Emerging Viral Diseases

Introduction

In recent years there has been increasing attention paid to the changing patterns of infectious diseases. In particular, the factors that lead to increases in the rates of these so-called ‘emerging infectious diseases’ (EIDs) has focused primarily on the role of human activities, such as land use changes, population growth, increased contacts with wild animal reservoirs and the degradation of health care resources. It is estimated that the majority – some estimates place it as high as 75%, of these emerging diseases are derived from animals. These zoonoses spill over from their natural reservoirs either through direct contact or indirectly through close contact with domestic animals and subsequently into human populations.

In this paper, we take the emphasis proposed by Childs (2004) that many of the key factors that receive the focus of researchers in EIDs are important ‘transmission modifiers’ They increase the potential for transmission. But the most important event in new disease emergence is genetic changes in the pathogen that make it possible to become established in a new host species, productively infect new individuals in the new hosts (typically humans) and create local, regional or world wide health threats.

We focus on six groups of viral pathogens and their diseases that have received increased attention during the past five years. We briefly outline some of their basic biology, epidemiology, clinical presentation and the bases, to the extent it is understood, of how these agents have emerged and the current risk they represent.

Influenza

Influenza viruses are segmented, negative sense, single stranded, RNA viruses (Orthomyxoviridae), including several genera that infect animals, as well as humans. Influenzavirus A has the widest host range and the segmented genome allows for the accumulation of genetic mutations as well as the reassortment of the various genomic segments between virus strains. Influenzavirus A subtypes are identified based on the antigenic characteristics of the surface glycoproteins hemagglutinin (16 subtypes of HA) and neuraminidase (9 subtypes of NA) (Alexander 2007).

Avian influenza

The major animal reservoirs of Influenza A are migratory birds and the majority of all the possible combinations of HA-NA subtypes have been isolated from them. Currently only subtypes H5 or H7 have caused significant disease in birds (highly pathogenic avian influenza; HPAI). This appears due to the genetic structure of the HA precursor that determines the cell tropism for successful replication. HPAI replicate in numerous cell types in various organs. This replication causes damage to vital organs in some species resulting in significant disease and increased death rates. The rates of disease for HPAI strains vary with the species of birds and other, poorly defined environmental conditions so not all H5 or H7 strains are HPAI (Alexander 2007). Contact between domestic and wild birds allows for the introduction of LPAI strains and subsequent mutations in the virus can lead to HPAI outbreaks. Infected domestic ducks, are often asymptomatic and are thought to serve as a primary conduit for the spread of HPAI H5N1 once established (Peiris et al 2007). Biosecurity of domestic flocks and associated materials to minimize contact with infectious wild species, active surveillance and aggressive flock eradication have been the only successful methods for eradication of virus once it is established (Peiris et al 2007).

In 1997 HPAI H5N1 virus was detected in domestic geese in Guangdong Province, China and spillover occurred to human poultry workers. These cases were some of the first in which a purely avian virus caused severe disease in humans, with 18 individuals infected, and six deaths. This outbreak also led to the slaughter of approximately 1.5 million poultry in Hong Kong to abort the outbreak – emphasizing the tremendous economic as well as health impacts of these outbreaks (Peiris et al 2007). The virus was a reassortant with the H5 derived from geese with the other genes apparently derived from viruses prevalent in quail from the region (Peiris et al 2007). In addition to H5N1, the direct transmission of avian influenza viruses of subtypes H7N7, H9N2, and H7N3 has been associated occasionally with human disease (Peiris et al 2007).

Continued surveillance through the early 2000’s indicated that the original virus lineage was lost, but the repeated appearance of other lineages that included the goose H5 was ongoing. By 2003 a specific genotype of H5N1, (Z genotype) became the dominant form. The establishment of regional H5N1 epidemics in human populations has not yet occurred, but transmission from infected birds to humans continues. The first cases in China after the 1997 outbreak were associated with the Z strain in 2003. These and most subsequent human cases have been linked with human exposure to poultry and poultry products.

