• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of pnasPNASInfo for AuthorsSubscriptionsAboutThis Article
Proc Natl Acad Sci U S A. Jul 27, 2004; 101(30): 11007–11012.
Published online Jul 19, 2004. doi:  10.1073/pnas.0401478101
PMCID: PMC503734

Genetic organization of the psbAD region in phages infecting marine Synechococcus strains


The discovery of the genes psbA and psbD, encoding the D1 and D2 core components of the photosynthetic reaction center PSII (photosystem II), in the genome of the bacteriophage S-PM2 (a cyanomyovirus) that infects marine cyanobacteria begs the question as to how these genes were acquired. In an attempt to answer this question, it was established that the occurrence of the genes is widespread among marine cyanomyovirus isolates and may even extend to podoviruses. The phage psbA genes fall into a clade that includes the psbA genes from their potential Synechococcus and Prochlorococcus hosts, and thus, this phylogenetic analysis provides evidence to support the idea of the acquisition of these genes by horizontal gene transfer from their cyanobacterial hosts. However, the phage psbA genes form distinct subclades within this lineage, which suggests that their acquisition was not very recent. The psbA genes of two phages contain identical 212-bp insertions that exhibit all of the canonical structural features of a group I self-splicing intron. The different patterns of genetic organization of the psbAD region are consistent with the idea that the psbA and psbD genes were acquired more than once by cyanomyoviruses and that their horizontal transfer between phages via a common phage gene pool, as part of mobile genetic modules, may be a continuing process. In addition, genes were discovered encoding a high-light inducible protein and a putative key enzyme of dark metabolism, transaldolase, extending the areas of host-cell metabolism that may be affected by phage infection.

Strains of unicellular cyanobacteria Synechococcus and Prochlorococcus are abundant in the world's oceans, and they dominate the prokaryotic component of the picophytoplankton. Together, they contribute 32–89% of primary production in oligotrophic regions of the oceans (14). The recent discovery (5) that a phage infecting marine Synechococcus strains encodes key photosynthetic genes has important implications for our understanding of the effect of phage infection on the photosynthetic picophytoplankton physiology and consequent impact on major biogeochemical cycles. The acquisition of photosynthesis genes by a phage, presumably by horizontal gene transfer, begs the questions of (i) how common the possession of photosynthesis genes by such phages is, (ii) where the acquired genes were from, and (iii) whether the acquisition was a single rare ancestral event or a common phenomenon in the oceans.

The cyanobacterial picoplankton, together with all organisms capable of oxygenic photosynthesis, possess two photosynthetic reaction centers, PSI and PSII (photosystems I and II). The PSII complex is a large protein–pigment assembly in the thylakoid membrane that catalyses the light-dependent oxidation of water to molecular oxygen. At the core of PSII lies a heterodimer of two related proteins, D1 and D2, which binds the pigments and cofactors necessary for primary photochemistry. During active photosynthesis, D1 and, to a lesser extent, D2 turn over rapidly as a result of photodamage and are replaced by newly synthesized polypeptides in a repair cycle. When the rate of photoinactivation and damage of D1 exceeds the capacity for repair, photo-inhibition occurs, resulting in a decrease in the maximum efficiency of PSII photochemistry (6). Environmental stresses, such as UV-B, may enhance the rate of damage such that photoinhibition may occur at light fluence rates that would not normally exert this effect (7). Fluence rates of both photosynthetically active radiation and UV light in oceanic ecosystems may be high enough, particularly in surface waters, to cause photoinhibition (8).

Viruses in general, and bacteriophages in particular, have been shown to be abundant in marine ecosystems and are thought to exert major biogeochemical and ecological effects (9). Viral infection of marine unicellular cyanobacteria was first reported in 1990 (10, 11), and isolates of these cyanoviruses were first characterized in 1993 (1214). Recently, it was shown that one of these phages, S-PM2, which is capable of infecting Synechococcus strains, carries copies of the psbA and psbD genes encoding the D1 and D2 proteins of PSII (5). It is speculated that the expression of phage-encoded D1 and D2 proteins in infected cells would permit a continued PSII repair cycle to operate after host protein synthesis had been shut down, thus maintaining the photosynthetic activity of the cells and concomitant oxygen evolution and ensuring the provision of energy for extended viral replication. The D1 protein encoded by S-PM2 is similar to the D1 proteins of marine Synechococcus sp. WH8102, and indeed, homology can be detected at the DNA sequence level, suggesting that S-PM2 acquired the gene horizontally from its Synechococcus host. By establishing whether the presence of psbA genes in cyanophage genomes is a widespread phenomenon and, if so, whether gene organization is similar in geographically distinct isolates, it becomes possible to establish whether the acquisition was a single rare ancestral event or is a common phenomenon in the oceans.

