Dale_Molecular Genetics of Bacteria 4th ed

(a) Integration

λ

att P

Chromosome

(b) Excision

Prophage

Int + Xis

λ

attP

att B

Figure 4.11 Integration and excision of l DNA. (a) Integration occurs by site-specific recombination, mediated by the Int protein. (b) Excision requires both Int and Xis

interacts with Int to modify its specificity so that it can carry out excision leading to the formation of an intact chromosome and a circular phage molecule.

It should be emphasized that integration is not an essential feature of lysogeny. Some lambda mutants that are unable to integrate can still form lysogens, in which the prophage is maintained as an extrachromosomal circular DNA molecule – in effect it is a plasmid. Such lambda lysogens are normally unstable, but some temperate phages normally form stable lysogens in which the prophage is a plasmid. The bacteriophage P1 (which we will encounter again later on as a transducing phage and as a cloning vector) is a good example.

Phage

attP

TTCAGCTTTTTTATACTAAGTTG

AAGTCGAAAAAATATGATTCAAC

GCCTGCTTTTTTATACTAACTTG

CGGACGAAAAAATATGATTGAAC

attB

Chromosome

(b) After integration

Chromosome

attL

GCCTGCTTTTTTATACTAAGTTG

CGGACGAAAAAATATGATTCAAC

Prophage

TTCAGCTTTTTTATACTAACTTG

AAGTCGAAAAAATATGATTGAAC

attR

Figure 4.12 Structure of the attachment sites of l. (a) Integration requires site-specific recombination between attP (on the phage) and attB (on the chromosome). (b) After integration, the sites at either end of the prophage (attL and attR) have the same core sequence (shaded) as attP and attB but different flanking sequences

4.3.3Lytic and lysogenic regulation of bacteriophage lambda

Regulation of bacteriophage is worth considering in more detail as it provides well understood examples of different regulatory mechanisms. In particular, we can look at the temporal control of the lytic cycle (i.e. how the sequential expression of different sets of genes is achieved) and the control of lysogeny. The latter

genes, concerned mainly with replication, are nearer to the control region, from where the major early transcripts originate. Later in infection, we find expression of the late genes responsible for synthesis of the head and tail proteins and production of the mature phage particle. An expanded representation of the control region is shown in Figure 4.14.

Soon after infection, transcription is initiated at two major promoters which are known as the major leftwards and rightwards promoters, PL and PR . These promoters are recognized very efficiently by E. coli RNA polymerase. Rightwards transcription from PR however ends at the transcriptional terminators marked tR1 and tR2 (tR1 is a weak terminator and some of the transcripts ignore it); the leftwards transcript ends at tL1. Thus only a comparatively small number of genes are expressed immediately, amongst which are the two genes N and cro.

The product of gene N has specific anti-terminator activity, i.e. it allows transcription to proceed through the tR1, tR2 and tL1 terminators. The N protein binds to specific sites on the DNA upstream from the terminators and interacts with the RNA polymerase to allow it to ignore these transcriptional termination sites. The extended transcripts include a further set of genes (the ‘delayed early’ genes). The rightwards transcript then extends as far as a strong terminator tR3 that is unaffected by the action of the N protein. The delayed early genes expressed from the rightwards transcript are involved in DNA replication; those from the leftwards transcript, although not essential for lytic growth, do include genes such as red and gam that make phage production more efficient. (In order to keep things reasonably simple, we will ignore anything else that happens to the leftwards transcript).

Anti-terminator

Late transcript

Anti-terminator

Figure 4.14 Regulation of the major lytic transcripts of bacteriophage lambda. Early transcription initiated at PL and PR terminates at tL1 and tR1=tR2. The product of gene N allows anti-termination and expression of delayed early genes, with the rightwards transcript terminating at tR3. The gene Q product is also an anti-terminator and allows expression of the late genes from PR0

further phage transcription can be prevented before it has gone past the point of no return. This is achieved by the use of PE, which is a high level promoter. On the other hand, once lysogeny is established, continued synthesis of repressor is required at only a comparatively low level, which occurs by the use of PM. The lysogenic decision is thus self-reinforcing, with the presence of the CI repressor on the OR site preventing transcription of other genes but also ensuring continued synthesis of the repressor.

If however the repressor is removed from the OR site (which occurs sometimes spontaneously and also as a consequence of various stresses that cause induction of the prophage), this not only relieves the repression of PR, but also switches off PM, thus preventing further synthesis of repressor to replace that which has been destroyed. This is therefore also a self-reinforcing switch mechanism.

Mechanism of binding of repressor and Cro proteins to the operator site

In the above description, the operator site OR was treated as though it were a simple DNA sequence to which either the repressor or Cro could bind. The nature of this operator site and of the binding of these proteins is worth considering further because of the information this system has supplied regarding the binding to DNA of regulatory proteins in general. Only the OR site will be considered, but there is considerable similarity in the binding to the OL site.

The OR operator in fact consists of three similar adjacent regions of DNA, OR1, OR2 and OR3 (Figure 4.17). The rightwards promoter PR overlaps with OR1 and the maintenance promoter PM overlaps with OR3. Both the CI repressor and the Cro protein contain a characteristic structure with a helix-turn-helix motif that is typical of many regulatory proteins (see Chapter 3). This is capable of interacting with each of these three sites although with different affinities and different consequences.

