<hr noshade="noshade" size="1"> A.L. Moxon Honorary Lectures
Special Circular 167-99
<hr noshade="noshade" size="1"> Involvement of Selenium in the Regulation of Viral Virulence
Orville A. Levander
U.S. Department of Agriculture, ARS
Beltsville Human Nutrition Research Center
Nutrient Requirements and functions Laboratory
Beltsville, MD 20705.
Keshan Disease: Deficiency vs Infection
Keshan disease (KD) is familiar to all scientists working in selenium nutrition. Discovery that this juvenile cardiomyopathy could be prevented by supplementing with selenium was one of the great triumphs of human trace element nutrition (Keshan Disease Research Group, 1979). Although there is no doubt about the role of selenium deficiency in causing this disease, certain epidemiological evidence indicates that there must be more to KD than an uncomplicated deficiency. For example, KD has a pronounced seasonal variation, which cannot be explained only on the basis of changes in selenium status: in North China, KD occurs mainly during the winter season, whereas in the South, the disease is seen primarily in the summer (Yang et al., 1985). This type of seasonal variation is much more typical of an infectious disease rather than a nutritional deficiency disease.
The Chinese scientists were well aware of the possibility that an infectious agent might be involved in the etiology of KD. In fact, they were able to isolate a number of viruses from KD patients. Among these viruses was included a coxsackie B4 virus. When this virus was tested for its ability to cause heart damage in selenium-deficient and selenium-adequate mice, it was found in the early 1980's that the virus was much more cardiotoxic and inflicted greater heart damage in the deficient mice (Bai et al., 1980). Some years later, Professor Melinda Beck of the University of North Carolina approached me about initiating a series of collaborative experiments to determine the influence of selenium status on viral virulence.
Selenium vs Virulence
Our first objective was to duplicate the earlier results of the Chinese scientists. Professor Beck had two different viral strains available for testing. The first strain, called coxsackievirus B3/20 (CVB3/20), causes heart damage even in animals fed normal diets. We were able to show that this virus caused increased heart damage in selenium-deficient mice compared to mice fed diets adequate in selenium (Beck et al., 1994a). These results essentially confirmed those reported earlier by the Chinese.
The second strain of virus, called coxsackievirus B3/0 (CVB3/0), does not cause heart damage in mice that are fed normal diets. That is, the virus is said to be avirulent in contrast to the CVB3/20 which is considered to be virulent. When the avirulent strain of coxsackievirus (CVB3/0) was inoculated into selenium-deficient mice, it caused a moderate degree of heart damage (Beck et al., 1994b). Thus, by feeding a diet deficient in selenium, we allowed an avirulent virus to express virulence. As far as we are aware, this was the first demonstration that the nutritional status of the host would permit an avirulent strain of virus to exhibit virulence. This was a novel finding, which could have significant implications for nutrition and infection interrelationships in the Third World, or wherever malnutrition was a problem.
Investigations into a variety of immune system functions indicated that mice infected with the coxsackievirus produced similar amounts of neutralizing antibody regardless of whether the animals were fed diets adequate or deficient in selenium. Therefore, the effects of selenium deficiency on viral virulence were probably not mediated through the antibody producing arm of the immune system. On the other hand, impediments in T cell function were noted in the selenium-deficient infected animals. Lymphocyte proliferation was somewhat reduced in response to either mitogen or antigen challenge. Natural killer (NK) cell activity was also mildly reduced due to selenium deficiency, but later experiments showed little correlation between NK cell activity and virulence in fish oil-fed mice infected with the coxsackievirus.
Anti-virulent Effects of Natural vs Synthetic Antioxidants
We felt it quite reasonable to test the effect of vitamin E deficiency on the virulence of the coxsackievirus because of the close metabolic and nutritional relationship between selenium and vitamin E. As with selenium deficiency, there was an increased virulence of the coxsackievirus in vitamin E-deficient mice, and this trend was exacerbated by feeding fish oil (Beck et al., 1994c). This was true for both the avirulent and virulent strains of the virus. As with selenium deficiency, B cell function did not appear to be impaired in the vitamin E-deficient mice, whereas T cell function was depressed in the deficient animals. We also found that feeding fish oil had a marked suppressive effect on NK activity regardless of the vitamin E status of the mice. Normally, NK activity is considered to play a crucial role in viral clearance. However, under the conditions of our coxsackievirus-infected mouse model, it was clear that we could suppress NK activity by approximately 90 percent by feeding fish oil to our vitamin E-supplemented mice without having any effect on the apparent virulence of the virus.