Materials and Methods

Isolation, Propagation, and Maintenance of Bacteriophage Strains. Phages were isolated and propagated by using Synechococcus sp. WH7803, grown in artificial sea water, as described by Wilson et al. (15). The isolation details of the phages characterized in detail are shown in Table 1, and the same information for phages screened only for the presence of psbA genes is provided in Table 2, which is published as supporting information on the PNAS web site.

Table 1.
Phages used in this study

PCRs, Southern Blotting, and DNA Sequencing. DNA from cyanophages was extracted by using the method described by Wilson et al. (14). Homologs of psbA in other cyanophages were initially detected by Southern blotting. A hybridization probe was prepared from a 525-bp PCR product from S-PM2, which was amplified by using the following primers: S-PM2F1S, 5′-GCTGCTTCTCTTGATGAGTG-3′; and S-PM2R2S, 5′-AGTGTAGCGAACGAGAGTTG-3′. The PCRs were carried out in a total volume of 50 μl, containing 200 μM dNTPs, 2 mM MgCl2, 50 nM primers, 2 μl of S-PM2 DNA, 2 units of Taq polymerase, and 1× enzyme buffer (Helena Biosciences). Amplification conditions were as follows: 92°C for 2 min, 30 cycles of 92°C for 30 sec, 47°C for 30 sec, 72°C for 30 sec, with a final extension of 4 min at 72°C. The PCR product was gel-extracted and labeled with [α-32P]ATP by using DNA polymerase I Klenow fragment in 1× labeling buffer (Promega) at 25°C for 2 h. Unincorporated nucleotides were removed by purification through a Sephadex G-25 column (16). The labeled S-PM2 psbA PCR product was used as a hybridization probe against cyanophage DNA that had been digested with EcoRI and run on a 0.75% agarose gel. Transfer to a nylon membrane (Amersham Biosciences) was achieved by the method described by Sambrook et al. (16). Blots were prehybridized in buffer (5× Denhardt's solution/6× 0.15 M NaCl/10 mM phosphate, pH 7.4/1 mM EDTA (SSPE)/0.5% SDS (wt/vol) for 60 min. Hybridization was at 65°C overnight in fresh buffer (5× Denhardt's solution/6× SSPE/0.5% SDS, wt/vol). Membranes were washed three times for 10 min in 2× SSC/0.1% (wt/vol) SDS at 63°C; twice in 1× SSC/0.1% (wt/vol) SDS at 63°C for 10 min; and four times in 0.1× SSC/0.1% (wt/vol) SDS at 63°C for 5 min.

The primers for used screening and sequencing the psbA regions of cyanophage genomes were based on those of Zeidner et al. (17). The reverse primer designed in this study spanned the intron-insertion site in the psbA gene of cyanophage S-PM2 and was shortened to omit the region corresponding to the first exon. Therefore, primers could amplify psbA genes whether or not they contained an intron. The following primers were used: psbAF, 5′-GTNGAYATHGAYGGNATHMGNGARCC-3′; and psbAR (2), 5′-GGRAARTTRTGNGC-3′. PCRs were carried in a total volume of 25 μl, containing 200 μM dNTP, 2 mM MgCl2/50 nM primers, 2 μl of template, 1.5 units of Taq polymerase, and 1× enzyme buffer. Amplification conditions were as follows: 92°C for 2 min, 10 cycles of 92°C for 30 sec, 64°C (–1°C per cycle) for 30 sec, and 68°C for 1 min. There was then an extension of 2 min at 68°C, followed by 25 cycles of 92°C for 30 sec, 56.5°C for 30 sec, and 68°C for 1 min. The final extension was at 68°C for 4 min. Cyanophage DNA was screened for the presence of contaminating Synechococcus DNA by the use of primers specific to 16S ribosomal DNA from oxygenic phototrophs. The following primers were used: CYA106F, 5′-CGGACGGGTGAGTAACGCGTG-3′ (18); CYA781R(A), 5′-GACTACTGGGGTATCTAATCCCATT-3′ (19); and OXY1313R(B), 5′-GACTACAGGGGTATCTA ATCCCTTT-3′ (18). The conditions used were the same as those used by Nubel et al. (19).

