Differential Bacteriophage Mortality on Exposure to Copper

APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Oct. 2011, p. 6878–6883 0099-2240/11/$12.00 doi:10.1128/AEM.05661-11 Copyright © 2011, American Society for Microbiology. All Rights Reserved.

Vol. 77, No. 19

Differential Bacteriophage Mortality on Exposure to Copper䌤 Jinyu Li1 and John J. Dennehy2* Biology Department, Queens College of the City University of New York, Flushing, New York,1 and The Graduate Center of the City University of New York, New York, New York2 Received 28 May 2011/Accepted 3 August 2011

Many studies report that copper can be used to control microbial growth, including that of viruses. We determined the rates of copper-mediated inactivation for a wide range of bacteriophages. We used two methods to test the effect of copper on bacteriophage survival. One method involved placing small volumes of bacteriophage lysate on copper and stainless steel coupons. Following exposure, metal coupons were rinsed with lysogeny broth, and the resulting fluid was serially diluted and plated on agar with the corresponding bacterial host. The second method involved adding copper sulfate (CuSO4) to bacteriophage lysates to a final concentration of 5 mM. Aliquots were removed from the mixture, serially diluted, and plated with the appropriate bacterial host. Significant mortality was observed among the double-stranded RNA (dsRNA) bacteriophages ⌽6 and ⌽8, the single-stranded RNA (ssRNA) bacteriophage PP7, the ssDNA bacteriophage ⌽X174, and the dsDNA bacteriophage PM2. However, the dsDNA bacteriophages PRD1, T4, and ␭ were relatively unaffected by copper. Interestingly, lipid-containing bacteriophages were most susceptible to copper toxicity. In addition, in the first experimental method, the pattern of bacteriophage ⌽6 survival over time showed a plateau in mortality after lysates dried out. This finding suggests that copper’s effect on bacteriophage is mediated by the presence of water.

veyed a broad range of bacteriophages (phages) to determine which molecular components rendered them most susceptible to copper inactivation. Included in our study were ssRNA (PP7), dsRNA (⌽6 and ⌽8), ssDNA (⌽X174), and dsDNA (␭, T4, PRD1, and PM2) phages. Of these, phages ⌽6, ⌽8, PRD1, and PM2 contain lipid envelopes. We determined that the non-lipid-containing dsDNA phages were most resistant to copper treatment. The lipid-containing phages, except for PRD1, were highly susceptible to copper. Also susceptible, but to a lesser degree, were the phages containing single-stranded RNA or DNA. These data should assist future studies seeking to determine the molecular targets of copper.

Although essential in many biological processes, copper has long been known to be toxic in high concentrations, particularly to microbes (5, 9). Its use in medicine dates back to the Egyptian, Greek, and Roman civilizations. In the 19th and early 20th centuries, copper preparations saw widespread use as antimicrobial agents prior to the discovery of antibiotics. Indeed, numerous studies have demonstrated the strong antimicrobial properties of copper solutions and of dry copper surfaces (9). Renewed interest in the use of copper as an antimicrobial agent stems from the rise of antibiotic-resistant bacteria and emerging viruses. In public facilities such as hospitals, schools, and nursing homes, antibiotic-resistant bacteria and viruses are often transferred between hosts via contact with surfaces such as countertops, railings, and doorknobs. Because of this, the application of copper to frequently contacted surfaces in public and commercial facilities has gained much traction (6, 19, 20). Despite its clear effectiveness, the mechanisms by which copper inactivates viruses remain opaque. Most reports suggest that virus inactivation is likely to result from the generation of hydrogen peroxide (H2O2) and/or reactive oxygen species (ROS) by redox cycling between the different copper species (9, 11, 16, 28, 36). These molecules are known to damage critical biological molecules, such as DNA (2, 27, 34, 35), proteins (13, 16, 17), and phospholipids (23, 31, 33). Viruses generally consist of nucleic acid (in either single-stranded [ss] or double-stranded [ds] RNA or DNA forms) surrounded by a protein coat. Some viruses also possess a lipid envelope that can be internal or external to the protein coat. We sur-