N,N'-diphenyl-p-phenylenediamine (DPPD) is a synthetic organic antioxidant that has powerful vitamin E-like activity but bears no relationship to the molecular structure of the tocopherols. Nonetheless, DPPD was highly effective in preventing the increased heart damage seen in vitamin E-deficient mice infected with the virulent coxsackievirus B3/20 (Beck, 1997). These results provided strong evidence that increased oxidative stress in the host milieu was responsible for the increased virulence of either the virulent or avirulent strains of the coxsackievirus that we observed in our selenium- or vitamin E-deficient mice.
What Changed? Host vs Pathogen
How does increased dietary oxidative stress translate into increased viral virulence and its attendant increased heart muscle damage? Two possibilities exist: such stress could either affect the ability of the host to resist the virus or, conversely, it could affect the ability of the virus to inflict damage on the host. The first possibility was consistent with the fact that coxsackievirus is known to exert a direct cytotoxic effect on heart muscle cells (Chow et al., 1992; Ruppert et al., 1994). One could imagine, therefore, that any interference with normal cellular metabolism might predispose cardiomyocytes to increased virus-induced damage. Heart muscle damage (i.e., nutritional muscular dystrophy) is a common feature of selenium/vitamin E deficiency per se (i.e., without viral involvement) in several species of animals, including turkeys, swine, sheep, cows and horses (Combs and Combs, 1986). Thus, heart muscle cells in selenium/vitamin E-deficient animals seem at greater risk of damage due to exposure to any of a variety of external insults (e.g., heavy metals, xenobiotics, viruses, etc.). This possibility could perhaps be tested by determining the effect of coxsackievirus on normal and deficient heart muscle cells cultured in vitro.
The second possibility was supported by much less experimental evidence, that is, the notion that the antioxidant- (selenium or vitamin E) deficient diet fed to the host might somehow change the virus itself in some way so that it became more virulent. In fact, one could reasonably argue that there was no precedent for such an idea. It seemed much more reasonable that the diet fed to the host should have an effect on the host rather than on the pathogen. Nonetheless, we carried out a passage experiment which consisted of two parts (Beck et al., 1995). In the first part, two groups of mice, one selenium-deficient, the other selenium-adequate, were inoculated with the benign form of the virus, CVB3/0. After 10 days, the hearts of the mice were harvested, and the virus was isolated from the tissue. Virus from the two groups of mice was cultured separately in HeLa cells and then inoculated into another group of mice -- all selenium adequate. The CVB3/0 that had been passed through the adequate mice in the first part of the experiment, so-called CVB3/0Se<sup>+</sup>, caused no heart damage when inoculated into the adequate mice in the second part. On the other hand, the CVB3/0 that had been passed through the deficient mice in the first part of the experiment, so-called CVB3/0Se<sup>-</sup>, did indeed cause heart damage when inoculated into the adequate mice in the second part. In other words, passage of the normally benign CVB3/0 through a selenium-deficient host, thereby allowing it to replicate under conditions of in vivo oxidative stress, altered the phenotype of the virus so that the previously avirulent CVB3/0 now became the virulent CVB3/0Se<sup>-</sup>.
