Population dynamics of Vibrio cholerae and its bacteriophage
1. CHAPTER 1. INTRODUCTION 1.1 Cholera : background and history / 1 1.2 Factors affecting cholera outbreaks / 3 1.3 Bacteriophage : history and the role in cholera epidemics / 6 1.4 Population dynamics of bacteria and bacteriophage : in the laboratory and in the environment & mechanisms of bacterial resistance to phage / 13 1.5 Summaries of Chapter 2,3&4 / 19 2. CHAPTER 2. AN EXPERIMENTAL STUDY OF THE POPUATION AND EVOLUTIONARY DYNAMICS OF VIBIRO CHOLERAE O1 AND THE BACTERIOPHAGE JSF4 2.1 Introduction / 23 2.2 Materials and Methods / 25 2.3 Results / 31 2.4 Discussion / 41 3. CHAPTER 3. THE POPULATION AND EVOLUTIONARY DYNAMICS OF VIBRIO CHOLERAE AND ITS BACTERIOPHAGE: CONDITIONS FOR MAINTAINING PHAGE-LIMITED COMMUNITIES 3.1 Introduction / 47 3.2 Materials and Methods / 50 3.3 Results / 54 3.4 Discussion / 68
4. CHAPTER 4. THE POPULATION AND EVOLUTIONARY DYNAMICS OF BACTERIA IN PHYSICALLY STRUCTURED HABITATS: THE ADAPTIVE VIRTUES OF MOTILITY 4.1 Introduction / 72 4.2 Results / 76 4.3 Discussion / 83 4.4 Methods / 87 5. CHAPTER 5. SUMMARY AND DISCUSSION Summary and Discussion / 91 6. BIBLIOGRAPHY Bibliography / 97 7. APPENDIX 7.1 Supplementary Information to chapter 2 / 109 7.2 Supplementary Information to chapter 3 / 119 7.3 Supplementary Information to chapter 4 / 131
List of Figures Figure 1.1 / 7 Figure 2.1 / 33 Figure 2.2 / 36 Figure 2.3 / 39 Figure 2.4 / 41 Figure 3.1 / 55 Figure 3.2 / 57 Figure 3.3 / 59 Figure 3.4 / 59 Figure 3.5 / 62 Figure 3.6 / 65 Figure 3.7 / 66 Figure 3.8 / 68 Figure 4.1 / 77 Figure 4.2 / 78 Figure 4.3 / 81 Figure 4.4 / 82
List of Tables Table 2.1 / 26
CHAPTER 1 Introduction 1.1 Cholera: background and history Cholera, a diarrheal disease, has been infecting humans for at least a thousand years based on descriptions of its unique symptoms in ancient Greek and Chinese literature (Mekalanos, Rubin and Waldor 1997). It is one of the most rapidly fatal human diseases known. Death can occur as fast as eight hours from the onset of symptoms. Before the development of intravenous and oral fluid replacement therapies, cholera was estimated to have killed over 100 000 people each year in India alone between 1900 and 1950. As the disease grew to pandemic proportions in the early 1800s, it killed millions worldwide (Mekalanos, Rubin and Waldor 1997). Although diarrhea can be caused by various bacterial pathogens, no other bacterium produces the repeated pandemics characteristic of Vibrio cholerae. V. cholerae, a gram-negative bacterium, was isolated as the causative organism of cholera by Robert Koch from rice water stools of patients in Egypt and India during the fifth pandemic (1883 - 1884). The bacteria exist as free-living organisms in coastal waters throughout the world. Most V. cholerae do not cause harm to humans. Out of more than 200 serogroups of V. cholerae reported so far, only two serogroups (O1 and O139) are toxigenic and are responsible for major pandemics and epidemics. Humans contract cholera by ingesting water or food contaminated with toxigenic V. cholerae. Two main weapons possessed by V. cholerae enable it to infect and grow viciously within human hosts: cholera toxin (CT) and toxin co-regulated pilus (TCP). Toxigenic V. cholerae carries one or more copies of CT genes that produce toxin causing secretion of
chloride ion into the lumen of the intestine while inhibiting sodium uptake (Mekalanos, Rubin and Waldor 1997). The result of this imbalanced secretion and uptake of ions occurring in the intestine is watery diarrhea that takes on the appearance of rice-water, a cloudy liquid with flecks of mucus. Up to 20 liters of diarrhea during a single infection have been reported in cholera infected patients and the fluid usually contains 10 8 bacteria per milliliter that will be disseminated to the aquatic environment and continue to infect more susceptible people in the absence of preventative measures. Without the help of TCP, a crucial colonization factor in the small intestine, CT alone does not cause extensive damage to the host. V. cholerae mutant strains without TCP are more than 1000-fold reduced in their capacity to colonize the infant mouse and essentially do not colonize the intestine of human volunteers (Taylor et al. 1987; Herrington et al. 1988; Mekalanos, Rubin and Waldor 1997).
