© 2005 Nature Publishing Group
Photosynthesis genes in marine viruses yield
proteins during host infection
Debbie Lindell1, Jacob D. Jaffe2†, Zackary I. Johnson1†, George M. Church2& Sallie W. Chisholm1,3
Cyanobacteria, and the viruses (phages) that infect them, are
significant contributors to the oceanic ‘gene pool’1,2. This pool is
dynamic, and the transfer of genetic material between hosts and
their phages3–6probably influences the genetic and functional
diversity of both. For example, photosynthesis genes of cyanobac-
terial origin have been found in phages that infect Prochlorococ-
in ocean ecosystems. These genes include psbA, which encodes the
photosystem II core reaction centre protein D1, and high-light-
inducible (hli) genes. Here we show that phage psbA and hli genes
are expressed during infection of Prochlorococcus and are co-
transcribed with essential phage capsid genes, and that the
amount of phage D1 protein increases steadily over the infective
period. We also show that the expression of host photosynthesis
genes declines over the course of infection and that replication of
the phage genome is a function of photosynthesis. We thus
propose that the phage genes are functional in photosynthesis
and that they may be increasing phage fitness by supplementing
the host production of these proteins.
Photosynthesis in cyanobacteria, algae and plants requires two
photosystems (denoted PSI and PSII). The D1 and D2 proteins
(encoded by psbA and psbD, respectively) form a heterodimer in
the reaction centre of PSII and bind the components required
for photochemistry. The D1 protein is turned over rapidly owing
to light-induced damage10; thus, its de novo synthesis is required
for sustained photosynthesis10. High-light-inducible proteins
(HLIPs) protect the photosynthetic apparatus from photodamage
by dissipating excess light energy11.
Numerous cyanophages contain photosynthesis genes (psbA and
at least one other)5,7–9with highly conserved amino acid
sequences7,12, suggesting that they encode functional proteins that
may be involved in maintaining host photosynthesis during infec-
tion. Here we used Prochlorococcus MED4 and the podovirus P-SSP7
(a T7-like phage5) as a model system to begin exploring this
hypothesis. We considered that if these genes are involved in host
photosynthesis, then the amount of phage production might be
dependent on photosynthetic performance and, conversely, host
photosynthesis might be compromised by phage infection. We
examined the first part of this hypothesis by inhibiting photosyn-
thesis with darkness or DCMU (3-(3,4-dichlorophenyl)-1,1-
dimethylurea), an inhibitor of electron flow from PSII to PSI,
which led to a respective four- or twofold reduction in replication
of the phage genome (Fig. 1a). Thus, as in other systems13–15,
continued photosynthesis is necessary for maximal phage replica-
tion. To determine the converse, that is, whether host photosynthesis
is influenced by phage infection, we measured PSII photochemical
conversion efficiency (Fv/Fm) and functional cross-sectional area
only slightly, whereas the latter was constant throughout this period
(Fig.1b, c),indicating thatphage infection doesnotleadtoamarked
decline inPSII performance,as occurs insome14–16, butnotother17–19
photosynthetic host–virus systems.
Thus, continued photosynthesis is required for maximum phage
production in our system. Moreover, photosynthesis is sustained
during infection when a decline in the transcription and translation
of host genes might be expected, suggesting that the expression of
phage photosynthesis genes might be supplementing host metabo-
lism. To address this hypothesis, we determined whether these phage
genes are expressed and, if so, how their expression relates to that of
the homologous host genes. Using probes specific for the phage and
host psbA and hli genes, we found that both of the phage genes were
transcribed (Fig. 2a, b and Supplementary Fig. 1). Using polymerase
chain reaction with reverse transcription (RT–PCR), we determined
Figure 1 | Photosynthesis and phage infection. a, Replication of the phage
genome was assessed by quantifying the phage gene encoding DNA
polymerase in host cells kept in the light, transferred to the dark or treated
with DCMU while in the light. Significantly lower (P , 0.05) phage DNA
was detected in dark- and DCMU-treated cells after 4h. b, c, PSII
photochemical conversion efficiency (Fv/Fm; b) and PSII functional
absorption cross-sectional area (jPSII; c) in infected and control cells. The
absorption cross-section remained constant, although there was a 10%
decline in PSII conversion efficiency. Error bars indicate the s.d. from
1Department of Civil and Environmental Engineering, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139, USA.2Department of Genetics, Harvard Medical
School, Boston, Massachusetts 02115, USA.3Department of Biology, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139, USA. †Present addresses: The
Broad Institute of Harvard and MIT, Cambridge, Massachusetts 02141, USA (J.D.J); Department of Oceanography, University of Hawaii, 1000 Pope Road, Honolulu, Hawaii
96822, USA (Z.I.J.).