Phenotypic Change vs Genotypic Change
The results of the passage experiment suggested that the phenotypic change observed in the properties of the CVB3/0 passed through a selenium-deficient host was actually due to a change in the genetic make-up of the virus. How else to account for the increased virulence of the CVB3/0Se- in the second part of the passage experiment? Proof of this contention, however, could only be obtained by sequencing the RNA from CVB3/0Se- and comparing the sequence with that of the parent CVB3/0 strain. When such sequencing analysis was performed, it was found that the nucleotide composition of the newly virulent CVB3/0Se<sup>-</sup> had changed to resemble that of the naturally virulent wild type strain CVB3/20 at 6 of 7 genomic sites thought to influence virulence (Beck et al., 1995). To our knowledge, this was the first report demonstrating that the nutritional status of the host could exert any control over the genetic nature of an invading pathogen. But how did the redox state of the host cell alter viral genetics?
Mutation vs Selection
Two biochemical mechanisms could account for the effect of oxidative stress on the viral genome: mutation of the viral RNA genetic material due to oxidative damage or selection of a pre-existing viral substrain because of a change in growth conditions in the host cell (i.e, shift in peroxidative tone).
Mutagenesis in mammalian cells as a result of oxidative DNA damage is a well-known and well-studied phenomenon (Beckman and Ames, 1997; Wang et al., 1998; Rossman and Goncharova, 1998). In contrast, relatively little is known about oxidative base damage in RNA, although 8-oxo-guanosine has been identified in RNA under oxidizing conditions (Schneider et al., 1993). Therefore, one could postulate that the RNA genome of the coxsackievirus had been altered via oxidative attack, thereby allowing the virus to mutate to another, more virulent form.
RNA viruses do not exist as a single discrete species but rather occur as a "swarm" of closely related entities (so-called "quasispecies") (Eigen, 1993). Such genetic heterogeneity is thought to be an advantage to the virus, which then can adapt quickly to changing environmental conditions. In the case of coxsackievirus B3, it seems that something quite like the CVB3/20 strain is favored to grow under conditions of increased oxidative stress. Thus, when selenium-deficient mice are inoculated with CVB3/0, a genomic microvariant, which is favored to replicate under such conditions, takes over and becomes the dominant viral strain and the "output" virus becomes different from the "input" virus.
Theoretical Considerations vs Practical Implications
Genomes of RNA viruses, such as coxsackievirus, have high mutation rates, perhaps as much as 10<sup>4</sup> to 10<sup>6</sup> the rate of a comparable DNA genome (Holland et al., 1982). This high rate of mutation is due to the lack of any proofreading capability to correct errors made during replication of the viral RNA genome. Such high rates of mutation lead to rapid evolution, which would be very advantageous to the virus in terms of coping with changes in its environment. Assuming conservatively that only 20% of the positions in a 10 kilobase RNA virus genome can be substituted (the coxsackievirus has a genome of about 7.5 kb), Holland (1990) pointed out that 4<sup>2000</sup> permutations are possible. Only a small fraction of these combinations is likely to be viable, but the net result is still a very large number. He went on to predict some unpleasant surprises in our future when some of these mutants "affect our plant crops, domestic animals, and burgeoning human population."
A simple fact of nature is that the majority of all plant, animal and human viruses are RNA viruses (Holland, 1993). For example, hepatitis, polio, foot-and-mouth disease, measles, Ebola, Lassa, Marburg, tobacco mosaic virus, tomato bushy stunt virus, Hantavirus, flu and HIV are all RNA viruses or RNA virus-caused diseases. What Professor Beck and I have managed to do is to show that the evolution of one of these RNA viruses, the coxsackievirus, can be impacted by host nutritional status (Figures 1 and 2). Until that discovery, the effect of host nutrition on the genetic drift of a pathogen was unrecognized and unstudied. Obviously, this research trail has just begun and much work is ahead, but even if just a small portion of the broader implications of these results proves to be correct, this discovery should lead to significant advances in our understanding of how viruses interact with their hosts.
Bai, J., S. Wu, K. Ge, X. Deng and C. Su. 1980. The combined effect of selenium deficiency and viral infection on the myocardium of mice. Acta Acad. Med. Sin. 2:29.
Beck, M.A., P.C. Kolbeck, Q. Shi, L.H. Rohr, V.C. Morris and O.A. Levander. 1994a. Increased virulence of a human enterovirus (coxsackievirus B3) in selenium-deficient mice. J. Infect. Dis. 170:351.