MATERIALS AND METHODS Growth media. Two types of growth media were used in our experiments: lysogeny broth (LB; sometimes incorrectly referred to as Luria-Bertani broth) (3, 4) and SB broth (8). LB consists of 10 g Bacto tryptone, 10 g NaCl, and 5 g Bacto yeast extract per liter of water. LB top and bottom agars contained 7 g and 15 g Bacto agar per liter, respectively. SB broth contained 8 g Difco nutrient broth, 26 g NaCl, 12 g MgSO4 䡠 7H2O, 1.5 g CaCl2 䡠 6H2O, and 0.7 g KCl per liter of water. SB top and bottom agars contained 7 g and 15 g Bacto agar per liter, respectively. Bacteriophage and bacterial host strains. All bacteriophages used in this study, their host bacteria, and their growth conditions are listed in Table 1. To obtain phage lysates, one colony of a given host was added to 10 ml of the specified growth medium and cultured for 18 h with rotary shaking (220 rpm). Stationary-phase culture (100 to 200 ␮l) was added to fresh medium along with 1 ␮l frozen phage stock. Following 18 h incubation, phages were purified by filtering culture through 0.22-␮m filters (Durapore; Millipore, Bedford, MA). Phage particles per milliliter were quantified via serial dilution and determination of titers. Determination of titers consists of adding diluted phage lysate and 100 to 200 ␮l of stationary-phase culture to 3 ml top agar (stored as liquid at 45°C; gels to solid at 25°C), vortexing, and pouring the mixture into a 100- by 15-mm petri dish containing 35 ml bottom agar (10). Plaques were counted on plates where 30 to 500 plaques were visible and used to estimate the PFU per milliliter of the original lysate by multiplying the number of plaques times the dilution factor.

* Corresponding author. Mailing address: Biology Department, Queens College of the City University of New York, 65-30 Kissena Blvd., Flushing, NY 11367. Phone: (718) 997-3411. Fax: (718) 9973445. E-mail: [email protected]. 䌤 Published ahead of print on 12 August 2011. 6878

VOL. 77, 2011

PHAGE INACTIVATION BY COPPER

6879

TABLE 1. Bacteriophages and bacterial host strains used in experimentsa Bacteriophageb

Host

Growth medium, temp (°C)

␭ ⌽6 ⌽8 ⌽X174 PRD1

Escherichia coli MG1655 Pseudomonas syringae pv. phaseolicola HB10Y Pseudomonas syringae pv. phaseolicola LM2489 E. coli C Salmonella enterica serovar Typhimurium LT2 (pLM2) E. coli B Pseudomonas aeruginosa PA01 Pseudoalteromonas espejiana BAL 31

LB, 37 LB, 25 LB, 25 LB, 37 LB supplemented with 25 mg/ml⫺1 kanamycin, 37 LB, 37 LB, 37 SB broth, 25

T4 PP7 PM2

Type of lipid membrane

Nucleic acid content

None External External None Internal

dsDNA dsRNA dsRNA ssDNA dsDNA

None None Internal

dsDNA ssRNA dsDNA

a

See Materials and Methods for medium formulations. ␭ is from Ing-Nang Wang, University at Albany, Albany, NY; ⌽6 and ⌽8 are from Leonard Mindich, Public Health Research Institute Center, Newark, NJ; ⌽X174 is from Holly Wichman, University of Idaho, Moscow, ID; and PRD1, T4, PP7, and PM2 are from Felix d’Herelle Reference Center for Bacterial Viruses, Quebec City, Canada. b