Vol 438|3 November 2005|doi:10.1038/nature04111
© 2005 Nature Publishing Group
that phage psbA mRNA made up 50% of the total (phage plus host)
psbA transcripts by 7–8h after infection. By 4–5h after infection,
mRNA from the single host psbA gene had dropped to 50–60% of
maximal levels (Fig. 2a and Supplementary Fig. 1). Transcription of
16 out of 22 of the host hli genes also declined significantly during
infection (hli12 is shown as a representative of this gene family in
Fig. 2b). Because the maximal level of first-round infection achieved
so far with this host–phage system is 50% (D.L. and S.W.C.,
unpublished data), the presence of only 50–60% of host psbA and
hli transcripts relative to the control suggests that transcription of
psbA transcripts in infected cells were transcribed from the phage
genome. The decline in host gene expression was not specific to
photosynthesis genes but was part of a general reduction in host
transcription subsequent to infection (D.L. and S.W.C., unpublished
The decline in host psbAtranscription should cause a reduction in
translation of the D1 protein and, because this protein is turned over
rapidly10, a decrease in host D1 titre. To verify this, we identified and
quantified peptides specific for the host and phage D1 proteins
(Fig. 3). The host D1 protein declined during infection to roughly
75% of maximal levels (Fig. 2c), reflecting a decrease of 50% in
infected cells. Accompanying this reduction was a steady increase in
the homologous protein encodedby thephage (Fig. 2c), which made
up about 10% of total D1 in infected cells by the end of the latent
Although these results are consistent with our hypothesis that the
expression of phage D1 protein helps to bolster host photosynthesis,
the amount of phage D1 did not quantitatively compensate for the
little during infection (Fig. 1b). This suggests that additional factors
may be involved in maintaining host photosynthesis. For example,
reduce photosystem damage as well as being involved in the re-
assembly of PSII20. Furthermore, phage D1 may be more efficient
than host D1 during infection6. If photosynthetic antennae are
shared among reaction centres, which is consistent with other
studies21, then fewer, more efficient functional reaction centres
could lead to an increase in the functional absorption cross-section;
however, this was not observed (Fig. 1c). Thus, although we are left
with an imperfect balance sheet for the host and phage D1 protein,
the inverse temporal expression of host and phage photosynthesis
genes is striking and suggests that there is a functional
A fitness advantage conferred by phage photosynthesis genes not
Figure 2 | Expression of phage and host photosynthesis genes. Temporal
patterns of psbA mRNA (a), hli12 mRNA (b) and D1 peptides (c).
Expression during infection was normalized to expression in uninfected
control cells and is presented relative to the maximal level for each
individual gene. Data in a and b were obtained from microarray analyses.
See Supplementary Fig. 1 for confirmation of the results shown in a by
RT–PCR.Host D1 peptideswere significantlylower in thesecond half ofthe
latent period (5–8h) than in the first half (0–4h); P , 0.01. Error bars
indicate the s.d. from biological replicates.