Considering our extensive knowledge on how V. cholerae sickens people and the fact that we are familiar with the details of transmission, it is a shame that cholera remains a major threat to human health worldwide. The distribution of epidemic cholera cases is directly related to public health measures taken to provide clean water. For example, In Europe and North America, epidemic cholera is essentially non-existent as a result of sewage systems which chlorinate water and prevent drinking water contamination. Elsewhere in the world, primarily Asian, Africa and South America, clean drinking water is not accessible to all population and sanitation can be poor. As a result, cholera is most prevalent in these regions of the world. The true burden of cholera is estimated to reach
several million cases per year, predominantly in Asia and Africa (Nelson et al.2009). From 1995 to 2005, Africa reported the largest and most frequent cholera outbreaks, followed by Asia (Griffith, Kelly-Hope and Miller 2006). Zimbabwe probably offers the most recent example of the tragedy that befalls a country and its people when cholera strikes. The 2008-2009 outbreak rapidly spread across every province and brought rates of mortality similar to those witnessed as a consequence of cholera outbreaks a hundred years ago (Nelson et al.2009). This enormous human suffering caused by cholera earned cholera the distinction of being one of the “emerging and reemerging infections” threatening many developing countries (Satcher 1995).
1.2 Factors affecting cholera outbreaks A distinctive epidemiological feature of cholera is its seasonal pattern in endemic regions, such as Bangladesh. Bangladesh is situated in the Ganges delta, the region where all except the seventh cholera pandemics started (Faruque, Albert and Mekalanos 1998). Epidemics usually occur twice a year, with the highest number of cases just after the monsoons from September to December. A smaller peak of cases also is observed during the spring, between March and May (Faruque et al. 2005.1 PNAS). Various factors, biological and environmental, have been hypothesized to shape the seasonal pattern of cholera outbreaks.
First, environmental factors, such as temperature or rainfall which can be broadly termed as “climate”, undoubtedly play a decisive role, not only in cholera outbreaks, but in many infectious diseases caused by bacteria pathogens (Colwell 1996; Pascual et al. 2000). V. cholerae is characterized by an increased growth rate at warm temperatures, which is evident in higher rates of isolation in the environment during warm months. Between epidemics in areas of endemicity, it is exceedingly difficult to isolate toxigenic V. cholerae; however, during periods of warmer water temperatures, success in isolation of toxigenic V. cholerae rises substantially (Lipp, Huq and Colwell 2002).
Climate also triggers a series of changes in other biological factors interacting with V. cholerae in the environment. Marine bacteria, including Vibrio spp., have been found to consume chitin as carbon and nitrogen source using their chitinase. Because seasonal changes in climate can influence populations of chitinous organisms, such as copepods, amphipods and other small crustaceans, climate may also influence the presence of cholera in the environment. Approximately 10 11 metric tons of chitin per year is produced in aquatic environments; >10 9 metric tons is produced by copepods alone (Lipp, Huq and Colwell 2002). Nalin first suggested that V. cholerae might use the strategy of its sister species, V. parahaemolyticus, and adsorb onto copepod zooplankton to survive the unfavorable environment between the epidemics (Nalin 1976). Nalin then demonstrated V. cholerae is able to adhere and grow on the chitin particles (Nalin et al. 1979). After this, many studies, both in the laboratory system and in the field, showed that the presence of crustacean copepods enhance the survival of V. cholerae (Huq et al. 1983;
Huq et al. 1984; Tamplin et al. 1990; Chiavelli, Marsh and Taylor 2001). Direct observation in Bangladesh strongly supports an important role for zooplankton in cholera outbreaks. During spring and late summer in Bangladesh, phytoplankton blooms occur, followed by zooplankton, with heaviest blooms occurring in September and October, which are then followed by cholera outbreaks (Colwell et al. 2003). Based on the association between zooplankton and V. cholerae, a simple procedure, using folded sari cloth to filter water before consuming, was developed in Bangladesh. Cholera bacteria associated with zooplankton copepods will be removed by simple filtration. Cholera cases in regions practicing water filtration were half as frequent compared with the regions where filtration was not performed (Colwell et al. 2003).
