontaneous streptomycin-resistant Escherichia coli mutant which is temperature-sensitive for suppression of a nonsense codon was studied for its ability to propagate phages T2, T4D, T5, phi K, f2, MS2, R17, Q beta, lambda as well as filamentous phages fl, fd and M13. Of all phages tested, only the growth of Q beta, lambda, and filamentous phages is inhibited in the mutant at 42 degree C. This selective inhibition suggests that, like Q beta, lambda and filamentous phages also require a read-through proten(s) which results from suppression of a termination codon.

Bacteriophage lambda mutants (lambdatp) that overproduce repressor

Lambda tp mutants, selected for their ability to form turbid plaques on lon hosts, overproduce repressor. The tp1 and tp2 mutations have been located within (or adjacent to) the cIII gene. The tp1 mutation reduced late gene expression, as measured by endolysin synthesis (in the absence of functional cI repressor) and progeny phage yield. The tp4 mutation was mapped in the cY-cII region, and complementation tests indicated that tp4 affects the diffusible product of the cII gene. The tp4 mutation also reduced progeny production, but did not markedly affect endolysin synthesis.

Anomalous behavior of bacteriophage lambda polypeptides in polyacrylamide gels: resolution, identification, and control of the lambda rex gene product

The resolution of lambia proteins was compared on the two types of sodium dodecyl sulfate-polyacrylamide gels commonly in use. The two kinds of gel differ essentially in the ratio of the cross-linker, N'-N-bismethylene-acrylamide (bisacrylamide), to acrylamide monomer. Several lambda proteins migrate relatively more slowly in gels with high bisacrylamide/acrylamide ratios (HB gels) than in gels with low ratios, although the two types of gel are of roughly equivalent porosity. This effect is illustrated by a change in relative position of both the Rex and Int proteins, with apparent increases in molecular weight of about 8 and 15%, respectively, in the HB gels. This work confirms that like repressor and Int, the 28.5-kilodalton protein, identified as Rex on HB gels, is postively regulated by the lambdacII and cIII products and negatively controlled Cro. An intact y site is required for Rex and repressor expression after infection, whereas their synthesis in a lysogen is dependent upon a functional maintenance promoter, Prm.

Regulation of the int gene of bacteriophage lambda: activation by the cII and cIII gene products and the role of the Pi and Pl promoters

The activation of the int gene by the cII and cIII gene products was studied by analysing int expression following infection of UV-irradiated cells by various phage mutants. Residual expression of int, probably from Pl, takes place in the absence of cII/cIII activation. Activation of the int gene, like that of the cI repressor gene, is poor at low multiplicities of infection. The mutation intC, which allows constitutive int expression in the lysogenic state, partially relieves the requirement for cII and cIII activation. The kinetics of Int synthesis after addition of the inhibitor rifampicin suggest that the activation occurs at the transcriptional level.

Nucleotide sequence of cro, cII and part of the O gene in phage lambda DNA

A nucleotide sequence comprising 960 base pairs of bacteriophage lambda DNA has been determined. The sequence includes the entire genes of the regulatory proteins cro and cII, and part of the O gene, together with control elements for their transcription and translation. The right-hand boundaries of the lambdaimm434 and lambdaimm21 substitutions and the cy42 mutation have been located.

A complex control circuit. Regulation of immunity in temperate bacteriophages

Temperate bacteriophages can display in a stable way two essentially different behaviours. In the immune state, a gene (cI) produces a repressor which prevents expression of all the other viral genes; in the non-immune state the typically viral functions are expressed. The choice between the two pathways and the establishment of one of them have much in common with cell determination and differentiation. This choice depends on a complex control system, in fact one of the most intricate nets of regulation known in some detail. Our paper provides a formal description and partial analysis of this regulatory net. It is shown that even for relatively simple known models, this kind of analysis uncovers predictions which had previously remained hidden. Some of these predictions were checked experimentally. The experimental part chiefly deals with the efficiency of lysogenization by thermoinducible lambda phage carrying mutations in one or more of the regulatory genes, N, cro and cII. Although N- mutations are widely known for preventing efficient integration, and both N- and cII mutations for preventing efficient establishment of immunity, it is shown that, as predicted by a simple model, both N- and cII- phage efficiently lysogenize at low temperature if they are in addition cro-. In contrast with lambda N- cro+, lambda N- cro- is not propagated as a plasmid at low temperature, precisely because it establishes immunity too efficiently. Genetic control circuits are described in terms of sets of logic equations, which relate the state of expression of genes or of chemical reactions (functions) to input (genetic and environmental) variables and to the presence of gene and reaction products (internal, or memorization varibles). From the set of equations, one derives a matrix which shows the stable stationary states (if any) of the system, and from which one can derive the pathways (temporal sequences of states) consistent with the model. This kind of analysis is complementary to the more widely used analysis based on differential equations; it allows one to analyze in less detail more complex systems. The language might be used as well, mutatis mutandis, in fields very different from genetics. The last part of the discussion deals with the role of positive feedback loops in our specific problem (establishment and maintenance of immunity in temperate bacteriophages) and in developmental genetics in general. As a generalization of an old idea, it is suggested that cell determination (for a given character) depends on a set of genes whose interaction constitutes a positive feedback loop. Such a system has two stable stationary states: which one is chosen will usually depend on additional controls grafted on the loop.

LambdacI mutants: intragenic complementation and complementation with a cI promoter mutant

Complementation for the maintenance of lysogeny was studied by superinfecting lambdacIts lysogens at 34 degrees C. and then heating to 43 degrees C. With certain exceptions, ts mutants with defects in the left half of the repressor complemented ts mutants with defects in the right half to produce a less heat-labile repressor (Fig. 3). All cI amber mutants failed to complement cIts mutants. The cI mutant c50 complements all ts mutants. Mutations in Pre (cy) or genes cII and cIII do not significantly affect the expression of cI by a superinfecting lambda genome in an immune lysogen. Mutants with very heat-labile repressors failed to complement lambdacy42 for the establishment of lysogeny at elevated temperatures, while those with less heat-sensitive repressors apparently did complement cy. According to a suggested model, the left side of the cI product is concerned primarily with subunit aggregation, while operator binding is the function of the right side of the oligomer.