Among the human H5N1 cases, there are family clusters but it is unclear if these represent infection from a common environmental source or limited human-to-human transmission. Viruses isolated from these clusters have not contained mutations that would indicate the virus has adapted to infect humans. However, increased exposure of humans to H5N1 infected animals increases the probability that mutations in the viral genome that can enhance infection of or transmission among humans will arise and be fixed in the virus strain.

Symptoms typically develop 2–4 days following exposure to sick poultry though incubation times of up to 8 days have been reported. Symptoms present as a flu-like illness with fever, cough, shortness of breath and radiological evidence of pneumonia. There is often bilateral diffuse patchy or interstitial infiltrates and segmental or lobular consolidation with air bronchograms (Peiris et al 2007). Conjunctivitis or upper respiratory symptoms are not prominent in H5N1 cases to date but diarrhea, vomiting and abdominal pain are frequently described. CNS involvement has rarely been described. There is a rapid progression in severe cases with a 4 day duration from time of onset of symptoms to hospitalization and a median time to death of 9 days. Laboratory results of severe cases include lymphopenia with an inverted ratio of CD4 positive: CD8 positive cells, thrombocytopenia and increased liver transaminase levels. The presence of H5N1 virus in tissues outside the respiratory tract is another factor that may contribute to disease severity (Peiris et al 2007). The number of asymptomatic or mild cases of H5N1 disease that occur is believed to be low but it is not clear how many of these cases are missed because of a lack of adequate surveillance and diagnostic testing.

There is a marked age bias in cases of severe disease with more than half the cases under age 20 and nearly 90% under the age of 40. The bias does not appear to be due to population-age structure. The lower case incidence and lower case fatality rates for H5N1 in those over 40 years of age remain unexplained. It appears that while exposure to a source of H5N1 infection is necessary, such exposure is not sufficient to explain the observed epidemiology of H5N1 disease. Other factors appear to be crucial in determining who gets infected and ill (Peiris et al 2007).

The adamantanes (amantadine and rimantadine) and the NA inhibitors (oseltamivir and zanamivir) are the two classes of drugs that are active against influenza viruses. The adamantanes inhibit the ion channel activity of the M2 membrane protein of influenza A viruses. The H5N1 virus strains that emerged in 1997 were amantadine sensitive but those isolated since 2003 are often resistant to these drugs. It appears that widespread prophylactic use of amantadine among poultry farmers contributed to the appearance of antiviral resistant virus strains. Adamantadine resistance has led to the widespread use of neuraminidase inhibitors (Tamiflu, Oseltamivir) to treat H5N1 infections. Clinical trials initially suggested that neuraminidase resistant influenza viruses rarely developed, but several drug resistant H5N1 viruses have been isolated from infected patients undergoing neuraminidase treatment. Whether this is due to the highly pathogenic nature of H5N1 viruses when compared to seasonal human influenza is not clear.

Human influenza

Human influenza usually refers to those subtypes that spread widely among humans. H1N1, H1N2, and H3N2 are the only known Influenzavirus A subtypes currently circulating among humans. The increased prevalence of amantadine resistance in circulating human influenza is of growing concern. Until recently, most resistance was identified among H3 subtypes in which from nearly ¾ of isolates from China and more than 90% of isolates in the U.S. were resistant (Barr et al 2007). Resistance of previously sensitive strains among patients can occur within days (Drinka and Haupt 2007) but rarely remains fixed in the virus population. In contrast amantadine remained effective against most H1 circulating subtypes until the 2006 influenza season. Evidence for increased levels of resistance (approximately 20% of isolates) has been recognized in countries throughout the world (Barr et al 2007). With this high level of adamantane-resistant H3 viruses persisting, the use of neuraminidase inhibitors is currently recommended for the treatment or prevention of influenza (Barr et al 2007). The intent is that by relieving the selective pressure induced by amantadines, it may be possible to favor sensitive strains so that in a few years amantadines will be useful again. However, it is not clear whether overuse of amantadines in the human population was the sole driving force behind the high prevalence of drug resistant virus strains. Genetic evidence suggests the amantadine resistance gene may be present in current H3N2 virus strains because it is a part of a very successful gene combination that allows for efficient virus spread unrelated to sensitivity.