Apart from the biological factors stated above, host immunity and bacterial virulence (hyper-infectious state) have also been suggested to be involved in shaping cholera outbreaks (Koelle et al. 2005; Hartley, Morris and Smith 2006; Nelson et.al 2009). All these factors are non-exclusive, interacting with each other and influencing cholera outbreaks on different scales and at different stages. Mathematical models accounting for extrinsic factors, such as temperature, as well as some biological factors, have shown that both non-biological and biological factors shape the epidemic cycle of cholera (Koelle et al. 2005; Hartley, Morris and Smith 2006). However, knowledge about one distinctive feature of cholera outbreaks, the rapid collapse of the epidemic, still remains elusive. In this dissertation, I will discuss one biological factor, the presence of phage, which could be responsible for the rapid collapse of cholera outbreaks. Interestingly, phage has been
associated with cholera since it was first observed in 1896 (Hankin 1896), but little is known about its role in bacterial population dynamics.
1.3 Bacteriophage : history and the role in cholera epidemics 1.3.1 General information about Bacteriophage Bacteriophage are small viruses that specifically infect and lyse bacteria. Although extremely diverse, all phage share several common steps during their life cycle: adsorption, separation of genetic materials from protein coat, expression and replication of the genetic material within bacterial host, virion assembly, release and transmission (Weinbauer 2004). Phage first adsorb to the surface of a bacterium through a two-step process. The first step of adsorption is reversible but specific to a particular cell surface component. This is followed by an irreversible binding to a receptor, through which the genetic material is injected into the cells. After the injection, genetic material of the phage is either integrated into host genome or remains free in the cytoplasm depending on the nature of phage and physiological state of the cells. The genomes of a particular group of phage, the lytic or virulent phage, always stay outside the bacterial genome and turn their bacterial hosts into a phage factory in which new progeny are produced. The new progeny phage are then released through the lysis of the host and seek new hosts to infect. Another phage group of phage, lysogenic or temperate phage, do not kill their hosts rapidly like lytic phage. Lysogenic phage instead replicate with the hosts by inserting their genome into the bacterial host genome. The harmonious state in which the
phage remains in the bacteria without killing it can be interrupted physically or chemically and lysogenic phage will then behave like lytic phage. Finally, pseudolysogenic phage are similar to lysogenic phage except their genomes remain in cytoplasm than being inserted into host genome (Weinbaur 2004). These pseudolysogenic phage are characterized by the ability of the phage to induce a chronic infection in the bacteria during which phage progeny are constantly released into the environment by budding or extrusion, without having to lyse the bacterial host cell (Weinbaur 2004). To illustrate life cycles of different bacteriophage, a figure (Figure 1.1) is presented below.
Figure 1.1 Types of viral life cycles
1.3.2 History of phage The scientific discovery of bacteriophage is attributed to Twort (1915) and d'Herelle (1917). However, the first observation of phage was made 20 years earlier by Ernest H. Hankin in 1896 (Hankin 1896). Intrigued by the difference between the cholera outbreaks along the Ganges and Jumna rivers in India, Hankin found an unknown source of antibacterial activity against V. cholerae in the river and then suggested that this unidentified substance, which passed through fine porcelain filters and was heat labile, was responsible for limiting the spread of cholera epidemics (Hankin 1896). Despite the capability for killing bacteria of this mysterious source, no one, including Hankin himself, used it for controlling bacterial populations until 1919 when d‟Herelle championed their use in controlling the spread of infectious diseases. D‟Herelle named them bacteriophages and eventually, he and his colleagues implemented phage therapy throughout the world, with major efforts in India, Egypt, the United States and the Soviet Union. Bacterial infections that have been successfully controlled using phage therapy include Shigella, Staphylococcus, Streptocooccus, Klebsiella, E. coli, Salmonella, V. cholerae, Pesudomonas and Proteus (Sulakvelidze, Alavidze and Morris 2001). Although most human phage therapy studies show positive results, research of phage therapy gradually declined in the west in the 1940s due to the failure to establish rigorous proof of efficacy and increasing availability of antibiotics (Levin and Bull 2004). The Republic of Georgia, however, survived the “anti-phage” times, and the George Eliava Institute of Bacteriophage, located in the capital city of Georgia, has actively employed phage therapy since the 1930s and is now the global center of phage therapy expertise.