Beck, M.A., P.C. Kolbeck, L.H. Rohr, Q. Shi, V.C. Morris and O.A. Levander. 1994b. Benign human enterovirus becomes virulent in selenium-deficient mice. J. Med. Virol. 43:166.
Beck, M.A., P.C. Kolbeck, L.H. Rohr, Q. Shi, V.C. Morris and O.A. Levander. 1994c.Vitamin E deficiency intensifies the myocardial injury of coxsackievirus B3 infection in mice. J. Nutr. 124:345.
Beck, M.A., Q. Shi, V.C. Morris and O.A. Levander. 1995. Rapid genomic evolution of a non-virulent coxsackievirus B3 in selenium-deficient mice results in selection of identical virulent isolates. Nature Med. 1:433.
Beck, M.A. 1997. Rapid genomic evolution of a non-virulent coxsackievirus B3 in selenium-deficient mice. Biomed. Environ. Sci. 10:307.
Beckman, K.B. and B.N. Ames. 1997. Oxidative decay of DNA. J. Biol. Chem. 272:19633.
Chow, L.H., K.W. Beisel and B.M. McManus. 1992. Enteroviral infection of mice with severe combined immunodeficiency. Evidence for direct viral pathogenesis of myocardial injury. Lab. Invest. 66:24.
Combs, G.F., Jr. and S.B. Combs. 1986. The Role of Selenium in Nutrition. Academic Press, Orlando.
Eigen, M. 1993. Viral quasispecies. Sci. Am. 269:42.
Holland, J.J. 1990. Continuum of change in RNA virus genomes. In: Concepts in Viral Pathogenesis. Notkins, A.L. and M.B.A. Oldstone, Eds., Springer-Verlag, New York, NY.
Holland, J. 1993. Replication error, quasispecies populations and extreme evolution rates of RNA viruses. In: Emerging Viruses. Morse, S.S., Ed. Oxford University Press, New York, NY.
Holland, J., K. Spindler, F. Horodyski, E. Grabau, S. Nichol and S. VandePol. 1982. Rapid evolution of RNA genomes. Science 215:1577.
Keshan Disease Research Group. 1979. Epidemiologic studies on the etiologic relationship of selenium and Keshan disease. Chin. Med. J. 92:477.
Levander, O.A. 1997. Nutrition and newly emerging viral diseases: an overview. J. Nutr. 127:948S.
Rossman, T.G. and E.I. Goncharova. 1998. Spontaneous mutagenesis in mammalian cells is caused mainly by oxidative events and can be blocked by antioxidants and metallothionein. Mutat. Res. 402:103.
Ruppert, M., V. Ruppert, B. Kuytz, H. Jomaa, I. Nakamura and B. Maisch. 1994. Coxsackievirus B3 infection leads to cell death of cardiac myocytes. J. Mol. Cell. Cardiol. 26:907.
Schneider, J.E., Jr., J.R. Phillips, Q. Pye, M.L. Maidt, S. Price and R.A. Floyd. 1993. Methylene blue and rose bengal photoinactivation of RNA bacteriophages: comparative studies of 8-oxoguanine formation in isolated RNA. Arch. Biochem. Biophys. 301:91.
Wang, D., D.A. Kreutzer and J.M. Essignmann. 1998. Mutagenicity and repair of oxidative DNA damage: insights from studies using defined lesions. Mutat. Res. 400:99.
Yang, G.Q. 1985. Keshan disease: an endemic selenium-related deficiency disease. In: Trace Elements in Nutrition and Children, Chandra, R.K., Ed., Raven Press, New York, NY, p. 273.
<hr noshade="noshade" size="1"> Back | Forward | Table of Contents
Special Circular 167-99
<hr noshade="noshade" size="1"> Involvement of Selenium in the Regulation of Viral Virulence
Orville A. Levander
U.S. Department of Agriculture, ARS
Beltsville Human Nutrition Research Center
Nutrient Requirements and functions Laboratory
Beltsville, MD 20705.
