Diel Infection of a Cyanobacterium by a Contractile Bacteriophage

Light was found to strongly influence the infection of a freshwatercyanobacterium (Synechococcus elongatus PCC 7942) by a contractileDNA phage named AS-1. Phage progeny production was correlatedwith the amount of light in the laboratory and occurred in adiel pattern under natural light. At least one effect of lighton AS-1 infection is at the level of adsorption.

INTRODUCTION

Top

Introduction

Viruses are present in marine and freshwater systems at morethan 10⁷ particles per milliliter (4, 18). This large numberargues strongly for an important role in global biological andgeochemical processes, including the recycling of nutrient storesdue to virus-induced lysis (4, 6, 13, 20). In addition, phylogeneticanalysis has revealed that viruses in aquatic ecosystems aregenetically diverse, suggesting that they are under significantselection pressure, perhaps in response to host defense systemsand/or variations in the environment (22). Understanding howenvironmental factors and/or host genetics affect the successof viral infection is therefore relevant to the understandingof aquatic ecosystems.

A previously characterized contractile DNA phage of Synechococcuselongatus named AS-1 is a member of the family Myoviridae, whichincludes the well-characterized T-even phages (11, 14-16) (Fig.1A). AS-1 infects the well-characterized freshwater obligatephototroph Synechococcus elongatus PCC 7942 (14). A key featureof S. elongatus biology is that its gene expression is regulatedby an intrinsic $~$ 24-h circadian clock (5, 7, 8). The productsof three key genes, kaiA, kaiB, and kaiC, are required for maintainingthe normal gene expression rhythm (7, 8). Deletion of any ofthese three genes will result in a loss of rhythmicity but willnot significantly affect growth rates (5). We were initiallyinterested in examining whether mutations in circadian rhythmregulators of S. elongatus would affect AS-1 infection. We foundthat deletions in the genes that regulate S. elongatus circadianrhythm had only minor effects on AS-1 progeny production, whereaslight had a profound influence on the AS-1 infection process.

View larger version (48K):
[in this window]
[in a new window]
FIG. 1. Bacteriophage AS-1 and progeny phage production in S. elongatus PCC 7942. (A) Electron micrographs of AS-1, showing an intact particle (left) and an empty one (right) that has contracted its injection tubule. Viruses were negatively stained with uranyl acetate. (B) Demonstration of the plaques produced by AS-1. (C) Comparison of the abilities of AS-1 to form plaques on a mutant with a deletion in the kaiB gene ( ${Delta}$ kaiB) and on its parental strain, AMC541, over time. Approximately 50 PFU was added to the cultures after they were released from entrainment (time zero), and the results were enumerated approximately 2 days later. (D) Comparison of the number of progeny phage released by AMC541 and circadian clock regulatory mutants in the light or the dark. Progeny phage titers were determined 18 h after the introduction of AS-1.

MATERIALS AND METHODS

Top

Materials and Methods

S. elongatus strain AMC541 is the parental strain for the circadianmutants. Strains AMC702, AMC703, and AMC704 have deletions inthe kaiA, kaiB, and kaiC genes, respectively. Additional geneticinformation concerning the strains used can be found in thereport of Ditty et al. (5). Cultures were usually grown underconstant illumination at 45 µmol photons m^–2 s^–1in a modified BG-11 medium (2). Light intensity was measuredwith a LiCor luminometer. The original preparation of AS-1 wasobtained from the American Type Culture Collection. A stockof at least 10⁹ phage per ml was prepared by infecting S. elongatusgrown in the presence of light for more than 18 h, followedby two clarifications of the supernatant of the culture lysateby centrifugation. The clarified lysate was stored at 4°Cwith a drop of chloroform. As previously determined by Saffermanet al. (14), the lysate was stable for a minimum of severalweeks at 4°C.

Phage titers were determined by a plaque assay that used BG-11agar (1.5% agar) overlaid with 4 ml of 0.75% agar in BG-11 medium,0.3 ml of log-phase S. elongatus AMC541, and 5 to 200 µlof diluted phage lysate. Entrainment of cultures to synchronizetheir circadian clocks was performed by completely wrappingan early-log-phase culture of S. elongatus with aluminum foilfor 6 h. Other incubation conditions (shaking speed, temperature,etc.) were not changed.