Figure 3 | Analysis of host and phage peptides. a, Extracted ion
chromatograms of endogenous host (1) and isotopically labelled synthetic
host (2) peptides, and endogenous phage (3) and isotopically labelled
synthetic phage (4) peptides. Endogenous and corresponding synthetic
peptides co-elute, whereas host and phage peptides have different retention
times and m/z values. The area under the peaks of the endogenous peptides
was used for quantification. b, Mass spectra taken at the apex of the
chromatographic peaks show that host and phage peptides have different
m/z values, as do the endogenous and isotopically labelled synthetic peptide
pairs. Note that the threonine residue in the third position of the synthetic
host and phage peptides was uniformly labelled with15N and13C isotopes,
adding 5Da to the mass of the peptides. c, Collision-induced dissociation
amino acid sequence in the partial peptide annotation is reversed because
the y-ion series is shown. The full sequence of each peptide is shown with
NATURE|Vol 438|3 November 2005
© 2005 Nature Publishing Group
supports the modular theory of phage evolution22, in which phages
evolve through the step-wise acquisition of genes from diverse
sources. According to this theory, acquired genes are initially
expressed autonomously and are integrated into the phage life
cycle if they provide a fitness advantage. The photosynthesis genes
in cyanophage originate from cyanobacteria7,9,12, and these phage
genomes also contain bacterial, and even archaeal and eukaryotic
genes5,6. Notably, the psbA and psbD genes in two Prochlorococcus
myoviruses7have putative promoter and transcriptional terminators
flanking the genes, suggesting that they are autonomously expressed.
By contrast, photosynthesis genes in the phage used here have
overlapping start and stop codons7and are co-transcribed with the
essential, highly expressed phage capsid genes surrounding the
photosynthesis genes (Fig. 4), suggesting that they have become an
integral part of the phage genome. Thus, the proposed fitness
advantage conferred by these photosynthesis genes, and other
genes acquired from their hosts5,6, may be a forerunner to their
becoming bona fide members of the cyanophage gene pool.
Although we have not proved that phage photosynthesis gene
products are participating in host photosynthesis, we favour this
hypothesis, first, because photosynthesis continues during infection
despite the decline in expression of host photosynthesis genes;
second, because maximal phage DNA replication is dependent on
this sustained photosynthesis; and last, because the high conserva-
tion of amino acid sequences7,12in the phage proteins suggests that
these proteins are functioning in the same role as the host proteins.
Nonetheless, we cannot rule out alternative functions. One could
arguethat D1,anefficientmanganese-binding protein10,isobtaining
this metal ion for phage enzymatic reactions. Or perhaps phage D1,
like the HLIP, is involved in dissipating excess light energy without
being actively involved in photosynthesis. Obviously, the next stages
in testing the ‘samefunction hypothesis’are to see whether phage D1
localizes to the host PSII complex and to study the behaviour of this
phage–host system using phage in which the photosynthesis genes
have been inactivated.
The dynamic nature of the oceanic gene pool has led to the genetic
diversification of donor and recipient genomes7and has apparently
guided their functional diversification as well. If phage photosyn-
thesis proteins do indeed function in host photosynthesis, this is a
striking example of the interaction of proteins encoded from two
distinct genomes in a single metabolic complex. Furthermore, the
abundance of cyanobacteria23and their phages16,24in the oceans
suggests that phage photosynthesis proteins have a small but signifi-
cant role in the conversion of light to chemical energy on a global
Experimental conditions. Prochlorococcus MED4 was grown at 218C under
continuous cool white light (25mmol photon m22s21) in Sargasso seawater
Pro99 medium25amended with 10mM HEPES (pH7.5) and 12mM sodium
bicarbonate. Cells were concentrated to 108cells per ml by centrifugation, and
triplicate cultures were infected with 3 £ 108infective phages per ml for a
multiplicityofinfection(MOI)of3 (exceptfor theexperimentshown inFig. 1a;
RNA and protein analyses were collected by centrifugation (12,400g for 15min
EDTA; pH5.2), snap frozen in liquid nitrogen and stored at 2808C. Sample
handling took 30min. For the experiment shown in Fig. 1a, Prochlorococcus at
108cells per ml were exposed to 107infective phages per ml for an MOI of 0.1.