Keshan Disease: Deficiency vs Infection
Keshan disease (KD) is familiar to all scientists working in selenium nutrition. Discovery that this juvenile cardiomyopathy could be prevented by supplementing with selenium was one of the great triumphs of human trace element nutrition (Keshan Disease Research Group, 1979). Although there is no doubt about the role of selenium deficiency in causing this disease, certain epidemiological evidence indicates that there must be more to KD than an uncomplicated deficiency. For example, KD has a pronounced seasonal variation, which cannot be explained only on the basis of changes in selenium status: in North China, KD occurs mainly during the winter season, whereas in the South, the disease is seen primarily in the summer (Yang et al., 1985). This type of seasonal variation is much more typical of an infectious disease rather than a nutritional deficiency disease.
The Chinese scientists were well aware of the possibility that an infectious agent might be involved in the etiology of KD. In fact, they were able to isolate a number of viruses from KD patients. Among these viruses was included a coxsackie B4 virus. When this virus was tested for its ability to cause heart damage in selenium-deficient and selenium-adequate mice, it was found in the early 1980's that the virus was much more cardiotoxic and inflicted greater heart damage in the deficient mice (Bai et al., 1980). Some years later, Professor Melinda Beck of the University of North Carolina approached me about initiating a series of collaborative experiments to determine the influence of selenium status on viral virulence.
Selenium vs Virulence
Our first objective was to duplicate the earlier results of the Chinese scientists. Professor Beck had two different viral strains available for testing. The first strain, called coxsackievirus B3/20 (CVB3/20), causes heart damage even in animals fed normal diets. We were able to show that this virus caused increased heart damage in selenium-deficient mice compared to mice fed diets adequate in selenium (Beck et al., 1994a). These results essentially confirmed those reported earlier by the Chinese.
The second strain of virus, called coxsackievirus B3/0 (CVB3/0), does not cause heart damage in mice that are fed normal diets. That is, the virus is said to be avirulent in contrast to the CVB3/20 which is considered to be virulent. When the avirulent strain of coxsackievirus (CVB3/0) was inoculated into selenium-deficient mice, it caused a moderate degree of heart damage (Beck et al., 1994b). Thus, by feeding a diet deficient in selenium, we allowed an avirulent virus to express virulence. As far as we are aware, this was the first demonstration that the nutritional status of the host would permit an avirulent strain of virus to exhibit virulence. This was a novel finding, which could have significant implications for nutrition and infection interrelationships in the Third World, or wherever malnutrition was a problem.
Investigations into a variety of immune system functions indicated that mice infected with the coxsackievirus produced similar amounts of neutralizing antibody regardless of whether the animals were fed diets adequate or deficient in selenium. Therefore, the effects of selenium deficiency on viral virulence were probably not mediated through the antibody producing arm of the immune system. On the other hand, impediments in T cell function were noted in the selenium-deficient infected animals. Lymphocyte proliferation was somewhat reduced in response to either mitogen or antigen challenge. Natural killer (NK) cell activity was also mildly reduced due to selenium deficiency, but later experiments showed little correlation between NK cell activity and virulence in fish oil-fed mice infected with the coxsackievirus.
Anti-virulent Effects of Natural vs Synthetic Antioxidants
We felt it quite reasonable to test the effect of vitamin E deficiency on the virulence of the coxsackievirus because of the close metabolic and nutritional relationship between selenium and vitamin E. As with selenium deficiency, there was an increased virulence of the coxsackievirus in vitamin E-deficient mice, and this trend was exacerbated by feeding fish oil (Beck et al., 1994c). This was true for both the avirulent and virulent strains of the virus. As with selenium deficiency, B cell function did not appear to be impaired in the vitamin E-deficient mice, whereas T cell function was depressed in the deficient animals. We also found that feeding fish oil had a marked suppressive effect on NK activity regardless of the vitamin E status of the mice. Normally, NK activity is considered to play a crucial role in viral clearance. However, under the conditions of our coxsackievirus-infected mouse model, it was clear that we could suppress NK activity by approximately 90 percent by feeding fish oil to our vitamin E-supplemented mice without having any effect on the apparent virulence of the virus.
