Citation
The inactivation of bacteria and viruses by hydrogen peroxide

Material Information

Title:
The inactivation of bacteria and viruses by hydrogen peroxide
Creator:
Asghari, Abdolkarim, 1959-
Publication Date:
Language:
English
Physical Description:
x, 145 leaves : ill. ; 29 cm.

Subjects

Subjects / Keywords:
Hydrogen peroxide ( lcsh )
Bactericides ( lcsh )
Viruses -- Inactivation ( lcsh )
Genre:
bibliography ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1993
Bibliography:
Includes bibliographical references (leaves 131-144).
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Abdolkarim Asghari.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
The University of Florida George A. Smathers Libraries respect the intellectual property rights of others and do not claim any copyright interest in this item. This item may be protected by copyright but is made available here under a claim of fair use (17 U.S.C. §107) for non-profit research and educational purposes. Users of this work have responsibility for determining copyright status prior to reusing, publishing or reproducing this item for purposes other than what is allowed by fair use or other copyright exemptions. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder. The Smathers Libraries would like to learn more about this item and invite individuals or organizations to contact the RDS coordinator (ufdissertations@uflib.ufl.edu) with any additional information they can provide.
Resource Identifier:
030158317 ( ALEPH )
30301415 ( OCLC )

Downloads

This item has the following downloads:


Full Text
THE INACTIVATION OF
BACTERIA AND VIRUSES BY HYDROGEN PEROXIDE
By
ABDOLKARIM ASGHARI
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 1993




To my beloved wife Zohreh and my two beautiful daughters Hamideh and Freshteh. And, finally, to my parents Javad and Atkeh.




ACKNOWLEDGEMENTS
I gratfully acknowledge the help and guidance given me by the members of my committee, Dr. Samuel R. Farrah, Dr. Lonnie 0. Ingram, Dr. Gabriel Bitton, Dr. Francis C. Davis, and Dr. Philip M. Achey. I am forever indebted to my committee chairman, Dr. Farrah, who showed me the true meaning of the relationship between the student and his major professor. His broad knowledge of science and expertise in my subject area were invaluable assets for the success of my work.
In addition, I extend thanks to the entire personnel of
the Department of Microbiology and Cell Science who have helped me in many ways to make my stay at the University of Florida an unforgettable experience.
Finally, I extend thanks to my family, especially my wife and my parents, for the love and support they have given me.




TABLE OF CONTENTS
ACKNOWLEDGEMENTS .
LIST OF TABLES vi
LIST OF FIGURES vii
ABSTRACT viii
CHAPTER 1 LITERATURE REVIEW . . . . . . . 1
Mechanism of Inactivation by Hydrogen Peroxide 4
Hydrogen Peroxide Damage to the Genetic
Materials .. 6
Role of Defence Mechanism Against Hydrogen
Peroxide. i. 8
Role of Fe and OH* in H 2 0 2- Induced
Cell Killing 13
Hydrogen Peroxide Effect on Other Cell
Components 15
Mechanism of Inactivation by Other Oxidizing Agents 18
Chlorine 18
Ozone 20
Mechanism of Inactivation by Ultraviolet Light . 22 Mechanism of Inactivation by Metal Ions . . . 24
Silver 24
Copper 25
Synergistic Effect of Hydrogen Peroxide and Other
Agents. 26 Ultraviolet Light 26
Iron 27
Copper 28
CHAPTER 2 INACTIVATION OF BACTERIA AND VIRUSES BY
HYDROGEN PEROXIDE 30
Mechanism of Inactivation by Hydrogen Peroxide . 30 Materials and Methods . . . . . . . 32
Results 35
Inactivation of Bacteria and Viruses with
Hydrogen Peroxide 35
iv




Factors Influencing Inactivation Rate by
Hydrogen Peroxide.................39
Effects of pH, temperature, and inactivation
media...........................39
Effects of chemical agents on hydrogen
peroxide toxicity.................40
Hydroxyl radical scavengers ...........41
Combined Inactivation of E. coli and MS2 by
Hydrogen Peroxide and Ultraviolet Light ... 42
Comparison Between Hydrogen Peroxide and
Chlorine Inactivation...............42
Combined Inactivation of E. coli and MS2 with
Hydrogen Peroxide and Silver Nitrate .. ......43
Discussion.......................43
Tables and Figures...................51
CHAPTER 3 SELECTIVE RECOVERY OF BACTERIOPHAGES BY
USING HYDROGEN PEROXIDE ...............85
Literature Review....................85
Materials and Methods..................88
Results........................110
Discussion......................113
Tables and Figures..................118
CHAPTER 4 CONCLUSION....................128
REFERENCE LIST........................131
BIOGRAPHICAL SKETCH....................145
V




LIST OF TABLES
Table 1 Inactivation of microorganisms with H202. 51 Table 2 Effects of H202 treatment on E. coli volume 52
Table 3 Lactate dehydrogenase leakage in the extracel
lular medium from E. coli cells exposed to
hydrogen peroxide ..... ............... 53
Table 4 Effects of pH, temperature, and medium on
inactivation of MS2 and E. coli with H202 54
Table 5 Effect of chemical agents on killing of
bacteriophages by hydrogen peroxide ........ 55 Table 6 Treatment of E. coli and MS2 with hydrogen
peroxide and ultraviolet light ......... 56
Table 7 Effect of hydrogen peroxide and chlorine on
biosynthesis and activity of beta-galactosidase enzyme in E. coli ... ............ 57
Table 8 Inactivation of MS2 and E. coli with
hydrogen peroxide silver nitrate,
and cupperic chloride .... ............. .. 58
Table 9 Inactivation of bacteria and bacteriophages
by HZ02 in wastewater samples (direct assay) 97
Table 10 Combined effects of crystal violet and H202
on the indigenous bacterial population in environmental samples with high levels of
bacteria ....... ................... 98
Table 11 Reductions in bacterial numbers using H202
treatment and/or crystal violet plate ..... .. 99 Table 12 Combined effects of crystal violet and
hydrogen peroxide on the indigenous
bacteriophage population in environmental
samples with high levels of bacteria ...... .100 Table 13 Comparison of other decontamination methods
with H202-crystal violet procedure ..... i. 101
Table 14 Influence of magnesium chloride concentration
on the magnesium peroxide content of modified
diatomaceous earth ..... .............. 118
Table 15 Changes in bacterial numbers on DE filters 119
Table 16 Stability of magnesium peroxide on DE
during daily filtration of tapwater ........ .120 Table 17 Release of oxidizing power from MgO . 121
Table 18 Influence of diatomaceous earth coated with
magnesium peroxide on bacterial growth . .. 122 Table 19 Influence of solids coated with magnesium
peroxide on bacterial growth in tapwater . 123
vi




LIST OF FIGURES
Figure 1 Inactivation of bacteria and viruses by H202 60
Figure 2 Inactivation of bacteria and viruses by H202 62
Figure 3 Inactivation of bacteria by H20 ........... ....64
Figure 4 Growth of E. coli in the presence of H 02 66
Figure 5 Effect of hydrogen peroxide on cell leakage 68
Figure 6 Effect of H202 treatment on oxygen
consumption by E. coli ... ............ 70
Figure 7 Effect of chemical agents on inactivation
of MS2 with hydrogen peroxide ........... 72
Figure 8 Effect of hydroxyl radical scavengers on
hydrogen peroxide inactivation of MS2 ....... 74 Figure 9 Effect of hydroxyl radical scavengers on
hydrogen peroxide inactivation of E. coli. 76
Figure 10 Relationship of riboflavin solubility
to inactivation of MS2 by H 02 ........ 78
Figure 11 Inactivation of E. coli with hydrogen
peroxide and chlorine ... ............. 80
-Figure 12 Inactivation of MS2 with H202 and chlorine. 82
Figure 13 Combined inactivation of E. coli with
hydrogen peroxide and silver nitrate ..... 84
Figure 14 Selective recovery of bacteriophages from
raw sewage using different decontamination
methods ....... .................... 103
Figure 15 Steps involved in the hydrogen peroxidecrystal violet (HPCV) phage assay procedure. 105
vii




Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy THE INACTIVATION OF BACTERIA AND VIRUSES BY HYDROGEN PEROXIDE By
Abdolkarim Asghari
May 1993
Chairman: Dr. Samuel R. Farrah Major Department: Microbiology and Cell Science
Studies on the inactivation of bacteria and viruses by hydrogen peroxide (H202) showed that bacteriophages are relatively more resistant to H202 than bacteria. Lipidcontaining phages are inactivated at a much faster rate by H202 than non-lipid containing bacteriophages.
Leakage of bacterial cell contents suggests that the integrity of the cell membrane is disrupted after treatment cells with H202. Significant amounts of lactate dehydrogenase and Mg2 ions were released into the extra cellular medium after treatment with H202. Oxygen consumption, redox potential, as well as cell volume decreased substantially upon treating E. coli cells with H202. Lower pH, a lack of organics in the inactivation medium, and a higher temperature enhanced the detrimental activity of H202.
viii




Chaotropic agents, which weaken hydrophobic associations, increased the rate of inactivation of bacteriophages. In contrast antichaotropic agents, which strengthen hydrophobic associations, decreased the lethal effect of H202. These results suggest that hydrophobic interaction between capsid proteins influence the rate of inactivation of viruses by H202.
The use of hydroxyl radical scavengers in inactivation media reduced the rate of inactivation of bacteria but not viruses, suggesting that different mechanisms are involved in inactivation of bacteria and viruses by H202. Unlike the killing of E. coli, OH are not the major cause of virucidal action of H202 The above results suggest that oxidizing power of H202 may be directly involved in the inactivation of viruses. This virucidal action was different when compared to virucidal action of chlorine. Inactivation of viruses by chlorine is much faster than inactivation by H202*
The use of H202 in recovery of bacteriophages from natural water samples with plates containing crystal violet is proposed. The selective inactivation of bacteria in environmental samples by H202 permits recovery of bacteriophages with minimal interference from indigenous bacteria. This method is either superior or as good as the currently available procedures. The use of H202 to coat diatomaceous earth and sand with magnesium peroxide for removal/inactivation of bacteria has also been proposed. Filters made with modified solid effectively removed and ix




inactivated bacteria as compared to the controls. Modified sand and diatomaceous earth added to unchlorinated tapwater kept the water free from bacteria for at least three weeks.
x




CHAPTER 1
LITERATURE REVIEW
The germicidal properties of aqueous hydrogen peroxide have been recognized for more than a century. Dr. Samuel S. Wallian praised hydrogen peroxide in an address
to the New York State Medical Society in 1892: "One can hardly refer to the medical journals without finding enthusiastic recommendations of it as a disinfectant of rare efficiency, an antiseptic of recognized merit and a germicide of decided potency" (Block, 1991; p167). Known as oxidized water then, hydrogen peroxide has had a rocky road in its acceptance as a disinfectant; first popular, then unpopular, and now finding special application where it serves particular functions of great value (Block, 1991).
The literature contains numerous accounts of the satisfactory applications of hydrogen peroxide as a
disinfectant for inanimate materials and inert surfaces in the aseptic packaging of food and dairy products (Toledo, 1975;
Naquib and Hussein, 1972). It is also used for processing systems and spacecraft hardware (Block, 1991). It has been recommended for the disinfection of surgical implant
1




2
components, contact lenses, and many other heat labile materials (DeRezende, 1968). The vapor phase of hydrogen peroxide provides the sole killing power of a newly patented sterilization system. This system provides a rapid, lowtemperature, low toxicity sterilant that could eliminate the toxic or carcinogenic gaseous sterilants such as formaldehyde and ethylene oxide (Caputo and Odlaug, 1983; Arlene-Klapes et al., 1990). Hydrogen peroxide has been used to remove the organic impurities in water. Along with ultraviolet light, hydrogen peroxide reduced the total organic carbon (TOC) content of distilled water samples by about 88% and of tapwater by 98% (Malaiyandi et al., 1980).
Hydrogen peroxide serves as a natural disinfectant and preservative and is naturally present in milk and honey. It is present in saliva and fights bacteria in the mouth. Hydrogen peroxide in phagocytes is one of the materials involved in the killing of the ingested bacteria, and it is a weak mutagen as tested by the standard Ames test (Kensese and Smith, 1989). It is also part of the defense system of some bacteria (Block, 1991; Dubreuil et al., 1984).
Hydrogen peroxide is produced in cells by the reduction of oxygen for respiration. Hydrogen peroxide, oxygen and its other by-products such as oxygen free radicals are harmful to living cells and must be removed constantly. Oxygen free radicals such as the superoxide anion 0 and the hydroxyl radical, OH-, are linked to many diseases such as rheumatoid




3
arthritis and cancer. Proteins, lipids, and nucleic acids are all at risk of being damaged by these oxidizing agents through redox reactions.
Aerobic cells have a system for disposing of the oxygen by reducing it to water through a series of enzymatic steps (Stoklasova, 1989). Catalase and peroxidase are responsible for removal of hydrogen peroxide and are the primary defense against this biocide (Fridovich, 1978). Products of recA and polA genes are also important in protection against hydrogen peroxide in E coli. The repair mechanism of defense is distinct from the enzymatic removal of the biocide by catalase. The following diagram illustrates the univalent pathway for oxygen reduction and the enzyme involved in the catalytic scavenging of intermediates (Fridovich, 1978).
02 ------ 02 ------ H202 -------------- OH -------H20
e e+2H e H20 e+H
02" + 02 + 2H+ --- H202 + 02 (Superoxide dismutase) 2H202 --- 2H20 + 02 (Catalase)
H202 + RH2 ---- 2H20 + R (Peroxidase)
Scavenging enzymes such as catalase, peroxidase, superoxide dismutase, and DNA repair enzymes that correct oxidative lesions are found throughout the aerobic organisms (Imlay et al., 1988; Lesko et al., 1980).
Although hydrogen peroxide has been used as a disinfectant for many years, its mechanism of action has not




4
been completely elucidated. It is proposed that hydrogen
peroxide derivatives such as hydroxyl radicals are responsible for cell damage inside the cell. Two modes of cell killing by
hydrogen peroxide can be distinguished using E. coli K-12. Mode one killing is accomplished with concentration of <5 mM (maximum inactivation at 23 mM) and mode two with >10 mM hydrogen peroxide. An intervening zone of partial resistance
is detectable between the two regions of killing. The two modes of cell killing are distinguishable kinetically and metabolically (Imlay and Linn, 1986; 1987; 1988; Imlay et al. 1988). Much of the data support the theory that damage to DNA is the major cause of inactivation by this biocide when low concentrations of the reagent are used. The lethal lesion of mode two killing is not known.
Mechanism of Inactivation by Hydrogen Peroxide
The characterization of cell damage caused by oxygen species has been complicated by the variety of direct and indirect effects observed, and the ultimate oxidant and the specific killing lesion in bacterial cell have not yet been
established. Oxidative damage to nucleic acids and other macromolecules such as lipids and proteins have been studied extensively (Imlay and Linn, 1986; 1988). DNA damage is shown by accumulation of single strand DNA fragments (Breimer and
Lindahl, 1985; Hoffman et al., 1984). Membrane damage from




5
oxygen species is observed as an accumulation of lipid peroxides, the loss of the diffusion barrier to membrane impermeable markers, and cell lysis (Seeberg and Steinum, 1980). A variety of enzymes have been shown to be inactivated by hydrogen peroxide (Aguirre and Hansberg, 1985; Hodgson, 1975; Kim et al.). These include superoxide dismutases from both eukaryotic and prokaryotic sources (Yamakura and Suzuki, 1986; Borders and Fridovich, 1985).
Exposure of logarithmical growing Escherichia coli to hydrogen peroxide leads to two kinetically distinguishable modes of cell killing. Mode one killing is pronounced at low (less than 5 mM) concentrations of hydrogen peroxide and is caused by DNA damage. Mutant strains that are defective in recombinational or excision DNA repair are especially sensitive to mode one killing. Indicators of DNA damage, mutagenesis and the induction of the SOS DNA repair regulon, accompany challenges of cells at low doses. (Imlay and Linn, 1986; 1988). Mode two killing requires a relatively high, higher than 10 mM dosage, and the site (sites) of the toxic injury has not been established. There is also an unexplained area of partial resistance to hydrogen peroxide between 5 and 10 mM concentrations.




6
H 202 Damage to the Genetic Materials
Genetic and physical effects of hydrogen peroxide and other oxidative stresses such as gamma radiation upon cellular DNA have been well studied. Oxygen radicals may attack DNA at either a sugar or a base giving rise to a large number of products. Attack at a sugar ultimately leads to sugar fragmentation, base loss and a strand break with a terminal fragmented sugar residue (Imlay and Linn, 1988). Such single strand breaks are accumulated during exposure of bacteria and mammalian cells to hydrogen peroxide, 02", gamma radiation, or ozone (Imlay and Linn, 1986; Seeberg and Steinum, 1980; Prise et al., 1989; Birnboim and Kanabus-Daminska, 1985; Sawadaishi et al., 1985; Hagensee and Moses, 1986).
The repair of apurinic/apyrimidinic (AP) sites in DNA is initiated by incisions catalyzed by specific endonucleases. In E. coli, exodeoxyribonuclease III, the product of xthA' gene, constitutes about 90% of the bacterial AP activity. E. coli mutants lacking exonuclease III (xth) are exceptionally sensitive to 5 mM of hydrogen peroxide. They are killed by hydrogen peroxide at 20 times the rate of wild type and at 3 to 4 times the rate of recA mutants (Demple et al., 1983b). Hydrogen peroxide induces single strand breaks in bacterial DNA and these breaks can be repaired by polymerase I-dependent and recA dependent pathways similar to those for the repair of X-ray induced single strand breaks (SS). recA recB, polAl, and




7
recA mutant strains of E. coli are not able to repair SS breaks efficiently, and are more sensitive to hydrogen peroxide than the wild type (Ananthaswamay, 1977). Strains deficient in RecA protein, exonuclease III, exonuclease V (RecBC enzyme), or DNA polymerase I are especially vulnerable to mode one killing and are not different from wild type in reacting to mode two killing (Imlay and Linn, 1987). Hagensee et al. (1987) showed the involvement of polymerase III in the H202-induced DNA damage repair during replication in Escherichia coli.
Prechallenging cell growth in anoxic medium significantly amplifies mode one killing of both repair deficient and repair proficient strains. Cells gain back their normal sensitivity to mode one killing upon dilution into oxygen saturated medium. As shown by chloramphenicol inhibition, under anaerobic condition cells synthesize proteins that enhance hydrogen peroxide mediated DNA damage (Imlay and Linn, 1986). Only actively metabolizing cells are subject to mode one killing, because starving cells lose their sensitivity to hydrogen peroxide. The cells regain their sensitivity upon addition of glucose to the medium (Imlay and Linn, 1986). The blocking of respiration chemically by cyanide or genetically by inactivating the NADH dehydrogenase gene dramatically sensitize E. coli to killing by H202. Desensitization rapidly follows after removal of cyanide from the medium (Imlay and Linn, 1986).




8
Bacterial cells behave differently upon challenge with concentrations of hydrogen peroxide representatives of the two modes of cell killing. Mode-two-killed cells remained permanently unit-sized, but mode-one-treated cells underwent extensive filamentation and did not septate (Brandi et al., 1989b; Imlay and Linn, 1986). A growth lag followed by a period of filamentation is observed with the recovery of the treated cell with low doses of hydrogen peroxide. The authors conclude that the lag is for repair of cell damage and filamentation is the result of DNA damage and subsequent block of septation which ends up with mode one killing (Imlay and Linn, 1986). Brandi et al. (1989b) observed similar results under light microscope and concluded that the size of filament increases with increase in exposure time (Brandi et al., 1989b). No such phenomenon is observed with mode two killing (Brandi et al., 1989b). Exposing the cells to H202 (mode two cell killing) resulted in decrease in cell volume as measured directly and by optical density (Brandi et al., 1989b).
Role of Defense Mechanism AQainst H202
Hydrogen peroxide induces SOS response in bacterial cells by increasing the rate of synthesis of recA protein (Imlay and Linn, 1987). SOS mutants are hypersensitive to hydrogen peroxide (Demple et al., 1983b; Chen and Bernstein, 1987). The sensitivity of recA strains to mode one killing suggest that




9
hydrogen peroxide activates the SOS response. This response protects the cells against mode one killing through an enhanced ability to carry out recombinational DNA repair (Chen and Bernstein, 1987; Demple and Halbrook, 1983a; Imlay and Linn, 1987). Salmonella typhimurium, Escherichia coli and other bacteria become resistant to killing by hydrogen peroxide and other oxidants when prechallenged with nonlethal dose of H202 (Chritsman et al., 1985; Bol and Yasbin, 1990; Demple and Halbrook, 1983a, Chen and Bernstein, 1987; Demple et al., 1983b; Plateaue et al., 1987). During adaptation to hydrogen peroxide, 30 proteins are induced. These investigators have identified a regulon under the control of the oxyR locus, which apparently encodes a positive effector of this response. Gene products are overproduced after exposure to inducing levels of hydrogen peroxide. Among the gene products overproduced are the scavengers of active oxygen species, catalase, superoxide dismutase and peroxidase. This regulon governs the protective response of E coli and S. typhimurium by adaptation to low levels of hydrogen peroxide and thus subsequent resistance to higher doses of the agent. Cells adapted to H202 are resistant to a variety of other agents causing oxidative damage, as well as to heat killing. Greenberg and Demple (1989) also reported that the redoxcycling agents menadione and paraquat (which generate superoxide) induced protection mechanism, which overlaps with one induced by H202. Jenkins et al. (1988) have shown that E.




10
coli K-12 responds to glucose or nitrogen starvation with synthesis of 30 proteins. Several of these proteins are also involved in protection against heat and hydrogen peroxide challenge. These investigators also showed that similar proteins are produced with heat and hydrogen peroxide adaptation. Unlike the SOS induction which enhances repair system, the oxyR regulon exerts its protective effect primary through an enhanced ability to scavenge partially reduced oxygen species. The two stress responses are induced by hydrogen peroxide and act by nonoverlapping processes to protect against lethal doses of hydrogen peroxide (Imlay and Linn 1986; 1987; Chritsman et al., 1985).
The enzymes catalase, peroxidase, and superoxide dismutase can be considered the primary defenses of the cell against H202 and other oxygen metabolites (Fridovich, 1978; Starke and Farber, 1985b; Hassan and Fridovich, 1979; Bayliss and Waites, 1981). The DNA repair systems also play a major role in protecting cells against hydrogen peroxide (Demple and Halbrook, 1983a). Carlssen and Carpenter (1980) and others (Ananthaswamy and Eisenstark, 1977; Winquist et al., 1984) studied the importance of different protection mechanisms against hydrogen peroxide. These studies showed that there is no correlation between the sensitivity to H202 and the cellular level of the enzymes catalase and superoxide dismutase (Carlssen and Carpenter, 1980). Miyasaki et al., (1985) reported that endogenous catalase activity is an important




11
determinant of resistance to bactericidal effects of H20. recA mutants were more sensitive than other mutants. These investigators concluded that a functional recA gene product is more important than the enzymes involved in scavenging and removing harmful oxygen species (Carlesson and Carpenter, 1980; Ananthaswamy and Eisenstark, 1977). Yet others (Yonei et al., 1987) have found that E. coli UM1, with no catalase activity is 20 fold more sensitive than wild type strains. Other researchers have also found that catalase negative mutants are 50-60 times more sensitive to H202 than their wild type counterparts (Loewen, 1984). Exposing the cells to nonlethal concentrations of hydrogen peroxide induces both DNA repair system and scavenging enzymes (Winquist et al., 1984; Demple and Halbrook, 1983a) which protect cells against higher doses of the biocide.
Catalase negative cells with normal DNA repair system and repair deficient mutants with normal catalase level were more sensitive to H202 than their respective wild type (Yonei et al., 1987) These results showed that catalase and DNA repair systems have distinct roles in protection against lethal damage of H202 to E. coli (Yonei et al., 1987) even though catalase activity is a major contributor to the cell's defense system (Miyasaki et al., 1985). Complex, nonsynthetic, growth medium such as heart infusion, but not defined medium, can change the concentration of catalase, and a high catalase




12
concentration protects the cells against higher doses of H202 (Bayliss and Waites, 1981).
Quin 2 (an intracellular calcium chelator) but not EGTA (an extracellular calcium chelator) prevents hydrogen peroxide induced DNA breakage and cytotoxicity in mammalian cells (Cantoni et al., 1989a). Quin 2 does not prevent formation of hydroxyl radicals. These authors suggest that the majority of DNA damage induced by hydrogen peroxide in intact cells at 370C are not caused by direct attack of hydroxyl or of secondary radicals on DNA. Instead, radicals elicit a complex cascade of metabolic secondary reactions which in part require Ca2 and ultimately leads to DNA breakage (Cantoni et al., 1989a)
The formation of a spectrin-hemoglobin complex following treatment of red cells with hydrogen peroxide is shown to be associated with alterations in cell shape and a decrease in membrane flexibility. SH groups of spectrin are involved in crosslinking since their blockage reduced the complex formation and decreased lipid peroxidation. Oxidized hemoglobin is also required for crosslinking (Snyder et al., 1988). These data indicate that lipid peroxidation and oxidation of sulfhydryl groups, induced by hydrogen peroxide treatment, resulted in deleterious effects on erythrocytes' membrane properties (Snyder et al., 1988).




13
Role of Fe3 and OH in H202-Induced Cell Killing
It has been shown that by increasing the iron content of Staphylococcus aureus and other bacterial cells, their susceptibility to hydrogen peroxide is dramatically enhanced (Repine et al., 1981b; Hoepelman et al., 1990; Sambri et al., 1991). Iron in microorganisms facilitates bactericidal mechanism by reacting with H202 to form more toxic hydroxyl radicals (-OH). Production of hydroxyl radicals occurs by the Fenton reaction
Fe2 + H202 ----- Fe3 + *OH + OH
Repine et al. (1981a) further showed that hydroxyl radical scavengers such as thiourea, dimethyl thiourea, sodium benzoate, and dimethyl sulfoxide inhibited hydrogen peroxide mediated killing of S. aureus. The role of hydroxyl radicals in bacterial cell killing by hydrogen peroxide has been studied by several investigators (Brandi et al., 1987; Brandi et al., 1989a; Hoepelman et al., 1990; Van Slays et al., 1986; Kleiman et al., 1990; Terada et al., 1991). They concluded that hydroxyl radicals are involved in production of mode two but not mode one killing by H202. These investigators found that hydroxyl radical scavengers, thiourea, ethanol and dimethyl sulfoxide and the iron chelator, desferrioxamine, did not affect the survival of cells exposed to 2.5 mM H202 (mode one killing). In addition, cell vulnerability to the same concentration of hydrogen peroxide was independent of the




14
intracellular iron content. In contrast, mode two lethality (i.e. cell killing generated by a concentration of H202 higher than 10 mM) was markedly reduced by -OH scavenger iron chelators and was augmented by increasing the intracellular iron content (Brandi et al., 1989a). These investigators proposed the following mechanism for H202 induced cytotoxicity in E. coli.
H202 --------- --- ? DNA ------------* MODE ONE KILLING
LIPIDS
H202 -------Fe--- OH" -- PROTEINS ------- MODE TWO KILLING
NUCLEIC ACIDS
Hydroxyl radicals seemed also to be the major lethal means against spores as hydroxyl radical scavengers inhibited spore lysis. Studies by Cantoni et al. (1989c) showed that wild type and superoxide dismutase mutants display a markedly different sensitivity to both modes of lethality produced by H202. They proposed a hypothetical enzyme that generates 02. This enzyme is active at H202 concentrations <5 mM but high concentrations of H202 (>5 mM) inactivate the enzyme (Cantoni et al., 1989c). Cantoni et al. (1989c) concluded that mode one cell killing is produced by superoxide anions whereas mode two cell killing is the consequence of the OH* attack. Brandi et al. (1988) suggested that superoxide ions are involved in regeneration of divalent iron (to allow further Fenton reactions). Starke and Farber (1985a) also show the need for ferric iron and superoxide ions for hepatocytes killing by




15
hydrogen peroxide. Again the free radical scavengers mannitol, thiourea, and benzoate protect cells against H202 (Starke and Farber, 1985a). Berglin et al. (1985; 1984) demonstrated that iron sulfide is more efficient than ferrous iron in catalyzing the formation of hydroxyl radicals similarly L-cysteine enhanced H202-induced killing by 100-fold.
H 202 Effects on other Cell Components
In addition to damaging DNA, hydrogen peroxide also damages other vital cell components such as lipids (Flenley, 1987) and proteins (Richards et al., 1988), especially in the cell membrane of both mammalian (Vander Zee et al., 1985) and bacterial (Brandi et al., 1989b) cells. This damage is usually followed by cell leakage and subsequent cell lysis. Several enzymes were shown to be inactivated by hydrogen peroxide (Yamakura and Suzuki, 1986; Steinman, 1982; Hodgson, 1975; Kim et al., 1985; Aguirre, 1986). One of the major targets of H202 attack is unsaturated lipids. These can undergo peroxidation which disrupts membrane structure and function (Girotti and Thomas, 1984; Schraufstatter et al., 1986). Girotti and Thomas (1984) have also made several observations in studying the lethal effects of H202 on erythrocytes. These workers observed the following: (a) efflux of low molecular weight molecules such as glucose-6-phosphate and Na (b) leakage was stimulated by Fe3 and chelating agents inhibited the efflux,




16
(c) both H202 and 02- were required since catalase;
(d)superoxide dismutase inhibited lipid peroxidation, and (e) hydroxyl radical scavengers e.g. ethanol, mannitol, and choline provided no protection against marker efflux and lipid peroxidation (Girotti and Thomas, 1984). Others have shown that oxidation of membrane unsaturated fatty acids is not an essential component of the toxicity of H202 to E. coli (Ohlrogge and Kernan, 1983). Vander Zee et al., (1985) also observed lipid peroxidation and K' leakage in erythrocytes. No correlation was found between lipid peroxidation and K leakage. These investigators showed that K leakage is due to the SH group oxidation, because diamide decreased the leakage by oxidizing the same SH groups (Vander Zee et al., 1985). Diamide oxidized the SH groups to disulfides but, H202 SH oxidation, in addition to disulfides, also yielded sulfenic and sulfonic acids. This further SH oxidation resulted in greater membrane permeability (Van der Zee 1985). Brandi et al. (1989b) found that E. coli cells exposed to high concentrations of H202 (>10 mM) show a reduction in cell volume as measured microscopically, and release of the enzyme lactate dehydrogenase into the culture medium.
Curran et al. (1940) found the greatest killing power of hydrogen peroxide against Bacillus spores at pH 3 and the least at pH 9. Baldry (1983) showed that varying the pH from 5-8 made no difference in treating several bacteria with hydrogen peroxide. Only one strain, Pseudomonas aeruginosa,




17
needed 10 times less amount to inhibit growth at pH 5 than pH 8 (Block, 1991). Brandi et al., (1987) and Cantoni et al., (1989a) studied the effects of temperature and anoxia on E. coli killing induced by hydrogen peroxide. They showed that low oxygen levels decrease the vulnerability to mode two treatment. Oxygen tension was not relevant as far as mode one killing is concerned (Brandi et al., 1987). Treating cells with H202 at 370C was more toxic than at 40C (Fiorani et al., 1990). These investigators concluded that damage at 370C may be indirectly mediated by temperature dependent metabolic events (Cantoni et al., 1989a; 1989b).
Domingue et al. (1988) compared the effects of the oxidizing power of chlorine, ozone, and hydrogen peroxide on Legionella pneumophilla. They found that in contrast to both ozone and chlorine, H202 inactivation required much higher concentrations. As with chlorine, there was a dose-response relationship between the concentration of H202 and the rate of inactivation by this biocide. These investigators found no decrease in oxidizing potential of H202 after 24 h (Dominique et al., 1988; Yoshpe-Purer and Eylan, 1968). Persistent killing effects of up to 13 d was observed in some cases (Yoshpe-Purer and Eylan, 1968).