After 1h in the light to allow adsorption, cultures were diluted 4,000-fold and
the MOI of phage stocks by the most probable number assay.
Quantitative detection of phage genomic DNA. Prochlorococcus cells were
collected on 0.2-mm pore-sized polycarbonate filters (Osmonics), washed with
sterile seawater followed by 3ml of preservation solution (10mM Tris, 100mM
method26. Phage genomic DNA was quantified by real-time PCR (see below),
targeting the phage DNA polymerase gene. Primer sequences are given in
Supplementary Table 1.
Photosynthesis measurements. A background irradiance gradient single-turn-
over fluorometer (BIG-STf) was usedto measurethe photosyntheticconversion
efficiency (Fv/Fm) and functional absorption cross-section area (jPSII) of PSII,
which measures the ability of PSII to absorb photons from antennae com-
plexes27. Triplicate samples were dark acclimated for 15–30min before single-
turnover fluorescence induction curve measurements. Fv/Fmand jPSIIwere
estimated by fitting standard models28to the data to determine Fo(initial
fluorescence), Fm(maximal fluorescence), Fv(Fm2 Fo) and jPSII.
RNA extraction and transcript analysis. Total RNAwas extracted by a mirVana
RNA isolation kit (Ambion). We removed DNA by a Turbo DNA-free kit
(Ambion). For microarray analysis, RNA was concentrated by ethanol precipi-
tation, and 2mg of total RNA was labelled and hybridized to custom-made
MD4-9313 arrays (Affymetrix) using the standard Affymetrix protocol for
Escherichia coli (http://www.affymetrix.com/technology/index.affx). This array
contains probe sets for genes and intergenic regions of both host and phage.
Standard affymetrix procedures were used for probe design and construction of
specific primers and 100U of SuperScript II (Invitrogen) in the presence of
200U of SuperaseIN (Ambion). Triplicate real-time PCR reactions were done
with a QuantiTect SYBR Green PCR kit (Qiagen) and primers at 0.3–1.0mM.
After 15min at 958C, 40 cycles of denaturation (958C, 15s), annealing (568C,
30s) and elongation (728C, 30s) were run on an DNA Engine Opticon (MJ
Research), which was followed by 5min at 728C and melt curve analysis.
Incorporation of SYBR stain into double-stranded DNA was determined
subsequent to elongation steps. Standard curves were generated with genomic
in Supplementary Table 1.
Protein preparation and analysis. Cells were lysed in 3M urea, 0.05% SDS and
50mM Tris-HCl (pH8). Proteins were digested with sequencing grade trypsin
(Promega) at a protein to trypsin ratio of 137.5:1, reduced with 10mM
dithiothreitol, alkylated with 50mM iodoacetamide and acidified to pH , 3.
Total protein was purified by solid-phase extraction using Oasis MCX (Waters),
concentrated by vacuum centrifugation, resuspended in 0.1% formic acid and
purified by solid-phase extraction using Oasis HLB (Waters). Peptides eluted
with 70% acetonitrile were concentrated by vacuum centrifugation to dryness
and resuspended in 5% acetonitrile and 5% formic acid. Tryptic digestions
of D1. Peptides with the sequences NH2-ETTETESQNYGYK-COOH and NH2-
ETTEDVSQNYGYK-COOH were observed by liquid chromatography mass
spectrometry (LC-MS; Fig. 3) and used as surrogates for the host and phage
Figure 4 | Co-transcription of phage photosynthesis and capsid genes.
a, Proposed operon of the photosynthesis gene region determined
bioinformatically7. b, Transcript levels of regions internal to the psbA and
g10 genes, and regions spanning from psbA to the capsid genes upstream
(g9 to psbA) and downstream (psbA to g10) of psbA at 4h after infection.
g9 encodes capsid assembly protein, g10 encodes major capsid protein.