Experimental procedures. We conducted two types of experiments in our study. In the first type, a total of 24 100-␮l aliquots of phage ⌽6 lysate were placed on coupons of copper or steel. Previous experiments reported that steel is relatively inert with respect to microbes; thus, it was used as a control in our experiments (21, 35). After 0, 60, 120, and 180 min, 3 coupons from each treatment were thoroughly rinsed with 1 ml LB, which was collected into sterile reagent reservoirs. The rinsing protocol consisted of holding the coupon over a reagent reservoir with sterile tweezers and repeatedly (5 times) pipetting 1 ml of the LB against the surface of the coupon so that it washed into the reservoir. The validity of this technique was demonstrated by comparing the titer of the source lysate with the titer of lysate placed on a coupon of steel, which was then immediately rinsed off, repeated 5 times. These values were not significantly different, suggesting that ⬎99.999% of phages were recovered by rinsing (F ⫽ 0.160, df ⫽ 1, P ⫽ 0.6993). It was noted that these 100-␮l aliquots dried out completely by 60 min. The surviving phages in each replicate were quantified by determining the titers on a Pseudomonas syringae pv. phaseolicola lawn with serial dilution as needed. Preliminary spot titers were used to determine the appropriate dilution factors to visualize 30 to 500 plaques on a lawn. In the second type of experiment, 5 ␮l of 1 M CuSO4 solution was added to 1 ml bacteriophage lysates to a final concentration of 5 mM. Control treatments received a similar-volume sham treatment of deionized water. At 60, 120, or 180 min, the lysate was serially diluted (if necessary) and plated on the appropriate host. Plaque counts were used to estimate titers in each sample. Titers at each time point are from independent experiments; i.e., samples were not repeatedly drawn from the same treatment.

copper and steel independently. For copper, significant differences in log10 titers were observed between time points 0 and 60, 0 and 120, and 0 and 180 min but not between time points 60 and 120, 60 and 180, and 120 and 180 min (statistics given in Table 2). Similarly significant but smaller declines were observed for the steel treatment (Table 2). Because steel was intended as a control treatment, these results seemed puzzling until we realized that the 100 ␮l of lysate placed on the coupons dried after 60 min. Therefore, we hypothesize that the phage inactivation observed in the steel treatments was due to desiccation, a known factor in

RESULTS Our initial experiments consisted of placing 100 ␮l bacteriophage ⌽6 lysate on coupons of copper or steel. After exposure to the metal for 0, 60, 120, or 180 min (replicated 3 times), coupons were rinsed with 1 ml LB, LB was serially diluted, and titers were determined on a P. syringae pv. phaseolicola lawn. Since aliquots were not repeatedly drawn from the same sample, each measurement of surviving phage is an independent experiment and a repeated-measures analysis was not required. Instead, the resulting data were analyzed using a full factorial, one-way analysis of variance (ANOVA) design. Here, the log10 titer was the dependent variable, and significance tests were performed for the factors of time, treatment, and their interaction. The data show that the log10 phage titer was significantly affected by time (F ⫽ 181.093, df ⫽ 3, P ⬍ 0.0001), treatment (F ⫽ 48.948, df ⫽ 1, P ⬍ 0.0001), and their interaction (F ⫽ 5.993, df ⫽ 3, P ⬍ 0.0076). Closer inspection of the data reveals that both copper and steel treatments showed declines in titers, but only over the first hour of exposure (Fig. 1). Unpaired t tests were used to compare declines in titers over time for

FIG. 1. Bacteriophage ⌽6 inactivation on copper or steel coupons. Log10 bacteriophage ⌽6 titers (PFUs/ml) are plotted against time (minutes) of exposure to copper or steel. One hundred microliters from a common high-titer lysate was placed on a coupon of either copper or steel (control). At the indicated times, the coupon was rinsed with 1 ml LB broth, and the titer of the rinse was determined on a P. syringae pv. phaseolicola lawn. Each exposure time for each treatment was replicated 3 times. Error bars represent 1 standard errors (SE). The data indicate that most inactivation occurs in the first hour, which coincides with the period when phages are still in solution. After the lysates dried out (⬃60 min), inactivation is significantly reduced, suggesting that water mediates copper activity. The difference between steel and copper killing is likely due to the effect of copper over that of desiccation.

6880

LI AND DENNEHY

APPL. ENVIRON. MICROBIOL.