18
Mechanism of Inactivation by Other Oxidizing Agents
Chlorine
Chlorine is an effective oxidizing agent and is the most widely used disinfectant. Dychdala (1991) listed several factors which determine the antimicrobial action of chlorine. pH has the greatest influence on the antimicrobial activity of chlorine in solutions. It is known that the disinfecting efficiency of chlorine decreases with an increase in pH and vice versa, which is parallel to the concentration of undissociated hypochlorous acid.
Sharp et al. (1980) suggested that the following
variables must be taken into consideration when studying the viral disinfection by chlorine, i) the time of contact of virus with the free chlorine, ii) the temperature of the reaction, iii) the total ionic strength of the reaction mixture, iv) the pH of the reaction mixture, v) the type of buffer used to maintain the pH, vi) the total chlorine concentration and the relative amounts of HOCI and OCI, and vii) the state of aggregation of the virus (Sharp et al., 1980).
Chlorine exists in solution as a mixture of hypochlorous acid and hypochlorite ion.
HOCIl < ---- > H + OCI"




19
The dissociation of hypochlorous acid depends on pH. The higher the pH, the greater the concentration of dissociated hypochlorous acid. Hypochlorous acid is much more potent than hypochlorite, so chlorine is more effective at lower pH. Higher temperatures increase the killing efficiency of chlorine (Dychdala, 1991). The presence of organic compounds reduces chlorine efficiency, which is defined as chlorine demand, and as the result chloramines are formed. This problem can be overcome by breakpoint chlorination, which is the application of a sufficient amount of chlorine to satisfy the initial chlorine demand. Adding more chlorine beyond the breakpoint increases free available chlorine species.
Dychdala (1991) proposed that chlorine rapidly inhibits some key enzyme systems essential to life by oxidizing the SH groups of these enzymes. Based on the studies of (Linquist et al., 1976) chlorine changes the membrane permeability of E. coli, and allows efflux of macromolecules such as protein and nucleic acid. Hurst et al. (1991) Observed that HOClpromoted inactivation was accompanied by extensive inhibition of respiration in E. coli and P. aeruginosa. Surface proteins could also be a major target for chlorine action (Roller et al., 1980). Bacteria are inactivated primarily through
irreversible sulfhydryl oxidation (Roller et al., 1980). Others supported this idea by showing that cleaved viral RNA is released from poliovirus capsid following treatment with chlorine (Taylor Butler, 1982; O'Brien and Newman, 1979). In




20
another study the viral sedimentation coefficient changed from 156 s for native poliovirus to 80 s after treatment with chlorine, due to release of its RNA (Alvarez and O'Brien, 1982b). Chlorine concentrations of less than 0.8 ppm resulted in inactivation of viruses without major structural changes, whereas chlorine concentrations in excess of 0.8 ppm resulted in leaks in the capsid protein and loss of RNA (Alvarez and O'Brien, 1982a).
Dychdala (1991) summarized the factors influencing the efficiency of chlorine as follows: pH, the concentration of free chlorine in the form of hypochlorous acid; temperature; and the presence of organic materials in the solution.
Ozone
Ozone is a strong and effective biocide. Lower concentrations of ozone and shorter contact times are required for treating samples than with chlorine and other disinfectants. Ozone is more effective than other oxidizing against resistant organisms such as amoebic cysts and viruses (Kim et al., 1980). Less than 1 ppm ozone inactivates 5-7 logs of virus in 5 s (Kim et al., 1980). Ozone was the most potent biocide against LeQionella pneumophilla when compared to chlorine and hydrogen peroxide (Domingue et al., 1988). The other investigators have found that RNA enclosed in the phage coat was inactivated less by ozone than were whole phages, but




21
more inactivated than naked RNA. They concluded that subcellular components of a microorganism can be more
resistant than the whole organism, and that the inactivation of a microorganism does not necessarily denature the genetic materials in it (Kim et al., 1980).
Kim et al. (1980) detected structural changes in the protein coat of bacteriophage f2 using electron microscopy after ozonation. They suggested that ozone breaks the capsid proteins into subunits and releases the RNA which then may become damaged afterward. As a result, the adsorption of virus particles to host pili was disrupted. Loss or severe reduction of L. pneumophilla unsaturated fatty acids was observed with ozone treatment (Roehm et. al., 1971). With its rapid rate of diffusion through bacterial cell walls and oxidizing power, ozone reacts with a wide range of organic compounds such as membrane-associated proteins containing sulfhydryl groups (Menzel, 1971).
Roy et al. (1980) observed that damage to the viral nucleic acid was the major cause of the inactivation of poliovirus 1. RNA was damaged by ozone concentration less than 0.3 ppm within 2 min of contact time. Furthermore, ozone treatment resulted in strand scission of supercoiled pBR322 plasmid (open the circular DNA) (Sawadaishi et al., 1985), and altered two of the four polypeptide chains present in the viral protein (Halliwell and Gutteridge, 1984).




22
Mechanism of Inactivation by Ultraviolet Light
Ultraviolet (UV) light includes electromagnetic radiations that fall in the wavelength band between 200 and 400 nm. It falls in between the energies of X-rays and the shortest wavelength of light visible to the human eye. The germicidal effects of UV light are limited to only a specific region of the UV light spectrum, with 265 nm being the most effective wavelength.
Studies of mutagenic effects and retardation of cell division suggested that these conditions are caused by the effect of UV on nucleic acids (Block, 1991). UV acts on cellular DNA primarily by producing links between adjacent pyrimidines on a DNA strand to form dimers. Dimers consist of primarily two thymine residues. Cytosine-thymine and cytosinecytosine dimers have also been identified in the cells exposed to UV-light, although these dimers are identified less frequently than thymine-thymine dimers. (Block, 1991). Dimers interfere with transforming ability of bacterial DNA and replication and also lead to cell death. According to Rahn and Landry, (1973), UV irradiation of DNA results in the formation of various kinds of photoproducts that may have a disruptive influence on the local integrity of the DNA structure. The cross-linking of DNA and protein plays a significant role in the killing of UV-irradiated cells (Block, 1991). Kelland et al. (1984) reported membrane damage, as shown by 8Rb leakage,




23
after treating E. coli K-12 cells with near UV irradiation. Shechmeister, (1991) reported that the survival curve of viruses with single-stranded DNA or RNA is different from the survival curve for double-stranded DNA.
Several factors influence bacterial sensitivity to UV. The more important are (1) pH; (2) bacterial growth stage; (the greatest sensitivity being in the logarithmic growth phase); (3) the presence of spores, which are about twice as resistant as vegetative cells and (4) presence of particles and turbidity of the liquid.
There are basically three different repair mechanisms. In excision repair in which, a damaged area of DNA is cleaved by an endonuclease and excised by an exonuclease, DNA polymerase then fills the gap. Photoreactivation repair involves an enzyme DNA photolyase. This enzyme, which is activated by visible light, is capable of binding to dimers and spliting dimers. During excision repair the UV-induced DNA damage is recognized and removed immediately by the enzyme. Postreplication repair, in contrast, is carried out after
replication. This is the least reliable repair mechanism, since the repair after replication often operates erroneously (Darnell et al., 1986).




24
Mechanism of Inactivation by Metal Ions
Silver
Silver compounds have been reported to bind with
bacterial DNA. Silver displaces the hydrogen bonds at specific sites and may prevent replication of the DNA and subsequent
cell division (Fox, 1978; Fox and Modak, 1974) .Richards (1981) studied the effect of silver nitrate on actively
dividing Pseudomonas aerucginosa cells and on the infectivity of T2 DNA. They concluded that silver caused cross linking of the bacteriophage DNA helix. Microbial DNA and envelope were damaged by silver ion (Richards, 1981). It is the
concentration of silver ions, not their physical nature, that is responsible for disinfecting capability of silver, because silver action is independent of the way in which silver is introduced into water, e. g. soluble silver salt, metallic silver, etc., (Modak and Fox, 1985; Just, 1936). The binding property of the silver ion is very important. These ions can
complex with proteins and nucleic acids by serving as an oxidative catalytic surface between them (Philips, 1958). silver ions can irreversibly bind to the bacterial cells (Thurman and Gerba, 1989). Silver ions attached to the surface of a container can keep the water in the container free of bacteria, because the silver ion would still be able to bind to bacterial surface and inactivate it (Just, 1936).




25
Three possible mechanisms for inhibition of bacterial cells by silver have been proposed: (1) interference with electron transport, (2) binding to the DNA, and (3) intercalation with the cell membrane (Tilton and Rosenberg, 1978). Silver ions can easily form insoluble compounds with anions, sulfhydryl groups, and many biological materials, such as enzymes which are responsible for disinfectant activity of silver. No protein or nucleic acid leakage has been reported upon treatment of bacterial cells with silver, which suggests that cell lysis did not occur. Higher doses of silver are needed to inactivate viruses, probably because it is harder to denature the viral protein coat than to oxidize the complexed sulfhydryl groups (Rahn, 1973).
Copper
Copper and other metal ions may function as either lewis acids or bases when present as metal complexes (Thurman and Gerba, 1989). Using their protons, they can facilitate hydrolysis or nucleophilic displacement and make a bond available for nucleophilic attack by hydroxyl radicals. Copper attacks respiratory enzymes in the cell membrane of E. coli by binding thiol or other groups on protein molecules. The injured cells decrease the oxygen use and increase use of fermentation pathways during the recovery (Plastourgen and Hoffman, 1984) Low level of Cu2+ in chlorine-free distribution




26
water caused injury of coliform population upto 64%. Copper concentrations 0.025 and 0.050 mg/l caused over 90% injury within 6 and 2 days, respectively. Studies of the metabolism of injured E. coli cells indicated that the respiratory chain is at least one site of the copper-induced damage in injured cells (Domek et al., 1984).
Synergistic Effect of Hydrogen Peroxide and other Agents
Hydrogen peroxide shows synergism with both physical and chemical factors. Its antimicrobial activity is enhanced when it is used with transition metals such as iron, copper, silver, and physical agents like ultraviolet light and ultrasonic energy (Block, 1991; Bayliss and Waites, 1980; 1979).
Ultraviolet Light
Hydrogen peroxide is highly effective against spores when used simultaneously with ultraviolet light (Bayliss and
Waites, 1980; 1979; Waites et al., 1979; Halliwell and Gutteridge, 1984). Bacillus subtilis spores were killed 2000 fold faster when treated with H202 and UV simultaneously (Bayliss and Waites, 1979). Irradiation of E. coli and Streptococcus faecalis with 254 nm UV light and incubation with 1% H202 bring 99.99% inactivation in just 30 s (Bayliss




27
and Waites, 1979). Waites et al. (1981) showed that the greatest kill of B. subtilis spores in the presence of H202 was accomplished with UV irradiation around 270 nm. Their results also showed that the action of UV light is not directly on the spore DNA but is related to the free hydroxyl radicals produced from H202 that are close to or within the spores (Waites et al. 1979).
Another physical agent that has shown synergism with hydrogen peroxide is ultrasonic energy. Ultrasonic energy is thought to disperse and agitate the cell aggregates, increasing surface contact with the disinfectant, increasing the permeability of the cell membrane to the disinfectant, and accelerating the interaction between the disinfectant and the cell components.(Ahmad and Russel, 1975).
Iron
Growing the cells in the presence of high concentration of iron made S. aureus cells more susceptible to H202 (Repine 1981). Hydroxyl radical formation was enhanced due to high iron content of these cells. Iron catalyzed the formation of hydroxyl radicals through Fenton reaction (Repine et al., 1981a ). This reaction has been discussed earlier in this section. These investigators further showed that hydroxyl radical scavengers inhibited the cell killing by H202.




28
Copper
Treating spores of Clostridium bifermentans with copper made the spores 3000 fold more susceptible to H202 (Bayliss, 76). Dithiothreitol (DTT) treated spores of Clostridium perfringens were shown to be inactivated by H202 500 fold more than untreated controls (Waites et al., 1979; Halliwell and Gutteridge, 1984). It was proposed that DTT removes the protein coat that protect the spores from H202, and that copper increases the rate of breakdown of H202 and the rate of cleavage of peptide bonds by H202 (Block, 1991).
Goals and Prospects of the Present Study
In this study the impact of hydrogen peroxide treatment on E. coli and bacteriophage MS2 under different experimental conditions has been investigated. The results indicate that bacteria and viruses are inactivated at different rates and probably different mechanisms are involved in inactivation of bacteria and viruses by hydrogen peroxide. The survey of treatment of microorganisms with H202 revealed that gram negative bacteria are relatively more susceptible to hydrogen peroxide than gram-positive ones. Inactivation of E. coli and MS2 by hydrogen peroxide and chlorine have been compared. The use of hydrogen peroxide in selective recovery of bacteriophages from natural water samples and in modification




29
of diatomaceous earth and sand for removal/ inactivation of bacteria were also investigated.




CHAPTER 2
INACTIVATION OF BACTERIA AND VIRUSES BY H202
Mechanism of Inactivation by H202
E. coli cells exposed to H202 are inactivated by at least two lethality modes distinguishable by metabolic, kinetic, and genetic criteria (Imlay and Linn, 1986). Mode one killing occurs at low (less than 5mM) concentrations of H202 and requires metabolically active cells. DNA damage appears to be the site for mode one killing, since strains deficient in RecA protein, exonuclease III, or RecBC enzyme are especially vulnerable to this mode of killing. Both mutagenesis and induction of SOS DNA repair systems accompany challenges at H202 concentrations representing mode one killing. Mode two cell killing doses not require metabolically active cells and occurs at H202 concentration of more than 10mM. DNA repair deficient mutant cells are inactivated at the same rate by mode two doses as wild type strains, which indicates that DNA damage is not the major target of H202 attack (Imlay and Linn, 1987).
There is a considerable amount of evidence indicating that hydroxyl radicals produced via the Fenton reaction are a
30




31
major cause of H202-induced lethality (Brandi et al., 1989b; Repine et al., 1981a). This has been demonstrated by the fact that both iron chelators and hydroxyl radical scavengers reduce the level of active hydroxyl radicals and protect cells from killing by H202. Brandi et al. (1989) showed that hydroxyl radicals are involved in the production of mode two but not mode one killing by H22*
In this study the impact of hydrogen peroxide treatment on laboratory microorganisms was surveyed. The results showed that gram negative bacteria are more susceptible to H202 treatment than gram positive bacteria, and bacteriophages are relatively more resistant than bacteria in general. Lipidcontaining phages were more susceptible to H202 than were nonlipid-containing bacteriophages. The inactivation of E. coli and bacteriophage MS2 by H202 was compared under different experimental conditions. The results indicate that E. coli is inactivated at a much faster rate than MS2. A survey of the effects of chemical agents on inactivation of bacteria and viruses with H202 was done. These results revealed that hydroxyl radicals are not the major cause of inactivation of bacteriophages, (as they are in killing of E. coli), and probably different mechanisms are involved in inactivation of the two organisms.
The effects of physical factors, type of inactivation media, and ultraviolet light on inactivation of microorganisms by H202 has been studied. The release of cytoplasmic components




32
(as the sign of cell membrane damage) into the culture medium and changes in cell volume upon treatment of E. coli cells with H202 was also investigated.
Inactivation of bacteria and viruses by two oxidizing compounds H202 and chlorine, was compared to investigate any differences between the mechanism of action of each biocide. Alasri et al. (1992) reported that E. coli, S. aureus, and P. aeruginosa are inactivated at faster rate by chlorine than hydrogen peroxide. The combined treatment of hydrogen peroxide with several other agents such as silver nitrate, cupperic chloride, and ultraviolet light are investigated. These studied were carried out to show any possible synergistic effects between these agents and hydrogen peroxide. There is synergistic effect between two chemicals if the combined treatment shows greater toxicity (more than the sum of the individual treatments (Marking and Dawson, 1973).
Materials and Methods
Materials; the following chemicals were purchased from Fisher Scientific Co. (Fairlawn, N.J.) and Sigma Co. (St. Louis, Mo) : Na2HPO4, NaSCN, sodium citrate, thiourea, urea, tween 80, dimethylsulfoxide, sodium chloride, sodium fluoride, sodium trichloroacetate, sodium hypochlorite, silver nitrate, and hydrogen peroxide.




33
Cultures and growth conditions; the following bacterial strains used in this study were obtained from the American Type Culture Collection; Escherichia coli C-3000 (ATCC
15597), Escherichia coli K-12 (ATCC 10240), recA', Staphylococcus aureus (ATCC 27660), Bacillus cereus (ATCC 11778), Streptococcus faecalis (ATCC 19344), Streptococcus saprophyticus (ATCC 15305), Mycobacterium smegmatis (ATCC 10143), Pseudomonas aeruqinosa (ATCC 10145), Vibrio cholerae (ATCC 14035). The bacteriophages used in this study and their host cultures were MS2, T2, OX174 (E. coli C-3000), P22, PRD-1 Salmonella typhimurium (ATCC 19585), and 06 (Pseudomonas syringae ATCC 21781). NBY medium (ATCC medium 815) was used for growing P. syringae cultures were incubated at 280C. All other bacteria were grown in 3% tryptic soy broth at 370C.
Inactivation studies; bacteria and viruses were suspended in either 0.05 M sodium phosphate, pH 7, or 3% tryptic soy broth, pH 7, for inactivation studies. Concentrated viruses were resuspended in 0.05M Na2HPO4 pH 7 and diluted to produce a final concentration of approximately l0s PFU/ml. Dilutions of hydrogen peroxide were prepared fresh from a 50% (w/v) stock solution. All experiments were carried out at room temperature and pH 7 unless specified otherwise. To remove the residuals of hydrogen peroxide excess catalase (0.05 mg/ml, final concentration) added to the inactivation medium and/or samples were diluted appropriately.




34
Cell leakage study; E. coli cells were treated in the presence of various concentrations of H202. Cells were pelleted by centrifugation for 2 min in an Eppendorf microcentrifuge tube, and the supernatants analyzed. The concentration of magnesium ions was measured using the "60-Second Magnesium" reagent system supplied by the American Monitor Corporation (Indianapolis, IN). Leakage values are expressed as a
percentage of that released into the supernatant by vortex mixing suspensions for 30 sec with chloroform (0.2ml per 2 ml of suspension).
Enzyme assays; beta-galactosidase activity and synthesis were measured according to Dutton et al. (1988). Lactate dehydrogenase activity was estimated according to Beutler (1975).
Oxygen consumption experiment; oxygen consumption was measured after treating E. coli cells with various concentrations of H202, washing the cells with growth medium, and resuspending the cells in fresh growth medium, by using a Gilson respirometer.
Measure of oxidizing power; the amount of oxidizing power in H202 and chlorine solution was measured using an iodometric method adopted from "Standard Methods for the Examination of Water and Wastewater" 17th edition (APHA-AWWA-WPCF, 1989).
Determination of riboflavin solubility; to determine the solubility of riboflavin, an excess amount of riboflavin was added to the test solutions. Undissolved riboflavin was




35
removed by centrifugation for 10 min at 3000 rpm. The solutions were then diluted in distilled water and dissolved riboflavin was determined by measuring the absorbance at 444 nm using spectrophotometer (Bausch & Lamb, Spectronic). Results were compared to the solubility of riboflavin in deionized water as the control.
Study of synergism; bacteria and viruses were tested with hydrogen peroxide or with hydrogen peroxide in the presence of ultraviolet light, cupperic chloride, or silver nitrate. For Ag tests the solutions were collected in Chambers' solution (Chambers et al., 1962).
Statistical analysis; each experiment was done at least three times in duplicate. The student t-test (p=0.05) was used for comparing the mean values. All statistical analysis were carried out using "instat" program for IBM.
Results
Inactivation of Bacteria and Viruses with H202
The effects of H202 treatment for 1 h at room temperature (0.1% H202) on different bacteria and viruses is presented in (Table 1). Non-spore forming Gram positive bacteria, acid fast bacteria, and bacteriophages are inactivated at a slower rate than gram negative bacteria. Bacteriophages are relatively more resistant to H202 than bacteria (Figure 1 and 2).




36
Bacteriophages had an average of 75% survival compare to only 5.3% survival for bacteria (Table 1). The lipid-containing phages such as 06 and PRD-l were much more susceptible to H202 than other bacteriophages (Figure 1 and 2). Their survival rate is intermediate between that of gram positive bacteria and other non-lipid phages, (Figures 1 and 2).
The toxicity of hydrogen peroxide to E. coli was investigated under different experimental conditions. Two modes of killing were apparent; mode one killing refers to lethality at lower concentrations of H202 (0.01% or less, >2.9 mM) and mode two lethality at which killing occurs at higher concentrations (0.05% or more). When repair deficient strains of E. coli K-12 were treated with H202, they were much more susceptible to mode one killing than wild type. H 202
concentration representing mode two lethality inactivated both wild type and repair deficient mutant strains at the same rate, (Figure 3).
In order to investigate the possible damage to bacterial cell membranes as a result of H202 treatment, the appearance of cytoplasmic components in the culture medium and loss of cell volume were studied. E. coli cells were allowed to grow either in the presence or absence of 0.01% or 0.1% (representative of mode one and mode two lethality respectively) H202 and, at various time intervals, the
absorbance at 550 nm was measured along with cell viability (Table 2). In addition cells were analyzed at different




37
intervals microscopically to monitor any morphological
changes. Changes were observed as regards the morphology of E. coli cells treated with H 2 0 2 The changes were dependent on the amount of oxidant used in the treatment. Cells remained unit sized 3 h after treatment with 0.1% H 2 0 21 (data not shown). Results depicted in Figure 4 shows treatment with 0.01% H 2 0 2
resulted in a growth lag followed by a partial recovery of cell growth. Exposure to 0.1% H 2 0 2 also produced a continuous, time dependent decrease in optical density (Table 2). In fact, OD values were reduced to about 50% of the initial OD (immediately before adding H 2 0 2 ) after 5 h treatment with 0.1% H 2 02 (Figure 4). The viable count indicated great initial loss in cell colony forming ability (>50%) after treatment with 0.01% H 2 0 2 which was followed by partial recovery of viable but damaged cells.
In similar experimental conditions, cells were centrifuged immediately after addition of H 2 0 2 and after 5 h treatment. Examination of pellet sizes visually revealed that the pellet size was substantially smaller after 5 h treatment with H 2 0 2 Results in Table 2 shows that the pellet volume reduced more than 30% after 5 h. This indicates the loss of cell components as the result of possible cell membrane damage.
In an attempt to further investigate the damage to cell membrane integrity, the leakage of molecules upon treatment with H 2 0 2 was studied. Mg2+ leakage happened only when cells




38
were treated with H2 02 concentrations representing mode two and killing was dependent on the exposure time and H 20 2
concentration (Figure 5) Studies on lactate dehydrogenase leakage revealed that the enzyme accumulated in a time and concentration dependent fashion in the supernatant (Table 3). The leakage of materials with low levels of H 2O02 treatment was negligible. One can conclude that even though there might be
some membrane damage with lower levels of the oxidant, the damage is repairable, or there is too little membrane damage
to cause leakage of detectable levels of cell components. When higher concentrations of H 20 2 were used the cells were not able to repair the initial damage and, the longer the cells were exposed to H 20 2, the more materials were leaked out.
The ability of E. coli to consume oxygen was tested in
different experimental conditions. Results in Figure 6 show cells treated with 0.01% H 20 2 partially recovered their oxygen consumption ability whereas the cells subjected to 0.1%H202
never regained their ability to consume oxygen. The same cultures were tested for the effects of H 20 2 on their electron transport system by their ability to reduce tetrazolium chloride before and after treatment with H 202.' Cells retained 51% of their ability to reduce tetrazolium chloride after treatment with 0.1% H2,2 compare to 79% reduction after treatment with 0.01% H 2 02 (Figure 6).




39
Factors Influencing Inactivation Rate by H 202
Effects of PH, temperature, and inactivation media
H2 02 toxicity is affected by the pH of the solution, with greater inactivation under acidic conditions. As it is shown on Table 4, H 20 2 is much more effective at pH 5 than at pH 7 and pH 9. Table 4 shows the effect of heat on inactivation of
E. coli and MS2 with hydrogen peroxide. More than 90% of E. coli and almost all of MS2 survived 0. 1% H 20 2 treatment for upto 1 h at 40C. At 25 and 370C, however, inactivation
patterns were similar for MS2 and not for E. coli. H202 was
more toxic for E. coli at 370C than 250C.
The oxidizing power of hydrogen peroxide in solution is
influenced by the amount of organic matter presents in the solution. There is a reciprocal relationship between the amount of oxidizing power and the inactivation efficiency of
hydrogen peroxide. Table 4 shows the relationship between the relative amount of oxidizing power in different solutions and
the survival of MS2. Phosphate buffer with no organics present generates the highest oxidizing power and greatest inactivation. It is quite possible that some of the oxidizing power is consumed by the organic matter present in the inactivation medium. So the consumed oxidizing power will not be available for inactivation activity.




40
Effects of chemical agents on H202 toxicity
The effects of chaotropic and antichaotropic agents on the lethality of H202 was studied using bacteriophages OX174, p6, and MS2. The list and concentrations of chemicals used in these experiments are given in figure 7. Results depicted in, Figure 7 show the survival of MS2 after treatment with 0.1% H202 in the presence of different chemical agents. The correlation between the inactivation rate and the degree of hydrophobicity (as measured by riboflavin solubility) is shown in, Figure 10. Sodium thiocyanate along with other chaotropic agents, urea, tween 80, and thiourea, which weaken hydrophobic association, significantly increased the rate of inactivation of MS2 by hydrogen peroxide noticeably. In contrast sodium citrate and other antichaotropic agents, which strengthen hydrophobic associations, decreased the lethal effect of H202. A very high correlation (r2=0.988) was observed between the effects of salts on inactivation of MS2 by H202 and their effects on hydrophobic interactions (Figure 10). However, this correlation was not as high when non-salt chemicals, urea, thiourea, and tween 80 were taken into account (0.550). Details of statistical analysis are discussed in the materials and methods section.
Results of testing other bacteriophages such as OX174 and 06 with hydrogen peroxide in the presence of salts and other agents is given in Table 5. pX174 (a single-stranded DNA




41
bacteriophage) and 06 (a lipid-containing virus) inactivation studies have similar profiles as MS2 which is a singlestranded RNA virus (Table 5).
Hydroxyl radical scavengers
The role of hydroxyl radicals in the inactivation of MS2 by H202 was studied using the hydroxyl radical scavengers thiourea, DMSO, and ethanol. Results shown in Figure 8 indicate that 35 mM thiourea significantly enhanced the virucidal action of H202 on MS2. 9% ethanol significantly decreased the inactivation of MS2 by H202. Ethanol at high concentration reacts with H202 and decreases the oxidizing power of H202. 50 mM DMSO did not have any significant effect on H202 inactivation. These results indicate that hydroxyl radicals are not the major lethal species against bacteriophages. The effects of DMSO and thiourea on riboflavin solubility (as the measure of their effects on hydrophobicity of the solution) were also studied (Figure 10). These results indicated that thiourea weakens the hydrophobic interactions and so increases H202 lethality. The addition of DMSO does not change the hydrophobicity of the solution, therefore it does not have any effect on H202 killing efficiency.