It was not possible to amplify the entire psbA and psbD gene regions of cyanophages by PCR directly because the N-terminal end of psbA is not sufficiently conserved to allow the design of degenerate primers. Therefore, to sequence this region and to identify genes upstream from the psbA and psbD regions, primer walking was used. The genomic DNA required for direct sequencing was produced by using the GenomiPhi DNA amplification kit (Amersham Biosciences) according to the manufacturer's instructions, yielding 4–7 μg of DNA from 15 ng of starting material. For the sequencing reaction, 1 μg of genomic DNA was added to 5 pmol of primer, and the total volume was made up to 6 μl with sterile water. Sequence data were imported into seqman (DNASTAR, Madison, WI). Primers specific to the new sequence were designed by using primer designer 3.0 (Scientific and Educational Software, Durham, NC), and the process was repeated until the sequence information for the genes upstream and downstream from the psbA and psbD regions had been obtained. Primers were synthesized commercially by TAG Newcastle (Newcastle, U.K.).

DNA Sequence and Phylogenetic Analysis. Analysis of DNA sequences was carried by the identification of ORFs by using orf finder (available at www.ncbi.nlm.nih.gov/gorf/gorf.html) and subsequent blast searching (20). The subcellular localization of proteins was predicted by using psort-b (21). Sequences were aligned by using clustalw (22), and the output was checked and corrected manually.

Phylogenetic trees were constructed by using paup* 4.0 (23) and mrbayes (24). In the case of mrbayes, trees were determined by using 500,000 iterations with a sample frequency of 100 with a burn-in of 50. All analyses in paup* were performed by using branch-and-bound searches, with the “collapse” option and “furthest” addition sequence selected. For nucleotide alignments, any gaps in the data matrix were treated as missing data, and indels (insertions or deletions) were coded separately and appended to the sequence data matrix. Coding of indels was binary (deletions, 0; insertions, 1). Data were analyzed by using parsimony and distance methods in paup* 4.0, with no weighting or ordering imposed on the characters. Support for clades was estimated by means of nonparametric bootstrap analyses, as implemented in paup* 4.0 by using 1,000 replicates.


Bacteriophages capable of infecting marine phycoerythrin-containing Synechococcus strains were first characterized in 1993, and most isolates were myoviruses with icosahedral heads and contractile tails (1214). The myovirus S-PM2, which was originally isolated from sea water from the English Channel by using Synechococcus sp. WH7803 as a host, has been shown to encode the psbA and psbD genes specifying the D1 and D2 proteins of PSII (5). To determine whether the acquisition of the psbA and psbD genes by myoviruses infecting marine Synechococcus strains was a rare or common event, it was decided to establish whether the psbA gene could be detected in other marine cyanophage isolates and, if so, to compare the genetic organization of the psbA region in geographically distinct isolates and multiple isolates from the same location.

The PCR product from phage S-PM2 was used initially as a hybridization probe in Southern blots of phage DNAs, including S-WHM1, S-BM4, and S-RSM2, which have geographically distinct isolation sites and gave positive signals for the presence of a psbA gene (data not shown), which was subsequently confirmed by DNA sequencing. A large number of myovirus isolates from the Red Sea were then screened for the occurrence of a psbA gene by using the degenerate PCR primers (data not shown). In total, the combination of Southern blotting and PCR screening indicated that 37 of 68 isolates (all myoviruses) carried a copy of the psbA gene, suggesting that the possession of psbA genes by marine cyanomyoviruses is a common phenomenon. Two phage isolates, in addition to S-PM2, yielded a larger PCR product, indicative of the presence of the putative intron. In the case of S-RSM88, this intron was confirmed by DNA sequencing. Five of the strains giving a positive signal for the presence of the psbA gene were selected for more detailed analysis. The S-RSM2, S-RSM28, and S-RSM88 strains are isolates obtained from the Gulf of Aqaba in the Red Sea, whereas S-BM4 and S-WHM1 were obtained from Bermuda and Woods Hole Harbor, respectively.

To investigate the genetic organization of the phages in the region of the psbA gene, the regions upstream and downstream were sequenced. Initially, the PCR products obtained by using the psbA primers were sequenced and the sequence was extended from the original PCR primers. Primers were then designed to sequence walk in either direction from psbA and to make the sequence double-stranded. All of the phages studied carried the psbD gene in addition to psbA. The genetic organization of the psbAD region of each of the phages is shown in Fig. 1. S-RSM88 exhibits an identical organization to S-PM2 with psbA and psbD being separated by two ORFs. Indeed, with the exception of a single-base-pair substitution in ORF 178, the DNA sequences of S-PM2 and S-RSM88 are identical in the 3,555 bp from the beginning of psbA to the end of psbD. This similarity extends to the presence of a group I intron in identical positions in the two psbA genes, inserted between codons 334 and 335. Although comprising only 212 nucleotides, the intron can be folded into a secondary structure containing all of the canonical group I structural features (Fig. 2) (25, 26). The similarity between S-PM2 and S-RSM88 also extends upstream from psbA. Thus, it would appear that the psbA and psbD genes in these two phages might occur as part of a conserved module that, given the considerable geographical separation of their isolation sites, is mobile.