Amplification from mRNA (‘RT–PCR’) was more than two orders of
magnitude greater than from genomic DNA (‘no RT control’). The lower
transcript abundance across gene boundaries may be due to
posttranscriptional processing or additional autonomous transcription of
psbA and g10. Error bars indicate the s.d. from biological replicates.
NATURE|Vol 438|3 November 2005
© 2005 Nature Publishing Group
peptides were used as mass spectrometric standards to verify the elution time
and tandem MS fragmentation patterns of the host and phage peptides (Fig. 3).
D1 peptides were identified and quantified by reversed-phase LC-MS as
described29from duplicate injections of 5.7mg of total peptides using a hybrid
linear ion trap, Fourier transform ion cyclotron resonance mass spectrometer
(ThermoElectron) with resolution set to 100,000 and a mass accuracy of
^8p.p.m. Extracted ion chromatograms were generated by monitoring the
mass/charge (m/z) values of 775.3363 ^ 0.006 and 767.3388 ^ 0.006 for the
host and phage peptides, respectively. We used the area under the peaks from
XCalibur Software for quantification (see also http://arep.med.harvard.edu/
mapquant.html for an alternative method).
Received 27 June; accepted 27 July 2005.
Published online 12 October 2005.
1. Breitbart, M. et al. Genomic analysis of uncultured marine viral communities.
Proc. Natl Acad. Sci. USA 99, 14250– -14255 (2002).
Venter, J. C. et al. Environmental genome shotgun sequencing of the Sargasso
Sea. Science 304, 66– -74 (2004).
Canchaya, C., Fournous, G., Chibani-Chennoufi, S., Dillmann, M. L. & Brussow, H.
Phage as agents of lateral gene transfer. Curr. Opin. Microbiol. 6, 417– -424
Palenik, B. et al. The genome of a motile marine Synechococcus. Nature 424,
1037– -1042 (2003).
Sullivan, M. B., Coleman, M. L., Weigele, P., Rohwer, F. & Chisholm, S. W.
Three Prochlorococcus cyanophage genomes: signature features and ecological
interpretations. PLoS Biology 3, e144 (2005).
Mann, N. H. et al. The genome of S-PM2, a ‘photosynthetic’ T4-type
bacteriophage that infects marine Synechococcus. J. Bacteriol. 187, 3188– -3200
Lindell, D. et al. Transfer of photosynthesis genes to and from Prochlorococcus
viruses. Proc. Natl Acad. Sci. USA 101, 11013– -11018 (2004).
Mann, N. H., Cook, A., Millard, A., Bailey, S. & Clokie, M. Bacterial
photosynthesis genes in a virus. Nature 424, 741 (2003).
Millard, A., Clokie, M. R. J., Shub, D. A. & Mann, N. H. Genetic organization of
the psbAD region in phages infecting marine Synechococcus strains. Proc. Natl
Acad. Sci. USA 101, 11007– -11012 (2004).
10. Adir, N., Zer, H., Shochat, S. & Ohad, I. Photoinhibition—a historical
perspective. Photosynth. Res. 76, 343– -370 (2003).
11.Havaux, M., Guedeney, G., He, Q. & Grossman, A. R. Elimination of high-light-
inducible polypeptides related to eukaryotic chlorophyll a/b-binding proteins
results in aberrant photoacclimation in Synechocystis PCC6803. Biochim.
Biophys. Acta 1557, 21– -33 (2003).
12. Zeidner, G. et al. Potential photosynthesis gene recombination between
Prochlorococcus and Synechococcus via viral intermediates. Environ. Microbiol. 7,
1505– -1513 (2005).
13. Adolph, K. W. & Haskelkorn, R. Photosynthesis and the development of blue– -
green algal virus N-1. Virology 47, 370– -374 (1972).
14. MacKenzie, J. J. & Haselkorn, R. Photosynthesis and the development of blue– -
green algal virus SM-1. Virology 49, 517– -521 (1972).
15. Sherman, L. A. Infection of Synechococcus cedrorum by the cyanophage AS-1M.
Virology 71, 199– -206 (1976).