42
Combined Inactivation of E. coli and MS2 by H202 and UV-Lictht
The combined effects of H202 and UV light on bacteriophage MS2 and E coli were studied using 366 nm UV irradiation and 0.01% H202 (w/v) for 30 min. Table 6 depicts the results of this study. While 0.01% H202 or UV treatment are almost ineffective against MS2, their combined treatment is significantly different (at p=0.05 level) from individual treatments against this bacteriophage. The combined treatment of E. coli with UV light and H202 is not greatly different from the individual treatment of bacterium, showing that the effects of UV and H202 are additive.
Comparison Between H202 and Chlorine Inactivation
The oxidizing power of H202 and chlorine were compared in their ability to inactivate E. coli and bacteriophage MS2. Concentrations of the biocides with equal oxidizing power as measured by an iodometric method were used in inactivation studies. The rate of inactivation of E. coli by chlorine was slightly faster than the inactivation by H202 (Figure 11). On the other hand the inactivation rate of bacteriophage MS2 was much faster with chlorine than H202 (Figure 12). The biosynthesis and activity of enzyme beta-galactosidase were measured (as described by Dutton et al., 1988) after treating E. coli cells with chlorine and hydrogen peroxide. Chlorine




43
decreased the enzyme biosynthesis 4 times more than hydrogen peroxide. At the same time the enzymatic activity of beta galactosidase went down 12-fold faster by hydrogen peroxide than chlorine (Table 7).
Combined Inactivation of E. coli and MS2 with H202 and AgNO3
In order to find any synergism between H 202 and silver nitrate E. coli and MS2 were treated with AgNO3 and in combination with H202. Table 8 and Figure 13 shows the results of these experiments. Statistical analysis of these data indicates that the treatment of both MS2 and E. coli with a combination of 0.25ppm AgNO3 and 0.01% H202 significantly reduced the viability of both cultures when compared with individual treatments. Comparison of the concentrations needed to produce 50% mortality (Marking and Dawson 1973) reveals that the effect of mixtures of AgNO3 and H202 is synergistic. This means that silver nitrate enhances antibacterial activity of hydrogen peroxide.
Discussion
A survey of microbial inactivation by H202 indicated that gram negative bacteria are more sensitive to H202 than gram positive bacteria (except B. cereus) and viruses. Lipid containing bacteriophages are inactivated at a faster rate




44
than non-lipid phages. Treatment of E. coli cells with increasing concentrations of hydrogen peroxide results in a bimodal pattern of lethality. Low concentrations (<0.01%) of the oxidant produce mode one killing which can be distinguished from mode two killing (generated by concentrations of H202 higher than 0.1%) since the first mode of lethality unlike the second, requires active cellular metabolism and is enhanced in DNA repair-deficient strains. A possible explanation for the differential inactivation can be due to the differences in diffusion rates of H202 through bacteria and viruses membrane barriers. Apparently lipids are a major factor in the susceptibility of an organism to an oxidizing agent.
In this study different components of E. coli were examined for possible damage as the result of H202 treatment. Leakage studies shows that the integrity of the cell membrane is interrupted after treating E. coli cells for more than 2 h. Significant amount of the enzyme lactate dehydrogenase and Mg2+ ions show up in the inactivation medium. Others have shown leakage of molecules such as glucose-6-phosphate, Na and K+ (Girotti and Thomas, 1984; Vander Zee et al., 1985). Cell volume decreased as measured by a reduction in dry weight and a decrease in absorbance at 550 nm. Brandi (1989a) reported a reduction in cell size after similar treatment. Oxygenic respiration is also disrupted by H202 treatment. The ability of E. coli cells to utilize oxygen as well as the




45
integrity of their electron transport systems are adversely affected after treatment with H202. Other enzymes are also targets of H202 treatment. The enzyme beta galactosidase activity as well as its synthesis are inhibited by the action of H202.
The toxicity of H202 to bacteria and viruses increases as the pH of the inactivation mixture decreases. However the pH factor is not as pronounced as it is with other disinfectants such as phenol and chlorine (Block, 1991). H202 is more effective against bacteria and viruses at 370C than 40C. The presence of organic materials in the inactivation media reduces the lethality by consuming part of the oxidizing power generated by H202. The above physical and chemical factors are also important in disinfection activity of other oxidizing agents such as chlorine (Block, 1991).
The inactivation of viruses by H202 is affected by the presence of chaotropic and antichaotropic salts, which indicates that compounds that affect hydrophobic interactions are important in the action of H202 against viruses. Chaotropic salts are likely to increase the lethality of H202 by weakening the hydrophobic associations between viral capsid proteins. On the other hand antichaotropic salts shield the virus against H202 by strengthening the hydrophobic associations. Hydrophobicity plays an important role in the interaction of viruses with other molecules as it gives the strongest




46
contribution to viral capsid protein stability (Farrah, 1981; 1982; Stryer, 1984).
The catalytic role of iron in bactericidal action of H202 through generation of hydroxyl radical scavengers has already been discussed (Repine et. al., 1981a). It has been shown that hydroxyl radicals promote E. coli cell killing when higher doses of H202 (>20 mM) are used (Brandi et. al., 1987). These investigators concluded that hydroxyl radicals are required for mode two cell killing by H202. Although hydroxyl radical scavengers decrease the toxicity of H202 to bacteria, they act differently for viruses. In fact thiourea, a hydroxyl radical scavenger, enhances the lethality of H202 against MS2, while it decreases the antibacterial action of hydrogen peroxide by as high as 96% (Repine et al., 1981b; Brandi et al., 1989a), Figure 9. DMSO reduced the antibacterial activity of H202 substantially (Repine et al., 1981b; Brandi et al., 1989a), Figure 9, but made no difference on inactivation of MS2 by this agent. These results show that different mechanisms are involved in the inactivation of viruses and bacteria. Unlike the killing action of H202 against E. coli, hydroxyl radicals are not the major cause of virucidal action of H202. Urea, a weak hydroxyl radical scavenger, inhibited the antibacterial action of H202 by 6% (Repine et al., 1981b). On the other hand, urea greatly enhanced the virucidal action of H202 on bacteriophages. These results suggest that the oxidizing power of H202 may be directly involved in the inactivation of




47
viruses, and there is (are) no intermediate species (hydroxyl radical) involved.
In contrast to the similarities between bactericidal activities of chlorine and H202, viruses are inactivated at different rate by these two biocides. It is not known why chlorine inactivates bacteria and viruses faster than H202. The rate of diffusion through natural barriers such as the cell
membrane or the virus capsid could be different for each biocide. Domingue et al. (1988) found that in contrast to ozone and chlorine, H202 inactivation requires much higher concentrations. Hydrogen peroxide inactivates gram positive bacteria, gram negative bacteria, and bacteriophages at different rates. It is possible that -the differential
inactivation could be the result of the reduced ability of hydrogen peroxide to penetrate the microorganism barriers. Hydrogen peroxide and chlorine affected enzyme synthesis and activity of E. coli at different rates which further suggests
that chlorine and hydrogen peroxide probably have different inactivation mechanisms.
Exactly how HOCl destroys microorganisms has never been
demonstrated experimentally. Oxygen liberated as the result of HOCi and other oxidizing agents such as H202 and KMnO4 activity is able to combine with components of cell protoplasm. Despite the production of large amounts of nascent oxygen by H202 and




48
KMnO4, they are not able to kill microorganisms as fast as chlorine.
Rudolph and Levine (1941) proposed that the bactericidal effect of hypochlorite is completed in two successive phases;
(1) the penetration of the germicide into the bacterial cells and (2) the chemical reaction of the germicide with protoplasm of the cell to form toxic complexes that destroy the organism. These phases could also be applied to H 202.* Similar killing mechanisms have been proposed for chlorine and H202.* These mechanisms of action include; (a) change in membrane
permeability, (b) inactivation of key enzymes required for metabolism, and (c) interference with genetic materials in the cell.
Combined treatment of E. coli and MS2 with H 2 02 and UTV light shows that UV light enhances H 2 02 inactivation of both cultures, but this synergistic effect is more pronounced for
the virus. A kill of 99.99% was produced in cell suspension of E. coli by UV irradiation at 254 nm for 30 s in the presence of 1.0% H 202 (Bayliss 1980-52). These results indicate that the synergistic inactivation of E. coli requires higher concentration of hydrogen peroxide.
Silver nitrate enhanced inactivation of bacteria and viruses by H 20 2 significantly. This effect is synergistic and not additive as analyzed according to the Marking and Dawson. (1973) method. Both silver and H 20 2 are known to attack DNA and cause subsequent inhibition of cell division (BrandI er: al.,




49
1989a; Imlay and Linn 1987; Richards, 1981). No cell lysis is observed with silver but silver ions can intercalate with the cell membrane. This intercalation may enhance the damage to the cell membrane by H202.
In summary, two modes of lethality postulated for the action of H202 on E coli with H202. H202 inactivates bacteria and viruses at different rates. Bacteria and viruses are apparently inactivated by different mechanisms as the proposed role of hydroxyl radicals in bactericidal action of hydrogen peroxide in not a major factor in inactivation of bacteriophages by hydrogen peroxide. Lipid is a major factor in the enhanced inactivation of lipid-containing bacteriophages as compared with the non lipid phages. It is concluded that combined damage to different components of E. coli occurs, which is followed by disruption of the cell membrane as the cause of death. Hydrophobic interactions influence the effects of H202 on MS2 phage. The stronger the hydrophobic association, the more resistant MS2 is against H202. It is the oxidizing power of H202 that is directly involved in inactivation of viruses and no intermediate species involved in the process.
Hydrogen peroxide shows synergism with uv-light and silver nitrate. Bayliss and Waites (1976) reported that combined treatment of Clostridium spores with copper and hydrogen peroxide is significantly more effective than treating spores with copper or hydrogen peroxide alone.




50
This work studied the inactivation of bacteria and
viruses with hydrogen peroxide. The data indicate that viruses are relatively more resistant to the action of hydrogen
peroxide than bacteria. The effects of salts and detergents as well as hydroxyl radical scavengers on the inactivation of viruses with hydrogen peroxide are studied as well. The combined treatment of bacteria and viruses with hydrogen
peroxide and other agents such as silver nitrate, cupperic chloride, and UV light is also studied. Finally the inactivation of bacteria and viruses with chlorine and hydrogen peroxide are compared.




51
Table 1. Inactivation of microorganisms with hydrogen
peroxidea
Type Strain % Survivors
Bacteria S. aureus 17.6 2.9b
S. faecalis 11.4 2.1
S. saprophyticus 6.6 1.0
M. smegmatis 6.2 2.4
E. coli 0.2 0.1
P. aeruginosa 0.1 0.1
V. cholerae < 0.1
B. cereus < 0.1
Mean 5.3
Bacteriophages MS2 90 3
T2 72 6
OX174 74 6
P22 90 2
06 43 5
PRD-1 46 6
Mean 69
a. Cells were treated with 0.1% hydrogen peroxide for
1 h; catalase was then added to stop inactivation.
b. Values represent the mean and standard deviation.




52
Table 2. Effects of H 20 2 treatment on E. coli volume
Time %Initial
(min) Volume' OD50 % Survival
75 95 0.57 7
150 92 0.50 0
225 74 0.36 0
300 67 0.29 0
a. Cells were allowed to grow in the presence of
0. 1% H2 0 and aliquots were removed at different
intervals. Volume of the pellet, optical density,
and viability were determined. Results are the
mean values two separate experiments.




53
Table 3. Lactate Dehydrogenase Leakage in the
Extracellular medium from E. coli Cells
Exposed to Hydrogen Peroxidea
Treatment OD 550 LDH
(Time) mUnits/ml/O.D
None
100 min 5.4 0.50
300 min 11.8 0.63
0.01% H202
100 min 0.55 9.75
300 min 2.90 6.90
0.1% H202.
100 min 0.60 38.50
300 min 0.32 71.30
a. Cells were grown either in the absence or
presence of 0.01% or 0.1% H202. At each time
interval OD values were measured and LDH activity in the supernatant was measured.
Results represent the mean from two separate
experiments.




54
Table 4. Effects of pH, Temperature, and Medium on
Inactivation of MS2 with Hydrogen Peroxidea
Inactivation %Oxidizing % Survival
Medium Powerb pH TempoC MS2 E. coli
50mM Na2HPO4 100 7 37 56 3 > 0.1
50mM Na2HPO4 NAc 7 25 70 5 0.1 0
50mM Na2HPO4 NA 7 4 92 2 90 3
50mM Na2HPO4 NA 5 25 0 0
50mM Na2HPO4 NA 9 25 95 4 NDd
3% Beef Extract 46 7 25 90 2 0.20.1
3% Tryptic Soy
Broth 81 7 25 84 4 > 0.1
a. Bacteriophages were treated with 0.1% H202 under the given conditions for 60 min.
b. Oxidizing power measured by iodometric methods as described in the text.
c. Not applicable. These values were not different from the oxidizing power of 50mM Na2HPO4 pH 7. d. Not done.




55
Table 5. Effect of Chemical Agents on Killing of
Bacteriophages by Hydrogen Peroxidea
Virus (Type) Salt (Type)b % Survivalc
MS2 Buffer (control) 70 5
MS2 100mM NaCitrate 90 4 *
MS2 100mM NaSCN 39 5 *
MS2 100mM NaF 75 1
OX174 Buffer (control) 68 3
pX174 100mM NaCitrate 79 5 *
OX174 100mM NaSCN 25 2 *
pX174 100mM NaF 80 6 *
6 Buffer (control) 29 8
p6 100mM NaCitrate 42 6 *
6 100mM NaSCN 6 3 *
6 100mM NaF 35 7
a. Bacteriophages were treated with 0.1% hydrogen
peroxide for lh at room temperature.
b. 50mMNa2HPO4 pH 7 buffer used as control buffer;
all other salts were dissolved in this buffer.
C. indicates values significantly different from the
buffer control at p=0.05 level.




56
Table 6. Treatment of E. coli and Bacteriophage
MS2 with H202 and Ultraviolet lighta.
Strain Treatment % Survivala
MS2 0.01% H202 95 3
MS2 366 nmUV 98 1
366 nm UV +
MS2 0.01% H202 32 6
E. coli 0.01% H202 75 10
E. coli 366 nm UV 85 4
366-nm UV
E. coli 0.01% H202 40 3
a. Cells and Phages were treated for 30 min in 0.05
mM Na2HPO4 pH 7 at room temperature.
b. Indicates values significantly different from
the individual treatments at p=0.05 level.




57
Table 7. Effect of Hydrogen Peroxide and Chlorine on
Biosynthesis and Activity of Beta- Galactosidase
Enzyme in E. coli
Units of Units of
Treatment Biosynthesis Activity % Viable
None 100 100 100
0.1% H202 13.5 2.7 0.1
0.05%
Chlorine 3.4 32.4 > 0.1
Bacterial cells were treated with hydrogen peroxide
and chlorine as described in text. Beta-galactosidase
activity was measured at 420 nm.




58
Table 8. Inactivation of Bacteriophage MS2 and E. coli
with H202, Silver Nitrate, and Cupperic Chloride
Percent Survival
Treatment E. coli MS2
0.01% H202 38 5 95 3a
0.25 ppm AgNO3 32 2 69 3
0.5 ppm CuCl2 95 1 NDb
0.01% H202
0.25 ppm AgNO3 3 2 21 3
0.01% H202
0.5 ppm CuCl2 68 6 ND
Bacteria and phage samples were treated at room
temperature. Samples were drawn after 30 min and diluted in Chamber's solution (for silver nitrate treatment) and 1% tryptic soy broth (for cupperic
chloride) treatment containing catalase. a. Values are mean standard deviation. b. Not Done.




Fig 1. Inactivation of bacteria and viruses was carried
out by treating cultures with 0.1% hydrogen peroxide in 0. 05M Na 2HPO 4 pH 7. Samples were drawn at different
intervals and assayed for survival.




60
Inactivation of Bacteria and Viruses by Hydrogen Peroxide
100
>
50
C
U
C)
PRD-1 c-
E. coli
0
0 30 60 90 120
Time (min)




Fig 2. Bacteriophage and bacterial cultures were treated
in the presence of increasing hydrogen peroxide
concentrations for 30 min in 0.05M Na2HPO4 pH 7.
Inactivation stopped either by dilution or addition of excess catalase. Results represent the means of
triplicate determinations.




62
Inactivation of Bacteria and Viruses by Hydrogen Peroxide
100
MS2
>g PRD-1
> Phi 6
O00 50
4-)
C
Q)
U
c-)
E. coil 01
0 .05 .10
Percent H202 (W/v)




Fig 3. Survival of E. coli C-3000, wild type, and
repair deficient mutant strain, K-12 cells
after treatment with hydrogen peroxide. Cells
were treated with increasing H 20 2 concentrations
for 15 min in 0.05 M Na 2HP041 pH 7, at 250C. Cell viability was measured using plate count agar. Results
represent the mean of 3-5 separate experiments.




64
Inactivation of Bacteria by Hydrogen Peroxide
100
(n
50
4-)
C
CL
(IJ
C-3000
K-12 (recA) 0,
0 .05 .10
Percent H202 (w/v)




Fig 4. Growth rates of E. coli cultures growing in 3% tryptic soy broth
containing various concentrations of hydrogen peroxide. Cells were
plated for survival test at different intervals on plate count
agar. Results represent the mean of triplicate determination.




Growth of E coli in the Presence of
Hydrogen Peroxide
2.0
aa control
0
1.5
XU
0
S-A 0 01%H202 o 1.0
A
05
+ 0.1%H202
0.0 I
0 60 120 180 240
Time (min)




Fig 5. Effects of hydrogen peroxide on cell leakage. E. coli
cells were grown in presence of various concentrations of
hydrogen peroxide for 1 h. Magnesium ions in the
extracellular medium was measured using 60 Second
Magnesium test as described in the text. Results
represent the mean and standard deviation of triplicate
experiments.




68
100
Q)
v
Q)
50
57
0
0
0.0 0.5 1.0
% H202 (W/V)




Fig 6. Effect of hydrogen peroxide treatment on the oxygen
consumption by E. coli. Cells were treated with either 0.01% or 0.1% H202. After centrifugation, washing, and resuspension in fresh growth medium (3% TSB), oxygen
consumption was measured.




70
Effects of H202 Treatment on Oxygen Consumption by E. coli
500
Control
0.01% H202
C
)
X
250
0.1% H202
0
0 30 60
Time (min)




Fig 7. Effects of Chemical agents on inactivation of
bacteriophage MS2 with hydrogen peroxide. Phages were treated with 0.1% hydrogen peroxide for 1 h in 0.05 mM Na2HPO4 pH 7.0 at room temperature in the presence or absence of different chemicals.
Inactivation reactions were stopped either by dilution or addition of excess amounts of catalase. Samples were
assayed for survival as described in the text.




Percent Survival
Ul 0
0 0 0
n (D
rn
QiNM Na-Citrate *
CD
0 0~
0C
OAM NaSC
(D
1)
0.2% Tweer2:
T(D
Qirvi~T- I'a~ T G
. . . . . . . . . . . .
.... .... .... .... ...




Fig 8. Effects of hydroxyl radical scavengers on the
inactivation of bacteriophage MS2 with hydrogen
peroxide. MS2 samples were treated with 0.1%
H202 in 0.05 mM Na2HPO4 pH 7.0 at room temperature
in the presence or absence of hydroxyl radical
scavengers and assayed for survival rates.




74
Effect of Hydroxyl Radical
Scavengers on H202Inactivation 100
50
-p,
T
LOO
0) C)
0)
--
00
-4
00
0




Fig 9. Prevention of hydrogen peroxide-induced cell killing
by thiourea, ethanol, or dimethyl sulfoxide. Ethanol,
, thiourea, ,dimethyl sulfoxide ,
and control, r Adopted from Brandi et al. (1989b).




76
100 A B
MODE ONE MODE TWO
KILLING KILLING
80 *
-.J1
60
(n 4020
2.5 25
H202 (mM)




Fig 10. Relationship between the effects of different
chemical agents on inactivation of bacteriophage
MS2 and their effects on hydrophobic interactions,
as measured by riboflavin solubility. Results depicted
here are adapted from Figures 7 and 8.




78
Relationship of Riboflavin
Solubility to Inactivation of MS2
150
b NaS
aTCA
Thiourea NaHPO4
- 100 _Urea Tween NaHC4
:3 Tween DMSO NC
O0
C Na-Cit e
(4
0
-Q 50
CY2
a. Salts only
b. All compounds
0
0 50 100
Percent Survival




Fig 11. Comparative inactivation of E. coli with H202
and chlorine. E. coli cells were treated with hydrogen peroxide or hypochlorous acid that had the same amounts of oxidizing power (as measured by an iodometric test) for 1 h. at room temperature in O.05M NaZHPO, pH 7.
Cells were plated on plate count agar for viability.




80
Inactivation of E. coli with
Hydrogen Peroxide and Chlorine
100
._>
50
-I-)
C
G)
U
Chlorine A H202
0*
0 3 6 9
Oxidizing Power (MEQ/L)




Fig 12. Comparative inactivation of bacteriophage MS2 with HO02
and chlorine. Virus samples were treated with hydrogen peroxide or hypochlorous acid that had the same amounts of oxidizing power (as measured by an iodometric test) for 1 h. at room temperature. Samples were assayed for phage viability as described in the
text.




82
Inactivation of MS2 with
Hydrogen Peroxide and Chlorine
100
---I H 0
50
4-J
c
C
L
(1)
Chlorine
0 '- _0 3 6 9 12 15
Oxidizing Power (MEQ/L)




Fig 13. Combined inactivation of E. coli with hydrogen peroxide
and silver nitrate. Samples were treated at room
temperature, and at different intervals aliquots were removed and diluted in Chambers solution containing
catalase.




84
Combined Inactivation of E. coli
with H202 and AgNO3
10
c( 50
4-j
C
1)
.01% H202 0.25 ppm
+ 0.25 ppm AgN03 0.01% H202
0 AqNO I I
0 50 100
Time (min)




CHAPTER 3
SELECTIVE RECOVERY OF BACTERIOPHAGES BY USING H202
Literature Review
Overgrowth of host bacteria by indigenous bacteria can interfere with the plaque assay of bacteriophages from such environmental materials as water, wastewater, fresh, and marine sediments (Goyal, 1987). It is often necessary to eliminate or inactivate the indigenous bacterial flora to prevent the interference with the host bacterial lawn and resolution of the plaques (Kennedy et al., 1985). Several procedures have been developed to reduce the contamination by bacteria without greatly reducing the number of bacteriophages in water samples. These methods include membrane filtration, incorporation of antibiotics into the assay media, the use of selective media, and chloroform pretreatment of the samples.
Membrane filtration through microporous filters with pore diameters larger than viruses but smaller than bacteria is a simple method for removing unwanted bacteria (Cornax, 1990; Tartera and Jofre, 1987). Tartera and Jofre (1987) used membrane filters with 0.22- Am pore size (Millipore Corp. Bedford, Mass.) to decontaminate samples such as ground water, raw sewage, and other sewage-contaminated waters.
85




86
The use of antibiotic-resistant host bacteria along with the incorporation of antibiotics into the assay medium can reduce the interference of the indigenous flora in the sample. Gerba et al. (1978) incorporated penicillin and streptomycin into media and used an antibiotic resistant strain of E. coli for assay of their isolates from natural sewage solids.
Lipid solvents, such as ether and chloroform, can be used to inactivate bacteria but not viruses that do not contain lipid. Treating samples such as activated sludge, polluted river water, and sewage effluent with chloroform produced large reductions of indigenous bacteria and enhanced plaque resolution in previous studies (Cornax, 1990; Clarke, 1983; Glass and O'Brien, 1980; Tartera and Jofre, 1987; Vaughn and Metcalf, 1975).
Another approach to control indigenous bacteria is the use of selective media which select for the phage host bacteria and inhibit growth of other bacteria (Goyal, 1987; Grunnet, 1977; Kennedy; 1985; Parker, 1981). Kennedy et al. (1985) compared selective media like EC medium, Gram Negative, and nutrient broth supplemented with sodium deoxycholate for enumeration of coliphages in activated sludge effluent and polluted lake water. These workers concluded that certain selective media, such as EC and Gram Negative media, allow the assay of environmental samples either directly or following concentration procedures by the agar overlay method without any decontaminating procedure or the use of antibiotics.




87
Other decontamination methods use detergents. Phages have been reported to be highly resistant to various detergents including saponin, sodium dodecyl sulfate, and sodium deoxycholate (Burnett, 1940). The treatment of samples with cationic detergents such as Emulsol-607, Zephiran, and cetylpyridinium chloride has been suggested for assay of coliphages in sewage (Kalter, 1946).
Our initial studies on the inactivation of laboratory strains of bacteria and phages by hydrogen peroxide revealed that bacteria are inactivated more rapidly than phages by this oxidant. This led to the use of hydrogen peroxide to treat natural samples to reduce the bacterial contamination. In some cases overgrowth of indigenous gram positive bacteria was a problem. It is known that gram positive bacteria are more sensitive to certain dyes, such as crystal violet, than gram negative ones (Bitton and Koopman, 1988). Therefore, crystal violet was incorporated into the bottom agar for standard phage assay.
This study shows that the assay of bacteriophage from environmental samples is improved by treating the samples with hydrogen peroxide, and by adding crystal violet to the agar plates. This procedure enables us to quantify bacteriophages in samples containing a relatively high number of indigenous bacteria.




88
Materials and Methods
Samples; raw sewage and trickling filter effluent samples were collected from the University of Florida sewage treatment plant. Lake water was collected from Lake Alice on the University of Florida campus. This lake receives the final effluent from the campus sewage treatment plant. Barnyard materials were from the Department of Agricultural Engineering research facilities at the University of Florida. Samples were either treated directly or 2 to 3 liters of the samples were concentrated by passing the samples through positively charged filters (Virosorb 1MDS [AMF Cuno, Inc., Meridian Conn.]) (Shields, 1986b). Adsorbed viruses were eluted by passing 10% beef extract, pH 9 through filters. The filter eluates were adjusted to pH 7 and mixed with an equal volume of saturated ammonium sulfate. The resulting floc was collected by centrifugation and resuspended in 0.05M Na2HPO4 pH 7 (Sheilds and Farrah 1986a). Some samples were concentrated by using magnetite-organic flocculation (MOF) method in which isoelectric casein and magnetite was added to samples and pH adjusted to 4.5. The resulting flocs were collected using a magnet and dissolved in 0.05M Na2HPO4 pH 7 as previously described (Bitton et al., 1979).
Cultures; the following bacterial strain used in this study as host was obtained from the American Type Culture




89
Collection; E. coli C-3000 (ATCC 15597). Cultures were grown in 3% tryptic soy broth pH 7.
BacteriophaQe assay; unless stated otherwise the host bacteria for enumeration and detection of bacteriophages was E. coli C-3000. EC medium and plate count agar with or without incorporation of 1 ppm crystal violet were used as bottom agar and 3% tryptic soy broth with 0.8% agar was used as the top agar for the standard agar overlay technique used in detection and enumeration of all phages. 0.1 ml sample and 0.1 ml host bacterium were mixed with 4 ml top agar in all assays. Water samples were either assayed directly or after treatment by one of the following methods: 1. H202, 2. Chloroform, 3. Filtration with a series of 0.45- and 0.2- Am Filterite filters (Filterite Corp., Timonium, MD) in 25-mm holders. Samples were treated with either 0.1 or 0.5% hydrogen peroxide for 1 h. Residual hydrogen peroxide was removed either by dilution or addition of catalase (140 units, 0.05 mg/ml w/v, final concentration). Chloroform treatment was as previously described (Kennedy et al., 1985).
The HPCV Phage Assay Procedure; the hydrogen peroxide crystal violet (HPCV) procedure is summarized as the following; Natural samples were treated, either directly or after concentration, with 0.1% hydrogen peroxide (v/v final concentration) on a shaker (800 rpm) at room temperature for 1 h. Samples were either diluted or catalase was added (0.05




90
mg/ml, final concentration). Standard agar overlay assay was performed using bottom agar containing 1 ppm crystal violet.
Chemicals and Media; EC medium and tryptic soy broth were from Difco Laboratories, Detroit, MI. Catalase was obtained from Sigma Chemical Co., St. Louis, MO. All other chemicals were obtained from Fisher Scientific Products.
Statistical analysis; each experiment was done at least twice in triplicate. The student t-test (p = 0.05) was used for comparing the mean values.
RESULTS
Initial studies revealed that Escherichia coli was inactivated faster with hydrogen peroxide than was bacteriophage MS2 (Figure 1 and 2). As it is shown in Figure 1, about 15% of MS 2 phage was inactivated by 0.1% hydrogen peroxide (v/v final concentration) in time it took to inactivate >99% of E. coli. MS2 was resistant to even higher concentrations of hydrogen peroxide (Figure 2). This pattern of inactivation was then confirmed with other phage types and bacterial species (Table 1). Except for vegetative cells of B. cereus, gram positive and acid fast bacteria were
relatively more resistant to hydrogen peroxide than were gram negative bacteria.
Treating samples of trickling filter effluent and raw sewage with 0.1% H202 for 1 h at room temperature reduced




Full Text

PAGE 1

THE INACTIVATION OF BACTERIA AND VIRUSES BY HYDROGEN PEROXIDE By ABDOLKARIM ASGHARI A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 1993

PAGE 2

To my beloved wife Zohreh and my two beautiful daughters Hamideh and Freshteh. And, finally, to my parents Javad and Atkeh.