TABLE 2. Unpaired t tests of phage ⌽6 log10 titers over time for copper and steel treatmentsa Treatment

Copper

Steel

Time points (min)

df

t value

P value

0, 60 0, 120 0, 180 60, 120 60, 180 120, 180

4 4 3 4 3 3

29.647 26.759 31.121 ⫺0.107 2.052 1.964

⬍0.0001 ⬍0.0001 ⬍0.0001 0.9198 0.1325 0.1443

0, 60 0, 120 0, 180 60, 120 60, 180 120, 180

4 4 3 4 3 3

12.059 8.593 12.203 0.305 1.066 0.563

0.0003 0.0010 0.0012 0.7755 0.3647 0.6128

a Phage lysate was placed on a copper or steel coupon for a given duration and then washed off with LB. Data show that a plateau was reached at approximately 60 min, which coincides with the time at which lysates dried out on copper or steel coupons.

bacteriophage mortality (1, 15). Moreover, since phage titer did not decline after the sample dried out, we further hypothesize that the presence of water mediates phage inactivation due to copper, at least for the phage ⌽6. Finally, we conclude that the difference in inactivation between the steel treatment and the copper treatment during the first hour of the experiment was due to the effect of copper. Thus, during the first hour, desiccation resulted in a 1-log decline in titers, whereas copper caused a 2-log decline in titers. Once lysates desiccated, no further inactivation was observed (Table 2). To eliminate desiccation as a factor, we conducted further experiments by adding 1 M copper sulfate (CuSO4) solution to high-titer phage lysates to a final concentration of 5 mM. We found that this treatment had varied effects on different phage strains. For example, the dsDNA phages T4, ␭, and PRD1 were unaffected or slightly affected by copper treatment (Table 3 and Fig. 2). In contrast, the dsDNA phage PM2, the dsRNA phages ⌽6 and ⌽8, the ssDNA phage ⌽X174, and the ssRNA phage PP7 all showed steep declines in log10 titers over time (Table 3 and Fig. 2). Generally, lipid-containing phages were more susceptible than non-lipid-containing phages, regardless of the type of genetic material they possessed. While dsDNA phages were generally not susceptible to this concentration of CuSO4, the lipid-containing dsDNA phage PM2 was highly vulnerable. Three lipid-containing phages (⌽6, ⌽8, and PM2) were inactivated by CuSO4, but a fourth lipid-containing phage (PRD1) was not. In addition, phages with single-stranded nucleic acid showed strong susceptibility to copper treatment. However, further experiments make clear that these differences are due largely to copper sulfate concentrations. In another experiment, log10 phage ␭ titers showed significant declines over time in 20 mM CuSO4 solution but not in controls (F ⫽ 166.609, df ⫽ 3, P ⬍ 0.0001) (Fig. 3). These declines in phage ␭ viability at this CuSO4 concentration were comparable to the responses showed by ⌽X174 at a much lower CuSO4 concentration.

TABLE 3. Full factorial ANOVA statistics for the effect of copper treatment on bacteriophage survival (log titer) over timea Factor

df value

␭

Strain

Time Treatment Interaction

3 1 3

1.123 6.697 2.108

0.3604 0.0164 0.1269

⌽6


3 1 3

10.831 53.315 8.832

⬍0.0001 ⬍0.0001 0.0001

PRD1


3 1 3

0.168 0.348 0.129

0.9043 0.5635 0.9413

⌽X174


3 1 3

30.424 233.370 87.762

⬍0.0001 ⬍0.0001 ⬍0.0001

T4


3 1 3

17.092 53.410 10.131

⬍0.0001 ⬍0.0001 0.0006

PP7


3 1 3

67.171 377.575 62.313

⬍0.0001 ⬍0.0001 ⬍0.0001

PM2


3 1 3

14.328 123.512 14.748

⬍0.0001 ⬍0.0001 ⬍0.0001

⌽8


3 1 3

296.707 76.041 33.054

⬍0.0001 ⬍0.0001 ⬍0.0001

a

F value

P value

See Materials and Methods for experimental details.