16. Suttle, C. A. & Chan, A. M. Marine cyanophages infecting oceanic and coastal
strains of Synechococcus—abundance, morphology, cross-infectivity and
growth-characteristics. Mar. Ecol. Prog. Ser. 92, 99– -109 (1993).
17. Ginzburg, D., Padan, E. & Shilo, M. Effect of cyanophage infection on CO2
photoassimilation in Plectonema boryanum. J. Virol. 2, 695– -701 (1968).
18. Rahoutei, J., Garcia-Luque, I. & Baron, M. Inhibitioin of photosynthesis by viral
infection: effect on PSII structure and function. Physiol. Plant. 110, 286– -292
19. Arias, M. C., Lenardon, S. & Taleisnik, E. Carbon metabolism alterations in
sunflower plants infected with the sunflower chlorotic mottle virus. J. Phytopath.
151, 267– -273 (2003).
20. Xu, H., Vavilin, D., Funk, C. & Vermaas, W. Multiple deletions of small Cab-like
proteins in the cyanobacterium Synechocystis sp. PCC 6803—consequences for
pigment biosynthesis and accumulation. J. Biol. Chem. 279, 27971– -27979
21. Bruyant, F. et al. Diel variations in the photosynthetic parameters of
Prochlorococcus strain PCC 9511: combined effects of light and cell cycle.
Limnol. Oceanogr. 50, 850– -863 (2005).
22. Hendrix, R. W., Lawrence, J. G., Hatfull, G. F. & Casjens, S. The origins and
ongoing evolution of viruses. Trends Microbiol. 8, 504– -508 (2000).
23. Partensky, F., Hess, W. R. & Vaulot, D. Prochlorococcus, a marine
photosynthetic prokaryote of global significance. Microbiol. Mol. Biol. Rev. 63,
106– -127 (1999).
24. Sullivan, M. B., Waterbury, J. B. & Chisholm, S. W. Cyanophages infecting the
oceanic cyanobacterium Prochlorococcus. Nature 424, 1047– -1051 (2003).
25. Moore, L. R., Post, A. F., Rocap, G. & Chisholm, S. W. Utilization of different
nitrogen sources by the marine cyanobacteria Prochlorococcus and
Synechococcus. Limnol. Oceanogr. 47, 989– -996 (2002).
26. Zinser, E. R. et al. Prochlorococcus ecotype abundance in the North Atlantic
Ocean revealed by an improved quantitative PCR method. Appl. Environ.
Microbiol. (in the press).
27. Johnson, Z. I. Development and application of the background irradiance
gradient– -single turnover fluorometer (BIG-STf). Mar. Ecol. Prog. Ser. 283,
73– -80 (2004).
28. Kolber, Z. S., Prasil, O. & Falkowski, P. G. Measurements of variable chlorophyll
fluorescence using fast repetition rate techniques—defining methodology and
experimental protocols. Biochim. Biophys. Acta 1367, 88– -106 (1998).
29. Gerber, S. A., Rush, J., Stemman, O., Kirschner, M. W. & Gygi, S. P. Absolute
quantification of proteins and phosphoproteins from cell lysates by tandem
MS. Proc. Natl Acad. Sci. USA 100, 6940– -6945 (2003).
Supplementary Information is linked to the online version of the paper at
Acknowledgements We thank T. Rector and R. Steen for doing the Affymetrix
GeneChip experiments; C. Steglich, M. Sullivan, M. Coleman, and E. Zinser for
discussions; and M. Sullivan for comments on the manuscript. This research was
supported by grants from the National Science Foundation (to S.W.C.), the
Gordon and Betty Moore Foundation’s Program in Marine Microbiology (to
S.W.C), and the Department of Energy Genomes to Life Program (to S.W.C and
Author Information Reprints and permissions information is available at
npg.nature.com/reprintsandpermissions. The authors declare no competing
financial interests. Correspondence and requests for materials should be
addressed to S.W.C. (email@example.com).
NATURE|Vol 438|3 November 2005