PAGE 3

ACKNOWLEDGEMENTS I gratfully acknowledge the help and guidance given me by the members of my committee, Dr. Samuel R. Farrah, Dr. Lonnie O. Ingram, Dr. Gabriel Bitton, Dr Francis C. Davis, and Dr. Philip M. Achey. I am forever indebted to my cornrni ttee chairman, Dr. Farrah, who showed me the true meaning of the relationship between the student and his major professor. His broad knowledge of science and expertise in my subject area were invaluable assets for the success of my work. In addition, I extend thanks to the entire personnel of the Department of Microbiology and Cell Science who have helped me in many ways to make my stay at the University of Florida an unforgettable experience. Finally, I extend thanks to my family, especially my wife and my parents, for the love and support they have given me. iii

PAGE 4

ACKNOWLEDGEMENTS. LIST OF TABLES. LIST OF FIGURES ABSTRACT. TABLE OF CONTENTS CHAPTER 1 LITERATURE REVIEW ... Mechanism of Inactivation by Hydrogen Peroxide. Hydrogen Peroxide Damage to the Genetic Materials ........... Role of Defence Mechanism Against Hydrogen Peroxide. . . . Role of Fe 3 + and OH in H 2 o 2 -Induced Cell Killing. . Hydrogen Peroxide Effect on Other Cell Components. . ..... Mechanism of Inactivation by Other Oxidizing Agents Chlorine. . Ozone . . Mechanism of Inactivation by Ultraviolet Light Mechanism of Inactivation by Metal Ions. Silver. Copper. Synergistic Effect of Hydrogen Peroxide and Other Agents ... Ultraviolet Light Iron .. Copper ... CHAPTER 2 INACTIVATION OF BACTERIA AND VIRUSES BY iii vi vii viii 1 4 6 8 13 15 18 18 20 22 24 24 25 26 26 27 28 HYDROGEN PEROXIDE. . . 30 Mechanism of Inactivation by Hydrogen Peroxide 30 Materials and Methods. . . 32 Results. . . . . 35 Inactivation of Bacteria and Viruses with Hydrogen Peroxide . . . 3 5 iv

PAGE 5

Factors Influencing Inactivation Rate by Hydrogen Peroxide. . .... Effects of pH, temperature, and inactivation media . . . . Effects of chemical agents on hydrogen peroxide toxicity ...... Hydroxyl radical scavengers ..... Combined Inactivation of E. coli and MS2 by Hydrogen Peroxide and Ultraviolet Light Comparison Between Hydrogen Peroxide and Chlorine Inactivation ......... Combined Inactivation of E. coli and MS2 with Hydrogen Peroxide and Silver Nitrate .. Discussion .... Tables and Figures .. CHAPTER 3 SELECTIVE RECOVERY OF BACTERIOPHAGES BY USING HYDROGEN PEROXIDE. Literature Review. Materials and Methods. Results. . Discussion ..... Tables and Figures CHAPTER 4 CONCLUSION. REFERENCE LIST ... BIOGRAPHICAL SKETCH V 39 39 40 41 42 42 43 43 51 85 85 88 110 113 118 128 131 145

PAGE 6

Table 1 Table 2 Table 3 Table 4 Table 5 Table 6 Table 7 Table 8 Table 9 Table 10 Table 11 Table 12 Table 13 Table 14 Table 15 Table 16 Table 17 Table 18 Table 19 LIST OF TABLES Inactivation of microorganisms with H 2 0 2 Effects of H 2 0 2 treatment on E. coli volume Lactate dehydrogenase leakage in the extracel lular medium from E. coli cells exposed to hydrogen peroxide. Effects of pH, temperature, and medium on inactivation of MS2 and E. coli with H 2 0 2 Effect of chemical agents on killing of bacteriophages by hydrogen peroxide .. Treatment of E. coli and MS2 with hydrogen peroxide and ultraviolet light. Effect of hydrogen peroxide and chlorine on biosynthesis and activity of beta-galactosi dase enzyme in E. coli Inactivation of MS2 and E. coli with hydrogen peroxide silver nitrate, and cupperic chloride ... Inactivation of bacteria and bacteriophages by H 4 0 2 in wastewater samples (direct assay) Combined effects of crystal violet and H 2 0 2 on the indigenous bacterial population in environmental samples with high levels of bacteria ... Reductions in bacterial numbers using H 2 0 2 treatment and/or crystal violet plate. Combined effects of crystal violet and hydrogen peroxide on the indigenous bacteriophage population in environmental samples with high levels of bacteria .. Comparison of other decontamination methods with H 2 0 2 -crystal violet procedure Influence of magnesium chloride concentration on the magnesium peroxide content of modified diatomaceous earth .. Changes in bacterial numbers on DE filters Stability of magnesium peroxide on DE during daily filtration of tapwater .. Release of oxidizing power from Mgo 2 Influence of diatomaceous earth coated with magnesium peroxide on bacterial growth Influence of solids coated with magnesium peroxide on bacterial growth in tapwater. vi 51 52 53 54 55 56 57 58 97 98 99 100 101 118 119 120 121 122 123

PAGE 7

Figure Figure Figure Figure Figure Figure Figure Figure Figure LIST OF FIGURES 1 Inactivation of bacteria and viruses by H 2 o 2 2 Inactivation of bacteria and viruses by H 2 0 2 3 Inactivation of bacteria by H 2 0 2 4 Growth of E. coli in the presence of H o 2 5 Effect of hydrogen peroxide on cell lea\age .. 6 Effect of H 2 o 2 treatment on oxygen consumption by E. coli .......... 7 Effect of chemical agents on inactivation of MS2 with hydrogen peroxide .. 8 Effect of hydroxyl radical scavengers on hydrogen peroxide inactivation of MS2. 9 Effect of hydroxyl radical scavengers on hydrogen peroxide inactivation of E. coli .. Relationship of riboflavin solubility Figure 10 Figure 11 Figure 12 Figure 13 Figure 14 Figure 15 to inactivation of MS2 by H~0 2 Inactivation of E. coli with hydrogen peroxide and chlorine .......... Inactivation of MS2 with H 2 0 2 and chlorine. Combined inactivation of E. coli with hydrogen peroxide and silver nitrate .... Selective recovery of bacteriophages from raw sewage using different decontamination methods. . . . . Steps involved in the hydrogen peroxide crystal violet (HPCV) phage assay procedure. vii 60 62 64 66 68 70 72 74 76 78 80 82 84 103 105

PAGE 8

Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy THE INACTIVATION OF BACTERIA AND VIRUSES BY HYDROGEN PEROXIDE By Abdolkarim Asghari May 1993 Chairman: Dr. Samuel R. Farrah Major Department: Microbiology and Cell Science Studies on the inactivation of bacteria and viruses by hydrogen peroxide (H 2 o 2 ) showed that bacteriophages are relatively more resistant to H 2 o 2 than bacteria. Lipid containing phages are inactivated at a much faster rate by H 2 o 2 than non-lipid containing bacteriophages. Leakage of bacterial cell contents suggests that the integrity of the cell membrane is disrupted after treatment cells with H 2 o 2 Significant amounts of lactate dehydrogenase and Mg 2 + ions were released into the extra cellular medium after treatment with H 2 o 2 Oxygen consumption, redox potential, as well as cell volume decreased substantially upon treating h coli cells with H 2 O 2 Lower pH, a lack of organics in the inactivation medium, and a higher temperature enhanced the detrimental activity of H 2 O 2 viii

PAGE 9

Chaotropic agents, which weaken hydrophobic associations, increased the rate of inactivation of bacter iophages. In contrast antichaotropic agents, which strengthen hydrophobic associations, decreased the lethal effect of H 2 0 2 These results suggest that hydrophobic interaction between capsid proteins influence the rate of inactivation of viruses by H 2 o 2 The use of hydroxyl radical scavengers in inactivation media reduced the rate of inactivation of bacteria but not viruses, suggesting that different mechanisms are involved in inactivation of bacteria and viruses by H 2 0 2 Unlike the killing of~ coli, OH are not the major cause of virucidal action of H 2 0 2 The above results suggest that oxidizing power of H 2 0 2 may be directly involved in the inactivation of viruses. This virucidal action was different when compared to virucidal action of chlorine. Inactivation of viruses by chlorine is much faster than inactivation by H 2 0 2 : The use of H 2 0 2 in recovery of bacteriophages from natural water samples with plates containing crystal violet is proposed. The selective inactivation of bacteria in environmental bacteriophages samples by with minimal interference recovery of from indigenous bacteria. This method is either superior or as good as the currently available procedures. The use of H 2 o 2 to coat diatomaceous earth and sand with magnesium peroxide for removal/inactivation of bacteria has also been proposed. Filters made with modified solid effectively removed and ix

PAGE 10

inactivated bacteria as compared to the controls. Modified sand and diatomaceous earth added to unchlorinated tapwater kept the water free from bacteria for at least three weeks. X

PAGE 11

CHAPTER 1 LITERATURE REVIEW The germicidal properties of aqueous hydrogen peroxide have been recognized for more than a century. Dr Samuel S Wallian praised hydrogen peroxide in an address to the New York State Medical Society in 1892: "One can hardly refer to the medical journals without finding enthusiastic recommendations of it as a disinfectant of rare efficiency, an antiseptic of recognized merit and a germicide of decided potency" (Block, 1991; pl67). Known as oxidized water then, hydrogen peroxide has had a rocky road in its acceptance as a disinfectant; first popular, then unpopular, and now finding special application where it serves particular functions of great value (Block, 1991). The literature contains numerous accounts of the satisfactory applications of hydrogen peroxide as a disinfectant for inanimate materials and inert surfaces in the aseptic packaging of food and dairy products (Toledo, 1975; Naquib and Hussein, 1972). It is also used for processing systems and spacecraft hardware (Block, 1991) It has been recommended for the disinfection of surgical implant 1

PAGE 12

2 components, contact lenses, and many other heat labile materials (DeRezende, 1968). The vapor phase of hydrogen peroxide provides the sole killing power of a newly patented sterilization system. This system provides a rapid, low temperature, low toxicity sterilant that could eliminate the toxic or carcinogenic gaseous sterilants such as formaldehyde and ethylene oxide (Caputo and Odlaug, 1983; Arlene-Klapes et al., 1990). Hydrogen peroxide has been used to remove the organic impurities in water. Along with ultraviolet light, hydrogen peroxide reduced the total organic carbon (TOC) content of dis'tilled water samples by about 88% and of tapwater by 98% (Malaiyandi et al., 1980). Hydrogen peroxide serves as a natural disinfectant and preservative and is naturally present in milk and honey. It is present in saliva and fights bacteria in the mouth. Hydrogen peroxide in phagocytes is one of the materials involved in the killing of the ingested bacteria, and it is a ~eak mutagen as tested by the standard Ames test (Kensese and Smith, 1989). It is also part of the defense system of some bacteria (Block, 1991; Dubreuil et al., 1984). Hydrogen peroxide is produced in cells by the reduction of oxygen for respiration. Hydrogen peroxide, oxygen and its other by-products such as oxygen free radicals are harmful to living cells and must be removed constantly. Oxygen free radicals such as the superoxide anion, 0 2 and the hydroxyl radical, OH, are linked to many diseases such as rheumatoid

PAGE 13

3 arthritis and cancer. Proteins, lipids, and nucleic acids are all at risk of being damaged by these oxidizing agents through redox reactions. Aerobic cells have a system for disposing of the oxygen by reducing it to water through a series of enzymatic steps (Stoklasova, 1989) Catalase and peroxidase are responsible for removal of hydrogen peroxide and are the primary defense against this biocide (Fridovich, 1978). Products of recA+ and polA+ genes are also important in protection against hydrogen peroxide in coli The repair mechanism of defense is distinct from the enzymatic removal of the biocide by ca ta lase. The following diagram illustrates the univalent pathway for oxygen reduction and the enzyme involved in the catalytic scavenging of intermediates (Fridovich, 1978). 0 2 ------->O ------->H O --------------OH ------->H 0 e 2 e +2H+ 2 2 e H 2 0 e+H+ 2 (Superoxide dismutase) (Catalase) (Peroxidase) Scavenging enzymes such as catalase, peroxidase, superoxide dismutase, and DNA repair enzymes that correct oxidative lesions are found throughout the aerobic organisms (Imlay et al., 1988; Lesko et al., 1980). Although hydrogen peroxide has been used as a disinfectant for many years, its mechanism of action has not

PAGE 14

4 been completely elucidated. It is proposed that hydrogen peroxide derivatives such as hydroxyl radicals are responsible for cell damage inside the cell. Two modes of cell killing by hydrogen peroxide can be distinguished using h coli K-12. Mode one killing is accomplished with concentration of <5 mM (maximum inactivation at 3 mM), and mode two with >10 mM hydrogen peroxide. An intervening zone of partial resistance is detectable between the two regions of killing. The two modes of cell killing are distinguishable kinetically and metabolically (Imlay and Linn, 1986; 1987; 1988; Imlay et al., 1988). Much of the data support the theory that damage to DNA is the major cause of inactivation by this biocide when low concentrations of the reagent are used. The lethal lesion of mode two killing is not known. Mechanism of Inactivation by Hydrogen Peroxide The characterization of cell damage caused by oxygen species has been complicated by the variety of direct and indirect effects observed, and the ultimate oxidant and the specific killing lesion established. oxidative in bacterial cell have not yet been damage to nucleic acids and other macromolecules such as lipids and proteins have been studied extensively (Imlay and Linn, 1986; 1988). DNA damage is shown by accumulation of single strand DNA fragments (Breimer and Lindahl, 1985; Hoffman et al., 1984). Membrane damage from

PAGE 15

5 oxygen species is observed as an accumulation of lipid peroxides, the loss of the diffusion barrier to membrane impermeable markers, and cell lysis ( Seeberg and Steinurn, 1980). A variety of enzymes have been shown to be inactivated by hydrogen peroxide (Aguirre and Hansberg, 1985; Hodgson, 1975; Kirn et al.). These include superoxide dismutases from both eukaryotic and prokaryotic sources (Yarnakura and Suzuki, 1986; Borders and Fridovich, 1985). Exposure of logarithmical growing Escherichia coli to hydrogen peroxide leads to two kinetically distinguishable modes of cell killing. Mode one killing is pronounced at low (less than 5 mM) concentrations of hydrogen peroxide and is caused by DNA damage. Mutant strains that are defective in recombinational or excision DNA repair are especially sensitive to mode one killing. Indicators of DNA damage, mutagenesis and the induction of the SOS DNA repair regulon, accompany challenges of cells at low doses. (Imlay and Linn, 1986; 1988). Mode two killing requires a relatively high, higher than 10 mM dosage, and the site (sites) of the toxic injury has not been established. There is also an unexplained area of partial resistance to hydrogen peroxide between 5 and 10 mM concentrations.

PAGE 16

6 H 2 Q 2 Damage to the Genetic Materials Genetic and physical effects of hydrogen peroxide and other oxidative stresses such as gamma radiation upon cellular DNA have been well studied. Oxygen radicals may attack DNA at either a sugar or a base giving rise to a large number of products. Attack at a sugar ultimately leads to sugar fragmentation, base loss and a strand break with a terminal fragmented sugar residue (Imlay and Linn, 1988). Such single strand breaks are accumulated during exposure of bacteria and mammalian cells to hydrogen peroxide, 0 2 -, gamma radiation, or ozone (Imlay and Linn, 1986; Seeberg and Steinum, 1980; Prise et al 1989; Birnboim and Kanabus-Daminska, 1985; Sawadaishi et al 1985; Hagensee and Moses, 1986). The repair of apurinic/apyrimidinic (AP) sites in DNA is initiated by incisions catalyzed by specific endonucleases. In h coli, exodeoxyribonuclease III, the product of xthA gene, constitutes about 90% of the bacterial AP activity. _h coli mutants lacking exonuclease III (xth ) are exceptionally sensitive to 5 mM of hydrogen peroxide. They are killed by hydrogen peroxide at 20 times the rate of wild type and at 3 to 4 times the rate of recA mutants (Demple et al., 1983b). Hydrogen peroxide induces single strand breaks in bacterial DNA and these breaks can be repaired by polymerase I-dependent and recA dependent pathways similar to those for the repair of X -ray induced single strand breaks (SS). recA recB, polAl, and

PAGE 17

7 recA mutant strains of h coli are not able to repair SS breaks efficiently, and are more sensitive to hydrogen peroxide than the wild type (Ananthaswamay, 1977). Strains deficient in RecA protein, exonuclease III, exonuclease V (RecBC enzyme), or DNA polymerase I are especially vulnerable to mode one killing and are not different from wild type in reacting to mode two killing (Imlay and Linn, 1987). Hagensee et al. (1987) showed the involvement of polymerase III in the H 2 o 2 -induced DNA damage repair during replication in Escherichia coli. Prechallenging cell growth in anoxic medium significantly amplifies mode one killing of both repair deficient and repair proficient strains. Cells gain back their normal sensitivity to mode one killing upon dilution into oxygen saturated medium. As shown by chloramphenicol inhibition, under anaerobic condition cells synthesize proteins that enhance hydrogen peroxide mediated DNA damage (Imlay and Linn, 1986). Only actively metabolizing eel ls are subject to mode one killing, because starving cells lose their sensitivity to hydrogen peroxide. The cells regain their sensitivity upon addition of glucose to the medium (Imlay and Linn, 1986). The blocking of respiration chemically by cyanide or genetically by inactivating the NADH dehydrogenase gene dramatically sensitize h coli to killing by H 2 0 2 Desensitization rapidly follows after removal of cyanide from the medium (Imlay and Linn, 1986)

PAGE 18

8 Bacterial cells behave differently upon challenge with concentrations of hydrogen peroxide representatives of the two modes of cell killing. Mode-two-killed cells remained permanently unit-sized, but mode-one-treated cells underwent extensive filamentation and did not septate (Brandi et al., 1989b; Imlay and Linn, 1986). A growth lag followed by a period of filamentation is observed with the recovery of the treated cell with low doses of hydrogen peroxide. The authors conclude that the lag is for repair of cell damage and filamentation is the result of DNA damage and subsequent block of septation which ends up with mode one killing (Imlay and Linn, 1986). Brandi et al. (1989b) observed similar results under light microscope and concluded that the size of filament increases with increase in exposure time (Brandi et al., 1989b). No such phenomenon is observed with mode two killing (Brandi et al., 1989b). Exposing the cells to H 2 o 2 (mode two cell killing) resulted in decrease in cell volume as measured directly and by optical density (Brandi et al., 1989b). Role of Defense Mechanism Against H 2 Q 2 Hydrogen peroxide induces SOS response in bacterial cells by increasing the rate of synthesis of recA protein (Imlay and Linn, 1987). SOS mutants are hypersensitive to hydrogen peroxide (Dernple et al., 1983b; Chen and Bernstein, 1987). The sensitivity of recA strains to mode one killing suggest that

PAGE 19

9 hydrogen peroxide activates the SOS response. This response protects the cells against mode one killing through an enhanced ability to carry out recombinational DNA repair (Chen and Bernstein, 1987; Demple and Halbrook, 1983a; Imlay and Linn, 1987). Salmonella typhimurium, Escherichia coli and other bacteria become resistant to killing by hydrogen peroxide and other oxidants when prechallenged with nonlethal dose of H 2 0 2 (Chritsman et al., 1985; Boland Yasbin, 1990; Demple and Halbrook, 1983a, Chen and Bernstein, 1987; Demple et al., 1983b; Plateaue et al., 1987). During adaptation to hydrogen peroxide, 30 proteins are induced. These investigators have identified a regulon under the control of the oxyR locus, which apparently encodes a positive effector of this response. Gene products are overproduced after exposure to inducing levels of hydrogen peroxide. Among the gene products overproduced are the scavengers of active oxygen species, catalase, superoxide dismutase and peroxidase. This regulon governs the protective response of coli and typhimurium by adaptation to low levels of hydrogen peroxide and thus subsequent resistance to higher doses of the agent. Cells adapted to H 2 0 2 are resistant to a variety of other agents causing oxidative damage, as well as to heat killing. Greenberg and Demple ( 1989) also reported that the redox cycling agents menadione and paraquat (which generate superoxide) induced protection mechanism, which overlaps with one induced by H 2 0 2 Jenkins et al. (1988) have shown that h

PAGE 20

10 coli K-12 responds to glucose or nitrogen starvation with synthesis of 30 proteins. Several of these proteins are also involved in protection against heat and hydrogen peroxide challenge. These investigators also showed that similar proteins are produced with heat and hydrogen peroxide adaptation. Unlike the sos induction which enhances repair system, the oxyR regulon exerts its protective effect primary through an enhanced ability to scavenge partially reduced oxygen species. The two stress responses are induced by hydrogen peroxide and act by nonoverlapping processes to protect against lethal doses of hydrogen peroxide (Imlay and Linn 1986; 1987; Chritsman et al., 1985). The enzymes ca ta lase, peroxidase, and superoxide dismutase can be considered the primary defenses of the cell against H 2 o 2 and other oxygen metabolites (Fridovich, 1978; Starke and Farber, 1985b; Hassan and Fridovich, 1979; Bayliss and Waites, 1981). The DNA repair systems also play a major role in protecting cells against hydrogen peroxide (Demple and Halbrook, 1983a). Carlssen and Carpenter (1980) and others (Ananthaswamy and Eisenstark, 1977; Winquist et al., 1984) studied the importance of different protection mechanisms against hydrogen peroxide. These studies showed that there is no correlation between the sensi ti vi ty to H 2 o 2 and the cellular level of the enzymes catalase and superoxide dismutase (Carlssen and Carpenter, 1980). Miyasaki reported that endogenous catalase activity et is al., (1985) an important

PAGE 21

11 determinant of resistance to bactericidal effects of H 2 0 2 recA mutants were more sensitive than other mutants. Thes e investigators concluded that a functional recA gene product is more important than the enzymes involved in scavenging and removing harmful oxygen species (Carlesson and Carpenter, 1980; Ananthaswamy and Eisenstark, 1977). Yet others (Yonei et al. 198 7) have found that h coli UMl, with no ca ta lase activity is 20 fold more sensitive than wild t y pe strains. Other researchers have also found that catalase negative mutants are 50-60 times more sensitive to H 2 o 2 than their wild type counterparts (Loewen, 1984). Exposing the cells to nonlethal concentrations of hydrogen peroxide induces both DNA repair system and scavenging enzymes (Winquist et a l ., 1984; Demple and Halbrook, 1983a) which protect cells against higher doses of the biocide. C atalase negative cells with normal DNA repair system and repair deficient mutants with normal catalase level were more sensitive to H 2 0 2 than their respective wild type (Yonei et al., 1987) These results showed that catalase and DNA repair systems have distinct roles in protection against lethal damage of H 2 0 2 to h coli (Yonei et al., 1987) even though catalase activity is a major contributor to the cell's defense system (Miyasaki et al., 1985). Complex, nonsynthetic, growth medium such as heart infusion, but not defined medium, can change the concentration of ca ta lase, and a high ca ta lase

PAGE 22

12 concentration protects the cells against higher doses of H 2 0 2 (Bayliss and Waites, 1981). Quin 2 (an intracellular calcium chelator) but not EGTA (an extracellular calcium chelator) prevents hydrogen peroxide induced DNA breakage and cytotoxicity in mammalian cells (Cantoni et al., 1989a). Quin 2 does not prevent formation of hydroxyl radicals. These authors suggest that the majority of DNA damage induced by hydrogen peroxide in intact cells at 37C are not caused by direct attack of hydroxyl or of secondary radicals on DNA. Instead, radicals elicit a complex cascade of metabolic secondary reactions which in part require ca 2 + and ultimately leads to DNA breakage (Cantoni et al., 1989a). The formation of a spectrin-hemoglobin complex following treatment of red cells with hydrogen peroxide is shown to be associated with alterations in cell shape and a decrease in membrane flexibility. SH groups of spectrin are involved in crosslinking since their blockage reduced the complex formation and decreased lipid peroxidation. oxidized hemoglobin is also required for crosslinking (Snyder et al., 1988). These data indicate that lipid peroxidation and oxidation of sulfhydryl groups, induced by hydrogen peroxide treatment, resulted in deleterious effects on erythrocytes' membrane properties (Snyder et al., 1988).

PAGE 23

13 Role of Fe~ and OH in H 2 Q 2 -Induced Cell Killing It has been shown that by increasing the iron content of Staphylococcus aureus and other bacterial cells, their susceptibility to hydrogen peroxide is dramatically enhanced (Repine et al., 1981b; Hoepelman et al., 1990; Sambri et al., 1991). Iron in microorganisms facilitates bactericidal mechanism by reacting with H 2 o 2 to form more toxic hydroxyl radicals (OH). Production of hydroxyl radicals occurs by the Fenton reaction Fe 2 + + H 2 0 2 ----+ Fe 3 + OH + OH" Repine et al. ( 1981a) further showed that hydroxyl radical scavengers such as thiourea, dimethyl thiourea, sodium benzoate, and dimethyl sulfoxide inhibited hydrogen peroxide mediated killing of~ aureus. The role of hydroxyl radicals in bacterial cell killing by hydrogen peroxide has been studied by several investigators (Brandi et al., 1987; Brandi et al., 1989a; Hoepelman et al., 1990; Van Slays et al., 1986; Kleiman et al., 1990; Terada et al., 1991). They concluded that hydroxyl radicals are involved in production of mode two but not mode one killing by H 2 o 2 These investigators found that hydroxyl radical scavengers, thiourea, ethanol and dimethyl sulfoxide and the iron chelator, desferrioxamine, did not affect the survival of cells exposed to 2.5 mM H 2 0 2 (mode one killing). In addition, cell vulnerability to the same concentration of hydrogen peroxide was independent of the

PAGE 24

14 intracellular iron content. In contrast, mode two lethality (i.e. cell killing generated by a concentration of H 2 o 2 higher than 10 mM) was markedly reduced by OH scavenger iron chelators and was augmented by increasing the intracellular iron content (Brandi et al., 1989a). These investigators proposed the following mechanism for H 2 0 2 induced cytotoxicity in h coli. H 2 0 2 ---------?-?---DNA----------MODE ONE KILLING I LIPIDS H 2 o 2 -------Fe-OH -j PROTEINS-----MODE TWO KILLING I NUCLEIC ACIDS Hydroxyl radicals seemed also to be the major lethal means against spores as hydroxyl radical scavengers inhibited spore lysis. Studies by Cantoni et al. ( 1989c) showed that wild type and superoxide dismutase mutants display a markedly different sensitivity to both modes of lethality produced by H 2 0 2 They proposed a hypothetical enzyme that generates 0 2 This enzyme is active at H 2 0 2 concentrations <5 mM but high concentrations of H 2 0 2 (>5 mM) inactivate the enzyme (Cantoni et al., 1989c). Cantoni et al. (1989c) concluded that mode one cell killing is produced by superoxide anions whereas mode two cell killing is the consequence of the OH attack. Brandi et al. (1988) suggested that superoxide ions are involved in regeneration of divalent iron (to allow further Fenton reactions). Starke and Farber (1985a) also show the need for ferric iron and superoxide ions for hepatocytes killing by

PAGE 25

15 hydrogen peroxide. Again the free radical scavengers rnannitol, thiourea, and benzoate protect cells against H 2 o 2 (Starke and Farber, 1985a). Berglin et al. (1985; 1984) demonstrated that iron sulfide is more efficient than ferrous iron in catalyzing the formation of hydroxyl radicals similarly L-cysteine enhanced H 2 o 2 -induced killing by 100-fold. H 2 Q 2 Effects on other Cell Components In addition to damaging DNA, hydrogen peroxide also damages other vital cell components such as lipids (Flenley, 1987) and proteins (Richards et al., 1988), especially in the cell membrane of both mammalian (Vander Zee et al., 1985) and bacterial (Brandi et al., 1989b) cells. This damage is usually followed by cell leakage and subsequent cell lysis. Several enzymes were shown to be inactivated by hydrogen peroxide (Yarnakura and Suzuki, 1986; Steinman, 1982; Hodgson, 1975; Kirn et al., 1985; Aguirre, 1986). One of the major targets of H 2 o 2 attack is unsaturated lipids. These can undergo peroxidation which disrupts membrane structure and function (Girotti and Thomas, 1984; Schraufstatter et al., 1986). Girotti and Thomas (1984) have also made several observations in studying the lethal effects of H 2 o 2 on erythrocytes. These workers observed the following: (a) efflux of low molecular weight molecules such as glucose-6-phosphate and Na, (b) leakage was stimulated by Fe 3 and chelating agents inhibited the efflux,

PAGE 26

( C) both o 2 were 16 required since catalase; (d)superoxide dismutase inhibited lipid peroxidation, and (e) hydroxyl radical scavengers e.g. ethanol, mannitol, and choline provided no protection against marker efflux and lipid peroxidation (Girotti and Thomas, 1984) Others have shown that oxidation of membrane unsaturated fatty acids is not an essential component of the toxicity of H 2 0 2 to h coli (Ohlrogge and Kernan, 1983). Vander Zee et al., (1985) also observed lipid peroxidation and K+ leakage in erythrocytes. No correlation was found between lipid peroxidation and K+ leakage. These investigators showed that K+ leakage is due to the SH group oxidation, because diamide decreased the leakage by oxidizing the same SH groups (Vander Zee et al., 1985). Di amide oxidized the SH groups to disulf ides but, H 2 o 2 SH oxidation, in addition to disulfides, also yielded sulfenic and sulf onic acids. This further SH oxidation resulted in greater membrane permeability (Van der Zee 1985). Brandi et al. (1989b) found that h coli cells exposed to high concentrations of H 2 o 2 (>10 mM) show a reduction in cell volume as measured microscopically, and release of the enzyme lactate dehydrogenase into the culture medium Curran et al. (1940) found the greatest killing power of hydrogen peroxide against Bacillus spores at pH 3 and the least at pH 9. Baldry (1983) showed that varying the pH from 5-8 made no difference in treating several bacteria with hydrogen peroxide. Only one strain, Pseudomonas aeruginosa,

PAGE 27

17 needed 10 times less amount to inhibit growth at pH 5 than pH 8 (Block, 1991) Brandi et al. ( 198 7) and Cantoni et al. (1989a) studied the effects of temperature and anoxia on.&.:.. coli killing induced by hydrogen peroxide. They showed that low oxygen levels decrease the vulnerability to mode two treatment. Oxygen tension was not relevant as far as mode one killing is concerned (Brandi et al., 1987). Treating cells with H 2 o 2 at 37C was more toxic than at 4C (Fiorani et al., 1990). These investigators concluded that damage at 37C may be indirectly mediated by temperature dependent metabolic events (Cantoni et al., 1989a; 1989b). Domingue et al. ( 1988) compared the effects of the oxidizing power of chlorine, ozone, and hydrogen peroxide on Legionella pneumophilla. They found that in contrast to both ozone and chlorine, H 2 o 2 inactivation required much higher concentrations. As with chlorine, there was a dose-response relationship between the concentration of H 2 O 2 and the rate of inactivation by this biocide. These investigators found no decrease in oxidizing potential of H 2 O 2 after 24 h (Dominique et al., 1988; Yoshpe-Purer and Eylan, 1968). Persistent killing effects of up to 13 d was observed in some cases (Yoshpe-Purer and Eylan, 1968).

PAGE 28

18 Mechanism of Inactivation by Other Oxidizing Agents Chlorine Chlorine is an effective oxidizing agent and is the most widely used disinfectant. Dychdala (1991) listed several factors which determine the antimicrobial action of chlorine. pH has the greatest influence on the antimicrobial activity of chlorine in solutions. It is known that the disinfecting efficiency of chlorine decreases with an increase in pH and vice versa, which is parallel to the concentration of undissociated hypochlorous acid. Sharp et al. (1980) suggested that the following variables must be taken into consideration when studying the viral disinfection by chlorine, virus with the free chlorine, reaction, iii) the total ionic i) the time of contact of ii) the temperature of the strength of the reaction mixture, iv) the pH of the reaction mixture, v) the type of buffer used to maintain the pH, vi) the total chlorine concentration and the relative amounts of HOCl and oc1, and vii) the state of aggregation of the virus ( Sharp et al., 1980). Chlorine exists in solution as a mixture of hypochlorous acid and hypochlorite ion. HOCl. <----> H + OCl.

PAGE 29

19 The dissociation of hypochlorous acid depends on pH. The higher the pH, the greater the concentration of dissociated hypochlorous acid. Hypochlorous acid is much more potent than hypochlorite, so chlorine is more effective at lower pH. Higher temperatures increase the killing efficiency of chlorine (Dychdala, 1991). The presence of organic compounds reduces chlorine efficiency, which is defined as chlorine demand, and as the result chloramines are formed. This problem can be overcome by breakpoint chlorination, which is the application of a sufficient amount of chlorine to satisfy the initial chlorine demand. Adding more chlorine beyond the breakpoint increases free available chlorine species. Dychdala (1991) proposed that chlorine rapidly inhibits some key enzyme systems essential to life by oxidizing the SH groups of these enzymes; Based on the studies of (Linquist et al., 1976) chlorine changes the membrane permeability of h coli, and allows efflux of macromolecules such as protein and nucleic acid. Hurst et al. ( 1991) Observed that Hoc1 promoted inactivation was accompanied by extensive inhibition of respiration in h coli and~ aeruginosa. Surface proteins could also be a major target for chlorine action (Roller et al., 1980). Bacteria are inactivated primarily through irreversible sulfhydryl oxidation (Roller et al., 1980). Others supported this idea by showing that cleaved viral RNA is released from poliovirus capsid following treatment with chlorine (Taylor Butler, 1982; O'Brien and Newman, 1979). In

PAGE 30

20 another study the viral sedimentation coefficient changed from 156 s for native poliovirus to 80 s after treatment with chlorine, due to release of its RNA (Alvarez and O'Brien, 1982b). Chlorine concentrations of less than 0.8 ppm resulted in inactivation of viruses without major structural changes, whereas chlorine concentrations in excess of 0.8 ppm resulted in leaks in the capsid protein and loss of RNA (Alvarez and O'Brien, 1982a). Dychdala (1991) summarized the factors influencing the efficiency of chlorine as follows: pH, the concentration of free chlorine in the form of hypochlorous acid; temperature; and the presence of organic materials in the solution. Ozone Ozone is a strong and effective biocide. Lower concentrations of ozone and shorter contact times are required for treating samples than with chlorine and other disinfectants. Ozone is more effective than other oxidizing against resistant organisms such as amoebic cysts and viruses (Kim et al., 1980). Less than 1 ppm ozone inactivates 5-7 logs of virus in 5 s (Kim et al., 1980) Ozone was the most potent biocide against Legionella pneumoohilla when compared to chlorine and hydrogen peroxide (Domingue et al., 1988). The other investigators have found that RNA enclosed in the phage coat was inactivated less by ozone than were whole phages, but

PAGE 31

more inactivated than naked RNA. 21 They concluded that subcellular components of a microorganism can be more resistant than the whole organism, and that the inactivation of a microorganism does not necessarily denature the genetic materials in it (Kirn et al 1980). Kirn et al. ( 198 o) detected structural changes in the protein coat of bacteriophage f2 using electron microscopy after ozonation. They suggested that ozone breaks the capsid proteins into subunits and releases the RNA which then may become damaged afterward. As a result, the adsorption of virus particles to host pili was disrupted. Loss or severe reduction of h pneurnophilla unsaturated fatty acids was observed with ozone treatment (Roehm et. al., 1971). With its rapid rate of diffusion through bacterial cell walls and oxidizing power, ozone reacts with a wide range of organic compounds such as membrane-associated proteins containing sulfhydryl groups (Menzel, 1971). Roy et al. (1980) observed that damage to the viral nucleic acid was the major cause of the inactivation of poliovirus 1. RNA was damaged by ozone concentration less than o. 3 ppm within 2 min of contact time. Furthermore, ozone treatment resulted in strand scission of supercoiled pBR322 plasmid (open the circular DNA) (Sawadaishi et al., 1985), and altered two of the four polypeptide chains present in the viral protein (Halliwell and Gutteridge, 1984).