DISCUSSION Copper has a long history of usage in sanitary and medical contexts because of its antimicrobial properties (5, 9). However, exactly how copper kills microbes is still a mystery. It has been suggested that copper damages lipids, proteins, and nucleic acids via the production of reactive hydroxyl radicals. In the presence of oxygen, many metals react with water to produce H2O2 (11). Although Grunberg reported that a smaller amount of H2O2 was recovered experimentally from copper treatments than from treatments with other metals (11), it is likely that the H2O2 was then depleted through a Fenton-type reaction to produce reactive hydroxyl radicals (9, 25). Copper can also generate H2O2 by reacting with sulfhydryls, such as cysteine and glutathione. Hydroxyl radicals are well known to damage cellular components via the oxidation of lipids and proteins (17, 37). Copper can also directly damage cellular components. For example, copper rapidly inactivates the catalytic clusters of dehydratases, a family of enzymes that are critical in central catabolic and biosynthetic pathways (17). In another study, copper was found to rapidly inactivate HIV protease (16). It is clear that copper can cause damage to many biologically critical molecules, but which type of molecule is the primary target of copper toxicity? Although DNA is susceptible to copper fragmentation (35), it is unlikely to be the primary target, because copper cannot readily access the DNA until after cell death (9, 18, 31, 35). A second line of thought is that,

VOL. 77, 2011


6881

FIG. 3. Phage ␭ inactivation in 20 mM copper sulfate solution. Log10 bacteriophage titers (PFUs/ml) are plotted against time (minutes) for ␭ phage exposed to either 20 mM copper sulfate or control treatments. Here, 10 ␮l of 1 M CuSO4 solution was added to 500 ␮l bacteriophage lysate (replicated 4 times per time point). At the appropriate time, lysate was serially diluted and plated on the appropriate host. Plaque counts were used to estimate titers in samples. Titers at each time point are from independent experiments, i.e., samples were not repeatedly drawn from the same treatment. Titers at time 0 represent four estimates from an untreated lysate. Error bars are 1 SE.

FIG. 2. Bacteriophage inactivation in 5 mM copper sulfate solution. Log10 bacteriophage titers (PFUs/ml) are plotted against time (minutes) for eight phage types exposed to either 5 mM copper sulfate or control treatments. In these experiments, 5 ␮l of 1 M CuSO4 solution was added to 1 ml bacteriophage lysate (replicated at least 3 times per time point). At the appropriate time, lysate was serially diluted and plated on the appropriate host. Plaque counts were used to estimate titers in samples. Titers at each time point are from independent experiments, i.e., samples were not repeatedly drawn from the same treatment. Titers at time 0 represent three or more estimates from an untreated lysate. For ⌽8, no phages were recovered after 60 min of exposure, and the plot is not in error. Error bars are 1 SE. Asterisks mark comparisons where CuSO4 and control treatment mean bacteriophage titers were found to be not significantly different (P ⬎ 0.05) based on unpaired, two-way Student’s t tests.

at high concentrations, copper influx overwhelms homeostatic copper extrusion mechanisms (24, 26, 32). Intracellular copper buildup then interferes with biologically critical metabolic pathways, leading to cell death (17). More recently, it has been proposed that copper directly damages cellular lipid membranes, causing membrane ruptures and the loss of the proton