Recent work has also implicated a heritable component to susceptibility to influenza virus infection. Data primarily from the 1918 influenza pandemic suggests certain families had a much higher incidence of fatalities from influenza infection than others. The factors mitigating this increased sensitivity have not been identified but identifying the genes involved may shed light onto the pathogenic mechanisms involved in severe influenza virus disease.

Adenovirus 14

Adenovirus 14 (Ad14) was recognized more than 50 years ago (Rowe et al 1953). It has been a sporadic cause of acute respiratory disease (ARDs) but an emerging serotype sometimes causes severe and occasionally fatal respiratory illness in patients of all ages, including healthy young adults (MMWR 2007). Although much of the current outbreak was initially detected in 2006, there are preliminary reports (Lewis et al 2007) of earlier (including fatal) cases in Oregon in 2005. An earlier appearance also is suggested by Metzgar and colleagues (2007) who monitored U.S. military recruits prospectively from 2002–2006 at training facilities. Prior to 2006 Ad 4 was the most common form of adenovirus causing disease. Within a year, Ad14 was geographically widespread and by the early spring of 2006, Ad 14 as well as two other adenoviruses (Ad21, and Ad7), were common among ill recruits at multiple localities within the U.S. This suggested that these emergences were driven by importation from the US civilian population from which the affected trainees were recruited.

Regardless, in late spring 2006 a 12 day old child died following a week-long illness. Ad14 was identified following isolation, PCR typing and serological assays. Sequence analyses demonstrated consistency with later outbreaks in the western and southern U.S., although no epidemiological linkage was found. There are several unique mutations associated with recent Ad14 isolates but it is not clear whether these genetic changes have enhanced disease pathogenesis or allowed for increased transmission between humans.

The Ad 14 cases are associated with a broad spectrum of clinical illness, including conjunctivitis, febrile upper respiratory illness, pneumonia, and gastrointestinal disease. Severe illness can occur in newborn or elderly patients or in patients with underlying medical conditions but is generally not life-threatening in otherwise healthy adults. The community outbreak recognized in Oregon best characterized this disease. The series included 31/50 adenovirus infected individuals who had Ad14 infection. Among 30 of these patients, 22 (73%) were male; the median age was 53.4 years (range: 2 weeks–82 years). Five cases (17%) occurred in patients aged less than 5 years, while 20 cases were aged more than 18 years. Twenty-two patients (73%) required hospitalization, sixteen (53%) required intensive care, and seven (23%) died from severe pneumonia. Median age of the patients who died was 63.6 years. One death occurred in an infant aged 1 month. No link was identified in hospitals or the community to explain transmission of Ad14 from one patient to another (MMWR 2007).

The control of adenovirus outbreaks is challenging because virus is shed in both respiratory secretions and feces and can persist for weeks on environmental surfaces. Adenovirus isolation and genetic characterization are not routinely performed which may lead to an underestimate of the number of Ad14 cases. Guidelines for the care of patients with pneumonia should be followed in cases of suspected adenoviral pneumonia. The CDC recommends that clinicians with questions related to testing of patients for adenovirus or Ad14 infection should contact their state health departments, which can provide assistance (MMWR 2007).

Human Polyomavirus

In 2007, a previously unrecognized polyomavirus was identified from respiratory secretions in a 3 year old child in Australia who was diagnosed with pneumonia. This virus was genetically distinct from other members of the genus and subsequent surveys of several thousand patients with acute respiratory tract infections found infection with this now recognized WU virus in approximately 2% of the patients in Australia and the mid-West of the U.S. (Gaynor et al. 2007; Le et al 2007). At nearly the same time, a second novel human polyomavirus, KI, was identified. This virus appears related to WU but distinct with respect to its genetic sequence. Additional surveys in Canada show a similar prevalence of infection (2.5%) in a smaller sample of young children co-infected with RSV and a somewhat higher occurrence (6%) in a small sample of asymptomatic children (Abed et al. 2007). Thus, the data indicate that infection with the viruses is geographically widespread, though at relatively low frequency and the extent of clinical disease currently appears relatively mild. Little is known about the epidemiology or methods of transmission of the viruses.