PAGE 32

22 Mechanism of Inactivation by Ultraviolet Light Ultraviolet (UV) light includes electromagnetic radiations that fall in the wavelength band between 200 and 400 nm. It falls in between the energies of X-rays and the shortest wavelength of light visible to the human eye. The germicidal effects of UV light are limited to only a specific region of the UV light spectrum, with 265 nm being the most effective wavelength. Studies of mutagenic effects and retardation of cell division suggested that these conditions are caused by the effect of UV on nucleic acids (Block, 1991). UV acts on cellular DNA primarily by producing links between adjacent pyrimidines on a DNA strand to form dimers. Dimers consist of primarily two thymine residues. Cytosine-thymine and cytosine cytosine dimers have also been identified in the cells exposed to UV-light, although these dimers are identified less frequently than thymine-thymine dimers. (Block, 1991). Dimers interfere with transforming ability of bacterial DNA and replication and also lead to cell death. According to Rahn and Landry, (1973), UV irradiation of DNA results in the formation of various kinds of photoproducts that may have a disruptive influence on the local integrity of the DNA structure. The cross-linking of DNA and protein plays a significant role in the killing of UV-irradiated cells (Block, 1991). Kelland et al. (1984) reported membrane damage, as shown by 86 Rb leakage,

PAGE 33

23 after treating h coli K-12 cells with near UV irradiation. Shechmeister, ( 1991) reported that the survival curve of viruses with single-stranded DNA or RNA is different from the survival curve for double-stranded DNA. Several factors influence bacterial sensitivity to UV. The more important are (1) pH; (2) bacterial growth stage; (the greatest sensitivity being in the logarithmic growth phase); (3) the presence of spores, which are about twice as resistant as vegetative cells and (4) presence of particles and turbidity of the liquid. There are basically three different repair mechanisms. In excision repair in which, a damaged area of DNA is cleaved by an endonuclease and excised by an exonuclease, DNA polymerase then fills the gap. Photoreactivation repair involves an enzyme DNA photolyase. This enzyme, which is activated by visible light, is capable of binding to dimers and spliting dimers. During excision repair the UV-induced DNA damage is recognized and removed immediately by the enzyme. Post replication repair, in contrast, is carried out after replication. This is the least reliable repair mechanism, since the repair after replication often operates erroneously (Darnell et al., 1986).

PAGE 34

24 Mechanism of Inactivation by Metal Ions Silver Silver compounds have been reported to bind with bacterial DNA. Silver displaces the hydrogen bonds at specific sites and may prevent replication of the DNA and subsequent cell division (Fox, 1978; Fox and Modak, 1974). Richards (1981) studied the effect of silver nitrate on actively dividing Pseudomonas aeruginosa cells and on the infectivity of T2 DNA. They concluded that silver caused cross linking of the bacteriophage DNA helix. Microbial DNA and envelope were damaged by silver ion (Richards, 1981). It is the concentration of silver ions, not their physical nature, that is responsible for disinfecting capability of silver, because silver action is independent of the way in which silver is introduced into water, e. g., soluble silver salt, metallic silver, etc., (Modak and Fox, 1985; Just, 1936). The binding property of the silver ion is very important. These ions can complex with proteins and nucleic acids by serving as an oxidative catalytic surface between them (Philips, 1958). Silver ions can irreversibly bind to the bacterial cells (Thurman and Gerba, 1989). Silver ions attached to the surface of a container can keep the water in the container free of bacteria, because the silver ion would still be able to bind to bacter i al surface and inact i vate it (Just, 1936 )

PAGE 35

25 Three possible mechanisms for inhibition of bacterial cells by silver have been proposed: (1) interference with electron transport, ( 2) binding to the DNA, and ( 3) intercalation with the cell membrane (Tilton and Rosenberg, 1978). Silver ions can easily form insoluble compounds with anions, sulfhydryl groups, and many biological materials, such as enzymes which are responsible for disinfectant activity of silver. No protein or nucleic acid leakage has been reported upon treatment of bacterial cells with silver, which suggests that cell lysis did not occur. Higher doses of silver are needed to inactivate viruses, probably because it is harder to denature the viral protein coat than to oxidize the complexed sulfhydryl groups (Rahn, 1973). Copper Copper and other metal ions may function as either lewis acids or bases when present as metal complexes (Thurman and Gerba, 1989). Using their protons, they can facilitate hydrolysis or nucleophilic displacement and make a bond available for nucleophilic attack by hydroxyl radicals. ~opper ,/ attacks respiratory enzymes in the cell membrane of h coli by binding thiol or other groups on protein molecules. The injured cells decrease the oxygen use and increase use of fermentation pathways during the recovery (Plastourgen and Hoffman, 1984) Low level of cu 2 in chlorine-free distr i bution

PAGE 36

26 water caused injury of coliform population upto 64%. Copper concentrations 0.025 and 0.050 mg/1 caused over 90% injury within 6 and 2 days, respectively. Studies of the metabolism of injured fu coli cells indicated that the respiratory chain is at least one site of the copper-induced damage in injured cells (Domek et al., 1984). Synergistic Effect of Hydrogen Peroxide and other Agents Hydrogen peroxide shows synergism with both physical and chemical factors. Its antimicrobial activity is enhanced when it is used with transition metals such as iron, copper, silver, and physical agents like ultraviolet light and ultrasonic energy (Block, 1991; Bayliss and Waites, 1980; 1979). Ultraviolet Light Hydrogen peroxide is highly effective against spores when used simultaneously with ultraviolet light (Bayliss and Waites, 1980; 1979; Waites et al., 1979; Halliwell and Gutteridge, 1984). Bacillus subtilis spores were killed 2000 fold faster when treated with H 2 o 2 and UV simultaneously (Bayliss and Waites, 1979). Irradiation of fu coli and Streptococcus faecalis with 254 nm UV light and incubation with 1% H 2 o 2 bring 99.99% inactivation in just 30 s (Bayliss

PAGE 37

27 and Waites, 1979). Waites et al. (1981) showed that the greatest kill of !L._ subtilis spores in the presence of H 2 0 2 was accomplished with UV irradiation around 270 nm. Their results also showed that the action of UV light is not directly on the spore DNA but is related to the free hydroxyl radicals produced from H 2 o 2 that are close to or within the spores (Waites et al. 1979). Another physical agent that has shown synergism with hydrogen peroxide is ultrasonic energy. Ultrasonic energy is thought to disperse and agitate the cell aggregates, increasing surface contact with the disinfectant, increasing the permeability of the cell membrane to the disinfectant, and accelerating the interaction between the disinfectant and the cell components. (Ahmad and Russel, 1975). Growing the cells in the presence of high concentration of iron made~ aureus cells more susceptible to H 2 0 2 (Repine 1981). Hydroxyl radical formation was enhanced due to high iron content of these cells. Iron catalyzed the formation of hydroxyl radicals through Fenton reaction (Repine et al., 1981a ) This reaction has been discussed earlier in this section. These investigators further showed that hydroxyl radical scavengers inhibited the cell killing by H 2 0 2

PAGE 38

28 Copper Treating spores of Clostridium bifermentans with copper made the spores 3000 fold more susceptible to H 2 0 2 (Bayliss, 76). Oithiothreitol (OTT) treated spores of Clostridium perfringens were shown to be inactivated by H 2 o 2 500 fold more than untreated controls (Waites et al., 1979; Halliwell and Gutteridge, 1984). It was proposed that OTT removes the protein coat that protect the spores from H 2 o 2 and that copper increases the rate of breakdown of H 2 o 2 and the rate of cleavage of peptide bonds by H 2 0 2 (Block, 1991). Goals and Prospects of the Present Study In this study the impact of hydrogen peroxide treatment on&..:.. coli and bacteriophage MS2 under different experimental conditions has been investigated. The results indicate that bacteria and viruses are inactivated at different rates and probably different mechanisms are involved in inactivation of bacteria and viruses by hydrogen peroxide. The survey of treatment of microorganisms with H 2 o 2 revealed that gram negative bacteria are relatively more susceptible to hydrogen peroxide than gram-positive ones. Inactivation of&..:.. coli and MS2 by hydrogen peroxide and chlorine have been compared. The use of hydrogen peroxide in selective recovery of bacteriophages from natural water samples and in modification

PAGE 39

29 of diatomaceous earth and sand for removal/inactivation of bacteria were also investigated.

PAGE 40

CHAPTER 2 INACTIVATION OF BACTERIA AND VIRUSES BY HzOz Mechanism of Inactivation by H 2 Q 2 h coli cells exposed to H 2 o 2 are inactivated by at least two lethality modes distinguishable by metabolic, kinetic, and genetic criteria (Imlay and Linn, 1986). Mode one killing occurs at low ( less than 5mM) concentrations of H 2 O 2 and requires metabolically active cells. DNA damage appears to be the site for mode one killing, since strains deficient in RecA protein, exonuclease III, or RecBC enzyme are especially vulnerable to this mode of killing. Both mutagenesis and induction of SOS DNA repair systems accompany challenges at H 2 o 2 concentrations representing mode one killing. Mode two cell killing doses not require metabolically active cells and occurs at H 2 O 2 concentration of more than lOmM. DNA repair deficient mutant cells are inactivated at the same rate by mode two doses as wild type strains, which indicates that DNA damage is not the major target of H 2 o 2 attack (Imlay and Linn, 1987). There is a considerable amount of evidence indicating that hydroxyl radicals produced via the Fenton reaction are a JO

PAGE 41

31 major cause of H 2 o 2 -induced lethality (Brandi et al., 1989b; Repine et al., 1981a). This has been demonstrated by the fact that both iron chelators and hydroxyl radical scavengers reduce the level of active hydroxyl radicals and protect cells from killing by H 2 0 2 Brandi et al. (1989) showed that hydroxyl radicals are involved in the production of mode two but not mode one ki 11 ing by H 2 0 2 In this study the impact of hydrogen peroxide treatment on laboratory microorganisms was surveyed. The results showed that gram negative bacteria are more susceptible to H 2 0 2 treatment than gram positive bacteria, and bacteriophages are relatively more resistant than bacteria in general. Lipid containing phages were more susceptible to H 2 o 2 than were non lipid-containing bacteriophages. The inactivation of h coli and bacteriophage MS2 by H 2 o 2 was compared under different experimental conditions. The results indicate that h coli is inactivated at a much faster rate than MS2. A survey of the effects of chemical agents on inactivation of bacteria and viruses with H 2 o 2 was done. These results revealed that hydroxyl radicals are not the major cause of inactivation of bacteriophages, (as they are in killing of h coli), and probably different mechanisms are involved in inactivation of the two organisms. The effects of physical factors, type of inactivation media, and ultraviolet light on inactivation of microorganisms by H 2 o 2 has been studied. The release of cytoplasmic components

PAGE 42

32 (as the sign of cell membrane damage) into the culture medium and changes in cell volume upon treatment of h coli cells with H 2 0 2 was also investigated. Inactivation of bacteria and viruses by two oxidizing compounds H 2 o 2 and chlorine, was compared to investigate any differences between the mechanism of action of each biocide. Alasri et al. (1992) reported that h coli,~ aureus, and..:.. aeruginosa are inactivated at faster rate by chlorine than hydrogen peroxide. The combined treatment of hydrogen peroxide with several other agents such as silver nitrate, cupperic chloride, and ultraviolet light are investigated. These studied were carried out to show any possible synergistic effects between these agents and hydrogen peroxide. There is synergistic effect between two chemicals if the combined treatment shows greater toxicity (more than the sum of the individual treatments (Marking and Dawson, 1973). Materials and Methods Materials; the following chemicals were purchased from Fisher Scientific Co. (Fairlawn, N.J.) and Sigma Co. (St. Louis, Mo): Na 2 HP0 4 NaSCN, sodium citrate, thiourea, urea, tween 80, dimethylsulfoxide, sodium chloride, sodium fluoride, sodium trichloroacetate, sodium hypochlorite, silver nitrate, and hydrogen peroxide.

PAGE 43

33 Cultures and growth conditions; the following bacterial strains used in this study were obtained from the American Type Culture Collection; Escherichia coli C-3000 (ATCC 15597) Escherichia K-12 (ATCC 10240), recA Staphylococcus aureus (ATCC 27660), Bacillus cereus (ATCC 11778), Streptococcus faecal is (ATCC 19344) Streptococcus saprophyticus (ATCC 15305), Mycobacterium smegmatis (ATCC 10143), Pseudomonas aeruginosa (ATCC 10145), Vibrio cholerae (ATCC 14035). The bacteriophages used in this study and their host cultures were MS2, T2, X174 (h coli C-3000), P22, PRD-1 Salmonella typhimurium (ATCC 19585), and (Pseudomonas syringae ATCC 21781). NBY medium (ATCC medium 815) was used for growing~ syringae cultures were incubated at 28C. All other bacteria were grown in 3% tryptic soy broth at 37C. Inactivation studies; bacteria and viruses were suspended in either 0. 05 M sodium phosphate, pH 7, or 3 % tryptic soy broth, pH 7, for inactivation studies. Concentrated viruses were resuspended in 0. 05M Na 2 HPO 4 pH 7 and diluted to produce a final concentration of approximately 10 5 PFU/ml. Dilutions of hydrogen peroxide were prepared fresh from a 50% (w/v) stock solution. All experiments were carried out at room temperature and pH 7 unless specified otherwise. To remove the residuals of hydrogen peroxide excess catalase (0.05 mg/ml, final concentration) added to the inactivation medium and/or samples were diluted appropriately.

PAGE 44

34 Cell leakage study; h coli cells were treated in the presence of various concentrations of H 2 0 2 Cells were pelleted by centrifugation for 2 min in an Eppendorf microcentrifuge tube, and the supernatants analyzed. The concentration of magnesium ions was measured using the "60-Second Magnesium" reagent system supplied by the American Monitor Corporation (Indianapolis, IN). Leakage values are expressed as a percentage of that released into the supernatant by vortex mixing suspensions for 30 sec with chloroform (0.2ml per 2 ml of suspension). Enzyme assays; beta-galactosidase activity and synthesis were measured according to Dutton et al. ( 1988) Lactate dehydrogenase activity was estimated according to Beutler (1975). Oxygen consumption experiment; oxygen consumption was measured after treating h coli cells with various concentrations of H 2 0 2 washing the cells with growth medium, and resuspending the cells in fresh growth medium, by using a Gilson respirometer. Measure of oxidizing power; the amount of oxidizing power in H 2 o 2 and chlorine solution was measured using an iodometric method adopted from "Standard Methods for the Examination of Water and Wastewater" 17th edition (APHA-AWWA-WPCF, 1989). Determination of riboflavin solubility; to determine the solubility of riboflavin, an excess amount of riboflavin was added to the test solutions. Undissol v ed riboflavin was

PAGE 45

35 removed by centrifugation for 10 min at 3000 rpm. The solutions were then diluted in distilled water and dissolved riboflavin was determined by measuring the absorbance at 444 nm using spectrophotometer (Bausch & Lamb, Spectronic). Results were compared to the solubility of riboflavin in deionized water as the control. Study of synergism; bacteria and viruses were tested with hydrogen peroxide or with hydrogen peroxide in the presence of ultraviolet light, cupperic chloride, or silver nitrate. For Ag+ tests the solutions were collected in Chambers' solution {Chambers et al., 1962). Statistical analysis; each experiment was done at least three times in duplicate. The student t-test {p=0.05) was used for comparing the mean values. All statistical analysis were carried out using "instat" program for IBM. Results Inactivation of Bacteria and Viruses with H 2 Q 2 The effects of H 2 o 2 treatment for 1 hat room temperature (0.1% H 2 0 2 ) on different bacteria and viruses is presented in {Table 1). Non-spore forming Gram positive bacteria, acid fast bacteria, and bacteriophages are inactivated at a slower rate than gram negative bacteria. more resistant to H 2 o 2 than Bacteriophages are relatively bacteria (Figure 1 and 2).

PAGE 46

36 Bacteriophages had an average of 75% survival compare to only 5.3% survival for bacteria (Table 1). The lipid-containing phages such as and PRD-1 were much more susceptible to H 2 0 2 than other bacteriophages (Figure 1 and 2). Their survival rate is intermediate between that of gram positive bacteria and other non-lipid phages, (Figures 1 and 2). The toxicity of hydrogen peroxide to h coli was investigated under different experimental conditions. Two modes of killing were apparent; mode one killing refers to lethality at lower concentrations of H 2 0 2 (0.01% or less, >2.9 mM) and mode two lethality at which killing occurs at higher concentrations (0.05% or more). When repair deficient strains of h coli K-12 were treated with H 2 o 2 they were much more susceptible to mode one killing than wild type. H 2 0 2 concentration representing mode two lethality inactivated both wild type and repair deficient mutant strains at the same rate, (Figure 3). In order to investigate the possible damage to bacterial cell membranes as a result of H 2 o 2 treatment, the appearance of cytoplasmic components in the culture medium and loss of cell volume were studied. h coli cells were allowed to grow either in the presence or absence of 0.01% or 0.1% (representative of mode one and mode two lethality respectively) H 2 0 2 and, at various time intervals, the absorbance at 550 nm was measured along with cell viability (T able 2). In add i tion cells were analyzed at differen t

PAGE 47

37 intervals microscopically to monitor any morphological changes. Changes were observed as regards the morphology of h coli cells treated with H 2 o 2 The changes were dependent on the amount of oxidant used in the treatment. Cells remained unit sized 3 h after treatment with 0.1% H 2 o 2 (data not shown). Results depicted in Figure 4 shows treatment with 0.01% H 2 o 2 resulted in a growth lag followed by a partial recovery of cell growth. Exposure to 0.1% H 2 o 2 also produced a continuous, time dependent decrease in optical density (Table 2). In fact, OD values were reduced to about 50% of the initial OD (immediately before adding H 2 O 2 ) after 5 h treatment with 0.1% H 2 o 2 (Figure 4). The viable count indicated great initial loss in cell colony forming ability (>50%) after treatment with O. 01% H 2 O 2 which was followed by partial recovery of viable but damaged cells. In similar experimental conditions, cells were centrifuged immediately after addition of H 2 O 2 and after 5 h treatment. Examination of pellet sizes visually revealed that the pellet size was substantially smaller after 5 h treatment with H 2 O 2 Results in Table 2 shows that the pellet volume reduced more than 30% after 5 h. This indicates the loss of cell components as the result of possible cell membrane damage. In an attempt to further investigate the damage to cell membrane integrity, the leakage of molecules upon treatment with H 2 o 2 was studied. Mg2+ leakage happened only when cells

PAGE 48

38 were treated with H 2 o 2 concentrations representing mode two and killing was dependent on the exposure time and H 2 0 2 concentration (Figure 5). Studies on lactate dehydrogenase leakage revealed that the enzyme accumulated in a time and concentration dependent fashion in the supernatant (Table 3). The leakage of materials with low levels of H 2 o 2 treatment was negligible. One can conclude that even though there might be some membrane damage with lower levels of the oxidant, the damage is repairable, or there is too little membrane damage to cause leakage of detectable levels of cell components. When higher concentrations of H 2 0 2 were used the cells were not able to repair the initial damage and, the longer the cells were exposed to H 2 0 2 the more materials were leaked out. The ability of h coli to consume oxygen was tested in different experimental conditions. Results in Figure 6 show cells treated with 0.01% H 2 0 2 partially recovered their oxygen consumption ability whereas the cells subjected to 0.1% H 2 0 2 never regained their ability to consume oxygen. The same cultures were tested for the effects of H 2 o 2 on their electron transport system by their ability to reduce tetrazolium chloride before and after treatment with H 2 o 2 Cells retained 51% of their ability to reduce tetrazolium chloride after treatment with O. 1% H 2 0 2 compare to 79% reduction after treatment with 0.01% H 2 o 2 (Figure 6).

PAGE 49

39 Factors Influencing Inactivation Rate by H 2 Q 2 Effects of pH, temperature, and inactivation media H 2 0 2 toxicity is affected by the pH of the solution, with greater inactivation under acidic conditions. As it is shown on Table 4, H 2 o 2 is much more effective at pH 5 than at pH 7 and pH 9. Table 4 shows the effect of heat on inactivation of h coli and MS2 with hydrogen peroxide. More than 90 % of h coli and almost all of MS2 survived 0.1% H 2 0 2 treatment for upto 1 h at 4C. At 25 and 37C, however, inactivation patterns were similar for MS2 and not for h coli. H202 was more toxic for h coli at 37C than 25C. The oxidizing power of hydrogen peroxide in solution is influenced by the amount of organic matter presents in the solution. There is a reciprocal relationship between the amount of oxidizing power and the inactivation efficiency of hydrogen peroxide. Table 4 shows the relationship between the relative amount of oxidizing power in different solutions and the survival of MS2. Phosphate buffer with no organics present generates the highest oxidizing power and greatest inactivation. It is quite possible that some of the oxidizing power is consumed by the organic matter present in the inactivation medium. So the consumed oxidizing power will not be available for inactivation activity.

PAGE 50

40 Effects of chemical agents on H 2 Q 2 toxicity The effects of chaotropic and antichaotropic agents on the lethality of H 2 o 2 was studied using bacteriophages X174, and MS2. The list and concentrations of chemicals used in these experiments are given in figure 7. Results depicted in, Figure 7 show the survival of MS2 after treatment with 0.1% H 2 0 2 in the presence of different chemical agents. The correlation between the inactivation rate and the degree of hydrophobicity (as measured by riboflavin solubility) is shown in, Figure 10. Sodium thiocyanate along with other chaotropic agents, urea, tween 80, and thiourea, which weaken hydrophobic association, significantly increased the rate of inactivation of MS2 by hydrogen peroxide noticeably. In contrast sodium citrate and other antichaotropic agents, which strengthen hydrophobic associations, decreased the lethal effect of H 2 o 2 A very high correlation (r 2 =0.988) was observed between the effects of salts on inactivation of MS2 by H 2 o 2 and their effects on hydrophobic interactions (Figure 10). However, this correlation was not as high when non-salt chemicals, urea, thiourea, and tween 80 were taken into account (0.550). Details of statistical analysis are discussed in the materials and methods section. Results of testing other bacteriophages such as X174 and with hydrogen peroxide in the presence of salts and other agents is given in Table 5. Xl 7 4 ( a single-stranded DNA

PAGE 51

41 bacteriophage) and (a lipid-containing virus) inactivation studies have similar profiles as MS2 which is a single stranded RNA virus (Table 5). Hydroxyl radical scavengers The role of hydroxyl radicals in the inactivation of MS2 by H 2 0 2 was studied using the hydroxyl radical scavengers thiourea, DMSO, and ethanol. Results shown in Figure 8 indicate that 35 rnM thiourea significantly enhanced the virucidal action of H 2 o 2 on MS2. 9% ethanol significantly decreased the inactivation of MS2 by H 2 o 2 Ethanol at high concentration reacts with H 2 o 2 and decreases the oxidizing power of H 2 o 2 50 rnM DMSO did not have any significant effect on H 2 0 2 inactivation. These results indicate that hydroxyl radicals are not the major lethal species against bacteriophages. The effects of DMSO and thiourea on riboflavin solubility (as the measure of their effects on hydrophobicity of the solution) were also studied (Figure 10). These results indicated that thiourea weakens the hydrophobic interactions and so increases H 2 o 2 lethality. The addition of DMSO does not change the hydrophobicity of the solution, therefore it does not have any effect on H 2 0 2 killing efficiency.

PAGE 52

42 Combined Inactivation of E. coli and MS2 by H 2 Q 2 and UV-Light The combined effects of H 2 o 2 and UV light on bacteriophage MS2 and~ coli were studied using 366 nm UV irradiation and 0.01% H 2 O 2 (w/v) for 30 min. Table 6 depicts the results of this study. While 0. 01% H 2 o 2 or UV treatment are almost ineffective against MS2, their combined treatment is significantly different ( at p=0. 05 level) from individual treatments against this bacteriophage. The combined treatment of h coli with UV light and H 2 o 2 is not greatly different from the individual treatment of bacterium, showing that the effects of UV and H 2 o 2 are additive. Comparison Between H 2 Q 2 and Chlorine Inactivation The oxidizing power of H 2 o 2 and chlorine were compared in their ability to inactivate h coli and bacteriophage MS2. Concentrations of the biocides with equal oxidizing power as measured by an iodometric method were used in inactivation studies. The rate of inactivation of h coli by chlorine was slightly faster than the inactivation by H 2 o 2 (Figure 11). On the other hand the inactivation rate of bacteriophage MS2 was much faster with chlorine than H 2 o 2 (Figure 12). The biosynthesis and activity of enzyme beta-galactosidase were measured (as described by Dutton et al., 1988) after treating h coli cells with chlorine and hydrogen peroxide. Chlorine

PAGE 53

43 decreased the enzyme biosynthesis 4 times more than hydrogen peroxide. At the same time the enzymatic activity of beta galactosidase went down 12-fold faster by hydrogen peroxide than chlorine (Table 7). Combined Inactivation of E. coli and MS2 with H 2 Q 2 and AgN0 3 In order to find any synergism between H 2 o 2 and silver nitrate h coli and MS2 were treated with AgN0 3 and in combination with H 2 0 2 Table 8 and Figure 13 shows the results of these experiments. Statistical analysis of these data indicates that the treatment of both MS2 and h coli with a combination of O 25ppm AgN0 3 and O. O 1% H 2 o 2 significantly reduced the viability of both cultures when compared with indi victual treatments. Comparison of the concentrations needed to produce 50 % mortality (Marking and Dawson 1973) reveals that the effect of mixtures of AgN0 3 and H 2 0 2 is synergistic. This means that silver nitrate enhances antibacterial activity of hydrogen peroxide. Discussion A survey of microbial inactivation by H 2 o 2 indicated that gram negative bacteria are more sensitive to H 2 0 2 than gram positive bacteria (except fL_ cereus) and viruses. Lipid containing bacteriophages are inactivated at a faster rate

PAGE 54

44 than non-lipid phages. Treatment of h coli cells with increasing concentrations of hydrogen peroxide results in a bimodal pattern of lethality. Low concentrations (<0.01%) of the oxidant produce mode one killing which can be distinguished from mode two killing (generated by concentrations of H 2 o 2 higher than 0.1%) since the first mode of lethality unlike the second, requires active cellular metabolism and is enhanced in DNA repair-deficient strains. A possible explanation for the differential inactivation can be due to the differences in diffusion rates of H 2 o 2 through bacteria and viruses membrane barriers. Apparently lipids are a major factor in the susceptibility of an organism to an oxidizing agent. In this study different components of h coli were examined for possible damage as the result of H 2 o 2 treatment. Leakage studies shows that the integrity of the cell membrane is interrupted after treating h coli cells for more than 2 h. Significant amount of the enzyme lactate dehydrogenase and Mg2+ ions show up in the inactivation medium. Others have shown leakage of molecules such as glucose-6-phosphate, Na, and K+ (Girotti and Thomas, 1984; Vander Zee et al., 1985). Cell volume decreased as measured by a reduction in dry weight and a decrease in absorbance at 550 nm. Brandi (1989a) reported a reduction in cell size after similar treatment. Oxygenic respiration is also disrupted by H 2 o 2 treatment. The abilit y of h coli cells to utilize oxygen as well as the

PAGE 55

45 integrity of their electron transport systems are adversely affected after treatment with H 2 0 2 Other enzymes are also targets of H 2 o 2 treatment. The enzyme beta galactosidase activity as well as its synthesis are inhibited by the action of H 2 0 2 The toxicity of H 2 0 2 to bacteria and viruses increases as the pH of the inactivation mixture decreases. However the pH factor is not as pronounced as it is with other disinfectants such as phenol and chlorine (Block, 1991). H 2 0 2 is more effective against bacteria and viruses at 37C than 4C. The presence of organic materials in the inactivation media reduces the lethality by consuming part of the oxidizing power generated by H 2 0 2 The above physical and chemical factors are also important in disinfection activity of other oxidizing agents such as chlorine (Block, 1991). The inactivation of viruses by H 2 o 2 is affected by the presence of chaotropic and antichaotropic salts, which indicates that compounds that affect hydrophobic interactions are important in the action of H 2 0 2 against viruses. Chaotropic salts are likely to increase the lethality of H 2 o 2 by weakening the hydrophobic associations between viral capsid proteins. On the other hand antichaotropic salts shield the virus against H 2 0 2 by strengthening the hydrophobic associations. Hydrophobicity plays an important role in the interaction of viruses with other molecules as it gives the strongest

PAGE 56

46 contribution to viral capsid protein stability (Farrah, 1981; 1982; Stryer, 1984). The catalytic role of iron in bactericidal action of H 2 0 2 through generation of hydroxyl radical scavengers has already been discussed (Repine et. al., 1981a). It has been shown that hydroxyl radicals promote h coli cell killing when higher doses of H 2 0 2 (>20 mM) are used (Brandi et. al., 1987). These investigators concluded that hydroxyl radicals are required for mode two cell killing by H 2 o 2 Although hydroxyl radical scavengers decrease the toxicity of H 2 0 2 to bacteria, they act differently for viruses. In fact thiourea, a hydroxyl radical scavenger, enhances the lethality of H 2 0 2 against MS2, while it decreases the antibacterial action of hydrogen peroxide by as high as 96% (Repine et al., 1981b; Brandi et al., 1989a), Figure 9. DMSO reduced the antibacterial activity of H 2 0 2 substantially (Repine et al., 1981b; Brandi et al., 1989a), Figure 9, but made no difference on inactivation of MS2 by this agent. These results show that different mechanisms are involved in the inactivation of viruses and bacteria. Unlike the killing action of H 2 o 2 against h coli, hydroxyl radicals are not the major cause of virucidal action of H 2 0 2 Urea, a weak hydroxyl radical scavenger, inhibited the antibacterial action of H 2 0 2 by 6% (Repine et al., 1981b). On the other hand, urea greatly enhanced the virucidal action of H 2 o 2 on bacteriophages. These results suggest that the oxidizing power of H 2 o 2 may be directly involved in the inactivation of

PAGE 57

47 viruses, and there is (are) no intermediate species (hydroxyl radical) involved. In contrast to the similarities between bactericidal activities of chlorine and H 2 0 2 viruses are inactivated at different rate by these two biocides. It is not known why chlorine inactivates bacteria and viruses faster than H 2 o 2 The rate of diffusion through natural barriers such as the cell membrane or the virus capsid could be different for each biocide. Domingue et al. ( 1988) found that in contrast to ozone and chlorine, H 2 0 2 inactivation requires much higher concentrations. Hydrogen peroxide inactivates gram positive bacteria, gram negative bacteria, and bacteriophages at different rates. It is possible that the differential inactivation could be the result of the reduced ability of hydrogen peroxide to penetrate the microorganism barriers. Hydrogen peroxide and chlorine affected enzyme synthesis and activity of h coli at different rates which further suggests that chlorine and hydrogen peroxide probably have different inactivation mechanisms. Exactly how HOCl destroys microorganisms has never been demonstrated experimentally. Oxygen liberated as the result of HOCl and other oxidizing agents such as H 2 0 2 and KMn0 4 activity is able to combine with components of cell protoplasm. Despite the production of large amounts of nascent oxygen by H 2 0 2 and

PAGE 58

48 KMn0 4 they are not able to kill microorganisms as fast as chlorine. Rudolph and Levine (1941) proposed that the bactericidal effect of hypochlorite is completed in two successive phases; (1) the penetration of the germicide into the bacterial cells and (2) the chemical reaction of the germicide with protoplasm of the cell to form toxic complexes that destroy the organism. These phases could also be applied to H 2 0 2 Si mi lar killing mechanisms have been proposed for chlorine and H 2 o 2 These mechanisms of act ion include; ( a) change in membrane permeability, (b) inactivation of key enzymes required for metabolism, and (c) interference with genetic materials in the cell. Combined treatment of h coli and MS2 with H 2 0 2 and UV light shows that UV light enhances H 2 0 2 inactivation of both cultures, but this synergistic effect is more pronounced for the virus. A kill of 99.99% was produced in cell suspension of h coli by UV irradiation at 254 nm for 30 sin the presence of 1.0% H 2 0 2 (Bayliss 1980-52). These results indicate that the synergistic inactivation of h coli requires higher concentration of hydrogen peroxide Silver nitrate enhanced inactivation of bacteria and viruses by H 2 0 2 significantly. This effect is synergistic and not additive as analyzed according to the Marking and Dawson. (1973) method. Both silver and H 2 0 2 are known t o att ac k D N A and cause subsequent inh i bition of ce ll div i s i o n ( B randi et a l

PAGE 59

49 1989a; Imlay and Linn 1987; Richards, 1981). No cell lysis is observed with silver but silver ions can intercalate with the cell membrane. This intercalation may enhance the damage to the cell membrane by H 2 0 2 In summary, two modes of lethality postulated for the action of H 2 o 2 on coli with H 2 o 2 H 2 o 2 inactivates bacteria and viruses at different rates. Bacteria and viruses are apparently inactivated by different mechanisms as the proposed role of hydroxyl radicals in bactericidal action of hydrogen peroxide in not a major factor in inactivation of bacteriophages by hydrogen peroxide. Lipid is a major factor in the enhanced inactivation of lipid-containing bacteriophages as compared with the non lipid phages. It is concluded that combined damage to different components of h coli occurs, which is followed by disruption of the cell membrane as the cause of death. Hydrophobic interactions influence the effects of H 2 0 2 on MS2 phage. The stronger the hydrophobic association, the more resistant MS2 is against H 2 0 2 It is the oxidizing power of H 2 0 2 that is directly involved in inactivation of viruses and no intermediate species involved in the process. Hydrogen peroxide shows synergism with uv-light and silver nitrate. Bayliss and Waites (1976) reported that combined treatment of Clostr idium spores with copper and hydrogen peroxide is significantly more effective than treating spores with copper or hydrogen peroxide alone.