Fig. 1.
Genetic organization of the psbAD region in the following five cyanomyoviruses infecting Synechococcus sp. WH7803: S-RSM2, S-BM4, S-WHM1, S-PM2, and S-RSM88 (GenBank accession nos. ...
Fig. 2.
Secondary structure of the group I intron in psbA. Exon sequences are given in lowercase letters, and intron sequences are given in uppercase letters. Arrows indicate splice sites (ss). Bold lines show connections between intron structure domains, with ...

Analysis of the genetic organization of the psbA region of phages S-RSM2, S-BM4, and S-WHM1, however, gives contrasting results (Fig. 1). In all these phages, the psbA and psbD genes are immediately adjacent to each other, but the psbA and psbD genes differ in nucleotide sequence from each other. Each phage is quite distinct in the ORFs adjacent to psbAD. In the case of S-RSM2, upstream from psbA and on the opposite strand are two genes, one of which is predicted by psort (21) to have the characteristics of membrane protein. Downstream of psbD is an ORF encoding a putative transaldolase, complete with MipB domain. However, the closest scoring match is to a transaldolase from Caulobacter crescentus (E value, 8e–51) and not to cyanobacterial transaldolases. It is worth noting that transaldolase is a key enzyme of the oxidative pentose phosphate pathway, which is the primary route for dark oxidative metabolism in cyanobacteria (27). The psbAD region in S-WHM1 is bounded by two ORFs encoding polypeptides with no significant similarities to any entries in the protein databases. The polypeptide encoded by the ORF adjacent to psbA has a predicted secretion signal sequence. Upstream of the psbA gene in phage S-BM4 is an ORF encoding a polypeptide with similarity to a high-light inducible protein (HLIP). HLIP genes are found in both marine Synechococcus and Prochlorococcus strains, and on the basis of cluster analysis, some of the hli genes may be specific to marine cyanobacteria (28). In the case of the freshwater cyanobacterium Synechocystis sp. PCC6803, HLIPs have been implicated in the adaptation to variations in light intensity (29).

Phylogenetic analysis of the psbA and psbD genes was carried out by the following three different methods: the maximum parsimony and distance methods of paup* and mrbayes, which employs a Bayesian approach. All three methods gave essentially the same results, and only results obtained with mrbayes are presented here. A phylogenetic analysis of the phage-encoded psbA genes and other cyanobacterial psbA genes at the amino acid level gives no resolution of taxa (tree not shown). However, analysis at the nucleotide level (Fig. 3A) shows that there is a major clade with 100% support, which contains all of the phage sequences and the sequences from marine Synechococcus and Prochlorococcus strains. The psbA sequences from the freshwater cyanobacteria Nostoc and Synechocystis form an outgroup. Within the marine clade, the phage psbA genes form several distinct groups. The first group contains the identical S-RSM88 and S-PM2, with a clade support value of 100%. Sister to this group, with 100% clade support, is the psbA gene from Prochlorococcus MED4. A further distinct phage group supported with a clade support value of 100% contains the S-RSM28 and S-WHM1 psbA genes. Sister to this group, but with low clade support, is the psbA gene from a putative podophage (BAC9D04) (30), identified as such by the similarity of genes adjacent to psbA to genes in the marine cyanopodovirus P60 (data not shown), although this phage does not possess a copy of psbA (31). The phage S-RSM2 occupies a group with 100% clade support with eBAC65-3, which encodes a psbA gene from an environmental bacterial artificial chromosome (BAC) library. However, the BAC library was prepared in such way that phages would not be excluded (Oded Béja, personal communication), and thus, the origin of the gene is unclear. The final phage involved in this study, S-BM4, represents a sister group of S-RSM28 and S-WHM1. The two Synechococcus strains form a group with the putative Synechococcus psbA eBAC65-10.

Fig. 3.
Phylogenetic analysis using mrbayes of the psbA genes (A) and psbD genes (B) from the following six marine cyanomyoviruses: S-RSM2, S-BM4, S-WHM1, S-PM2, S-RSM28, and S-RSM88 (GenBank accession nos. ...