motive force (9, 31, 33). Evidence supporting this conjecture is that the speed of killing seems to preclude intracellular copper activity (31). Our experiments using phages removed intracellular damage as a consideration. Most phages consist of a nucleic acid genome surrounded by a protective shell consisting of repeating protein subunits (capsomers). Rarely, a phage may have an envelope consisting of phospholipids. We ascertained whether lipid-containing phages were more sensitive to copper than phages consisting of a protein capsid alone. In addition, we determined the susceptibility to copper of phages containing single- or double-stranded RNA or DNA. Our first experiments involved placing phage ⌽6 lysate directly on copper or steel (control) coupons and then assaying phage survival. Our results showed that there was a 3-log decline in phage viability following exposure to copper but only a 1-log decline on steel (Fig. 1). However, this decline was not linear over time; rather, it occurred in the first hour of exposure (Table 2). We later realized that our lysates dried out in the first 60 min; thus, it appears that liquid mediates copper killing, at least for phage ⌽6. Given this observation, our interpretation of the results is that two phenomena affected phage survival: desiccation and copper. Desiccation was responsible for a 1-log decline in viability on steel and copper, while copper exposure added an additional 2-log decline in the copper treatments. Because of these results, we concluded that future experiments should be done in aqueous solutions containing copper sulfate. In subsequent experiments, a broad range of phages were exposed to copper at a concentration of 5 mM, a level previously shown to kill a wide range of microbes (7, 13, 22). Our results showed that lipid-containing phages were more susceptible to copper than non-lipid-containing phages (Fig. 2 and Table 3). The dsDNA phages, ␭ and T4, were relatively unaffected by copper treatment, whereas the lipid-containing phages (⌽6, ⌽8, and PM2) showed marked declines over time. The one exception was phage PRD1, but PRD1’s lipid membrane is internal to a protective protein coat (30). How-

6882

LI AND DENNEHY

ever, so too is phage PM2’s lipid membrane. This result is puzzling, as PM2 and PRD1 share the same general architecture. Why, then, should PM2 be highly susceptible but not PRD1? Two non-lipid-containing phages, ⌽X174 and PP7, did not fit the observed pattern and showed considerable mortality, although not quite to the same degree. We note that these phages do not have dsDNA genomes (⌽X174 and PP7 have ssDNA and ssRNA genomes, respectively). Interestingly, two phages, ⌽6 and ⌽8, both members of the Cystoviridae, showed divergent mortality patterns. ⌽8 showed an almost 8-log decline in 60 min, whereas ⌽6 showed a more modest 4-log decline over 3 h. Although these phages are distantly related to each other among the few Cystoviridae known, they do share a number of similar features, including segmented dsRNA genomes, protein capsid structures, and lipid envelopes. They differ due to membrane structure and highly divergent protein sequences (12). Anecdotally, ⌽8 is said to be a more fragile phage and requires frequent replenishment of frozen stocks. It would be interesting to determine how differences between the two contribute to differences in copper survival. Additional experiments were done with the ␭ phage in higher concentrations of copper. These experiments showed that this dsDNA phage was indeed sensitive, showing a 4-log decline over several hours. It would seem that most, if not all, viruses are sensitive to copper, at least at high concentrations. Our results are qualitatively similar to those of other studies of copper-induced virus mortality. Yamamoto et al. were the first to observe phage inactivation by copper when an aliquot of MS2, an ssRNA phage, was left in a metal container (36). After a short time, Yamamoto et al. were unable to recover viable phage from the container. Subsequent experiments showed an almost 4-log decline in MS2 titers over 3 h, with similar results for another ssRNA phage, f2. An ssDNA phage, S13, was sensitive but not quite to the same degree. The dsDNA phages, T1 and T4, were resistant to copper exposure. Hwang et al. reported similar results for the MS2 phage, but like the Yamamoto experiments, the concentration of Cu2⫹ ions was not determined, because the assays were done on copper coupons (14). Sagripanti et al. reported that lipid-containing viruses (⌽6, Junin virus, herpes simplex virus) were more sensitive to copper solutions than non-lipid-containing viruses (T7, ⌽X174) (29). In addition, the RNA viruses, ⌽6 and Junin virus, were more sensitive than the DNA viruses, T7 and herpes simplex virus. Interestingly, in higher copper concentrations, T7 was also susceptible, much like ␭ was in our studies. Another study reported that the enveloped H9N2 influenza virus showed steep declines in viability in copper solutions (13). Taken together, our results and those of others show that viruses are sensitive to copper in solution but not to the same degree. Viruses with RNA genomes and lipid envelopes tend to be most sensitive to copper, followed by ssDNA viruses. dsDNA viruses tended to be more resilient, except if they too possessed a lipid envelope. In addition, it seems that the activity of copper against viruses is mediated by the presence of water.