RESULTS AND DISCUSSION

Top

Results and Discussion

Circadian rhythm mutants and AS-1 infection.
To examine the effects of mutations in the regulators of thecircadian clock, we entrained early-log-phase cultures of wild-typeS. elongatus and mutants with mutations in the kaiA, kaiB, andkaiC genes (5, 8, 10, 12). Upon release from a 6-h entrainmentperiod, aliquots of the cultures were taken over time and platedin plaque assays with approximately 50 PFU of AS-1. The plateswere incubated in continuous light until the number of plaqueswas enumerated approximately 2 days later. AS-1 plaques weregenerally uniform and were easily distinguished from the background(Fig. 1B). The number of plaques produced by the mutant deletedfor kaiB was not significantly different from that formed onthe parental strain, AMC541 (Fig. 1C). Similar results wereobtained with cultures containing mutations in the kaiA andkaiC genes and with unentrained cultures that were grown undercontinuous light of $~$ 45 µmol m^–2 s^–1 (datanot shown). These results suggest that deletions in the kaigenes do not significantly influence AS-1 infection.

To examine further the effects of deletion of kaiA and kaiBon AS-1 infection, we quantified progeny phage production incultures of S. elongatus inoculated with 3 x 10⁶ PFU of AS-1and incubated overnight for 18 h (sufficient time to allow approximatelytwo cycles of infection). Each mutant and its maternal strainwere grown with light at both 3 and 45 µmol m^–2s^–1. Infections of ${Delta}$ kaiA and ${Delta}$ kaiB mutants and the maternalstrain AMC541 all produced significant bursts of AS-1 at 45µmol m^–2 s^–1, and the final burst was morethan 3 log units higher than the input (Fig. 1D). However, at3 µmol m^–2 s^–1, the burst size was only ninefoldhigher than the input amount of AS-1 for AMC541 and was notsignificantly different from input for the ${Delta}$ kaiA and ${Delta}$ kaiB mutants.While infected AMC541 produced more phage than did the ${Delta}$ kaiAand ${Delta}$ kaiB mutants, we note that all AS-1 progeny production wasseverely reduced at low light. We interpret these results toindicate that the circadian regulatory proteins had only minoreffects on AS-1 infection. Furthermore, we hypothesize thatAS-1 infection of S. elongatus is correlated with light intensity,likely due to changes in the physiology of the host cell.

Light and AS-1 infection.
To confirm that phage infection is dependent on light, we inoculatedan early-log-phase culture of S. elongatus grown at 45 µmolm^–2 s^–1 with AS-1 at a multiplicity of infection(MOI) of 5. The culture was divided into two aliquots 30 minlater. One was maintained at 45 µmol m^–2 s^–1,and the second was covered with aluminum foil. The optical densitiesof the two cultures at 595 nm were monitored over time. Samplingof the covered culture was done in a darkened room. The coveredculture stopped increasing in its optical density almost immediatelyand maintained a fairly constant density throughout the experiment.Even after 18 h, the cells retained their characteristic greencolor and showed no visible signs of lysis. In contrast, theculture grown in light increased in density for several hours,then lysed between 6.5 and 9.5 h. The timing of the lysis wasconsistent with previous reports of AS-1 infection (14, 15,16). This result indicates that cells unable to photosynthesizewill not support normal lysis by AS-1.

Next, we tested the effects of different light intensities onAS-1 progeny production. S. elongatus was grown at 45 µmolm^–2 s^–1, infected with AS-1 for 1 h, and then dividedinto aliquots that were incubated at light intensities rangingfrom no detectable light to 45 µmol m^–2 s^–1.The yield of phage titers after 10 h was quantified and normalizedas a ratio of the input (Fig. 2B). Increased light intensityresulted in a corresponding increase in the production of progenyphage. These results provide further evidence that light directlyinfluences the success of AS-1 infection.

View larger version (33K):
[in this window]
[in a new window]
FIG. 2. Light affects lysis and progeny formation by AS-1. (A) Cell lysis curve for S. elongatus AMC541 infected with AS-1 (MOI, 5) under high and low light conditions. The optical density of the cell culture at 595 nm (OD₅₉₅) was determined on a ThermoSpectronic instrument by using the medium as a blank for each measurement. (B) PFU of phage produced by cultures grown at various intensities of light. The final titer (T_f) was divided by the titer in the initial inoculum (T₀) to determine the phage burst size (T_f/T₀). Each light condition was tested in three independent cultures, and the error bar indicates 1 standard deviation. (C) Effect of incubation time in the light on the ability to complete the infection process in the dark. Numbers to the left and right of the slash indicate the amount of time (in hours) for which a culture was grown in the light or the dark, respectively. Each condition was tested in two independent cultures with highly similar results.