A phylogenetic analysis of psbD genes from the same sources (with the exception of those from BAC libraries) shows an only slightly different picture of the relationships from groupings obtained with psbA. The first thing to note is that there is again a major clade with clade support of 100% that contains all of the phage, Synechococcus, and Prochlorococcus psbD genes. There is also good support (78%) both for a clade containing Synechococcus and the phages and for a subgroup containing all of the phages (100%). However, the S-PM2 and S-RSM88 phages no longer form a sister group with Prochlorococcus MED4 as they do for psbA. Within the phage, clades S-BM4, S-RSM2, S-WHM1, and S-RSM28 form a group with 100% support, and within this group, the two phages S-WHM1 and S-RSM28 form a well supported (93%) subgroup as they did for the psbA analysis.


The phage psbA genes fall into a clade that includes the psbA genes from their potential Synechococcus and Prochlorococcus hosts, supporting the idea that these genes were indeed acquired horizontally from their hosts and this is in agreement with the results obtained by Lindell et al. (32) for Prochlorococcus phages. A similar observation was made for psbD. However, the phage psbA genes form distinct subclades within this lineage, which suggests that their original acquisition was not very recent or that there is a very strong selection pressure on the psbA gene in the phage as compared with the organismal context. Another possibility, albeit less likely only because of a number of Synechococcus psbA genes that have now been sequenced, is that representative psbA genes from the actual cyanobacterial sources have not been characterized. The psbD genes show much better resolution than the psbA genes, which may suggest that psbD has a more ancient ancestry within the phage lineage, leading to two well defined radiations. An alternative explanation might be that there are tighter functional constraints on the evolution of psbA than there are on the evolution of psbD.

The different patterns of genetic organization of the psbAD region in the different phages suggest either that the psbA and psbD genes were acquired more than once by cyanomyoviruses or that horizontal transfer of these genes among phages through a common cyanomyoviral gene pool may be a common process. Indeed, these two options are not mutually exclusive. Thus, the original acquisition of the genes by phages may not have been recent, but when acquired, transfer of the genes among phages may be commonplace. An alternative explanation might be that the psbA and psbD genes were acquired in a single event and that subsequent horizontal transfer events have disrupted the linkage and might also have replaced the original alleles by homologous recombination. In this context, it is worth noting that it has been suggested that all of the double-stranded DNA-tailed phages might share common ancestry and that all double-stranded DNA phage genomes are mosaics with access, by horizontal exchange, to a large common genetic pool (33). It has been shown that phage S-PM2 shares a genetic module with the ecologically and evolutionarily distant coliphage T4 (34), and the fact that a putative podovirus (BAC9D04) (30) carries a copy of the psbA gene tends to confirm the breadth of this gene pool. The presence of psbA genes in cyanopodoviruses is confirmed by sequence analysis of a cyanopodovirus infecting Prochlorococcus strains (32). The difference in genetic organization between these phages, in which psbA and psbD are immediately adjacent to each other, and their hosts, in which these two genes are widely separated (3537), also suggests that the two genes may have been acquired independently.

Another observation that supports the idea of a dynamic gene pool, which includes photosynthesis genes, is the near identity of the psbAD region in two geographically distinct cyanomyovirus isolates, S-PM2 and S-RSM88. This result is consistent with the spread of a mobile “psbAD module” through the common phage gene pool and is supported by the fact that the two phages have quite different genome sizes (A. Millard, personal communication). Although no introns were observed in the Prochlorococcus phages studied by Lindell et al. (32), the psbA genes of these two phages form a single clade with Prochlorococcus MED4, raising the possibility that, although these two phages will infect Synechococcus strains, they may have ultimately acquired their psbA genes from Prochlorococcus.

The origin of the group I intron in the psbA gene of S-PM2 and S-RSM88 is not clear. Group I introns have been observed in many cyanobacteria but so far only in tRNA genes. Furthermore, these tRNA introns belong to subfamily IC3 and show no sign of relationship to the subfamily IA1 phage psbA introns (26). Interestingly, however, the psbA gene of Chlamydomonas moewusii chloroplasts contains two group I introns and the psbA gene of C. reinhardtii has four group I introns: intron 1 of C. moewusii and intron 4 of C. reinhardtii (both closely related to group IA1) sharing the same insertion site (38, 39). However, none of the five sites of group I intron insertion correspond to the placement of the intron in the phage psbA genes. Although capable of self-splicing in vitro, splicing of the introns in psbA of C. reinhardtii is stimulated by light in vivo (40), raising the interesting possibility that they participate in posttranscriptional regulation of gene expression.

Remarkably, some of the chloroplast introns are highly similar to introns in bacteriophages, including phage T4 (39), and the cyanophages described here bear a striking resemblance, both in morphology and in aspects of genome composition, to phage T4 and its relatives (34). Collectively, these correlations suggest a long-standing trafficking of introns between phages and psbA genes.