The most common clinical findings in the patients with WU include cough, upper respiratory tract symptoms, tachypnea and infiltrates or consolidation on radiography. The most frequent diagnoses were pneumonia (although a large portion had positive bacterial cultures – probably due in part to the sampling schemes), bronchiolitis, and upper respiratory tract infections. However, it is not clear as yet, what disease symptoms, if any, are associated with KI polyomavirus infection.

It may be possible, as with other human polyomaviruses, BK and JC viruses, that a persistent infection can develop. At least two patients had virus recovered up to six weeks after initial infection.

SARS-CoV

Severe acute respiratory syndrome (SARS) appeared late in 2002 in southern China. Initial cases were recognized as atypical pneumonia characterized by high fever, shortness of breath, cough, and pneumonia (Li 2004; Feng and Gao 2007). Most early cases were associated with people in the wild animal trade and their contacts (Peiris et al 2004). Serological surveys showed that a number of workers in these facilities had evidence of exposure without becoming ill. A physician from the region who visited Hong Kong was identified as the source of 16 cases which led to the world-wide outbreak. The most heavily involved countries were those with direct air travel links to Hong Kong and other urban areas in southern China. By the time the epidemic was contained in August 2003 in excess of 8400 cases and more than 900 fatalities were identified. Mortality was strongly age-dependent, with children and young adults rarely developing fatal disease, while more than half of the clinical cases over the age of 65 years died. Approximately ¼ to 1/3 of cases were health care workers lacking access to appropriate barrier protection. Transmission was believed primarily by droplet spread, and less frequently by direct contact or fomites. Viral shedding in feces also has been reported (Li et al 2004).

The etiologic agent (SARS-CoV) was identified as a previously unrecognized Coronavirus (Family Coronaviridae), an enveloped single stranded positive-sense RNA virus. Although SARS-like viruses were isolated from various animal species in the markets, it appears that SARS-CoV or closely related viruses circulate naturally in bats of the genus Rhinolophus and spill-over to other animals (Li et al 2005).

By analyzing the replication and genomic sequence of virus isolates collected during the outbreak from various animals and humans, a good understanding of how the virus became adapted to efficiently infect humans has been uncovered. For SARS-CoV, attachment to cells of the respiratory tract appears to be the primary step that needs to be overcome to have a productive infection in humans. When virus isolates collected from animals, humans early in the epidemic and humans late in the epidemic are compared, specific genetic changes in the spike protein of the virus are observed and these are associated with a tighter binding of virus to human respiratory epithelial cells and more efficient virus replication in the respiratory tract. The mutations that promote efficient virus replication in humans also change the antigenic nature of the protein so antibodies generated to human isolates of SARS-CoV block the infection of human SARS-CoV isolates but not that of animal isolates. This observation has important implications for understanding how viruses can adapt to new animal hosts but also has significant implications for the choice of potential SARS-CoV vaccine strains.

Since the original outbreak four additional cases of SARS have occurred. Three appear to be laboratory accidents, while the fourth has no clear link to laboratory exposure, suggesting that exposures to the wild-type virus continues (Peiris et al 2004). The apparent molecular changes that led to adaptation and effective human-to-human transmission do not appear to have been repeated. However, the outbreak has led to further research on Coronaviruses, more generally, and subsequent studies show additional, previously unrecognized viruses are linked to community acquired pneumonia and upper respiratory disease in the same region (Lau et al 2006).

Clinically, cases present following a 2–10 day incubation with fever (> 38°C). Frequently there is malaise, nonproductive cough, dyspnea, chills, rigors, and headache. Rhinorrhea and a sore throat are rare. Radiological signs after the onset of fever show consolidation that increases progressively in size, predominantly in the lower lung fields but pleural effusions are absent. Biopsy shows interstitial inflammation and Oxygen saturation is decreased in about half of patients. Laboratory tests show leucopenia, lymphocytopenia and thrombocytopenia (Li et al 2004). Diagnosis can be done by viral isolation and characterization, RT-PCR or serology (ELISA or IFA). However, the duration of detectable viremia or viral shedding is unknown. Currently, paired sera and a seroconversion or a rising titer is considered confirmation of suspected cases but is not used clinically. Diagnosis is based on clinical findings and exclusion of other causes of pneumonia (Li et al 2004; Peiris et al 2004).