PAGE 60

50 This work studied the inactivation of bacteria and viruses with hydrogen peroxide. The data indicate that viruses are relatively more resistant to the action of hydrogen peroxide than bacteria. The effects of salts and detergents as well as hydroxyl radical scavengers on the inactivation of viruses with hydrogen peroxide are studied as well. The combined treatment of bacteria and viruses with hydrogen peroxide and other agents such as silver nitrate, cupperic chloride, and UV light is also studied. Finally the inactivation of bacteria and viruses with chlorine and hydrogen peroxide are compared.

PAGE 61

51 Table 1. Inactivation of microorganisms with hydrogen peroxidea Type Strain % Survivors Bacteria s. aureus 17.6 2. 9b s. faecalis 11. 4 2. 1 s. saprophyticus 6.6 1.0 M. smegma tis 6.2 2.4 E. coli 0.2 0.1 P. aeruginosa 0.1 0.1 V. cholerae < 0.1 B. cereus < 0.1 Mean 5.3 Bacteriophages MS2 90 3 T2 72 6 X174 74 6 P22 90 2 43 5 PRD-1 46 6 Mean 69 a Cells were treated with 0.1% hydrogen peroxide for 1 h; catalase was then added to stop inactivation. b. Values represent the mean and standard deviation.

PAGE 62

Table 2. Effects of H 2 0 2 treatment on .E...:_ coli volume Time % Initial (min) Volumea 00 550 CJD % Survival 75 95 0.57 7 150 92 0.50 0 225 74 0.36 0 300 67 0.29 0 a. Cells were allowed to grow in the presence of 0.1% H 2 o 2 and aliquots were removed at different intervals. Volume of the pellet, optical density, and viability were determined. Results are the mean values two separate experiments. 52

PAGE 63

Table 3. Lactate Dehydrogenase Leakage in the Extracellular medium from fu coli Cells Exposed to Hydrogen Peroxidea Treatment (Time) None 100 min 300 min 0. 01% H 2 o 2 100 min 300 min o. 1% H 2 0 2 100 min 300 min OD 550 5.4 11.8 0.55 2.90 0.60 0.32 LOH mUnits/ml/0.D 0.50 0.63 9.75 6.90 38.50 71. 30 a. Cells were grown either in the absence or presence of 0.01% or 0.1% H 2 0 2 At each time interval OD values were measured and LOH activity in the supernatant was measured. Results represent the mean from two separate experiments. 53

PAGE 64

Table 4. Effects of pH, Temperature, and Medium on Inactivation of MS2 with Hydrogen Peroxidea %Oxidizing % Survival Inactivation Medium Powerb pH TempC MS2 fu coli 50mM Na 2 HPO 4 100 7 37 56 3 > 0.1 50mM Na 2 HPO 4 NA C 7 25 70 5 0.1 50mM Na 2 HPO 4 NA 7 4 92 2 90 3 50mM NazHPO4 NA 5 25 0 0 50mM NazHPO4 NA 9 25 95 4 NDd 54 0 3% Beef Extract 46 7 25 90 2 0.2.1 3% Tryptic Soy Broth 81 7 25 84 4 > 0.1 a. Bacteriophages were treated with 0.1% H 2 O 2 under the given conditions for 60 min. b. Oxidizing power measured by iodometric methods as described in the text. C. Not applicable. These values were not different from the oxidizing power of 50mM NazHPO4 pH 7 d. Not done.

PAGE 65

a. b. C. Table 5. Effect of Chemical Agents on Killing of Bacteriophages by Hydrogen Peroxidea 55 Virus (Type) Salt (Type)b % Survivalc MS2 Buffer (control) 70 MS2 l00rnM NaCitrate 90 MS2 l00rnM NaSCN 39 MS2 l00rnM NaF 75 X174 Buffer (control) 68 X174 l00rnM NaCitrate 79 X174 l00rnM NaSCN 25 X174 l00rnM NaF 80 Buffer (control) 29 l00rnM NaCitrate 42 l00rnM NaSCN 6 l00rnM NaF 35 Bacteriophages were treated with 0.1% hydrogen peroxide for lh at room temperature. 50rnMNa 2 HPO 4 pH 7 buffer used as control buffer; all other salts were dissolved in this buffer. 5 4 5 1 3 5 2 6 8 6 3 7 indicates values significantly different from the buffer control at p=0.05 level. *

PAGE 66

a b Table 6. Treatment of h coli and Bacteriophage MS2 with H 2 O 2 and Ultraviolet lighta. Strain Treatment % Survival a MS2 0.01% H2O2 95 3 MS2 366 nmUV 98 1 366 nm UV + MS2 0.01 % HzOz 32 6 h coli 0.01% HzO2 75 10 h coli 366 nm UV 85 4 366-nm UV h coli 0.01% H2O2 40 3 Cells and Phages were treated for 30 min in 0.05 mM Na 2 HPO 4 pH 7 at room temperature. Indicates values significantly different from the individual treatments at p=0.05 level. 56

PAGE 67

Table 7. Effect of Hydrogen Peroxide and Chlorine on Biosynthesis and Activity of BetaGalactosidase Enzyme in h coli Treatment None 0.05% Chlorine Units of Biosynthesis 100 13.5 3.4 Units of Activity 100 2.7 32.4 % Viable 100 0.1 > 0.1 Bacterial cells were treated with hydrogen peroxide 57 and chlorine as described in text. Beta-galactosidase activity was measured at 420 nm.

PAGE 68

Table 8. Inactivation of Bacteriophage MS2 and h coli with H 2 O 2 Silver Nitrate, and Cupperic Chloride a. b. Percent Survival Treatment MS2 0.01% HzOz 38 5 95 3a 0.25 ppm AgNO 3 32 2 69 3 0.5 ppm CuC1 2 95 1 NDb 0.01% HzOz 0.25 ppm AgNO 3 3 2 21 3 0.01% H2O2 0.5 ppm CuC1 2 68 6 ND Bacteria and phage samples were treated at room temperature. Samples were drawn after 30 min and diluted in Chamber's solution (for silver nitrate treatment) and 1% tryptic soy broth (for cupperic chloride) treatment containing catalase. Values are mean standard deviation. Not Done. 58

PAGE 69

Fig 1. Inactivation of bacteria and viruses was carried out by treating cultures with 0.1% hydrogen peroxide in 0. 05M Na 2 HP0 4 pH 7. Samples were drawn at different intervals and assayed for survival.

PAGE 70

(IJ > > L :J CJ) .... C Q) u L Q) Q_ so Inactivation of Bacteria and Viruses by Hydrogen Peroxide 60 OL..._ __ ___.-====-__.L-. __ __..___---=:::::s;;a 0 30 60 Time (min) 90 120

PAGE 71

Fig 2. Bacteriophage and bacterial cultures were treated in the presence of increasing hydrogen peroxide concentrations for 30 min in 0.05M Na 2 HP0 4 pH 7. Inactivation stopped either by dilution or addition of excess catalase. Results represent the means of triplicate determinations.

PAGE 72

cu > > L ::J (/) .. C QJ u L QJ 0... so Inactivation of Bacteria and Viruses by Hydrogen Peroxide o.____ __ ..._ __ ..._ __ ..____ ___ 0 OS .10 Percent H 2 0 2 (w/v) 62

PAGE 73

Fig 3. Survival of h coli C-3000, wild type, and repair deficient mutant strain, K-12 cells after treatment with hydrogen peroxide. Cells were treated with increasing H 2 o 2 concentrations for 15 min in 0.05 M Na 2 HPO 4 pH 7, at 25C. Cell viability was measured using plate count agar. Results represent the mean of 3-5 separate experiments.

PAGE 74

Inact i vat i on of Bacter i a by Hydrogen Perox i de 100 ------------------. (U > > L ::J c.n +-J C (1.) u L (1.) Q_ so K1 2 0 "--------'-------'----__.___ __ ___, 0 OS 10 Percent H 2 0 2 (w / v ) 6 4

PAGE 75

Fig 4. Growth rates of E. coli cultures growing in 3% tryptic soy broth containing various concentrations of hydrogen peroxide. Cells were plated for survival test at different intervals on plate count agar. Results represent the mean of triplicate determination.

PAGE 76

,,.----.. 0 ..-i X 0 L() L() 0 0 '--../ 0 .---t 0) 0 _J 2 0 15 10 05 Growth of E coli in the Presence of Hydrogen Peroxide ----control ----/. / ~.._ 0 01%H 2 0 2 /. -----/_ ____ ... ... -- --------01%H 2 0 2 o o ..__-~ _... __ .....___ ___._ __ ....___ ___._ __ .__ _. 0 60 120 Time (min) 180 240 O'I O'I

PAGE 77

Fig 5. Effects of hydrogen peroxide on cell leakage. h coli cells were grown in presence of various concentrations of hydrogen peroxide for 1 h. Magnesium ions in the extracellular medium was measured using 60 Second Magnesium test as described in the text Results represent the mean and standard deviation of triplicate exper!ments.

PAGE 78

Q) 0) cu _y cu Q) _J 0) L 50 0 0 0 68 .---... 1.0

PAGE 79

Fig 6. Effect of hydrogen peroxide treatment on the oxygen consumption by h coli. Cells were treated with either 0.01% or 0.1% H 2 o 2 After centrifugation, washing, and resuspension in fresh growth medium (3% TSB), oxygen consumption was measured.

PAGE 80

C Q) 0) >. X 0 >. L 0 :1 Effects of H 2 0 2 Treatment on Oxygen Consumption by E coli soo.-----------.--------, Control 250 0 ~::::::::.:__J__ __ _L_ __ ___,1....._ __ _L_ __ __J 0 30 60 Time (min) 70

PAGE 81

Fig 7. Effects of Chemical agents on inactivation of bacteriophage MS2 with hydrogen peroxide. Phages were treated with 0.1% hydrogen peroxide for 1 h in 0.05 mM Na 2 HP0 4 pH 7.0 at room temperature in the presence or absence of different chemicals. Inactivation reactions were stopped either by dilution or addition of excess amounts of catalase. Sample~ were assayed for survival as described in the text.

PAGE 82

cu > > L :J (/) ... C Q) u L Q) 0... Effects of Chemical Agents on Inactivation of MS2 w i th H 2 0 2 100.---------------, ............. ~ .:::: :. :::: :. tttttt.:t} : ............. :. tttt:ttjj tttt:tJ:jj ...... ... .. tttt:tJ:jj """"" tttt:tJ:jj """ "" 50 ............................ .... .... .-:::::.. tttt:tJ:jj """ "" ttttttti :::::::::: :. t:ttt.ttit :::: ::::::. ........... .. I-+-.-+~ ....... .... 14-+-+......,...: : : ::: : .:: : ............................ ... .. ..... ..... .. .... 14-+-+......,... ........... ............................ ::: : : : : .. :. 14-+-+......,... : :. ........................ ...... .. .. t:t:tttJ ......... .. t:t:tttJ ': .. : : : : : : : :: : t:t:tttJ ........... :. .... : .. 0 L
PAGE 83

Fig 8. Effects of hydroxyl radical scavengers on the inactivation of bacteriophage MS2 with hydrogen peroxide. MS2 samples were treated with 0.1% H 2 o 2 in 0.05 mM Na 2 HP0 4 pH 7.0 at room temperature in the presence or absence of hydroxyl radical scavengers and assayed for survival rates.

PAGE 84

ro > > L :J en .. C (1) u L (1) Q_ Effect of Hydroxyl Radical Scavengers on H202rnactivation 100 so 0 I._ Q) (/) 4L 4:J 0 CD L t..n 0 L 0 (J\ M 0 0 74

PAGE 85

Fig 9. Prevention of hydrogen peroxide-induced cell by thiourea, ethanol, or dimethyl sulfoxide. thiourea, .. ,dimethyl sulfoxide and control, CJ. Adopted from Brandi et al. killing Ethanol, !"'i 1 (1989b).

PAGE 86

76 100 A B MODE ONE MODE TWO KILLING KILLING 80 -' 60 > 0:: :::> Cf) 40 20

PAGE 87

Fig 10. Relationship between the effects of different chemical agents on inactivation of bacteriophage MS2 and their effects on hydrophobic interactions, as measured by riboflavin solubility. Results depicted here are adapted from Figures 7 and 8.

PAGE 88

_o ::J 0 Ul C > cu 4-0 _o er Relationship of Riboflavin Solubility to Inactivation of MS2 150 .----------------, 100 so b T h1ourea Urea a Salts only b All compounds 0 .___ __ _.__ __ _.__ __ ..___ ______. 0 so Percent Survival 100 78

PAGE 89

Fig 11. Comparative inactivation of h coli with H 2 O 2 and chlorine h coli cells were treated with hydrogen peroxide or hypochlorous acid that had the same amounts of oxidizing power (as measured by an iodometric test) for 1 h at room temperature in 0. 05M Na~HPO 4 pH 7. Cells were plated on plate count agar for viability.

PAGE 90

(U > > L :J (/) +> C Q) u L Q) o_ Inactivation of E coli with Hydrogen Peroxide and Chlorine 100------------~ 50 Chi :> ... H2D2 0 -----0 3 6 g Oxidizing Power (MEQ/L) 80

PAGE 91

Fig 12. Comparative inactivation of bacteriophage MS2 with H<0 2 and chlorine. Virus samples were treated with hydrogen peroxide or hypochlorous acid that had the same amounts of oxidizing power (as measured by an iodometric test) for 1 h. at room temperature. Samples were assayed for phage viability as described in the text.

PAGE 92

cu > > L :J Ul .. C Q) u L Q) o_ Inactivation of Hydrogen Peroxide MS2 with and Chlorine 100 '-----.. l: ....... -~ ---i-----I~ 50 Chlorine 0 L__ __ ..L..__ __ ..-=====------------__J 0 3 6 g 12 15 O x idiz in g Power ( MEQ/L ) 82

PAGE 93

Fig 13. Combined inactivation of h coli with hydrogen peroxide and silver nitrate. Samples were treated at room temperature, and at different intervals aliquots were removed and diluted in Chambers solution containing catalase.

PAGE 94

cU > > L :::J U) .. C Q) u L Q) Q_ 10 50 0 Combined Inactivation of E coli with H 2 0 2 and AgN0 3 50 Time (min) 100 84

PAGE 95

CHAPTER 3 SELECTIVE RECOVERY OF BACTERIOPHAGES BY USING H 2 0 2 Literature Review Overgrowth of host bacteria by indigenous bacteria can interfere with the plaque assay of bacteriophages from such environmental materials as water, wastewater, fresh, and marine sediments (Goyal, 1987). It is often necessary to eliminate or inactivate the indigenous bacterial flora to prevent the interference with the host bacterial lawn and resolution of the plaques (Kennedy et al., 1985). Several procedures have been developed to reduce the contamination by bacteria without greatly reducing the number of bacteriophages in water samples. These methods include membrane filtration, incorporation of antibiotics into the assay media, the use of selective media, and chloroform pretreatment of the samples. Membrane filtration through microporous filters with pore diameters larger than viruses but smaller than bacteria is a simple method for removing unwanted bacteria (Cornax, 1990; Tartera and Jofre, 1987). Tartera and Jofre (1987) used membrane filters with 0.22m pore size (Millipore Corp. Bedford, Mass.) to decontaminate samples such as ground water, raw sewage, and other sewage-contaminated waters. 85

PAGE 96

86 The use of antibiotic-resistant host bacteria along with the incorporation of antibiotics into the assay medium can reduce the interference of the indigenous flora in the sample. Gerba et al. (1978) incorporated penicillin and streptomycin into media and used an antibiotic resistant strain oft. coli for assay of their isolates from natural sewage solids. Lipid solvents, such as ether and chloroform, can be used to inactivate bacteria but not viruses that do not contain lipid. Treating samples such as activated sludge, polluted river water, and sewage effluent with chloroform produced large reductions of indigenous bacteria and enhanced plaque resolution in previous studies (Cornax, 1990; Clarke, 1983; Glass and O'Brien, 1980; Tartera and Jofre, 1987; Vaughn and Metcalf, 1975). Another approach to control indigenous bacteria is the use of selective media which select for the phage host bacteria and inhibit growth of other bacteria (Goyal, 1987; Grunnet, 1977; Kennedy; 1985; Parker, 1981). Kennedy et al. (1985) compared selective media like EC medium, Gram Negative, and nutrient broth supplemented with sodium deoxycholate for enumeration of coliphages in activated sludge effluent and polluted lake water. These workers concluded that certain selective media, such as EC and Gram Negative media, allow the assay of environmental samples either directly or following concentration procedures by the agar overlay method without any decontaminating procedure or the use of antibiotics.

PAGE 97

87 Other decontamination methods use detergents. Phages have been reported to be highly resistant to various detergents including saponin, sodium dodecyl sulfate, and sodium deoxycholate (Burnett, 1940). The treatment of samples with cationic detergents such as Emulsol-607, Zephiran, and cetylpyridinium chloride has been suggested for assay of coliphages in sewage (Kalter, 1946). Our initial studies on the inactivation of laboratory strains of bacteria and phages by hydrogen peroxide revealed that bacteria are inactivated more rapidly than phages by this oxidant. This led to the use of hydrogen peroxide to treat natural samples to reduce the bacterial contamination. In some cases overgrowth of indigenous gram positive bacteria was a problem. It is known that gram positive bacteria are more sensitive to certain dyes, such as crystal violet, than gram negative ones (Bitton and Koopman, 1988). Therefore, crystal violet was incorporated into the bottom agar for standard phage assay. This study shows that the assay of bacteriophage from environmental samples is improved by treating the samples with hydrogen peroxide, and by adding crystal violet to the agar plates This procedure enables us to quantify bacteriophages in samples containing a relatively high number of indigenous bacteria.

PAGE 98

88 Materials and Methods Samples; raw sewage and trickling filter effluent samples were collected from the University of Florida sewage treatment plant. Lake water was collected from Lake Alice on the University of Florida campus. This lake receives the final effluent from the campus sewage treatment plant. Barnyard materials were from the Department of Agricultural Engineering research facilities at the University of Florida. Samples were either treated directly or 2 to 3 liters of the samples were concentrated by passing the samples through positively charged filters (Virosorb lMDS [AMF Cuno, Inc., Meridian Conn.]) (Shields, 1986b). Adsorbed viruses were eluted by passing 10% beef extract, pH 9 through filters. The filter eluates were adjusted to pH 7 and mixed with an equal volume of saturated ammonium sulfate. The resulting floe was collected by centrifugation and resuspended in 0.05M Na 2 HPO 4 pH 7 (Sheilds and Farrah 1986a). Some samples were concentrated by using magnetite-organic flocculation (MOF) method in which isoelectric casein and magnetite was added to samples and pH adjusted to 4. 5. The resulting f locs were collected using a magnet and dissolved in 0.05M Na 2 HPO 4 pH 7 as previously described (Bitton et al., 1979). Cultures; the following bacterial strain used in this study as host was obtained from the American Type Culture

PAGE 99

89 Collection; h coli C-3000 (ATCC 15597). Cultures were grown in 3% tryptic soy broth pH 7. Bacteriophage assay; unless stated otherwise the host bacteria for enumeration and detection of bacteriophages was h coli C-3000. EC medium and plate count agar with or without incorporation of 1 ppm crystal violet were used as bottom agar and 3% tryptic soy broth with 0.8% agar was used as the top agar for the standard agar overlay technique used in detection and enumeration of all phages. 0.1 ml sample and 0.1 ml host bacterium were mixed with 4 ml top agar in all assays. Water samples were either assayed directly or after treatment by one of the following methods: 1. H 2 0 2 2. Chloroform, 3. Filtration with a series of 0.45and 0 2m Filterite filters (Filterite Corp., Timonium, MD) in 25-mm holders. Samples were treated with either 0.1 or 0.5% hydrogen peroxide for 1 h. Residual hydrogen peroxide was removed either by dilution or addition of catalase (140 units, 0.05 mg/ml w/v, final concentration). Chloroform treatment was as previously described (Kennedy et al., 1985). The HPCV Phage Assay Procedure; the hydrogen peroxide crystal violet (HPCV) procedure is summarized as the following; Natural samples were treated, either directly or after concentration, with 0.1% hydrogen peroxide (v/v final concentration) on a shaker (800 rpm) at room temperature for 1 h. Samples were either diluted or catalase was added (0.05

PAGE 100

90 mg/ml, final concentration). Standard agar overlay assay was performed using bottom agar containing 1 ppm crystal violet. Chemicals and Media; EC medium and tryptic soy broth were from Difeo Laboratories, Detroit, MI. Catalase was obtained from Sigma Chemical Co., St. Louis, MO. All other chemicals were obtained from Fisher Scientific Products. Statistical analysis; each experiment was done at least twice in triplicate. The student t-test (p = 0.05) was used for comparing the mean values. Initial inactivated studies faster RESULTS revealed that with hydrogen Escherichia coli peroxide than was was bacteriophage MS2 (Figure 1 and 2). As it is shown in Figure 1, about 15% of MS 2 phage was inactivated by 0.1% hydrogen peroxide (v/v final concentration) in time it took to inactivate >99% of h coli. MS2 was resistant to even higher concentrations of hydrogen peroxide (Figure 2). This pattern of inactivation was then confirmed with other phage types and bacterial species (Table 1). Except for vegetative cells of IL_ cereus, gram positive and acid fast bacteria were relatively more resistant to hydrogen peroxide than were gram negative bacteria. Treating samples of trickling filter effluent and raw sewage with O. 1% H 2 O 2 for 1 h at room temperature reduced

PAGE 101

91 bacterial numbers on plate count agar by approximately 3 log 10 Plating the samples on crystal violet plates led to approximately one additional log 10 reduction in bacterial numbers (Table 9) For phage assays, plates of untreated initial samples were read after 4-6 h incubation; further incubation led to overgrowth of indigenous bacteria and difficulty in plaque visualization. All other plates were read after 24 h incubation. Plaque numbers showed no significant change but plaque visualization was greatly improved when treated samples were plated on plates containing crystal violet. We used the hydrogen peroxide-crystal violet (HPCV) technique for decontamination of samples that have been prepared with different concentration procedures (Bitton et al. 1981; Bitton et al., Sheilds et al., 1986b). 1979; Sheilds and Farrah, 1986a; sample (barnwash) and For a more heavily the samples obtained contaminated using virus concentration procedures, two different H 2 0 2 concentration ( 0. 1 and 0. 5%) were tested. As shown in Table 10 bacterial reduction obtained for unchlorinated effluent was lower with 0.1% H 2 0 2 than the reduction obtained following treatment of unconcentrated samples (Table 9) An additional 0. 6 to 3 log 10 reduction in bacterial numbers was obtained when a higher concentration ( 0. 5%) of hydrogen peroxide was used. The combination of 0.1 % hydrogen peroxide treatment and plating on crystal violet plates reduced the number of total bacteria

PAGE 102

92 by an average of 3 log 10 when compared to the controls (Table 11) As shown in Table 12, increasing the concentration of hydrogen peroxide to 0.5% reduced the total number of bacteriophages. The use of 0.1% hydrogen peroxide and crystal violet plates permitted visualization of plaques that were not detectable after plating of the untreated samples. The HPCV assay procedure was compared with some of the currently used decontamination methods (Table 13). Samples of raw sewage and trickling water effluent were used directly. Samples were assayed for bacteriophage after treatment with chloroform or hydrogen peroxide, or after filtration, and assayed using EC or plate count agar with or without crystal violet as bottom agar. Plates of untreated samples were heavily contaminated with indigenous bacteria and could not be scored for plaque enumeration following 24 h incubation, however, plaques could be read after 4-6 h incubation. All other plaques were counted after 24 h incubation. Samples plated on EC (without any prior treatment) were small and the cloudy appearance of the plates made it hard to read the plaques. Pretreatment of the samples with chloroform significantly reduced the number of plaques and permitted growth of a small number of indigenous bacterial colonies on the plates. Filtration of the samples with 0.25 and 0.45 um filters resulted in plates free of indigenous bacterial contamination but this method was good only for small volume

PAGE 103

93 samples since the filters tend to clog rapidly. The student t-test was used to compare the results on Table 13. Only treatment with chloroform and plating on EC medium gave statistically significant lower numbers of plaques. Some examples of assays of treated and untreated samples are shown in Figure 14. Pretreatment of samples with O. 1% H 2 o 2 and subsequent assay on crystal violet plates gave easily readable plaques with little contamination by indigenous bacteria {Figure 14). Overgrowth of indigenous bacteria on plates with untreated samples and a reduction in plaque numbers following treatment of samples with chloroform were observed {Figure 14a and 14c) DISCUSSION Reduction or elimination of the indigenous bacterial flora in environmental samples is sometimes needed to prevent overgrowth of the phage host bacteria and to permit an accurate estimation of the phage population in these samples. The currently available methods have certain limitations. For example pretreatment of samples with chloroform generally reduces the number of bacter iophages recovered from environmental samples and in some cases more than 60% inactivation of phages has been reported (Kennedy et al. 1985). Also, it has been suggested that the use of chloroform should be reduced since it is toxic and is a known carcinogen

PAGE 104

in test animals (Clarke 1983). 94 Although EC and other selective media prevent the overgrowth of indigenous bacteria (Kennedy 1985), EC medium gave small and indistinct plaques in our experiments. The selective nature of these plates also limits their use to certain hosts such as h coli. Furthermore, both chloroform treatment and plating the samples on EC medium substantially lowered enumeration of plaques recovered from natural environments (Table 13). Filtration is another effective method for removing bacterial contamination from water samples. Ideally, filtration should remove bacteria while letting viruses pass through. However, filters tend to clog very fast with some samples, and may retain some of the particle-associated viruses. Antibiotics may be added to the plates used for phage assay to control indigenous bacteria. This procedure is limited to assays employing resistant host bacteria. The summary of the hydrogen peroxide-crystal violet (HPCV) phage assay procedure is given in Figure 15 and in the methods section. This procedure does not have some of the limitations associated with the current methods. Although the procedure caused the inactivation of some bacteriophages (upto 50% loss of bacteriophage T7), no significant loss of viral inactivation was observed with the natural samples. Therefore, the proposed method provides an alternative choice for the decontamination of natural samples. The procedure gave the same or more plaques than other methods for every sample

PAGE 105

95 tested. Very few bacteria escape the treatment so a healthy lawn is formed and this greatly improves the visualization and subsequent enumeration of plaques. The hydrogen peroxide used in this process is not toxic to humans at the level at which it is used, since the recommended concentration of hydrogen peroxide for mouthwash solution is 1. 5%. Also, residual hydrogen peroxide can be removed easily using catalase. Hydrogen peroxide was found to inactivate bacteriophages and bacteria at different rates. A concentration of o. 1% hydrogen peroxide reduced the numbers of several bacteria by an average of 94% but caused an average of 25% inactivation in the numbers samples with of bacteriophages hydrogen peroxide tested. Treating natural selectively reduced the indigenous bacterial flora and permitted better visualization of plaques of lawns of Escherichia coli strain C-3000. In some cases, indigenous gram positive bacteria were relatively resistant to the action of hydrogen peroxide but their growth could be limited by incorporation of crystal violet into the bottom agar used for plaque assays. The use of hydrogen peroxide treatment and crystal violet-containing plates permitted recovery of more phages from natural samples than other procedures, such as the use of chloroform to pretreat samples or the use of selective plating agar such as EC medium. In summary an improved procedure for bacteriophage assay of water samples from different sources has been proposed. The

PAGE 106

96 method is easy to perform, safe, inexpensive, and most importantly i t is better or at least as good as the currently used procedures for recovery of bacteriophages from natural samples. This procedure can be used for any water sample and may be applicable in assay of bacteriophages of other gram negative bacteria (as it is sometimes used for evaluation of water quality) (Wentsel et al., 1982) The procedure was found useful for detecting bacteriophages in direct and concentrated environmental samples containing relatively high levels of bacteria.