N,N'-diphenyl-p-phenylenediamine (DPPD) is a synthetic organic antioxidant that has powerful vitamin E-like activity but bears no relationship to the molecular structure of the tocopherols. Nonetheless, DPPD was highly effective in preventing the increased heart damage seen in vitamin E-deficient mice infected with the virulent coxsackievirus B3/20 (Beck, 1997). These results provided strong evidence that increased oxidative stress in the host milieu was responsible for the increased virulence of either the virulent or avirulent strains of the coxsackievirus that we observed in our selenium- or vitamin E-deficient mice.
What Changed? Host vs Pathogen
How does increased dietary oxidative stress translate into increased viral virulence and its attendant increased heart muscle damage? Two possibilities exist: such stress could either affect the ability of the host to resist the virus or, conversely, it could affect the ability of the virus to inflict damage on the host. The first possibility was consistent with the fact that coxsackievirus is known to exert a direct cytotoxic effect on heart muscle cells (Chow et al., 1992; Ruppert et al., 1994). One could imagine, therefore, that any interference with normal cellular metabolism might predispose cardiomyocytes to increased virus-induced damage. Heart muscle damage (i.e., nutritional muscular dystrophy) is a common feature of selenium/vitamin E deficiency per se (i.e., without viral involvement) in several species of animals, including turkeys, swine, sheep, cows and horses (Combs and Combs, 1986). Thus, heart muscle cells in selenium/vitamin E-deficient animals seem at greater risk of damage due to exposure to any of a variety of external insults (e.g., heavy metals, xenobiotics, viruses, etc.). This possibility could perhaps be tested by determining the effect of coxsackievirus on normal and deficient heart muscle cells cultured in vitro.
The second possibility was supported by much less experimental evidence, that is, the notion that the antioxidant- (selenium or vitamin E) deficient diet fed to the host might somehow change the virus itself in some way so that it became more virulent. In fact, one could reasonably argue that there was no precedent for such an idea. It seemed much more reasonable that the diet fed to the host should have an effect on the host rather than on the pathogen. Nonetheless, we carried out a passage experiment which consisted of two parts (Beck et al., 1995). In the first part, two groups of mice, one selenium-deficient, the other selenium-adequate, were inoculated with the benign form of the virus, CVB3/0. After 10 days, the hearts of the mice were harvested, and the virus was isolated from the tissue. Virus from the two groups of mice was cultured separately in HeLa cells and then inoculated into another group of mice -- all selenium adequate. The CVB3/0 that had been passed through the adequate mice in the first part of the experiment, so-called CVB3/0Se<sup>+</sup>, caused no heart damage when inoculated into the adequate mice in the second part. On the other hand, the CVB3/0 that had been passed through the deficient mice in the first part of the experiment, so-called CVB3/0Se<sup>-</sup>, did indeed cause heart damage when inoculated into the adequate mice in the second part. In other words, passage of the normally benign CVB3/0 through a selenium-deficient host, thereby allowing it to replicate under conditions of in vivo oxidative stress, altered the phenotype of the virus so that the previously avirulent CVB3/0 now became the virulent CVB3/0Se<sup>-</sup>.
Phenotypic Change vs Genotypic Change
The results of the passage experiment suggested that the phenotypic change observed in the properties of the CVB3/0 passed through a selenium-deficient host was actually due to a change in the genetic make-up of the virus. How else to account for the increased virulence of the CVB3/0Se- in the second part of the passage experiment? Proof of this contention, however, could only be obtained by sequencing the RNA from CVB3/0Se- and comparing the sequence with that of the parent CVB3/0 strain. When such sequencing analysis was performed, it was found that the nucleotide composition of the newly virulent CVB3/0Se<sup>-</sup> had changed to resemble that of the naturally virulent wild type strain CVB3/20 at 6 of 7 genomic sites thought to influence virulence (Beck et al., 1995). To our knowledge, this was the first report demonstrating that the nutritional status of the host could exert any control over the genetic nature of an invading pathogen. But how did the redox state of the host cell alter viral genetics?