The high degree of sequence identity of the psbAD cassettes of S-PM2 and S-RSM88 suggests a fairly recent lateral transfer. Many group I introns contain ORFs that encode endonucleases with specificity for the intronless DNA of genes that are homologous to their sites of insertion. By cleaving these intronless genes, the introns become duplicatively inserted into vacant sites during DNA repair, which is a gene conversion event (with coconversion of flanking DNA) that has been called “homing” (41), and the enzymes that initiate this process are called “homing endonucleases.” Intron homing, with coconversion of flanking DNA, would provide a ready explanation for the near identity of the psbAD regions of S-PM2 and S-RSM88, except that the minimal phage introns described here do not contain any ORFs capable of encoding an endonuclease. How could these introns be mobile?

The answer may reside in the ORF178 immediately downstream of psbA in both phages, whose deduced protein product closely resembles another optional phage ORF, gene 13.5 of Yersinia phage ΦYe03–12 (42), and other phage proteins of unknown function, all of which show some similarity to a portion of phage T4 gp49 (endonuclease VII), which is involved in recombination and packaging of phage DNA. Although it is not related in sequence to any of the established homing endonucleases, the active center of gp49 has been reported to be structurally homologous to members of one family of homing endonucleases (43).

Interestingly, a homing endonuclease does not need to reside in an intron to promote homing. In some bacteriophages, genes whose products are related to intronic endonucleases are inserted intercistronically, between conserved genes that are adjacent in closely related phages (4446). Several of these homologs have been shown to have endonuclease activity, cleaving other phage DNAs close to the endonuclease gene-insertion site. Repair of cleaved DNA inserts the endonuclease gene by gene conversion, which is exactly analogous to intron homing. In this “intronless homing,” the endonuclease gene retains its function as a mobile element without being associated with an intron, and the coconversion tract of flanking DNA can extend several kilobases from the site of cleavage (47, 48). It is possible that the protein encoded by ORF178 next to the phage psbA gene functions in this way, by mobilizing the entire psbAD cassette for lateral transfer between related bacteriophages.

The idea that the acquisition by marine cyanophages of genes encoding components of the photosynthetic reaction center confers an advantage gains further support from the discovery of an HLIP-encoding gene, hli, in phage S-BM4. HLIPs have been implicated in the adaptation to variations in light intensity (29). Furthermore, multiple hli genes and other photosynthetic genes have been discovered in the genomes of phages infecting Prochlorococcus strains (32). It has been suggested (5) that the expression of virus-encoded D1 and D2 proteins in infected cells would permit a continued PSII repair cycle to operate after host protein synthesis had been shut down, thus maintaining the photosynthetic activity of the cells and concomitant oxygen evolution, as well as ensuring the provision of energy for extended viral replication. However, high light intensity leading to photoinhibition is not likely to be an important factor in the lower levels of the euphotic zone; therefore, any fitness benefits conferred on phage by carriage of the psbA gene are only likely to be significant in the surface layers of the oceans. These considerations would apply to the other photosynthesis genes. Cyanobacteria employ the oxidative pentose phosphate pathway as their source of maintenance energy in the dark (27), and the discovery of a gene encoding a putative transaldolase, a key enzyme of this pathway, suggests that some phages may be capable of influencing dark metabolism as well. Interestingly, it has been proposed (49) that lytic phages infecting marine Synechococcus strains may be capable of establishing a pseudolysogenic state in cells that are nutrient-stressed. It should not be assumed that these genes play a role solely in the infection–lysis pathway but that they also might be expressed on a longer-term basis during the quasistable relationship of pseudolysogeny established in response to the nutrient-limited conditions, which apply in the oligotrophic central regions of the oceans.

The patchiness of the distribution of the psbA gene in cyanophages poses a question regarding ideas about its contribution to phage fitness. The single most important selection pressure on the evolution of phages is the density of infectable hosts. This factor, in the long term, will determine the length of the latent period and also whether lysogenic or lytic phages are likely to predominate. The current hypothesis regarding the contribution of the psbA gene to phage fitness assumes that the latent period is long enough for photoinhibition to restrict phage replication in the absence of a repair cycle. There is no direct method, to our knowledge, of determining what proportion of a natural cyanobacterial assemblage is susceptible to infection by cooccurring phages. It is very likely that different ecotypes within the assemblage will be susceptible to infection by different subsets of phages. Thus, one cyanobacterial ecotype constituting a significantly higher proportion of the assemblage might be susceptible to infection by phages with a short latent period and no psbA gene, whereas another ecotype at lower abundance may be infected by phages with a longer latent period, which would benefit from the carriage of the psbA gene. The situation is further complicated by the possibility of pseudolysogeny. Furthermore, it is quite conceivable that the host range of some cyanophages may extend to members of the heterotrophic bacterial community where possession of psbA would confer no benefit.