PAGE 107

Table 9. Inactivation of bacteria and bacteriophages by hydrogen peroxide in wastewater samples (direct assaysa) Initial H.2Q2 Treated Reduction Log 10 Sample Organism ( PCA} b (PCA) ( CVA) C (PCA} (CVA) Raw Aerobic Sewage Bacteria 3.2.lxl0 6 4. 8. 4xl0 3 6. 6. 4xl0 2 -2.82 -3.69 Bacteriad phage 4.5.4xl0 2 3.0.4xl0 2 5. 2. lxl0 2 -0.07 -0.06 Unchlorinated Effluent Aerobic Bacteria l.5.lxl0 5 9.8.7xl0 1 9. 7. 2xl0o -3.19 -4.19 a b C d Bacteriophage 5.6.8xl0 2 d4.6.5xl0 2 5.3.4xl0 2 -0.08 -0.02 Untreated samples and samples that had been treated with 0.1% hydrogen peroxide for 1 h were plated on plate count agar or plate count agar with 1 ppm crystal violet. Untreated samples assayed for bacteriophages using plate count agar were counted after 4 to 6 h incubation; all other samples were counted after 24 incubation. Plate count agar Plate count agar containing 1 ppm crystal violet Values represent the mean and standard deviation for triplicate samples. Underlined values are not significantly different at the p = 0.05 level.

PAGE 108

Table 10. Combined effects of crystal violet and hydrogen peroxide on the indigenous bacterial population in environmental samples with high levels of bacteria. Bacteria CFU/L Sample %H 0 2 PCAC CVAd Barnwasha 0 2. 5. lxl0 9 6. 9. 2xl0 6 Barnwash 0.1 8.1.lxl0 7 7. 4. lxl0 5 Barnwash 0.5 2. 0. 5xl0 7 6. 0. 3xl0 5 Unchlorinated Effluentb 0 2.6.7xl0 7 1. 8. 6xl0 6 Unchlorinated Effluent 0.1 1. 2. 4xl0 6 1. 7. 6xl0 5 Unchlorinated Effluent 0.5 3. 5. 4xl0 3 < 5 Lake Aliceb 0 4. 5. 3xl0 7 3. 6. 6xl0 6 Lake Alice 0.1 4. 1. 7xl0 6 1. 0. 7xl0 4 Lake Alice 0.5 2.6.4xl0 4 2.3.3xl0 3 a. Barnwash materials were treated directly. b Trickling filter effluent (unchlorinated) and Lake Alice water samples were concentrated by filter adsorption/elution and magnetic organic flocculation, respectively, as described in the text. C. Plate count agar d. Plate count agar containing 1 ppm crystal violet

PAGE 109

Table 11. Sample Barnwash Unchlorinated Effluent Lake Alice Meanc Reductions in bacterial numbers using H 2 0 2 treatment and/or crystal violet plate a Reduction(log 10 ) between treated and untreated samples -1. 49 -2.56 -3.53 -1.34 -1.16 -2.18 -1.04 -1.10 -3.65 -1. 39+0. 19A -1.61+0.67A -3. 12+0. 67 8 a Data from Table 10 b. c Plate count agar containing 1 ppm crystal violet Mean values with the same superscript are not different at the p = 0.05 level using the t-test I.D I.D

PAGE 110

Table 12 a b C. d e f. Combined effects of crystal violet and hydrogen peroxide on the indigenous bacteriophage population in environmental samples with high levels of bacteria. Phages PFU/L Sample %H202 PCAC CVAd Barnwash 8 0 NCe 1. 8. Oxl0 5 Barnwash 0.1 2. ll. lxl0 5 5. 8. lxl0 5 Barnwash 0.5 NDf 1. 1. 3xl0 5 Unchlorinated Effluentb 0 NC 1.l.7xl0 4 Unchlorinated Effluent 0.1 2. 0. lx10 4 2. 4. Jx10 4 Unchlorinated Effluent 0.5 8. 6. Jx10 3 1. 6. lxl0 4 Lake Aliceb 0 NC 0. 9. lxl0 4 Lake Alice 0.1 NC 1. 4. Oxl0 4 Lake Alice 0.5 NC 0. 1. Oxl0 2 Barnwash materials were treated directly Trickling filter effluent (unchlorinated) and Lake Alice water samples were concentrated by filter adsorption/elution and magnetic organic flocculation respectively as described in the text. Plate count agar Plate count agar containing 1 ppm crystal violet Not able to count because of high level of indigenous bacteria Not done t-' 0 0

PAGE 111

Table 13. Comparison of other decontamination methods with the Hydrogen peroxide-Crystal Violet Procedurea a. b. c. d. Treatment Untreatedb Untreated-plated on CVAd Filtered Chloroform (5%} Plated on EC agar 0. 1% H 2 O 2 0.1% HzOz-plated on CVA Bacteriophage detected (PFU/ML} in: Unchlorinated Effluent 5. 5. 2xl0 2 c 4. 6. 5xl0 2 6. 6. 2xl0 2 0. 8. 3xl0 2 2. 7. 6xl0 2 4. 6. 4xl0 2 5. 3. 4xl0 2 Raw Sewage 5. 1. 6xl0 2 c 4. 5. 6xl0 2 4. 1. 2xl0 2 l.4.4xl0 2 3. 1. lxl0 2 3. 8. 1x10 2 5. 2. lxl0 2 Samples were plated in triplicates using h coli C-3000 as the host following the indicated treatment. Unless otherwise stated, plate count agar was used as the bottom agar. Plaques were counted after 4 to 6 h incubation; all other samples were counted after 24 h incubation. Plate count agar containing 1 ppm crystal violet I-' 0 I-'

PAGE 112

Fig. 14. Equal volumes of raw sewage samples plated without treatment (A), following treatment with 0.1% hydrogen peroxide (B), and following treatment with 5% chloroform (C) as described in the text.

PAGE 113

I-' 0 w

PAGE 114

Fig 15. Steps involved in the hydrogen peroxide-crystal violet (HPCV) phage assay procedure.

PAGE 115

Treat sample or concentrate with O i"Mt (v/v) hydrogen peroxide + Shake for 1 hr at room temperature + Add 0 05 mg/ml catalase + Perform the Agar Overlay Assay using bottom agar containg 1 ppm crystal violet 105

PAGE 116

CHAPTER 4 MODIFICATION OF SOLIDS WITH HYDROGEN PEROXIDE Literature Review Microporous filters remove microorganisms mechanically to provide high purity water for the pharmaceutical and other industries. Major problems with these filters are a tendency for rapid clogging and bacterial growth on the filters (Levy, 1991). For example microporous filters used in industry face pore blockage and subsequently slow flow and short life, in part because of bacterial growth on the filters. Filters with small pores may be useful for small scale filtration of water over short times but are impractical for the large-scale or long-term treatment of drinking water (Levy, 1991; Gerba et al., 1988). An ideal filter would be the one that inactivates microorganisms as it mechanically removes them and is able to treat large volumes of water. These filters may remove microorganisms without the need for the addition of oxidizing agents to water samples. Gerba et al. ( 1988) studied the removal of bacteria and viruses from water by filters containing magnesium oxide or magnesium peroxide. Filters 106

PAGE 117

107 with magnesium oxide removed bacteria and viruses from water but permitted survival of the viruses and bacterial survival In contrast, filters with magnesium peroxide both removed and inactivated bacteria and viruses (Gerba et al., 1988). Previous studies have shown that microporous filters, diatomaceous earth, and sand can all be modified by in situ precipitation of metallic salts (Farrah and Preston, 1991). This modification increased the concentration of metals on the solids and increased virus absorption in water (Farrah et al., 1991; Farrah and Preston, 1991). This work presents another procedure for modifying heat-stable solids such as sand and diatomaceous earth to produce solids coated with magnesium peroxide. The solids can be used to make filters that inactivate bacteria or as a source of oxidizing power that can be added to solutions to control bacterial growth. Materials and Methods Chemicals; diatomaceous earth (Grade 1), sodium iodide, soluble starch, sodium thiosulfate, and magnesium chloride were obtained from Sigma Chemical Co., St. Louis, MO.; hydrogen peroxide (50%), was obtained from Fisher Scientific Corp. Modification of diatomaceous earth; diatomaceous earth was mixed with 2, 4, or 6 M magnesium chloride (2 ml/g) for 30 min. The excess solution was decanted off and the material

PAGE 118

dried at 37c. 108 The diatomaceous earth with a dried layer of magnesium chloride was then heated in a porcelain crucible with a bunsen burner until the odor of hydrochloric acid could no longer be detected. After cooling, the diatomaceous earth (with a presumed coating of magnesium oxide) was mixed with 50% hydrogen peroxide ( 2 ml/g) with overnight agitation. Since the reaction between hydrogen peroxide and magnesium oxide is exothermic, the reaction vessel was cooled by immersing in a water bath (25C). After overnight incubation, the material was dried and stored at room temperature until used. The amount of magnesium peroxide associated with the diatomaceous earth was determined by measuring production of iodine from iodide (Clesceri, 1989). Bacteria; the following bacteria were used: Salmonella typhimurium (ATCC 19585); Escherichia coli (ATCC 15597); Pseudomonas aeruginosa (ATCC 10145); and Staphylococcus aureus (ATCC 27660). Bacteria were grown overnight in 3% trypticase-soy broth before use. All bacterial assays were performed using plate-count agar (Difeo Lab). Filtration experiments Suspensions of diatomaceous earth or diatomaceous earth coated with MgO 2 in deionized water were collected in AP20 fiberglass prefilters in 25-mm holder (Millipore Corp, Milldale, CN). Approximately 0.5 g was held by a filter. The tap water was dechlorinated (as checked by tolidine) by exposure to UV light and seeded with approx i mately 10 6 CFU/ml of bacteria from an overnight

PAGE 119

culture. 109 Ten ml of the seeded water was then passed through several filters. Bacteria in the initial samples and filter effluents were measured. Bacteria removed by the filters were recovered by mixing the filter contents with 10 ml of 3% beef extract (Difeo), pH 7. Bacteria were recovered from filters immediately after filtration or after 24 hrs incubation at room temperature. Experiments with indigenous bacteria and chlorinated tap water were carried out by passing 50 ml of tap water through the filters every 24 hrs. The aerobic bacteria present in the filter effluent were counted on plate count agar. Batch experiments; fifty ml volumes of nutrient broth (Difeo), unchlorinated effluent from the University of Florida wastewater treatment plant, or dechlorinated tap water were incubated at room temperature with no treatment or after mixing with 0.5 g of untreated diatomaceous earth or diatomaceous earth coated with magnesium peroxide. Either 5 or 15 g of treated diatomaceous earth was added to 2 1 of dechlorinated tap water in 4 1 beakers. The total aerobic bacteria in the samples was measured initially and every 48 h. Similar experiments were carried out using modified sand and dechlorinated tap water.

PAGE 120

110 Results Increasing the concentration of magnesium chloride used for the initial treatment increased the percent of magnesium peroxide on the coated diatomaceous earth (Table 14). Since increasing the concentration of magnesium chloride over 2 M only produced marginal increases in the final concentration of magnesium peroxide, 2 M magnesium chloride was used to treat diatomaceous in subsequent studies. Similar results were obtained with sand. The magnesium peroxide coating was fairly stable at room temperature, but did decline with prolonged storage. Approximately 40% of the activity was lost after six months storage at room temperature in tightly closed container. Filters containing either untreated diatomaceous earth or diatomaceous earth coated with magnesium peroxide removed greater than 99.9% of bacteria from water (data not shown). Initially, approximately the same number of seeded bacteria could be recovered from both types of filters (Table 15) After incubation at room temperature for 24 hrs, the numbers of bacteria on untreated filters increased but the numbers on filters with magnesium peroxide declined by approximately 4 log ,o. Filters made with untreated diatomaceous earth and diatomaceous earth coated with magnesium peroxide were used to filter tapwater. The effluent from filters made with untreated

PAGE 121

111 diatomaceous earth initially had few bacteria (Figure 16). Daily filtering of chlorinated tap water with from 1 to 100 CFU/ml of aerobic bacteria through the filters led to the presence of relatively high numbers of bacteria in the effluent after two days. Filters made with diatomaceous earth coated with magnesium peroxide delayed the appearance of bacteria in the effluent by only one or two days. Although the number of bacteria in the effluent from filters containing modified diatomaceous earth was lower than the number in the effluent from filters containing untreated diatomaceous earth, it was still significantly higher than that of the tapwater before filtering. About 86% of the oxidizing power initially present on coated diatomaceous earth remained on the material removed from filters after 10 days of passing tapwater through filters and leaving the filters at room temperature. However, the indigenous bacteria were allowed to grow on the filters and more than 10 4 CFU/ml were present in the effluent (Table 16). Therefore, the presence of bacteria in the filter effluent was not associated with a significant reduction in oxidizing power on filters. Successive washes of o. 5 g unused diatomaceous earth coated with magnesium peroxide led to release of small amount of oxidizing power (approximately 2% of the total activity for each wash) as compared to total oxidizing power present on coated material. Most of the oxidizing activity (87%) remained

PAGE 122

112 on coated diatomaceous earth after 4 consecutive washes, (Table 17). Nutrient broth left at room temperature for 48 hrs became turbid and heavily contaminated with bacteria (Table 18). Nutrient broth incubated in a similar way with diatomaceous earth coated with magnesium peroxide remained clear and had relatively few bacteria. The number of bacteria in unchlorinated effluent declined after 48 hrs incubation at room temperature. The addition of diatomaceous earth coated with magnesium peroxide to unchlorinated effluent resulted in a greater decline in bacterial numbers over the same period. The addition of diatomaceous earth or sand coated with magnesium peroxide to tap water produced a low level of hydrogen peroxide activity. The level equivalent to approximately 0.06% and 0.16% hydrogen peroxide, for sand and diatomaceous earth respectively, changed little between 15 min and 10 days of incubation. This relatively low level of hydrogen peroxide was sufficient to control the bacterial levels in tap water for at least two weeks. The results of studies using 0.5 g of modified sand and 50 ml of dechlorinated tapwater are shown in Figure 17. In this case, the ratio of water to sand was 100 ml/g. Increasing the volume of water to 2 1 and the amount of sand to 5 g (400 ml/g) led to the growth of bacteria in a few days (data not shown). As shown in Table 19, fifteen grams of modified diatomaceous materials in 2 1 of water (133 ml/g) was

PAGE 123

113 sufficient to maintain bacterial counts of< 10 4 /ml for 6 days (sand) or 8 days (diatomaceous earth). After 14 days incubation at room temperature, the level of bacteria was greater than 10 6 /ml in all samples. Salmonella typhimurium (10 4 /ml) was added to the samples after 14 days. typhimurium added to the samples with modified diatomaceous earth or sand declined by 3 log 10 in 48 hrs. In contrast, typhimurium added to the samples without modified materials increased by 1 log 10 (Table 19) Discussion Oxidizing agents, such as chlorine and ozone, have been used for many years to inactivate bacteria in water distribution systems. Although they are effective in controlling pathogens in water, the use of these agents is associated with problems such as high cost, a possible health hazard due to the production of undesirable disinfection by products, and an absence of residual activity in the treated solution (Clarke, 1983). For example, ozone must be generated at the point of treatment and does not persist long in water. Chlorine residuals are maintained during distribution of tap water but are dissipated when water is stored exposed to UV light. The dissipation of chlorine in swimming pools is one example of this. In this study, we investigated the use of magnesium peroxide coated solids as a source of oxidizing

PAGE 124

114 power (hydrogen peroxide) to control the growth of bacteria in water. Inactivation of bacteria by hydrogen peroxide has been examined in several previous studies (Asghari et al., 1992; Imlay and Linn, 1986; 1988; Diguiseppi and Fridovich, 1982; Repine et al., 1981a). Two different mechanisms of bacterial inactivation has been proposed (Imlay and Linn, 1988). DNA has been reported to be the primary target when less than 5 rnM hydrogen peroxide is used in the treatment. The mechanism of hydrogen peroxide killing at higher concentration has not been explained in detail. Membrane lipids, proteins, nucleic acids, and other vital components of the microorganisms could be the target of the oxidizing power of hydrogen peroxide. There are two problems associated with the use of aqueous solutions of hydrogen peroxide for disinfection. One is that high concentration of hydrogen peroxide is hazardous. If lower and safer concentrations are used to treat water, then relatively large volumes of hydrogen peroxide are needed. The use of magnesium peroxide-coated solids as a source of hydrogen peroxide in water reduces these problems. They provide a concentrated source of oxidizing power and can be handled safely. Magnesium peroxide has been incorporated into filters that inactivate bacteria and viruses removed from water (Gerba et al., 1988). These filters inactivated bacteria faster than viruses. We have previously found that diatomaceous earth can be coated with metallic hydroxides,

PAGE 125

115 metallic oxides and magnesium peroxide (Imlay and Linn, 1988; Diguiseppi and Fridovich, 1982). In this study, we examined the effect of magnesium peroxide coated solids on bacteria. Filters made with diatomaceous earth coated with magnesium peroxide removed bacteria from water and inactivated them over a 2 4 hr period. However, the f i 1 ters were not effective in maintaining low levels of bacteria in the filter effluent for longer than two to three days. Bacteria were present in high numbers in the filter effluents even though most of the oxidizing power initially associated with the diatomaceous earth remained on the coated materials even after 10 days of incubation. The addition of solids coated with magnesium peroxide to aqueous solutions was an effective method for keeping residual oxidizing activity in the solutions. In this study, diatomaceous earth and sand coated with magnesium peroxide provided a way to treat solutions with hydrogen peroxide for days to weeks. The slow release of hydrogen peroxide controlled the bacterial populations, even though the solutions were exposed to the air during the incubation period. A ratio of water to solids of approximately 100 ml/g was required to control bacterial growth in solutions. Much higher ratios permitted rapid bacterial growth in solutions. Even after indigenous bacterial levels had risen to over 10 6 CFU/ml, added~ typhimurium declined in samples containing coated sand and diatomaceous earth.

PAGE 126

116 Powdered magnesium peroxide can also be used as a source of hydrogen peroxide in solutions. However, coated diatomaceous earth or sand settle out of solutions and can be more easily removed from solutions than powered magnesium peroxide. These inactivate modified materials were able to remove and/or bacteria in solutions or on filters. The slow release of oxidizing power from magnesium peroxide coated sand or diatomaceous earth provides antibacterial activity in solutions for several days or weeks. Filters made with magnesium peroxide coated diatomaceous earth are however effective in controlling bacteria for only a few days. The reasons for the successful control of bacteria in batch systems but not on filters are not known. The oxidizing power existed on solids lasts relatively for a long time in closed container on shelf. In summary eating diatomaceous earth or sand that has been soaked with a solution of magnesium chloride results in coating of the materials with magnesium oxide. The addition of hydrogen peroxide converts this coating to magnesium peroxide. The magnesium peroxide coating on sand or diatomaceous earth is relatively stable during storage. Greater than 50% of the oxidizing power remained after 6 months storage at room temperature. The addition of diatomaceous earth or sand coated with magnesium peroxide to aqueous solutions led to release of hydrogen peroxide into the solution, reduced the number of

PAGE 127

bacteria, controls. and prevented bacterial growth as Filters made with modified 117 compared with and untreated diatomaceous earth removed bacteria from water equally well. However, bacteria survived or grew on filters made with untreated materials, but were greatly reduced in numbers on filters made with modified diatomaceous earth.

PAGE 128

Table 14. Influence of magnesium chloride concentration on the magnesium peroxide content of modified diatomaceous eartha. Concentration of magnesium chloride in modifying solution Percent magnesium peroxide a. 2 M 4 M 6 M 8.4 0.9 9.4 3.5 12.3 0.6 0.5 g of modified diatomaceous earth dissolved in 50 ml deionized water and the total oxidizing power calculated using iodometric method as described in the text.

PAGE 129

TABLE 15. Changes in bacterial numbers on diatomaceous earth filtersa a Bacteria (CFU/ml) recovered from filters Untreated Diatomaceous earth Diatomaceous earth with magnesium peroxide Bacteria Initial 24 hrs Initial 24 hrs Salmonella 5.1 3. 5 X 10 6 5.5 3.8 X 10 7 7.2 3. 3 X 10 6 4.7 3.3 xl0 2 ty:Qhimurium Escherichia 3.8 2. 3 X 10 6 7.4 9.1 X 10 7 1. 7 0.9 X 10 6 5.6 2.8 xl0 2 coli Pseudomonas 9.3 2. 4 X 10 6 1.2 1. 3 X 10 8 6.6 2.4 X 10 6 2.8 2.7 x10 2 aeruginosa Sta2hyloco8.3 1. 6 X 10 6 8. 1 4.1 X 10 7 7.1 0.2 X 10 6 7.3 0.9 x10 1 ccus aureus Ten ml of tapwater with approximately 10 6 of the indicated bacteria was passed through filters containing diatomaceous earth or diatomaceous earth that had been coated with magnesium peroxide. Bacteria removed from the water by the filters were recovered in 10 ml of 3% beef extract from three filters immediately and from three filters after 24 hrs incubation at room temperature (approximately 25C).

PAGE 130

Table 16. Stability of magnesium peroxide on diatomaceous earth during daily filtration of tapwater a. Time (Day) 1 10 MEQ/L Oxidizing Powera 8.4 0.9 7.2 0.2 CFU/Ml Bacteria in Effluenta <5 1.4 0.9 X 10 4 Peroxide activity in diatomaceous earth coated with MgO 2 content of 25mm filters was measured before and after ten days of daily passing of 50 ml tapwater through filters. The effluents were collected and tested for total bacterial count as described in the text. ..... [\J 0

PAGE 131

Table 17. Release of oxidizing power from magnesium peroxidea a. b. Sample MEQ/L Oxidizing Powerb Initial 8.59 0.62 Wash #1 0.17 0.02 Wash #2 0.16 0.02 Wash #3 0.16 0.01 Wash #4 0.15 0.03 Final 7.47 0.53 The oxidizing power associated with 0.5 g of unused diatomaceous earth coated with MgO~ (initial) and 0.5 g of coated DE that had been washed with four 50 ml volumes of deionized water (final) was determined. The oxidizing power in the 50 ml water washes was also determined. The values represent the mean and standard deviation for triplicate determinations.

PAGE 132

TABLE 18. Influence of diatomaceous earth coated with magnesium peroxide (DE-Mg0 2 ) on bacterial growtha Bacteria (CFU/ml)b Sample Initial After 48 hrs Nutrient broth 3.2 0.1 X 10 3 7.4 0.4 X 10 8 Nutrient broth 3.2 0.1 X 10 3 7.5 2.5 X 10 1 + DE-Mg0 2 Unchlorinated effluent 1. 4 0.2 X 10 6 5.5 2.3 X 10 5 Unchlorinated effluent 1. 4 0.2 X 10 6 1. 4 0.6 X 10 4 + DE-Mg0 2 a. Fifty ml of the indicated solution was left untreated or mixed with 0.5 g of diatomaceous earth coated with magnesium peroxide (DE-Mg0 2 } and incubated at room temperature for 48 h b. Total bacteria on plate-count agar, mean values and range per duplicate determinations. I-' N N

PAGE 133

Table 19.Influence of solids coated with magnesium peroxide on bacterial growth in tapwatera Bacteria (CFU/ML) in: Diatomaceous Days Control earth Sand 0 1. 2 0.5 X 10 1 1. 2 0.5 X 10 1 1. 2 0.5 X 10 1 4 1. 6 0.0 X 10 4 < 5 < 5 6 1.1 2.7 X 10 6 0.9 0.0 X 10 1 3. 6 1.2 X 10 3 8 7.4 5.2 X 10 6 1. 2 1. 6 X 10 3 4 8 0.8 X 10 5 14 2.5 3.4 X 10 7 2.8 3.9 X 10 6 3 .1 3.5 X 10 6 16 5.0 7.0 X 10 7 7.0 9.9 X 10 6 4. 2 5.7 X 10 7 16 s. 1. 5 0.0 X 10 5 3.5 4.0 X 10 1 2 4 1. 3 X 10 1 tygh i muriumb a b. Two thousand ml of dechlorinated tap water was incubated without (control) or with 15 grams of MgO 2 coated sand or diatomaceous earth. Total aerobic bacteria were assayed on plate count agar at the indicated intervals. Mean values and the range per duplicate determinations are given. tyghimurium (2.9 x 10 4 CFU/ml) was added to all water samples at 14 d.

PAGE 134

Fig 16. Bacteria in the effluents from filters made with diatomaceous earth (DE) was investigated by passing fifty ml tapwater through filters containing either untreated DE or DE coated with magnesium peroxide every 24 h. The total aerobic bacteria present in the filter effluents were determined using plate count agar. Symbols: o untreated DE; o treated DE. Bars indicate the Standard deviation for triplicate determinations.

PAGE 135

n hi bi t i on 0 f Bacterial Growth on F i I t e r s Made w i t h Mg0 2 Coat e d DE 6 --0 5 'Q) ..__, ...c u -0 4 3: co T r e o I e d DE 0 '<..:> <-> ..0 3 0 0 'Q)
PAGE 136

Fig 17. The influence of sand coated with magnesium peroxide on growth of bacteria in tapwater. Fifty mi of tapwater was dechlorinated by exposure to ultraviolet light and incubated with or without 0.5 g coated sand at room temperature. Total aerobic bacteria present were determined using plate count agar. Values are the mean and standard deviation for triplicate determinations. Symbols: 0 untreated sand; 0 treated sand.

PAGE 137

nl I uence ol Sand Coot cd w1 l h MgO 2 on the Growth of 0oclerio ,n Topwoler 8 ..-Unlrtoltd Sond 0 7 '(l) 6 ...c u 3: 0 CT) 5 0 H~~-0 '(__') u Ll 4 0 0 '
PAGE 138

CHAPTER 5 CONCLUSION There has been an upsurge of interest in hydrogen peroxide in the last few decades. For example there has been a renewed interest in the using of H 2 o 2 and other disinfectants in drinking water sterilization, since public conscience is geared towards the ban of the chlorinated compounds in drinking water. Use of H 2 o 2 as a disinfectant for inanimate materials and inert surfaces in the aseptic packaging of food products is very well documented. New applications for H 2 o 2 are presented continuously. In this study mechanisms of inactivation of bacteria and viruses by H 2 o 2 were investigated. Bacteria are more sensitive to hydrogen peroxide than viruses, and lipid containing bacteriophages are more sensitive than non-lipid containing phages. Exposure of h coli cells O .1% hydrogen peroxide produces a time-dependent reduction in optical density of the culture, viability, and cellular volume. Studies on the effects of salts on the inactivation of bacteriophage MS2 with hydrogen peroxide indicate that weakening the hydrophobic associations of the inactivation mixture, resulted in enhanced killing of MS2 by the biocide. 128

PAGE 139

129 A high correlation was observed between the effects of chemical agents on inactivation of MS2 by hydrogen peroxide and their effects on hydrophobic interaction (as measured by riboflavin solubility). This study shows that hydrophobic interactions influence the rate of inactivation of MS2 by hydrogen peroxide. Hydroxyl radicals are the major cause of killing in h coli cells treated with hydrogen peroxide concentrations representing mode two. However, hydroxyl radicals do not play a similar role in inactivation of bacteriophages with hydrogen peroxide, as hydroxyl radical scavengers failed to protect MS2 against the virucidal action of hydrogen peroxide. In fact thiourea, a hydroxyl radical scavenger enhances the MS2 killing by hydrogen peroxide. From the data presented in this paper, it is concluded that membrane damage may be the target for mode two killing, and membrane permeability to hydrogen peroxide may determine the inactivation rate by this biocide. The lethal damage can not be determined by these data, but the results presented here show that enzymes, electron transport systems, cell membrane, and the genetic materials sustained damages due to hydrogen peroxide treatment. The chlorine inactivation rate of bacteria and viruses differs from that of hydrogen peroxide, which could be the indication of presence of two different inactivation mechanisms. There are synergistic relationships between

PAGE 140

130 hydrogen peroxide and other agents such as silver nitrate and ultraviolet light. They enhance biocidal activity of each other. The use of H 2 o 2 in selective recovery of bacteriophages from water samples is discussed in this paper. Treating water samples with hydrogen peroxide greatly reduced or inhibited growth of indigenous bacteria in natural water samples. By adding crystal violet to the agar plates growth of more indigenous bacteria is inhibited, which allows the formation of a healthy bacterial lawn. This procedure permits enumeration of bacteriophages in samples containing a relatively large number of indigenous bacteria. Hydrogen peroxide was used for modification of sand and diatomaceous earth for removal and/or inactivation of bacteria from contaminated samples. The above solids were coated with magnesium oxide which then was converted to magnesium peroxide. These modified solids prevent bacterial growth on filters made with these materials. Addition of sand or diatomaceous earth to water samples inactivated the indigenous bacteria and kept the water samples free from bacteria for several weeks.