Mutation vs Selection
Two biochemical mechanisms could account for the effect of oxidative stress on the viral genome: mutation of the viral RNA genetic material due to oxidative damage or selection of a pre-existing viral substrain because of a change in growth conditions in the host cell (i.e, shift in peroxidative tone).
Mutagenesis in mammalian cells as a result of oxidative DNA damage is a well-known and well-studied phenomenon (Beckman and Ames, 1997; Wang et al., 1998; Rossman and Goncharova, 1998). In contrast, relatively little is known about oxidative base damage in RNA, although 8-oxo-guanosine has been identified in RNA under oxidizing conditions (Schneider et al., 1993). Therefore, one could postulate that the RNA genome of the coxsackievirus had been altered via oxidative attack, thereby allowing the virus to mutate to another, more virulent form.
RNA viruses do not exist as a single discrete species but rather occur as a "swarm" of closely related entities (so-called "quasispecies") (Eigen, 1993). Such genetic heterogeneity is thought to be an advantage to the virus, which then can adapt quickly to changing environmental conditions. In the case of coxsackievirus B3, it seems that something quite like the CVB3/20 strain is favored to grow under conditions of increased oxidative stress. Thus, when selenium-deficient mice are inoculated with CVB3/0, a genomic microvariant, which is favored to replicate under such conditions, takes over and becomes the dominant viral strain and the "output" virus becomes different from the "input" virus.
Theoretical Considerations vs Practical Implications
Genomes of RNA viruses, such as coxsackievirus, have high mutation rates, perhaps as much as 10<sup>4</sup> to 10<sup>6</sup> the rate of a comparable DNA genome (Holland et al., 1982). This high rate of mutation is due to the lack of any proofreading capability to correct errors made during replication of the viral RNA genome. Such high rates of mutation lead to rapid evolution, which would be very advantageous to the virus in terms of coping with changes in its environment. Assuming conservatively that only 20% of the positions in a 10 kilobase RNA virus genome can be substituted (the coxsackievirus has a genome of about 7.5 kb), Holland (1990) pointed out that 4<sup>2000</sup> permutations are possible. Only a small fraction of these combinations is likely to be viable, but the net result is still a very large number. He went on to predict some unpleasant surprises in our future when some of these mutants "affect our plant crops, domestic animals, and burgeoning human population."
A simple fact of nature is that the majority of all plant, animal and human viruses are RNA viruses (Holland, 1993). For example, hepatitis, polio, foot-and-mouth disease, measles, Ebola, Lassa, Marburg, tobacco mosaic virus, tomato bushy stunt virus, Hantavirus, flu and HIV are all RNA viruses or RNA virus-caused diseases. What Professor Beck and I have managed to do is to show that the evolution of one of these RNA viruses, the coxsackievirus, can be impacted by host nutritional status (Figures 1 and 2). Until that discovery, the effect of host nutrition on the genetic drift of a pathogen was unrecognized and unstudied. Obviously, this research trail has just begun and much work is ahead, but even if just a small portion of the broader implications of these results proves to be correct, this discovery should lead to significant advances in our understanding of how viruses interact with their hosts.
<center> <table border="0"> <tbody><tr> <th>Figure 1. Traditional view of nutrition-infection interactions. In this scheme, host nutritional status exerts its influence only via effects on the host immune system's ability to counteract the pathogenic agent. No consideration is given to effects on the pathogen itself. Figure reproduced with permission from Levander (1997).</th> </tr> <tr> <th> 
</th> </tr> </tbody></table> </center>
</th> </tr> </tbody></table> </center> <center> <table border="0"> <tbody><tr> <th>Figure 2. Modified view of nutrition-infection interactions. According to this scheme, not only can host nutritional status affect the host's immune response to the agent but also can influence the genetic make-up of the pathogen itself. Figure reproduced with permission from Levander (1997). </th> </tr> <tr> <th> 
</th> </tr> </tbody></table> </center>
References</th> </tr> </tbody></table> </center> Bai, J., S. Wu, K. Ge, X. Deng and C. Su. 1980. The combined effect of selenium deficiency and viral infection on the myocardium of mice. Acta Acad. Med. Sin. 2:29.