Next, we examined whether S. elongatus infected during growthin light could complete the infection in the dark. One flaskof S. elongatus grown at 45 µmol m^–2 s^–1 oflight was infected with AS-1. Aliquots were then removed at0.5, 1, and 3 h postinoculation and transferred to flasks coveredwith aluminum foil. The original sample and the aliquots wereincubated for a total of 10 h, after which the burst size wasdetermined. The cultures transferred to darkness after 0.5 and1.0 h were unable to produce a normal burst, while the cultureincubated in light for 3 h released a significant amount ofprogeny phage (Fig. 2C). These results indicate that shouldAS-1 infection proceed for a critical period of time in light,the remainder of the infection process can take place whilelight is limited.

The amount of light required for successful AS-1 infection canbe related to the progression of events during AS-1 infection(15, 16). AS-1 actively degrades the host DNA within 1 h afterinfection and will inhibit host protein synthesis 3 to 4 h afterinfection (16). AS-1 DNA could be detected from between 2.5and 3 h until the end of the infection, with the packaging ofphage particles taking place by 8 h. Interestingly, treatmentof AS-1-infected cells with the RNA polymerase inhibitor rifampinbefore 2 h completely stopped AS-1 DNA synthesis, while treatmentat 3 h after infection or later had only modest effects on DNAsynthesis (16). These observations indicate that some changesin the infection process occur near the 3-h time point. Whetherthese changes can be related to the requirement for light inAS-1 infection remains to be determined.

Diel infection by AS-1.
The correlation between light intensity and AS-1 burst sizeunder laboratory conditions (Fig. 2B) suggests that AS-1 infectionoccurs in a diel pattern. To determine if this is true for AS-1infection of S. elongatus under natural light, cultures grownby a west-facing window were inoculated every 4 h for a 36-hspan. Aliquots were collected at 4 and 8 h after infection andwere kept in the dark to allow time for the completion of oneround of infection. Cultures inoculated during the day producedup to 2 log units more phage than those inoculated in the eveningor night (Fig. 3). Phage production increased again at the startof the second day. Furthermore, of the cultures that were inoculateda few hours before dawn, those collected at 4 h, before therewas significant light, produced fewer progeny than the samesamples collected after 8 h, when there was more light. Thus,AS-1 infection appears to cycle in a diel manner.

View larger version (26K):
[in this window]
[in a new window]
FIG. 3. Time course examining AS-1 progeny production in cultures exposed to natural light. The horizontal axis indicates the timing of the initial introduction of AS-1 phage (MOI, ca. 1). Aliquots of the culture were harvested at 4 and 8 h after inoculation and were then kept in the dark for 8 to 16 h to allow completion of that round of infection. The burst size is calculated from the PFU at the end of the experiment (T_f) divided by the PFU in the initial inoculum (T₀).

Light and AS-1 adsorption.
Virus infection can be functionally separated into specificstages, starting with adsorption. Cseke and Farkas (3) had reportedthat adsorption of AS-1 to Anacystis nidulans is modulated bylight and sodium ions. We examined whether light affects AS-1adsorption of S. elongatus (Fig. 4). Cultures were grown inthe presence of light and aliquoted, and a portion was transferredto light at 3 µmol m^–2 s^–1 for 15 min priorto the addition of AS-1 to both the light and dark cultures.Thirty, 60, and 90 min later, aliquots were removed and centrifugedfor 1 min at 10,000 x g to remove the cells. The number of PFUpresent in the supernatant was quantified. Figure 4A shows theresults from two independent experiments, each with two replicatesthat yielded consistent results (data not shown). Cultures at3 µmol m^–2 s^–1 had approximately four- tosixfold-higher PFU in the supernatant fraction than culturesgrown at 45 µmol m^–2 s^–1 light, indicatingthat light could affect AS-1 adsorption. There was a generaldecrease in the total number of PFU in both cultures over time,suggesting a gradual loss of AS-1 infectivity in this experiment,possibly due to nonspecific binding by AS-1 or inactivationby UV light (19).