Chikungunya virus

Chikungunya virus (CHIKV) was first isolated in 1952 from humans, and Aedes aegypti and Culex pipiens mosquitoes during an epidemic of a non-fatal dengue-like illness on and near the Makonde plateau in south-eastern Tanzania (McCrae et al 1971). This Alphavirus (Family Togaviridae), is considered to be transmitted principally by various members of the mosquito genus Aedes (Diallo et al 1999) in a cycle between mosquitoes and various wildlife species. Sporadic spillover to humans has been characterized by mostly small, localized outbreaks.

Until recently it has been primarily restricted to countries surrounding or near the Indian Ocean, (McCrae et al 1971). Beginning in 2004 in East Africa, and continuing across many islands in the Indian Ocean and the Indian subcontinent in 2005–2006 an epidemic thought to affect more than 1.5 million people has occurred (Sourisseau et al 2007). Of special significance are hundreds of cases in holiday and business travelers from Europe and North America who returned to their homes incubating the virus (Lanciotti et al 2007; Taubitz et al 2007). Virus may have become established in southern Europe following the arrival of an incubating human visitor and was transmitted in 2007 by a previously introduced Aedes (Ae. albopictus) mosquito species in the region, potentially creating a new endemic focus (Rezza et al 2007).

Clinical signs and symptoms of CHIKV in adults and children are preceded by an incubation period of 4–7 days. It usually presents with sudden onset of high fever, fatigue and disabling joint and muscle pain and may be associated with a maculopapular rash and gastrointestinal complaints (Taubitz et al. 2007). A generally mild hemorrhagic syndrome also has been described, consisting of petechial purpura, epistaxis, or bleeding gums. The case fatality rate is low (< 5/1000 cases). The illness usually resolves within a week but chronic disease is reported in a large minority of patients. In these individuals severe arthralgias may persist for up to six months, though most resolve within two months (Taubitz et al 2007). Other complications reported in adults and children are neurologic (meningoencephalitis, polyneuropathy), hemorrhagic, and cardiac (pericarditis, myocarditis, cardiac arrhythmias) involvement. Some studies indicate a very high prevalence of respiratory virus infections in patients diagnosed with CHIKV. The size and careful documentation of the recent outbreak has made it possible to characterize a number of rare events associated with infection including mother-child transmission (Ramful et al 2007) as well as ocular involvement (Lalitha et al. 2007), which typically resolved.

Laboratory diagnosis of CHIKV infection involves serologic methods, virus isolation, or RT-PCR. Serological testing involves testing acute- and convalescent-phase serum specimens for IgM and IgG antibody, followed by a plaque reduction neutralization test (PRNT) to rule out other alphaviruses. Virus isolation and RT-PCR can be used with early acute-phase specimens (before day 5 post-onset) because duration of viremia is typically short (Lanciotti et al. 2007).

Analysis of recent virus isolates indicates that a single genetic change may be responsible for the large increase in cases observed recently. One mutation in the E or envelope protein of the virus, drastically increased the efficiency by which the virus could infect various mosquito species. Mosquitoes infected with these viruses were more efficient at transmitting virus to mammalian hosts, indicating that the virus had increased its vector competence through this single genetic change. The introduction and spread of both CHIKV and competent mosquito vectors in their traditional range, as well as to other parts of the globe, coupled with recreational and business travel of large numbers of people favor the epidemic spread of these agents, when environmental conditions are suitable for the initial emergence of the virus. Thus, health care providers must be aware of the potential role of travel activities to areas such as Africa, India and southern Europe, where this infectious agent is active when examining patients.