Beck, M.A., P.C. Kolbeck, Q. Shi, L.H. Rohr, V.C. Morris and O.A. Levander. 1994a. Increased virulence of a human enterovirus (coxsackievirus B3) in selenium-deficient mice. J. Infect. Dis. 170:351.
Beck, M.A., P.C. Kolbeck, L.H. Rohr, Q. Shi, V.C. Morris and O.A. Levander. 1994b. Benign human enterovirus becomes virulent in selenium-deficient mice. J. Med. Virol. 43:166.
Beck, M.A., P.C. Kolbeck, L.H. Rohr, Q. Shi, V.C. Morris and O.A. Levander. 1994c.Vitamin E deficiency intensifies the myocardial injury of coxsackievirus B3 infection in mice. J. Nutr. 124:345.
Beck, M.A., Q. Shi, V.C. Morris and O.A. Levander. 1995. Rapid genomic evolution of a non-virulent coxsackievirus B3 in selenium-deficient mice results in selection of identical virulent isolates. Nature Med. 1:433.
Beck, M.A. 1997. Rapid genomic evolution of a non-virulent coxsackievirus B3 in selenium-deficient mice. Biomed. Environ. Sci. 10:307.
Beckman, K.B. and B.N. Ames. 1997. Oxidative decay of DNA. J. Biol. Chem. 272:19633.
Chow, L.H., K.W. Beisel and B.M. McManus. 1992. Enteroviral infection of mice with severe combined immunodeficiency. Evidence for direct viral pathogenesis of myocardial injury. Lab. Invest. 66:24.
Combs, G.F., Jr. and S.B. Combs. 1986. The Role of Selenium in Nutrition. Academic Press, Orlando.
Eigen, M. 1993. Viral quasispecies. Sci. Am. 269:42.
Holland, J.J. 1990. Continuum of change in RNA virus genomes. In: Concepts in Viral Pathogenesis. Notkins, A.L. and M.B.A. Oldstone, Eds., Springer-Verlag, New York, NY.
Holland, J. 1993. Replication error, quasispecies populations and extreme evolution rates of RNA viruses. In: Emerging Viruses. Morse, S.S., Ed. Oxford University Press, New York, NY.
Holland, J., K. Spindler, F. Horodyski, E. Grabau, S. Nichol and S. VandePol. 1982. Rapid evolution of RNA genomes. Science 215:1577.
Keshan Disease Research Group. 1979. Epidemiologic studies on the etiologic relationship of selenium and Keshan disease. Chin. Med. J. 92:477.
Levander, O.A. 1997. Nutrition and newly emerging viral diseases: an overview. J. Nutr. 127:948S.
Rossman, T.G. and E.I. Goncharova. 1998. Spontaneous mutagenesis in mammalian cells is caused mainly by oxidative events and can be blocked by antioxidants and metallothionein. Mutat. Res. 402:103.
Ruppert, M., V. Ruppert, B. Kuytz, H. Jomaa, I. Nakamura and B. Maisch. 1994. Coxsackievirus B3 infection leads to cell death of cardiac myocytes. J. Mol. Cell. Cardiol. 26:907.
Schneider, J.E., Jr., J.R. Phillips, Q. Pye, M.L. Maidt, S. Price and R.A. Floyd. 1993. Methylene blue and rose bengal photoinactivation of RNA bacteriophages: comparative studies of 8-oxoguanine formation in isolated RNA. Arch. Biochem. Biophys. 301:91.
Wang, D., D.A. Kreutzer and J.M. Essignmann. 1998. Mutagenicity and repair of oxidative DNA damage: insights from studies using defined lesions. Mutat. Res. 400:99.
Yang, G.Q. 1985. Keshan disease: an endemic selenium-related deficiency disease. In: Trace Elements in Nutrition and Children, Chandra, R.K., Ed., Raven Press, New York, NY, p. 273.
<hr noshade="noshade" size="1"> Back | Forward | Table of Contents