Supplementary Material

Supporting Table:


This work was supported by a Natural Environment Research Council grant (to N.H.M.) and National Institutes of Health Grant AI 57158 (to D.A.S.). A.M. was supported by a Natural Environment Research Council Studentship.


This paper was submitted directly (Track II) to the PNAS office.

Abbreviations: HLIP, high-light inducible protein; BAC, bacterial artificial chromosome.

Data deposition: The sequences reported in this article have been deposited in the EMBL database (accession nos. AJ628768, AJ628769, AJ628858, AJ629075, AJ629221, and AJ630128).


1. Goericke, R. & Welschmeyer, N. A. (1993) Deep Sea Res. Part I Oceanogr. Res. Pap. 40, 2283–2294.
2. Li, W. K. W. (1995) Mar. Ecol. Prog. Ser. 122, 1–8.
3. Liu, H. B., Nolla, H. A. & Campbell, L. (1997) Aquat. Microb. Ecol. 12, 39–47.
4. Veldhuis, M. J. W., Kraay, G. W., VanBleijswijk, J. D. L. & Baars, M. A. (1997) Deep Sea Res. Part I Oceanogr. Res. Pap. 44, 425–449.
5. Mann, N. H., Cook, A., Millard, A., Bailey, S. & Clokie, M. (2003) Nature 424, 741–741. [PubMed]
6. Aro, E. M., Virgin, I. & Andersson, B. (1993) Biochim. Biophys. Acta 1143, 113–134. [PubMed]
7. Campbell, D., Eriksson, M. J., Oquist, G., Gustafsson, P. & Clarke, A. K. (1998) Proc. Natl. Acad. Sci. USA 95, 364–369. [PMC free article] [PubMed]
8. Piazena, H., Perez-Rodrigues, E., Hader, D. P. & Lopez-Figueroa, F. (2002) Deep-sea Res. II, Top. Stud. Oceanogr. 49, 3513–3528.
9. Fuhrman, J. A. (1999) Nature 399, 541–548. [PubMed]
10. Proctor, L. M. & Fuhrman, J. A. (1990) Nature 343, 60–62.
11. Suttle, C. A., Chan, A. M. & Cottrell, M. T. (1990) Nature 347, 467–469.
12. Suttle, C. A. & Chan, A. M. (1993) Mar. Ecol. Prog. Ser. 92, 99–109.
13. Waterbury, J. B. & Valois, F. W. (1993) Appl. Environ. Microbiol. 59, 3393–3399. [PMC free article] [PubMed]
14. Wilson, W. H., Joint, I. R., Carr, N. G. & Mann, N. H. (1993) Appl. Environ. Microbiol. 59, 3736–3743. [PMC free article] [PubMed]
15. Wilson, W. H., Carr, N. G. & Mann, N. H. (1996) J. Phycol. 32, 506–516.
16. Sambrook, J., Fritsch, E. F. & Maniatis, T. (1989) Molecular Cloning: A Laboratory Manual (Cold Spring Harbor Lab. Press, Plainview, NY).
17. Zeidner, G., Preston, C. M., Delong, E. F., Massana, R., Post, A. F., Scanlan, D. J. & Beja, O. (2003) Environ. Microbiol. 5, 212–216. [PubMed]
18. West, N. J., Schonhuber, W. A., Fuller, N. J., Amann, R. I., Rippka, R., Post, A. F. & Scanlan, D. J. (2001) Microbiology 147, 1731–1744. [PubMed]
19. Nubel, U., GarciaPichel, F. & Muyzer, G. (1997) Appl. Environ. Microbiol. 63, 3327–3332. [PMC free article] [PubMed]
20. Altschul, S. F., Madden, T. L., Schaffer, A. A., Zhang, J. H., Zhang, Z., Miller, W. & Lipman, D. J. (1997) Nucleic Acids Res. 25, 3389–3402. [PMC free article] [PubMed]
21. Gardy, J. L., Spencer, C., Wang, K., Ester, M., Tusnady, G. E., Simon, I., Hua, S., deFays, K., Lambert, C., Nakai, K. & Brinkman, F. S. L. (2003) Nucleic Acids Res. 31, 3613–3617. [PMC free article] [PubMed]
22. Thompson, J. D., Gibson, T. J., Plewniak, F., Jeanmougin, F. & Higgins, D. G. (1997) Nucleic Acids Res. 24, 4876–4882. [PMC free article] [PubMed]