PAGE 141

REFERENCE LIST Ahmad, F. I. K. and C. Russell. 1975. Synergism between Ultrasonic waves and hydrogen peroxide in the killing of microorganisms. J. Appl. Bacterial. 39:31-40. Alasri, A., C. Roques, and M. Georges. 1992. Bactericidal properties of peracetic acid and hydrogen peroxide, alone and in combination, and chlorine and formaldehyde against bacterial water strains. Can. J. Microbiol. 38:635-642. Alvarez, M. E. and R. T. O'Brien. 1982a. Effects of chlorine concentration on the structure of poliovirus. Appl. Environ. Microbiol. 43:237-239. Alvarez, M. E. and R. T. O'Brien. 1982a. Mechanism of inactivation of poliovirus by chlorine dioxide and iodine. Appl. Environ. Microbiol. 44:1064-1071. Ananthaswamy H. N. and A. Eisnstark. 1977. Repair of hydrogen peroxide-induced single strand breaks in Escherichia coli deoxyribonucleic acid. J. Bacteriol. 130:187-191. Aguirre, J. and W. Hansberg. 1985. Oxidation of Neurospora crassa glutamine synthetase. J. Bacter iol. 166: 104 0-1045. Arlene Klapes, N. and D. Vesley. 1990. Vapor-phase hydrogen peroxide as a surface decontaminant and sterilant. Appl. Environ. Microbial. 56:173-181. Asghari, A., S. R. Farrah, and G. Bitton. 1992. Use of hydrogen peroxide treatment and crystal violet agar plates for selective recovery of bacteriophages from natural environments. Appl. Environ. Microbiol. 58:1159-1163. Babich, H. and G. Stotzky. 1980. Environmental factors that influence the toxicity of heavy metals and gaseous pollutants to microorganisms. CRC Critic. Rev. Microbiol. 8:99-118. Baldry, M. G. C. 1983. The Bactericidal, fungicidal and sporicidal properties of hydrogen peroxide and peracetic acid. J. Appl. Bacterial. 54:417-423. 131

PAGE 142

132 Bayliss, C. E. and W. M. Waites. 1976. The effect of hydrogen peroxide and ultraviolet light on spores of Clostridium bifermentans. J. Gen. Microbial. 96:401-407. Bayliss, C. E. and W. M. Waites. 1979. The Combined effect of hydrogen peroxide and ultraviolet irradiation on bacterial spores. J. Appl. Bacteriol. 47:263-269. Bayliss, C. E. and W. M. Waites. 1980. The Effect of hydrogen peroxide and ultraviolet irradiation on non sporing bacteria. J. Appl. 48:417-422. Bayliss, C. E. and W. M. Waites. 1981. Resistance of Serratia marcescens to hydrogen peroxide. J. Appl. Bacterial. 50:131-137. Berglin, E. H. 1984. Hydrogen peroxide killing induced by cysteine in a growing culture of Escherichia coli. J. Bacterial. 152:81-88. Berglin, E. H. and J. Carlsson. 1985. Potentiation by sulfide of hydrogen peroxide-induced killing of Escherichia coli. Infec. Immun. 49:538-543. Beutler, E. 1975. Red Cell Metabolism. Grune and Stratton, New York, N.Y. Birnboim, H. C. and M. Kanabus-Daminska. 1985. The production of DNA strand breaks in human leukocytes by superoxide anion may involve a metabolic process. Proc. Natl. Acad. Sci. U. S. A. 82:6820-6824. Bitton, G., L. T. Chang, S. R. Farrah, and K. Clifford. 1981. Recovery of coliphages from wastewater effluent and polluted lake water by the magnetite organic flocculation method. Appl. Environ. Microbial. 41:93-96. Bitton, G., B. N. Feldberg, and S. R. Farrah. 1979. Concentration of enteroviruses from seawater and tapwater by organic flocculation using non-fat dry milk and casein. Water, Air, and Soil Pollution. 12:185-195. Bitton, G. and B. Koopman. 1988. Cell permeability to toxicants: an important parameter in toxicity tests using bacteria. CRC Critical Rev. in Environ. Contr. 18:177-188. Block, S. S. 1991. Preservation, pp Philadelphia, Pa. Disinfection, 171. Lea and Sterilization, and Febiger Publishers,

PAGE 143

133 Bol, D. K. and R. E. Yasbin. 1990. Characterization of an inducible oxidative stress system in Bacillus subtilis. J. Bacterial. 172:3503-3506. Borders, C. L. and I. Fridovich. 1985. A comparison of the effects of cyanide, hydrogen peroxide, and phenylglyoxal on eukaryotic cu, Zn superoxide dismutases. Arch. Biochem. Biophys. 241:472-476. Brandi, G., G. Schiavano, M. Magnani, A. Albano, F. Cattabeni, and G. Cantoni. 1988. Superoxide anions are required for the inactivation of Escherichia co 1 i induced by high concentration of hydrogen peroxide. Curr. Microbial. 17:117-120. Brandi G., M. Fiorani, C. Pierotti, A. Albano, F. Cattabeni, and o. Cantoni. 1989a. Morphological changes in Escherichia coli cells exposed to low or high concentration of hydrogen peroxide. 33:991-1000. Brandi, G. F. Role of induced 6:47-55. Cattabeni, A. Albano, and O. Cantoni. 1989b. hydroxyl radicals in Escherichia coli killing by hydrogen peroxide. Free Rad. Res. Comms. Breimer, L. H. and T. Lindahl. 1 985. enzymatic excision of DNA bases damaged by exposure to ionizing radiation or oxidizing agents. Mut. Res. 150:85-89. Burnett, F. M., and D. Lush. 1940. Action ~f certain surface active agents on viruses. Aust. J. Exp. Biol. Med. Sci. 18:41-150. Cantoni, 0., G. Brandi, G. F. Schiavano, A. Albano, and F. Cattabeni. 1989a. Lethality of hydrogen peroxide wild type and superoxide dismutase mutants of Escherichia goli ( a hypothesis on the mechanism of hydrogen peroxide induced inactivation of Escherichia coli). Chem. Biol. Interaction. 70:281-288. Cantoni, 0., P. Sestili, F. Cattabeni, G. Bellomo, s. Ru, K Cohen, and P. Cerutti. 1989b. Calcium chelator quin 2 prevents hydrogen peroxide-induced DNA breakage and cytotoxicity. Eur. J. Biochem. 58209-212. Cantoni, o., F. Cattabeni, V. Stocchi, R. E. Meyn, P. Cerutti, and D. Murray. 1989c. Hydrogen peroxide insult in cultured mammalian cells: relationships between DNA single strand breakage, poly(ADP-ribose) metabolism and cell killing. Biochim. Biophis. Act. 1014:1-7.

PAGE 144

134 Carlsson, J. and V. S. Carpenter. 1980. The recA+ gene product is more important than catalase and superoxide dismutase in protecting Escherichia coli against hydrogen peroxide toxicity. 142:319-321. Caputo, R. A. and T. E. Odlaug. 1983. Sterilization with ethylene oxide and other gases. In S. S. Block ( ed) Disinfection, Sterilization, and Preservation. Lea and Febiger. Philadelphia, Pa. Chambers, C. W., C. M. Proctor, and P. W. Kabler. 1962. Bactericidal efffect of low concentration of silver. J. Am. Water Works Assoc. 54:208-214. Chen, D. and C. Bernstein. 1987. Recombinational repair of hydrogen peroxide-induced damages in DNA of phage T4. Mutat. Res. 184:87-98. Christman, M .F., R. W. Morgan, F. S. Jacobsen, and B. N. Ames. 1985. Positive control of a regulon for defenses against oxidative stress and some heat-shock proteins in Salmonella typhimurium. Cell. 41:753-762. Clarke, N. A., and D. Berman. 1983. Disinfection of drinking water, swimming-pool water, and treated sewage ef fluents. In S.S. Block (ed.), Disinfection, Sterilization and Preservation. Lea & Febiger, Philadelphia, Pa. Clesceri, L. s., A. E. Greenberg, and R. R. Trussel. 1989. Standard methods for the examination of water and wastewater. APHA AWWA APCF. 17:448-451. Cornax, R., M.A. Morinigo, I G. Paez, M.A. Munoz, and J. J. Borrego. 1990. Application of direct plaque assay for detection and enumeration of bacteriophages of Bacteroides fragilis from contaminated-water samples. Appl. Environ. Microbiol. 56:3170-3173. Curran, H. R., F. R. Fuans, and A. Leviton. 1940. The sporicidal action of hydrogen peroxide and the use of crystalline catalase to dissipate residual peroxide. J. Bacteriol. 40:423-434. Darnell, J., H. Lodish, and B. Baltimore. 1986. Molecular Cell Biology. Scientific American Books. New York, N. Y. De Mello Filho, A. C., and R. Meneghini. 1985. Protection of mammalian cells by 0-phenanthroline from lethal and DNA damaging effects produced by active oxygen species. Biochem. Biophys. Acta. 847:82-88.

PAGE 145

135 Demple, B. F. and J. Halbrook. 1983. Inducible repair of oxidative DNA damage in h coli. Nature. 304:466-468. Demple, B. F., J. Halbrook, ands. Linn. 1983. Escherichia coli xth mutants are hypersensitive to hydrogen peroxide. 153:1079-1082. DeRezende, J. R. V. 1968. Disinfection of acrylic resin sections for use as surgical implants. Rev. Fae. Odontol. Univ. Sao Polo 6:95-102. DiGuiseppi, J. and I. Fridovich. 1984. The Toxicity of molecular oxygen. CRC Crit. Rev. Toxicol. 12:315-342. Domek, M. J., M. W. LeChevallier, S. C. Cameron, and G. A. McFeters. 1984. Evidence for the role of copper in the injury process of coliform bacteria in drinking water. Appl. Environ. Microbial. 45:68-73. Domingue, E. L., R. L. Tyndall, W.R. Mayberry, and 0. C. Pancorbo. 1988. Effects of three oxidizing biocides on Legionella pneumophila serogroup 1. Appl. Environ. Microbial. 54:741-747. Dubreuil, D., J. G. Bisaillon, and R. Beaudet. 1984. Inhibition of Neisseria gonorrhaeae growth due to hydrogen peroxide production of urogenital streptococci. Microbios. 39:159-167. Dutton, R. J., G. Bitton, and B. Koopman. 1983. Malachite green-INT (MINT) method for determining active bacteria in sewage. Appl. Environ. Microbial. 46:1263-1267. Dutton, R. J., G. Bitton, and B. Koopman. 1986. Rapid test for toxicity in wastewater systems. Toxicity Assessment. 2:147-158. Dutton, R. J., G. Bitton, an B. Koopman. 1988. Enzyme biosynthesis versus enzyme activity as a basis for microbial toxicity testing. Toxicity. Assessment. 3:245-253. Dychdala, G. R. 1991. Chlorinated compounds. In Disinfection, Sterilization, and Preservation, s. s. Block, ed. Lea and Febiger, Philadelphia, Pa. Farrah, S. S. and G. Bitton. 1978. Elution of poliovirus adsorbed to membrane filters. Appl. Environ. Microbial. 36:982-984.

PAGE 146

136 Farrah, s. R., D. O. Shah, and L. 0. Ingram. 1981. Effects of chaotropic and antichaotropic agents on elution of poliovirus adsorbed on membrane filters. Proc. Natl. Acad. Sci. U. S. A. 78:1229-1232. Farrah, s. R. 1982. Chemical factors influencing adsorption of bacteriophage MS2 to membrane filters. Appl. Environ. Microbiol. 43:6598-663. Farrah, S. R., M. A. Girard, G. A. Toranzos, and D. R. Preston. 1988. Adsorption of viruses to diatomaceous earth modified by in situ precipitation of metallic salts. z. Gesamte. Hyg. 34:520-521. Farrah, S. R., D. R. Preston. 1991. Adsorption of viruses by diatomaceous earth coated with metallic oxides and metallic peroxides. Wat. Sci. Tech. 24:235-240. Farrah, S. R., D.R. Preston, G.A. Toranzos, M. Girard, G. A. Erdos, and V. Vasuhdivan. 1991. Use of modified diatomaceous earth for removal and recovery of viruses in Water. Appl. Environ. Microbiol. 57:2502-2506. Fiorani, M., M. Furno, P. Sestili, F. Catabeni, and O. Cantoni. 1990. Inhibition of chinese hamster ovary cell DNA synthesis by hydrogen peroxide. Chem. Biol. Interact. 76:129-139. Flenley, D. C. 1987. What should an ideal antioxidant Do (and do not). Bull. Physiophathol. Repair. 23:279-285. Fox, C. L. Jr. 1978. Pharmacodynamics of sulphadiazine and related topical antimicrobial In recent advances in dermatopharmacology. P. and E. C. Gomez, eds. Spectrum Publications, New silver agents. Frost, York. Fox, c. L. Jr. and S. M. Mordak. 1974. Mechanisms of silver sulphadiazine action on burn wound infections. Antimicrob. Ag. Chemother. 5:528-588. Fridovich, I. 1978. The Biology of Oxygen Radicals. Sci. 201-875-879. Gerba, C. P. C. H. Stagg, and M. G. Abadie. 1978. Characterization of sewage solid-associated viruses and behavior in natural waters. Wat. Res. 12:805812. Gerba, C. P., K. Hou, and M. D. Sobsey. 1988. Microbial removal and inactivation from water by filters containing magnesium peroxide. J. Environ. Sci. Health. A23:41-58.

PAGE 147

Girotti, A. W. and J. P. Thomas. oxygen radicals on resealed Biol. Chem. 259:1744-1752. 137 1984. Damaging effect of erythrocyte ghosts. J. Glass, J. s. and R. T. O'Brien. 1980. Enterovirus and coliphage inactivation during activated sludge treatment. Wat. Res. 14:877-882. Greenberg, J. T. and B. Demple. 1989. A Global response induced in Escherichia coli by redox-cycling agents overlaps with that induced by peroxide stress. J. Bacterial. 171:3933-3939. Goyal, S. M. 1987. Methods in Phage ecology. In S. M. Goyal, C. P. Gerba, and G. Bitton, eds. Phage ecology. John Wiley & Sons, Inc., New York, N. Y. Grunnet, K. 1977. Comparison of methods for demonstrating coliphages in waters. Rev. Int. Oceanogr. Med. 48:17-19. Hagensee, M. E. and R. E. Moses. 1986. Repair response of Escherichia coli to hydrogen peroxide DNA damage. J. Bacterial. 168:1059-1065. Hagensee, M. E., S. K. Bryan, and R. E. Moses. 1987. DNA poymerase III requirement for repair of DNA damage caused by methyl methanesulfonate and hydrogen peroxide. J. Bacterial. 169:4608-4613. Halliwell B. and J. M. C. Gutteridge. 1984. Oxygen toxicity, oxygen radicals, transition metals and disease. Biochem. J. 219:1-14. Hassan, H. M. and I. Fridovich. 1979. Superoxide, hydrogen peroxide, and oxygen tolerance of oxygen-sensitive mutants of Escherichia coli. Rev. Infec. Dis. 1:357-369. Hoepelman, I. M., W. A. Bezemer, C. M. Vanderbroucke-Grauls, J. J. Max, and J. Verhoef. 1990. Bacterial iron enhances oxygen radical-mediated killing of Staphylococcus aureus by phagocytes. Infect. Immun. 58:26-31. Hodgson, E. K. and I. Fridovich. 1975. The interaction of bovine erythrocyte superoxide dismutase with hydrogen peroxide: inactivation of the enzyme. Biochemistry. 14:5294-5301. Hoffman, M. E., A. C. De Mello-Filho and R. Meneghini. 1984. Correlation between cytotoxic effects of hydrogen peroxide and the yields of DNA strand breaks in cell of different species. Biochem. Biophys. Acta. 781:234-238.

PAGE 148

138 Hurst, J. K., W. C. Barrette, Jr., B. R. Michel and H. Roem. 1991. Hypochlorous acid and myeloperoxidase-catalyzed oxidation of iron-sulfur clusters in bacterial respiratory dehydrogenases. Eur. J. Biochem. 202:1275-1282. Imlay, J. A. and S. Linn. 1986. Bimodal pattern of killing of DNA-repair-defective or anoxically grown Escherichia coli by hydrogen peroxide. J. Bacterial. 166:519-527. Imlay, J. A. and S. Linn. 1987. Mutagenesis and stress responses induced in Escherichia coli by hydrogen peroxide. J. Bacterial. 169:2967-2976. Imlay, J. A. and S. Linn. 1988. DNA damage and oxygen radical toxicity. Sci. 240-1302-1310. Imlay, J. A., S. M. Chin, and S. Linn. 1988. Toxic DNA damage by hydrogen peroxide through the fen ton reaction in vivo and in vitro. Sci. 240:640-642. Jenkins, D. E., J. E. Schultz, and A. Matin. 1988. starvation-induced cross protection against heat or hydrogen peroxide challenge in Escherichia coli. J. Bacterial. 170:3910-3914. Just, J., and A. Szniolis. 1936. Germicidal properties of silver in water. J. Am. Water Works Assoc. 28:492-505. Kalter, s. S., V. D. Mordaunt, and O. D. Chapman. 1946. The isolation of Escherichia coli phages by means of cationic detergents. J. Bacterial. 52:237-240. Kelland, L. R., S. H. Moss, and D. J. G. Davies. 1984. Leakage of 86 Rh'" after ultraviolet irradiation of Escherichia coli K-12 Photochem. Photobiol. 39:329-335. Kennedy, J.E., G. Bitton, and J. L. Oblinger. 1985. Comparison of selective media for assay of coliphages in sewage effluent and lake water. Appl. Environ. Microbial. 49:33-36. Kensese, S. M. and L. L. Smith. 1989. Hydrogen peroxide mutagenicity toward Salmonella typhimurium. Teratogenesis Carcinog. Mutagen. 9:211-218. Kim, C. K., D. M. Gentile, and Otis J. Sproul. 1980. Mechanism of ozone inactivation of bacteriophage f2. Appl. Environ. Microbial. 39:210-218.

PAGE 149

139 Kim, K., s. G. Rhee, and E. R. Stadman. 1985. Nonenzymatic cleavage of proteins by reactive oxygen species generated by dithiothreitol and iron. J. Biol. Chem. 260:15394-15399. Kleiman, N. J., R. R. Wang, and A. Spector. 1990. Hydrogen peroxide-induced DNA damage in bovine lens epithelial cells. Mutat. Res. 240:35-45. Lesko, S. A., R. L. Lorentzen, and P. o. P. Ts'o. 1980. Role of superoxide in deoxyribonucleic acid strand scission. Biochem. 19:3023-3028. Levy, R. V. and T. J. Leahy. 1991. Sterilization by filtration. In S. S. Block, ed. Sterilization, Disinfection, and Preservation. Lingquist, S., K. Lindhal, and P. Howard-Flanders. 1976. Methyl methane sulfonate-sensitivity mutants of ]h coli defective in an endonuclease specific for apurinic sites in decoxyribonucleic acid. J. Bacterial. 126:646-653. Link, E M. 1988. The Mechanism of pH-dependent hydrogen peroxide cytotoxicity in vitro. Arch. Biochem. Biophys. 265:362-372. Loewen, P. C. 1984. Isolation of catalase-def icient Escherichia coli mutants and genetic mapping of KatE, a locus that affects catalase activity. J. Bacterial. 157:622-626. Marking, L. L. and V. K. Dawson. 1973. Methods for assessment of toxicity or efficacy of mixtures of chemicals. Fish Control Laboratory, La Cross, Wisconsin. Marx, J. L. 1987. Oxygen free radicals linked to many diseases. Sci. 235:529-531. Malaiyani, M., M. Husain Sader, P. Lee, and R. O'Grady. 1980. Removal of organics in water using hydrogen peroxide in presence of ultraviolet light. Wat. Res. 14:1131-1135. Menzel, D. B. 1971. Oxidation of biologically active reducing substances by ozone. Arch Environ. Health. 23:149-153. Miyasaki, K. T., M. E. Wilson, J. J. Zambon, and R. J. Genco. 1985. Influence of endogenous catalase activity on the sensi ti vi ty of the oral bacterium Actinobacillus actinomycetemcomitans and the oral Haernophili to the bactericidal properties of hydrogen peroxide. Arch. Oral. Biol. 30:843-848.

PAGE 150

140 Modak, s. and C. L. Fox. Jr. 1985. Synergistic action of silver sulfadiazine and sodium piperacillin on resistant~ aeruginosa in vitro and in experimental burn wounds infections. J. Trauma. 25:27-32. Naguib, K. and L. Hussein. 1972. The Effect of H 2 0 2 on the bacteriological quality and nutritive value of milk. Milchwissencheft. 27:758-762. O'Brien, R. T. and J. Newman. 1979. Structural and compositional changes associated with chlorine inactivation of polioviruses. Appl. Environ. Microbiol. 38:1034-1039. Ohlrogge, J.B. and T. P. Kernan. 1983. Toxicity of activated oxygen lack of dependence on membrane unsaturated fatty acid composition. Biochern. Biophys. Res. Commun. 113:301-308. Parker, W. F. 1981. A Note on the use of membrane faecal coliform medium for enhancing resolution and accuracy when enumerating a small plaqueing coliphage. J. Appl. Bacterial. 51:81-84. Plastourgon, M. and M. R. Hoffman. 1984. Transformation and fate of organic esters in layered flow system. The role of trace metal catalysis. Environ. Sci. Technol. 18:756-763. Plateau, P., M. Frornant, and S. Blanquet. 1987. Heat shock and hydrogen peroxide response of Escherichia coli arenot changed by dinucleic tetraphosphatehydrolase overproduction. J. Bacterial. 169:3817-3320. Philip, C. R. and B. Warshoesky. 1958. Chemical. disinfectants. Arn. Rev. Microbiol. 12:525-532. Primrose, S. B., N. D Seeley, K. Nicolson. 1982. Methods bacteriophage ecology. Appl. 43:694-701. B. Logan, and J. W. for studying aquatic Environ. Microbial. Prise, K. M., S. Davis, and B. D. Michael. 1989. Cell killing and DNA damage in chinese hamster V79 cells treated with hydrogen peroxide. Int. Radiat. Biol. Relat. Stud. Phys. Chern. Med. 55:583-592. Rahn, R. 0., J. K. Setlow, and L. C. Landry. 1973. Ultraviolet irradiation of nucleic acids complexed with heavy atoms. II phosphorescence and photodirnerization of DNA complexed with Ag. Photochern. Photobiol. 18:29-36.

PAGE 151

141 Repine, J. E., O. W. Pfenninger, D. W. Talmage, Elaine M. Berger, and D. E. Pettijohn. 1981a. Dimethyl sulfoxide prevents DNA nicking mediated by ionizing radiation or iron/hydrogen peroxide-generated hydroxyl radicals. Proc. Natl. Acad. Sci. U. S. A. 78:1001-1003. Repine, J. E., R. B. Fox, and E. M. Berger. 1981b. hydrogen peroxide kills Staphylococcus aureus by reacting with staphylococcal iron to form hydroxyl radical J. Biol. Chem. 256:7094-7096. Richards, D. M., R. T. Dean, and W. Jessup. 1981. Membrane proteins are critical targets in free radical mediated cytolysis. Biochem. Biophys. Acta. 946:281-288. Richards, R. M. E. 1981. Antimicrobial action of silver nitrate. Microbios. 31:83-91. Roehm, J N J. G. Hadley, and D. B. Menzel 1971. Oxidation of unsaturated fatty acids by ozone and nitrogen dioxide. Arch. Environ. Health. 23:142-148. Roller, s. D., V. P. Oliveri, and K. Kawata. 1980. Mode of bacterial inactivation by chlorine dioxide. Wat. Res. 14:635-641. Rudolph, A. S. and Levine. 1940. Factors affecting the germicidal efficiency of hypochlori te solutions. Bull 150 Eng. Exp. Sta. Iowa State College. Roy, D., P. K. Y. Wong, R. S. Engelbrecht, and E. S. K. Cian 1981. Mechanism of enteroviral inactivation by ozone. Appl. Environ. Microbial. 41:718-723. Sambr i V., R. Cevenini, and M. La Placa. 1991. Susceptibility of iron-loaded Borrelia burgdorferi to killing by hydrogen peroxide and human polyrnorphonuclear leukocytes. FEMS Microbial. Lett. 65:67-71. Sawadaishi, K., K. Miura, E. Ohtsuka, T. Ueda, K. Ishizaki, and N. Shinriki. 1985. Ozonolysis of supercoiled pBR322 DNA resulting in strand scission to open circular DNA. Nucleic Acid Res. 13:7183-7194. Schraufstatter, I. U., P. A. Hyslop, D. B. Hinshaw, R. G. Spragg, L.A. Sklar, and C. G. Cochrane. 1986. Hydrogen peroxide-induced injury of cell and its prevention by inhibitors of poly(ADP-ribose) polymerase. Proc. Natl. Acad. Sc i U. S. A. 83:4908-4912.

PAGE 152

142 Seeberg, E. and A. L. Steinurn. 1980. Repair of X-ray-induced deoxyribonucleic acid single-strad breaks in xth mutants of Escherichia coli. J. Bacteriol.141:1424-1432. Sharp, D. G., D. C. Young, R. Floyd, and J. D. Johnson. 1980. Effect of ionic environment on the inactivation of poliovirus in water by chlorine. Appl. Environ. Microbiol. 39:530-534. Shields, P. A. and S. R. Farrah. 1986. Concentration of viruses in beef extract by flocculation with ammonium sulfate. Appl. Environ. Microbiol. 51:211-213. Shields, P.A., T. F. Tjatha. D. 0. Shah, and S. R. Farrah. 1986. Comparison of positively charged membrane filters and their use in concentrating bacteriophages in water. Wat. Res. 20:145-151. Snyder, L. M., N. L. Fortier, L. Leb, J. McKenney, J. Trainor, H. Sheerin, and N. Mohandas. 1988. The Role of membrane protein sulfhydryl groups in hydrogen peroxide mediated membrane damage in human erythrocytes. Biochirn. Biophys. Acta. 937:229-240. Starke, P. E. and J. L. Farber. 1985a. Ferric iron and superoxide ions are required for the killing of cultured hepatocytes by hydrogen peroxide. J. Biol. Chern. 260:10099-10104. Starke, P. E. and J. L. Farber. 1985b. Endogenous defenses against the cytotoxicity of hydrogen peroxide in cultured rat hapatocytes. J. Biol. Chern. 260:86-92. Steinman, H. 1982. Superoxide disrnutases; Protein chemistry and structure-function relationships, in superoxide disrnutase. Oberley, L. W. ed. CRC Press, Boca Raton, Fl. Stevenson, K. E. and B. D. Shafer. 1983. Bacterial spores resistance to hydrogen peroxide. Food Technol. 37:111-126. Stoklasova, A. 1989. Defence against oxygen. Sb. Ved. Pr. Lek. Fak. Hardci. Kralove. 32:157-168. Stryer, A. 1984. Biochemistry. the toxic action of Karlovy. Univerzity. Tarrny, E. M. and N. 0. Kaplan. 1968. Chemical characterization of D-Lactate dehydrogenase from Escherichia coli B. J. Biol. Chern. 243:2570-2586.

PAGE 153

143 Tartera, C. and J. Jofre. 1987. Bacteriophages active against Bacteroides fragilis in sewage-polluted waters. Appl. Environ. Microbial. 53:1632-1637. Taylor, G. R. and M. Butler. 1982. A Comparison of the virucidal properties of chlorine dioxide, bromine chloride and iodine. J. Hyg. 89:321-328. Terada, L. S., J. A. Leff, D. M. Guidot, I. R. Willingham, and J. E. Repine. 1991. Inactivation of xanthine oxidase by hydrogen peroxide involves site-directed hydroxyl radical formation. Free Radie. Biol. Med. 10:61-68. Thurman, R. B. and C. P. Gerba. 1989. The molecular mechanism of copper and silver ion disinfection on bacteria and viruses. CRC Crit. Rev. Environ. Cont. 18:295-315. Thurman, R. B. and C. P. Gerba. 1990. Molecular mechanisms of viral inactivation by water disinfectants. Adv. Appl. Microbial. 33:75-105. Tilton, R. C. and B. Rosenberg. 1978. Reversal of the silver inhibition of microorganisms by agar. Appl. Environ. Micorbiol. 35:1116-1120. Toledo, R. T. 1975. Chemical sterilants for aseptic packaging. Food Technol. 29:102-112. Tsuzuki, Y. and O.T. Tsuzuki. 1986. The effect of hydrogen peroxide on spores of Clostr idium per fr ingenes. Lett. Appl. Microbial. 2:65-68. van der Zee, J., T. M.A. R. Dubbelman, and J. van Steveninck. 1985. peroxide-induced membrane damaged in human erythrocytes. Biochim. Biophys. Acta. 818:38-44. Van Slays, M. A., R. Costa-de-Oliveira, I. Felzenszwalb, and C. F. Menck. 1986. Ascorbate-copper induced DNA lesions and repair in Escherichia coli: K-12 Cells. Carcinogenesis. 7:197-200. Vaughn, J. M. and T. G. Metcalf. 1975. Coliphages as indicators of enteric viruses in shellfish and shellfish rasing estuarine waters. Water Res. 9:613-616. Waites, W. M., C. E. Bayliss, N. R. King, and A. M. Davies. 1979. The Effects of transition metal ions on the resistance of bacterial spores to hydrogen peroxide and to heat. J. Gen. Microbial. 112:225-233.

PAGE 154

144 Waites, W. M., S. E. Harding, D.R. Fowler, S. H. Jones, D. Shaw, and M. Martin. 1988. The destruction of SI,X)res of~ subtilis by the combined effects of H 2 O 2 and UV light. Lett. Appl. Microbial. 7:139-140. Wentsel, R. S., P.A. O'Neil, and Evaluation of coliphage indicator of water quality. 43:430-434. J F. Kitchens. 1982. detection as a rapid Appl. Environ. Microbial. Winquist, L., U. Rannug, A. Rannug, and c. Ramel. 1984. Protection from toxic and mutagenic effects of hydrogen peroxide by catalase induction in Salmonella typhimurium. Mut. Res. 141:145-147. Yamakura, F. and K. Suzuki. 1986. Inactivation of Pseudomonas iron-superoxide dismutase by peroxide. Biochim. Biophys. Acta. 874:23-29 hydrogen Yamada, Y., H. Sato, H Tomioka, Relationship between the bacteria to active intracellular killing Microbial. 133:2015-2021. and J. Jidoi. 1987. susceptibility of various oxygen species and to by macrophages. J. Gen. Yonei, S., R. Yokota, and Y. Sato. 1987. The Distinct role of catalase and DNA repair systems in protection against hydrogen peroxide in Escherichia coli. Biochem. Biophys. Res. Commun. 143:638-644. Yoshpe-Purer, Y., and E. Eylan. 1968. Disinfection of water by hydrogen peroxide. Health Lab. Sci. 5:233-238.

PAGE 155

145 BIOGRAPHICAL SKETCH Abdolkarim Asghari was born in Persian Gulf port city of Bandar-E-Daylam, Iran, in 1959, and was raised in the capital city of Tehran, Iran. A few years after graduation from high school, he came to this country to pursue higher education degrees. He attended the University of Tennessee, where he got his BS degree in Microbiology in 1986. He then moved to Denton, Texas to continue his studies at the University of North Texas. He was awarded the degree of Master of Science in Molecular Biology in 1988. He finally moved to Gainesville, Florida to attend the Graduate program of the Department of Microbiology and Cell Science. He currently studies toward the degree of Doctor of Philosophy in the area of environmental microbiology.

PAGE 156

I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Samuel R. Farrah, Chairman Associate Professor of Microbiology and Cell Science I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. / ~C /j('l\__,1r ~ ( \:; _, Lonnie o. Ingra~ Professor of Microbiology and Cell Science I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philoso hy. el Bitton Pr fessor of Environmental Engineering Science

PAGE 157

I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Phillip j M. Achey / Professor of v Microbiology and Cell Science I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. cL / / 1 C < ~~:r>', ; _.. ~ > .. < Francis C. Davis Associate Professor of Microbiology and Cell Science This dissertation was submitted to the Graduate Faculty of the College of Agriculture and to the Graduate School and was accepted as partial fulfillment of the requirements for the degree of Doctor of Philosophy. May 1993 Dean o riculture Dean, Graduate School