The affinity of the repressor is greatest for OR1, but binding is cooperative; the repressor bound to OR1 stimulates binding of a second molecule to OR2. Binding of RNA polymerase to PR is thus prevented, while stimulation of transcription from PM occurs by protein–protein interactions between the repressor bound to OR2 and the RNA polymerase. In the normal lysogenic state, OR3 is unoccupied, but very high concentrations of repressor can result in binding to OR3 as well. This will switch off PM until the level of repressor falls below that needed to saturate OR3. Thus not only is the expression of all the genes other than the repressor, turned off, but this also ensures that expression of cI is maintained at a low level which is an advantage for the stable maintenance of the phage as there is very little burden on the cell.

Although Cro can bind to all three sites, its relative affinity for these sites is different from that of the repressor: it will bind first of all to OR3. This prevents RNA polymerase binding to the maintenance promoter PM, thus preventing the

Figure 4.17 Interaction of Cro and CI with the operator site OR

phage entering the lysogenic state. Only subsequently, as the level of Cro builds up, will it bind to sites 1 and 2; by this time the phage is committed to the lytic mode and sufficient of the transcripts from PL have been produced. The role of Cro now is to turn off the expression of the early genes that are no longer needed and thus to direct gene expression towards the late genes that are needed for production of phage particles.

In the light of this knowledge, we can refine the model for the control of lambda and the lytic–lysogenic decision. The key events are the binding of the CI repressor to OR1, which will prevent the lytic cycle, versus the binding of Cro to OR3, which will prevent the establishment of lysogeny by repressing PM. The binding of Cro to the operator site does not prevent synthesis of repressor from PE, as long as active CII protein remains. Mutations that increase the stability of CII and host mutations (hfl, for high frequency of lysogenization) that reduce the activity of proteolytic enzymes (and hence prolong the activity of CII), will therefore result in an increased frequency of lysogenization.

Superinfection immunity and zygotic induction

A lysogenic cell is resistant to infection by further particles (or related phages). This is called superinfection immunity. The reason for this should now be clear:

such a cell already contains pre-formed repressor protein which is therefore able to bind immediately to the OL and OR sites on the incoming DNA. Consequently, the incoming phage DNA will be repressed before transcription can even start.

The converse phenomenon is seen when a lysogen is used as a donor to transfer DNA to another cell by conjugation (see Chapter 6). At a certain time after the start of conjugation, the prophage (which is inserted into the chromosome) is transferred just like any other part of the chromosome. The recipient cell however, if it is not lysogenic, contains no repressor. The transferred prophage is therefore able to escape from repression and initiate a lytic cycle in the recipient, leading to a sudden fall in the number of viable recipients. This phenomenon is known as zygotic induction.

4.4 Restriction and modification

As described earlier in this chapter, the number of bacteriophages in a preparation is usually assayed by counting the number of phage plaques produced using a sensitive bacterium as an indicator. The assumption behind this procedure is that there is a direct correspondence between the number of phage particles and the number of plaques, i.e. that every phage particle gives rise to a lytic infection. In technical terms, it is said that the efficiency of plating (e.o.p) is 1. Sometimes that is not true and the e.o.p. is much less than 1. This could happen when the indicator strain is different from the one used to grow the phage. This may of course mean that the new indicator is not sensitive to the bacteriophage. The few plaques obtained may be due to spontaneous mutation in the bacteriophage, yielding host range mutants that are now able to infect the previously resistant organism.

However, a reduction in the number of phage plaques may be due to a different phenomenon known as host-controlled restriction and modification. Consider the situation shown in Figure 4.18. A suspension of bacteriophage particles has been obtained by growth on a host E. coli strain known as E. coli C; these are referred to as .C. If this preparation is assayed using E. coli C and a different strain (E. coli K) as the indicator organisms, for every 10 000 plaques obtained on E. coli C (which is assumed to have an e.o.p. of 1), there is only one plaque on E. coli K (e.o.p. ¼ 10 4). If a plaque from the E. coli K plate is picked by stabbing it with a toothpick and resuspending the material recovered in a small amount of buffer, the resulting phage preparation (now referred to as .K) has an e.o.p of 1 using either strain as the indicator. All the phage are now capable of growing on both strains.

This might be taken to indicate that a mutant phage which was altered in its host range had been selected, i.e. the original population contained a small number of mutants that were able to grow on strain K. However, if these apparent mutants are used to infect strain C again, the phage will again show its original

λ.C

Test for plaque formation

on two indicator strains

Restricted

E.coli C E.coli K growth

e.o.p. = 10−4

e.o.p. = 1

λ.K

Figure 4.18 Host-controlled restriction and modification of bacteriophages

e.o.p. of 10 4 on strain K. It behaves in the same way as the original .C population. It is highly unlikely that this behaviour could be caused by mutation in the normal sense of the word.

All this becomes clearer once the true explanation is known. Strain K produces an enzyme (endonuclease) that is capable of recognizing foreign DNA and degrading it. Such an enzyme is referred to as a restriction endonuclease, since it results in a restriction of the growth of the foreign bacteriophage. There are a large number of such restriction enzymes now known. Most of those commonly encountered are Type II restriction endonucleases which recognize specific DNA sequences and cut the DNA strands at that point (see Box 4.1). They have a