West Nile Virus

West Nile virus (WNV;, Flaviviridae), is a mosquito-borne, positive sense RNA virus that is transmitted between vertebrate animals and mosquitoes. Various bird species, as well as mosquito species differ in their ability to maintain and transmit the virus (Kramer et al 2007). This can produce markedly differing levels of risk depending on geographic locale, time of year and landscape patterns (Reisen and Brault 2007). Patterns of transmission may differ strikingly between urban and rural landscapes in the same region because of different susceptibilities of the mosquitoes and birds that occupy these habitats.

WNV was first isolated in the late 1930’s in north-western Uganda from a fever patient (Smithburn et al 1940). The virus appeared restricted to the Old World in Africa and the Middle East until its introduction into North America – first being detected in New York in 1999. Since then, the virus has spread rapidly throughout most of North America and into the Caribbean, Central and South America (Bertolotti et al 2007). However, there is little evidence of disease either in humans or domestic animals in these latter regions. The basis for this is unclear (Kramer et al 2007). In North America, large numbers of cases continue to be reported. Regions of the Upper mid-West and the Southeast in the U.S. as well as western provinces in Canada have reported thousands of human cases during the past two years (Reisen and Brault 2007). The size of the outbreaks appears to reflect, in part, the pattern of weather that either favor or inhibit various mosquito vector populations (Shone et al 2006; Kramer et al 2007).

The WNV strains that predominate in North America contain one particular genetic change that appears to increase the ability of the virus to replicate and cause disease in birds, in particular the North American crow. Virus replication in crows resulted in a 10–100 fold increase in the amount of infectious virus than that found in crows infected with viruses that did not contain the mutation. Increased viremia could increase the number of mosquitoes acquiring the virus during feeding, setting up a positive feedback system leading to virus spread.

Transmission to humans is predominantly through mosquito bite. Cases of transmission associated with blood transfusion, transplantation, breast milk and transplacental exposure also have been reported (Kramer et al 2007). Transmission among other animals has been reported by direct, physical contact between infectious and susceptible individuals (Austgen et al 2004).

Most people infected with WNV are asymptomatic. Symptoms may develop in 20–40% of people with WNV infection. The incubation period is 2–14 days before the onset of symptoms characterized by fever, headache, malaise, myalgia, fatigue, skin rash, lymphadenopathy, vomiting, and diarrhea. Most patients that are symptomatic present with flu-like symptoms (West Nile fever). However a small fraction of cases (< 1%) develop severe neuroinvasive diseases that can be classified into three clinical syndromes: West Nile meningitis, West Nile encephalitis, and acute flaccid paralysis. Whether these syndromes are different aspects of a continuous clinical spectrum or distinct entities is unknown. A recent study reported that most patients with neuroinvasive disorders can be classified as either having West Nile meningitis or West Nile encephalitis. Patients with the latter have a higher mortality rate and more severe complications (Bode et al 2006; Kramer et al 2007). In fatal cases viral antigens can be found inside neurons and neuronal processes in the brainstem and anterior horns. In general, the antigens are focal and sparse; but in severely immunosuppressed patients, viral antigens were seen extensively throughout the CNS (Guaner et al 2004). There may be a human genetic factor that plays a role in increased WNV disease. Several studies have shown that individuals possessing a gene called CCR5delta 32 have an increased risk of developing severe WNV disease. Interestingly, the presence of the CCR5delta32 gene is associated with a decreased risk of acquiring HIV.

Conclusions

As in the past, patterns of infectious disease transmission continue to change. As these examples demonstrate, human activities (travel, agriculture, land use changes, introduction of exotic species) do play a role in the changing levels of risk. In some situations, the relative lack of severe clinical disease and non-specific presentations make identifying the impacts these agents have on human health difficult to assess. However, in each of the cases where pathogens that cause significant morbidity and mortality emerge genetic changes can be identified that made transmission to humans easier or pathology more significant. Thus, to a great degree the continued emergence of new, highly pathological agents is beyond the scope of our ability to specifically predict or to prevent their occurrence. Rather, it is the careful monitoring of human populations and the recognition of unusual patterns of disease that will provide us with that important initial clue of the appearance of the next EID. At as basic science level, a better understanding of the basic requirements that lead to increased virus transmission between species and better viral replication in new species is needed to accurately assess the threat of a newly identified virus to the human population.