APPL. ENVIRON. MICROBIOL. ACKNOWLEDGMENTS We thank Corinne Michels, James Carpino, Nidhi Gadura, and two anonymous reviewers for critically reading the manuscript. Leonard Mindich, Holly Wichman, Ing-Nang Wang, and the Felix d’Herelle Reference Center for Bacterial Viruses supplied phage and host strains. Funding was provided by the National Science Foundation (grants DEB-0804039 and MCD-0918199 to J.J.D.) and the Professional Staff Congress of the City University of New York. Some of the work described here was conducted in part with equipment in the Core Facility for Imaging, Cellular and Molecular Biology at Queens College. REFERENCES 1. Abad, F. X., R. M. Pinto, and A. Bosch. 1994. Survival of enteric viruses on environmental fomites. Appl. Environ. Microbiol. 60:3704–3710. 2. Aruoma, O. I., B. Halliwell, E. Gajewski, and M. Dizdaroglu. 1991. Copper ion dependent damage to the bases in DNA in the presence of hydrogen peroxide. Biochem. J. 273:601–604. 3. Bertani, G. 2004. Lysogeny at mid-twentieth century: P1, P2, and other experimental systems. J. Bacteriol. 186:595–600. 4. Bertani, G. 1951. Studies on lysogenesis. 1. The mode of phage liberation by lysogenic Escherichia coli. J. Bacteriol. 62:293–300. 5. Borkow, G., and J. Gabbay. 2004. Putting copper into action: copper-impregnated products with potent biocidal activities. FASEB J. 18:1728. 6. Casey, A. L., et al. 2010. Role of copper in reducing hospital environment contamination. J. Hosp. Infect. 74:72–77. 7. Elguindi, J., J. Wagner, and C. Rensing. 2009. Genes involved in copper resistance influence survival of Pseudomonas aeruginosa on copper surfaces. J. Appl. Microbiol. 106:1448–1455. 8. Espejo, R. T., and E. S. Canelo. 1968. Properties of bacteriophage PM2—a lipid-containing bacterial virus. Virology 34:738. 9. Grass, G., C. Rensing, and M. Solioz. 2011. Metallic copper as an antimicrobial surface. Appl. Environ. Microbiol. 77:1541–1547. 10. Gratia, A. 1936. Numeric relations between lysogenic bacteria and bacteriophage particles. C. R. Seances Soc. Biol. Fil. 122:812–813. 11. Grunberg, L. 1953. The formation of hydrogen peroxide on fresh metal surfaces. Proc. Phys. Soc. Lond. Sec. B 66:153–161. 12. Hoogstraten, D., et al. 2000. Characterization of Phi 8, a bacteriophage containing three double-stranded RNA genomic segments and distantly related to Phi 6. Virology 272:218–224. 13. Horie, M., et al. 2008. Inactivation and morphological changes of avian influenza virus by copper ions. Arch. Virol. 153:1467–1472. 14. Hwang, J. Y., T. H. Ryu, Y. D. Lee, and J. H. Park. 2009. Viability loss of bacteriophage MS2 exposed to bronze alloy Yugi. Food Sci. Biotechnol. 18:1022–1026. 15. Iriarte, F. B., et al. 2007. Factors affecting survival of bacteriophage on tomato leaf surfaces. Appl. Environ. Microbiol. 73:1704–1711. 16. Karlstrom, A. R., and R. L. Levine. 1991. Copper inhibits the protease from human immunodeficiency virus-1 by both cysteine-dependent and cysteineindependent mechanisms. Proc. Natl. Acad. Sci. U. S. A. 88:5552–5556. 17. Macomber, L., and J. A. Imlay. 2009. The iron-sulfur clusters of dehydratases are primary intracellular targets of copper toxicity. Proc. Natl. Acad. Sci. U. S. A. 106:8344–8349. 18. Macomber, L., C. Rensing, and J. A. Imlay. 2007. Intracellular copper does not catalyze the formation of oxidative DNA damage in Escherichia coli. J. Bacteriol. 189:1616–1626. 19. Marais, F., S. Mehtar, and L. Chalkley. 2010. Antimicrobial efficacy of copper touch surfaces in reducing environmental bioburden in a South African community healthcare facility. J. Hosp. Infect. 74:80–82. 20. Mikolay, A., et al. 2010. Survival of bacteria on metallic copper surfaces in a hospital trial. Appl. Microbiol. Biotechnol. 87:1875–1879. 21. Noyce, J. O., H. Michels, and C. W. Keevil. 2007. Inactivation of influenza A virus on copper versus stainless steel surfaces. Appl. Environ. Microbiol. 73:2748–2750. 22. Patikarnmonthon, N., et al. 2010. Copper ions potentiate organic hydroperoxide and hydrogen peroxide toxicity through different mechanisms in Xanthomonas campestris pv. campestris. FEMS Microbiol. Lett. 313:75–80. 23. Quaranta, D., et al. 2011. Mechanisms of contact-mediated killing of yeast cells on dry metallic copper surfaces. Appl. Environ. Microbiol. 77:416–426. 24. Rensing, C., and G. Grass. 2003. Escherichia coli mechanisms of copper homeostasis in a changing environment. FEMS Microbiol. Rev. 27:197–213. 25. Rodriguez-Montelongo, L., L. C. De La Cruz-Rodriguez, R. N. Farias, and E. M. Massa. 1993. Membrane-associated redox cycling of copper mediates hydroperoxide toxicity in Escherichia coli. Biochim. Biophys. Acta 1144:77–84. 26. Rouch, D., J. Camakaris, B. T. O. Lee, and R. K. J. Luke. 1985. Inducible plasmid-mediated copper resistance in Escherichia coli. J. Gen. Microbiol. 131:939–943. 27. Sagripanti, J. L., P. L. Goering, and A. Lamanna. 1991. Interaction of