View larger version (26K):
[in this window]
[in a new window]
FIG. 4. Effect of light on the initiation of adsorption by AS-1. (A) Number of PFU present in the culture supernatant after 1 h of adsorption with cultures grown at 45 µmol m^–2 s^–1 or 3 µmol m^–2 s^–1. Each bar represents an independent culture inoculated with AS-1 for the amount of time indicated on the horizontal axis. Gray bars, PFU for cultures incubated in the dark; white bars, PFU for cultures inoculated in the light. (B) Progeny phage production of cultures inoculated for 1 h at either 45 µmol m^–2 s^–1 or 3 µmol m^–2 s^–1 of light. After the initial inoculation, cells in the culture were washed to remove unbound phages and allowed to resume infection in the light at 45 µmol m^–2 s^–1, as shown in the flow diagram. The results of the plaque-forming assays after resumption of infection (shown at the bottom) are expressed as PFU at the specified time postinfection (PI), with the ratio to the input PFU given in parentheses. The initial inoculum contained 3 x 10⁶ PFU.

Since the number of unadsorbed AS-1 phage changed by only afew fold in the previous experiment, we examined whether AS-1adsorption is affected by light by measuring progeny phage production.Cells were first grown with normal light (45 µmol m^–2s^–1) and then divided into two cultures, one of whichwas shifted to 3 µmol m^–2 s^–1 for 15 min priorto addition of AS-1 to both cultures. After a 60-min incubationperiod to allow for possible adsorption, the cells in the twocultures were pelleted by centrifugation and then washed threetimes with the culture medium in an attempt to remove unboundphage. The cells were then resuspended in the medium and grownin the light. Aliquots of the cells were removed at varioustimes and analyzed for plaque formation (Fig. 4B). The samplesremoved immediately after the washes had low PFU, providinga starting point for quantification of phage production. Overthe next 8 to 25 h, an abundance of progeny phage was releasedfrom the culture in which AS-1 could adsorb in the presenceof normal light. In contrast, the culture that had the initialphage adsorption period at 3 µmol m^–2 s^–1produced fewer progeny phage (Fig. 4B). These results confirmthat light had an effect very early in the AS-1 infection process,most likely at the level of adsorption.

Summary and conclusions.
Viruses need to be well tuned to the physiology of the hostcell in order to maximize their fitness. S. elongatus gene expressionis known to be controlled by a circadian clock, a process basedon the diurnal light/dark cycle on Earth. We observed that infectionby the contractile DNA phage AS-1 is more affected by the availabilityof light than by the circadian cycle.

Light appears to be intimately linked to the success of AS-1infection. This is not surprising, since the physiology of S.elongatus and other obligately photosynthetic bacteria willchange significantly in response to the diel cycle (see reference1 for a review). However, the fact that AS-1 could produce anormal burst after infection was allowed to proceed in the lightfor 3 h suggests that AS-1 could overcome the normal restrictionsafter a certain stage in the infection process. Being able tocomplete the infection process in the dark could also providean adaptive advantage to the phage, since progeny release couldoccur in the dark, thus decreasing exposure to UV light, whichcould deactivate phage (reference 19 and references within).An alternative interpretation of the results, which we considerless likely, is that light had a direct effect on phage proteinsthat could trigger the ability to infect S. elongatus.

Light is often limiting in aquatic systems, and the light/darkcycle is known to affect a number of cellular and ecologicalprocesses, including hormonal regulation in vertebrates (21),carbon partitioning in plants (23), and the migration of phytoplankton(9, 17). Should these results from AS-1 reflect the requirementsof additional photosynthetic microbes, then the regenerationof nutrients in aquatic ecosystems mediated by viral lysis couldbe correlated with light intensity.

ACKNOWLEDGMENTS

Helpful discussions with Ry Young, members of his lab, and theTexas A&M Cereal Killers are greatly appreciated.

C.C.K. acknowledges start-up funds from Texas A&M Universityand funding of his lab by the National Science Foundation. S.G.acknowledges funding from the National Institutes of Health(PO1 NS39546).

This work is dedicated to Rex Lowe in appreciation of his enthusiasmfor teaching and life.

FOOTNOTES

* Corresponding author. Mailing address: Department of Biochemistry and Biophysics, Texas A&M University, College Station, TX 77843. Phone: (979) 458-2235. Fax: (979) 845-9274. E-mail: ckao{at}tamu.edu.

REFERENCES

Top