As mentioned above, phage therapy was also used to fight cholera. The first therapy trial compared 244 untreated cholera patients with 219 patients who were treated with vibriophage; the untreated group had a 20% mortality rate whereas mortality in the treated group was 6.8% (Nelson et al. 2009). Several other studies showed similar results. Although suffering from the same limitations as other early phage therapy studies did (poor controls and inconsistent therapeutic results), these studies demonstrated enough success to continue the use of phage to treat cholera on a large scale at a later time. From 1928 to 1934, over a million vibriophage doses were prepared and disseminated in specific study communities in India. This application was novel because, for the first time, vibriophages were also disseminated into drinking water sources as a means of prophylaxis. The triennial death rates from cholera fell from 30 to 2 per 10,000 in communities that were treated with phage (Nelson et al. 2009).
At the same time, an interesting relationship between V. cholerae and vibriophages in the environment was also reported. Influenced by d‟Herelle and Malone„s suggestion that the cessation of cholera epidemic was due to the spread of bacteriophage from convalescent cases, Pasricha et al. studied the prevalence of vibriophage in nature and its relationship to cholera in Calcutta (Pasricha, de Monte and Gupta 1931). They found that vibriophage in nature vary with the incidence of the disease and that the mortality rate, which is high at the beginning of the cholera season, falls rapidly once vibriophage have become widely distributed in nature. It was therefore concluded that bacteriophage plays an important role in lowering mortality and in bringing an epidemic to a close (Pasricha, de
Monte and Gupta 1931). It is understandable that the phage biology in these early experiments were rather crude considering that phage had only been discovered 15 years ealier. However, this is the first scientific study, to our knowledge, clearly reporting the relationship between cholera and vibirophages from the natural environment and also from the patients. Subsequently, advances in rehydration and antibiotic therapy made cholera phage therapy, either as individual treatment or in the environment as a prophylaxis, insignificant.
The story of cholera and phage certainly did not end. In the 1990s, as more and more efforts emphasized the molecular pathology of V. cholerae, John Mekalanos and colleagues found that cholera toxin (CT), the key virulent factor, was actually carried by a filamentous phage (Waldor and Mekalanos 1996). They showed experimentally that a toxigenic V. cholerae strain under certain conditions, is able to transmit genetic elements containing CT genes to other strains which lack CT. This transfer, which does not require cell-cell contact, could be accomplished by co-culturing recipient cells with filtered culture supernatant from the donor. Electron microscopic studies later showed that culture supernatants from the donor strain contained structures similar in morphology to filamentous bacteriophage (Waldor and Mekalanos 1996). The phage was then designated CTX phage. What is more surprising is that CTX phage uses TCP as a receptor to infect naïve bacteria; as previously mentioned, TCP is another virulence factor in cholera pathogenesis (Waldor and Mekalanos 1996). This finding illustrates how
a non-pathogenic bacterium becomes a highly virulent pathogen by horizontal transfer mediated by lysogenic bacteriophage.