VOL. 77, 2011

28.

29.

30.

31. 32.

copper with DNA and antagonism by other metals. Toxicol. Appl. Pharmacol. 110:477–485. Sagripanti, J. L., L. B. Routson, A. C. Bonifacino, and C. D. Lytle. 1997. Mechanism of copper-mediated inactivation of herpes simplex virus. Antimicrob. Agents Chemother. 41:812–817. Sagripanti, J. L., L. B. Routson, and C. D. Lytle. 1993. Virus inactivation by copper or iron ions alone or in the presence of peroxide. Appl. Environ. Microbiol. 59:4374–4376. San Martin, C., et al. 2002. Minor proteins, mobile arms and membranecapsid interactions in the bacteriophage PRD1 capsid. Nat. Struct. Biol. 9:756–763. Santo, C. E., et al. 2011. Bacterial killing by dry metallic copper surfaces. Appl. Environ. Microbiol. 77:794–802. Solioz, M., H. K. Abicht, M. Mermod, and S. Mancini. 2010. Response of Gram-positive bacteria to copper stress. J. Biol. Inorg. Chem. 15:3–14.


6883

33. Suwalsky, M., B. Ungerer, L. Quevedo, F. Aguilar, and C. P. Sotomayor. 1998. Cu2⫹ ions interact with cell membranes. J. Inorg. Biochem. 70:233– 238. 34. Toyokuni, S., and J. L. Sagripanti. 1996. Association between 8-hydroxy-2⬘deoxyguanosine formation and DNA strand breaks mediated by copper and iron. Free Radic. Biol. Med. 20:859–864. 35. Warnes, S. L., S. M. Green, H. T. Michels, and C. W. Keevil. 2010. Biocidal efficacy of copper alloys against pathogenic enterococci involves degradation of genomic and plasmid DNAs. Appl. Environ. Microbiol. 76:5390–5401. 36. Yamamoto, N., W. Haller, and C. W. Hiatt. 1964. Mechanism of inactivation of bacteriophages by metals. Biochim. Biophys. Acta 91:257. 37. Yoshida, Y., S. Furuta, and E. Niki. 1993. Effects of metal chelating agents on the oxidation of lipids induced by copper and iron. Biochim. Biophys. Acta 1210:81–88.