The lambda - P22 problem

Abstract

Lambda and P22 are members of 2 families of tailed phages and have limited genomic relationships. Both form hybrids with many phages. P22 appears as a hybrid of mixed ancestry. Despite their similarities, lambda and P22 and their relatives form 2 distinct lineages and must be classified separately.

Siphovirus lambda and podovirus P22 have been said to be close relatives because they form hybrids. Both λ and P22 have genomic maps with similar gene order. They and a host of similar phages have been called “lambdoid,” a poor and catch-all term for a wide selection of enterobacteria phages with noncontractile tails.¹ The situation was summarized in 2005² and needs to be reconsidered as Grose and Casjens recently classified tailed enterobacteriophages into clusters.³

Botstein and Herskovitz (1974) created viable hybrids of λ and P22, in which the immunity, early control and replication genes are substitued for the analogous λ genes.⁴ Susskind and Botstein (1978) noted a similar genome organization and lysogeny regulation genes in λ and P22.⁵ Despite many differences, they considered P22 as a member of the “λ family.” Indeed, many phage biologists perceive λ and P22 as closely related.¹ This can be used by some as an argument against the ICTV phage classification;¹ however, the basis of this argument is thin.

1. The exchanged gene region is small and limited to the immC region.^1,5 The λ and P22 genomes have only 13.5% total homology.⁶ Similarities are limited to a few genes (cI, c2, c3, ead, nin, ral, 13, 23), most of which have a very high (up to 96%) identity with λ genes.⁷ Similar gene order (synteny) is common in tailed phages and not a proof of close relationships.

2. Hybridization is a basic feature in biology. In the laboratory, both λ and P22 form hybrids with each other and various, often morphologically unrelated enterobacteria or Bacillus phages (λ with Mu, P1, P4, T4, SPP1; P22 with P1, Fels 1, Fels2, and P221).² In addition, there exist hybrids between myovirus coliphages Mu and P1. It does not matter if some of these hybrids result from illegitimate recombination. In famed Pseudomonas phage ϕKZ, we find genes occurring in 22 bacterial species and 6 phages, plus some protozoa, nematods, insects, and eukaryote viruses.⁸ Worse, even humans carry some T4 genes, e.g., the T4 lysozyme. This does not invalidate the independent nature of phage ϕKZ and humans.

3. Therefore, P22 appears as a hybrid of mixed ancestry, which may join with λ to form more hybrids. The distinctiveness of λ and P22 is indeed recognized by Grose and Casjens when they classify λ and P22 into different subclusters within a λ supercluster.³ Thus, if λ and P2 are related, this is not a close relationship. Rather, it can be said that they became related by “marriage” of unrelated ancestors and fall presently into 2 distinct clades.

4. We should consider that nothing in biology is absolute, that “bridging groups” exist everywhere, and that tailed phages are all related. This is a case of “much ado about nothing.” In conclusion, λ and P22-like phages should be treated and classified as separate entities which may be called “genera.”

5. Similar problems exist in other phages. Namely, Forstova et al.⁹ described in 1982 viable hybrids of Bacillus licheniformis phages LP52 (isometric head) and θ (theta, elongated head). Both are siphoviruses. The DNAs in both phages share 62% DNA homology. This seems to be a case of extensive hybridization and certainly not a good reason against ICTV phage classifieation.

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.