23. Swofford, D. L. (1998) paup*, Phylogenetic Analysis Using Parsimony (and Other Methods) (Sinauer, Sunderland, MA).
24. Huelsenbeck, J. P. & Ronquist, F. (2001) Bioinformatics 17, 754–755. [PubMed]
25. Cech, T. R., Damberger, S. H. & Gutell, R. R. (1994) Nat. Struct. Biol. 1, 273–280. [PubMed]
26. Michel, F. & Westhof, E. (1990) J. Mol. Biol. 216, 585–610. [PubMed]
27. Schmetterer, G. (1994) in The Molecular Biology of Cyanobacteria, ed. Bryant, D. A. (Kluwer, Dordrecht, The Netherlands), pp. 409–435.
28. Bhaya, D., Dufresne, A., Vaulot, D. & Grossman, A. (2002) FEMS Microbiol. Lett. 215, 209–219. [PubMed]
29. Havaux, M., Guedeney, G., He, Q. F. & Grossman, A. R. (2003) Biochim. Biophys. Acta 1557, 21–33. [PubMed]
30. Zeidner, G. & Beja, O. (2004) Environ. Microbiol. 6, 528–534. [PubMed]
31. Chen, F. & Lu, J. R. (2002) Appl. Environ. Microbiol. 68, 2589–2594. [PMC free article] [PubMed]
32. Lindell, D., Sullivan, M. B., Johnson, Z. I., Tolonen, A. C., Rohwer, F. & Chisholm, S. W. (2004) Proc. Natl. Acad. Sci. USA 101, 11013–11018. [PMC free article] [PubMed]
33. Hendrix, R. W., Smith, M. C. M., Burns, R. N., Ford, M. E. & Hatfull, G. F. (1999) Proc. Natl. Acad. Sci. USA 96, 2192–2197. [PMC free article] [PubMed]
34. Hambly, E., Tetart, F., Desplats, C., Wilson, W. H., Krisch, H. M. & Mann, N. H. (2001) Proc. Natl. Acad. Sci. USA 98, 11411–11416. [PMC free article] [PubMed]
35. Dufresne, A., Salanoubat, M., Partensky, F., Artiguenave, F., Axmann, I. M., Barbe, V., Duprat, S., Galperin, M. Y., Koonin, E. V., Le Gall, F., et al. (2003) Proc. Natl. Acad. Sci. USA 100, 10020–10025. [PMC free article] [PubMed]
36. Rocap, G., Larimer, F. W., Lamerdin, J., Malfatti, S., Chain, P., Ahlgren, N. A., Arellano, A., Coleman, M., Hauser, L., Hess, W. R., et al. (2003) Nature 424, 1042–1047. [PubMed]
37. Palenik, B., Brahamsha, B., Larimer, F. W., Land, M., Hauser, L., Chain, P., Lamerdin, J., Regala, W., Allen, E. E., McCarren, J., et al. (2003) Nature 424, 1037–1042. [PubMed]
38. Turmel, M., Boulanger, J. & Lemieux, C. (1989) Nucleic Acids Res. 17, 3875–3887. [PMC free article] [PubMed]
39. Holloway, S. P., Deshpande, N. N. & Herrin, D. L. (1999) Curr. Genet. 36, 69–78. [PubMed]
40. Deshpande, N. N., Bao, Y. J. & Herrin, D. L. (1997) RNA 3, 37–48. [PMC free article] [PubMed]
41. Dujon, B. (1989) Gene 82, 91–114. [PubMed]
42. Pajunen, M. I., Kiljunen, S. J., Soderholm, M. E. L. & Skurnik, M. (2001) J. Bacteriol. 183, 1928–1937. [PMC free article] [PubMed]
43. Raaijmakers, H., Toro, I., Birkenbihl, R., Kemper, B. & Suck, D. (2001) J. Mol. Biol. 308, 311–323. [PubMed]
44. Sharma, M., Ellis, R. L. & Hinton, D. M. (1992) Proc. Natl. Acad. Sci. USA 89, 6658–6662. [PMC free article] [PubMed]
45. Gorbalenya, A. E. (1994) Protein Sci. 3, 1117–1120. [PMC free article] [PubMed]
46. Shub, D. A., Goodrichblair, H. & Eddy, S. R. (1994) Trends Biochem. Sci. 19, 402–404. [PubMed]
47. Belle, A., Landthaler, M. & Shub, D. A. (2002) Genes Dev. 16, 351–362. [PMC free article] [PubMed]
48. Liu, Q. Q., Belle, A., Shub, D. A., Belfort, M. & Edgell, D. R. (2003) J. Mol. Biol. 334, 13–23. [PubMed]
49. Mann, N. H. (2003) FEMS Microbiol. Rev. 27, 17–34. [PubMed]

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...