1.3.3 Rediscovery of the important relationship between cholera and bacteriophage About 60 years after Patrisha‟s report of the intriguing relationship between cholera and vibriophages, Shah and colleagues reported a similar association, but with a more systematic sampling method and much better knowledge of both bacteria and phage (Faruque et al. 2005.1). Over a three-year period, they systematically analyzed water samples collected from two major rivers and a lake in Dhaka. The majority of water samples suggested an inverse relationship between the presence of vibriophages capable of lysing a given serogroup of V. cholerae and the presence of a strain of that same serogroup (Faruque et al. 2005.1). Further, the number of cholera patients varied seasonally during the study period and frequently coincided with the presence of pathogenic V. cholerae strains in water samples that lacked detectable vibriophages (Faruque et al. 2005.1). During interepidemic periods, water samples were found to contain vibriophages but no bacteria (Faruque et al. 2005.1). In another smaller scale study, the environmental prevalence of the epidemic V. cholerae O1 strain and a particular vibriophage JSF4 was monitored during a local outbreak. In addition, excretions of the same phage were also monitored from cholera patients during the study period (~17 weeks) (Faruque et al. 2005.2). This closer look at the cholera-phage dynamics showed that the peak of epidemic was followed by high JSF4 levels in the environment. Furthermore, the buildup of the phage coincided with increasing excretion
of the same phage (JSF4) from the cholera patients (Faruque et al. 2005.2). Thus, these systematically collected data supported the original hypothesis presented by Patrisha and suggested that vibriophage play an important role in cholera outbreaks, particularly in ending them. To quantify this hypothesis, Jensen et.al developed a mathematical model combining the epidemiology of cholera and V. cholerae population dynamics in the presence of its bacteriophage (Jensen et al. 2006). The model predicts that under reasonable biological parameters, vibriophage can ameliorate cholera outbreaks. When the outbreak is initiated by an interruption of bacteria-phage equilibrium in the reservoir, the density of phage remaining in the reservoir affects the severity of the outbreak. Even if the outbreak is initiated directly by increased bacterial growth, in the absence of phage in the reservoir, the introduction of phage will reduce the severity of the outbreak and promote its decline. The major limitation of this model is bacterial resistance to phage, if developed, would essentially render phage useless on limiting the density of bacteria, producing no effect on the cholera epidemic.
Are populations of V. cholerae limited by bacteriophage? Or do V. cholerae develop resistance to phage? These questions, though important, were ignored or unanswered in the published reports speculating on the role of vibriophage in limiting cholera outbreaks. The goal of my dissertation is to fully characterize the effect of vibriophage on population dynamics of V. cholerae. Based on the laboratory results, which will be reviewed later, phage-resistance can easily develop in many other bacteria, including E.
coli, Pseudomoas, Salmonella etc. In natural environments that are of real interest to many scientists, data are quite scarce.
1.4 Population dynamics of bacteria and phage 1.4.1 Population dynamics of bacteria and phage in the laboratory Bacteria-phage systems have been of long-standing interest to many biologists: not only to microbiologists but also to ecologists and evolutionary biologists. Microbial research systems are appealing because microbes can be routinely grown in the laboratory, have short generation times and large populations that can be easily maintained, allowing for rapid evolutionary change and the chance to study communities on both ecological and evolutionary time scales (Bohannan and Lenski 2000). As such, bacteria-phage systems provide an ideal model system for the study of processes in ecology and evolution. Most of the studies in this field investigate population dynamics and interactions between one bacterial species and one phage. Almost all of these studies have shown that phage- resistant bacteria evolve and replace the majority of sensitive ancestors, regardless of the bacterial or phage strain under study (Chao, Levin and Stewart 1977; Lenski and Levin 1985; Middelboe 2000; Mizoguchi et al. 2003). In some studies, phage have also evolved mechanisms to infect resistant bacteria. However, in this evolutionary arms-race, the bacteria have always overcome the phage (Lenski and Levin 1985; Chao, Levin and Stewart 1977).
In laboratory studies, as well as mathematical models, the fate of the phage-bacteria culture is greatly influenced by the fitness costs of resistance mutations to the bacteria. Populations of bacteria are only temporarily controlled by phage and later controlled by resources (limiting nutrients) after the emergence of phage-resistant bacteria. One study is an exception. While studying the long-term co-evolutionary arms race between Pseudomonas fluorescens and one of its naturally associated phage, Buckling & Rainey found that the phage persisted after 300 generations of co-cultivation with bacteria (Buckling and Rainey 2002). Furthermore, they also could infect the co-existing bacterial populations, indicating that phage-resistant bacteria never evolved or never dominated the population. This study failed to report bacterial or phage density, thus, it is hard to make the conclusion that it is a phage-limited condition, though the resistance data suggests that this is the case.