Recognition and cleavage of the bacteriophage P1 packaging site ( pac) : I. Differential processing of the cleaved ends in vivo

Abstract

The packaging of bacteriophage P1 DNA into viral capsids is initiated at a specific DNA site called pac. During packaging, that site is cleaved and at least one of the resulting ends is encapsidated into a P1 virion. We show here that pac is located on a 620 base-pair fragment of P1 DNA ( EcoRI-20). When that fragment is inserted into the chromosome of cells that are then infected with P1, packaging of host DNA into phage particles is initiated at pac and proceeds down the chromosome, unidirectionally, for about five to ten P1 “headfuls” (about 5 × 10 5 to 10 × 10 5 bases of DNA). Using an assay for pac cleavage that does not depend on DNA packaging, we have identified a set of five amber mutations that are mapped adjacent to pac, and that define a gene (gene 9) essential for pac cleavage. Amber mutations that are located in genes necessary for viral capsid formation (genes 4, 8 and 23), or in a gene necessary for “late” protein synthesis (gene 10), do not affect pac cleavage. The latter result suggests that the synthesis of the pac cleavage protein is not regulated co-ordinately with other phage morphogenesis proteins. The products of pac cleavage were analyzed using two different DNA substrates. In one case, a single copy of pac was placed in the chromosome of P1-sensitive cells. When those cells were infected with P1, we could detect the cleavage of as much as 70% of the pac-containing DNA. The pac end destined to be packaged in the virion was detected five to 20 times more efficiently than was the other end. Since this result is obtained whether or not the infecting P1 phage can encapsidate the cut pac site, the differential detection of pac ends is not simply a consequence of one end being packaged and the other not. In a second case, pac was located in cells on a small (5 × 10 3 bases) multicopy plasmid. When those cells were infected with P1, neither pac end was detected efficiently after P1 infection, unless the cells carried a recBCD − mutation. In recBCD − cells, the results with plasmid- pac substrates were similar to those obtained with chromosomally integrated pac substrates. We interpret these results to mean that, following pac cleavage, the end destined to be packaged is protected from cellular nucleases while the other end is degraded by the action of at least two nucleases, one of which is the product of the host recBCD gene. If the two ends are within several thousand base-pairs, then degradation by the RecBCD nuclease proceeds rapidly through the entire molecule and neither end is detected efficiently. If the two ends are separated by an Escherichia coli chromosome, then degradation by the RecBCD nuclease from the unprotected end does not affect recovery of the protected end. In the absence of the RecBCD nuclease, a second nuclease degrades DNA from the unprotected end. That degradation, however, rarely proceeds far enough to affect recovery of the protected pac end.