Polioviruses and other Enteroviruses

Enteroviruses and new genera within the family Picornaviridae

1.0 Epidemiology of the Disease and Pathogen

Human EVs, like all viruses, are obligate intracellular parasites that are dependent upon living host cells for the essential elements of replication (Lees, 2000). EVs including poliovirus (PV) and non-polio enteroviruses (NPEVs) (i.e., coxsackieviruses, echoviruses, EVs) are among the most common genera of the Picornaviridae family that infect humans and therefore are known to circulate widely in human populations throughout the world (Pallansch et al., 2013). Picornaviridae is considered one of the oldest and most diversified virus families, presenting a non-enveloped, single, positive-stranded RNA (ssRNA) genome packed in a ~30 nm icosahedral capsid (Racaniello, 2013). This family includes many important enteric human and animal pathogens, such as poliovirus, hepatitis A virus, and foot-and-mouth-disease virus (Pallansch et al., 2013; Racaniello, 2013; Tapparel et al., 2013). Most EVs and other genera of enteric picornaviruses (Hepatovirus, Parechovirus, Cardiovirus, Kobuvirus, Salivirus, and Cosavirus) replicate in the epithelial cells of the gastrointestinal tract and their progeny are in fecal matter excreted into the environment.

1.1 Global Burden of Disease

1.1.1 Global distribution

Although enterovirus infections are a significant cause of morbidity and mortality throughout the world, in particular for infants and young children, reliable worldwide estimates of EV-related mortality are not currently available. Since enterovirus infections are not nationally notifiable and health authorities in different countries admit that these infections are never treated or tested, the number of enterovirus detections is likely a considerable underestimate of the true burden of disease. In the developed world, enterovirus infections result in hundreds of thousands of hospitalizations per year, with aseptic meningitis accounting for the vast majority of these cases. Enteroviral infections are more common in most developing countries, so it is reasonable to assume that significant morbidity can be attributed to these viruses globally (Pallansch et al., 2013). The disease burden in the developing world of the tropics is poorly estimated, with the exception of poliomyelitis. Indeed, the focus for the worldwide EV surveillance network is primarily on neurological presentation such as acute flaccid paralysis (AFP), to support poliovirus eradication programs. In general, national laboratories throughout the world report enterovirus typing results from clinical specimens to exclude poliovirus and establish the epidemiology of NPEVs. The World Health Organization (WHO) and the U.S Centers for Disease Control and Prevention (CDC) have guidelines for epidemiologic, clinical, and laboratory investigations of AFP to rule out poliovirus infection (WHO, 2004).

In the United States of America, the National Enterovirus Surveillance System (NESS) is a voluntary, passive surveillance system that has monitored trends in circulating EVs since 1961. Participating laboratories (state public health laboratories and local diagnostic laboratories willing to participate) report to CDC monthly summaries of enterovirus detections by virus type (i.e., serotype), with specimen type, collection date and demographic information. CDC summarizes the data and disseminates the results as shown in Table 1. Serotype-based EV surveillance provides a mechanism for determining patterns of enterovirus circulation and for identifying predominant serotypes for a given season. NPEVs are quite common, causing an estimated 10 – 15 million or more symptomatic infections per year in the US alone (Pallansch et al., 2013).

The WHO maintains an integrated global alert and response system for epidemics and other public health emergencies based on strong national public health systems and capacity and an effective international system for coordinated response (www.who.int). Although enterovirus-associated fatalities involving poliomyelitis have been more consistently investigated, WHO continues to track the evolving infectious disease situation for other enterovirus diseases including acute flaccid paralysis or “poliomyelitis-like” infections associated with NPEVs. Among these newer enterovirus infections are those associated with enterovirus-71 (EV-71) and enterovirus D68 (EV-D68). EV-71 infection induces mild symptoms (e.g., hand-foot-mouth disease [HFMD] herpangina, or fever) in young children that can also lead to severe neurological and systemic complications, including meningitis, encephalitis, and central nervous system (CNS) involvement with cardiopulmonary failure. EV-71 is emerging as the most important virulent neurotropic EV during the poliomyelitis eradication period and countries of the Asian Pacific Rim in particular have been affected by large outbreaks of EV-71-associated HFMD with severe neurologic complications or systemic disease as described in Tables 2 and 3 (Cardosa et al., 2003; Huang et al., 2014; Pallansch et al., 2013; Ryu et al., 2010). The transmission of EV-71 infections in other geographical areas is poorly understood.

EV-D68 infection has been associated with acute respiratory illness of differing severities ranging from mild upper respiratory tract infections to severe pneumonia including fatal cases in pediatric and adult patients (Imamura and Oshitani, 2015; Oberste et al., 2004). EV-D68 has been recognized as a reemerging pathogen due to its high frequency of detection in different parts of the world since 2008 – 2009 (Table 4). In the United States, the National Enterovirus Surveillance System received 79 EV-D68 reports during 2009 – 2013 (Midgley et al., 2014). Cases of respiratory syndrome with neurological complications, including acute flaccid myelitis, have been reported in the USA, Canada, Norway, and France (Khan, 2015; Lang et al., 2014; Milhano et al., 2015). However recent investigations indicate that EV-D68 infection is most commonly associated with respiratory illness and no definitive connection between EV-D68 and acute flaccid myelitis has been established (http://www.cdc.gov/ncird/investigation/viral/2014-15/investigation.html).

In temperate regions of the world, the majority of EV infections (70 to 80%) occur in summer and fall, whereas in the tropics, EV infections occur year-round or with increased incidence during the rainy season. Unexplained cycles of infection (i.e., 2, 3, 4 or 5 years) of some human EVs have been identified (Pallansch et al., 2013; Stellrecht et al., 2010).

1.1.2 Symptomology

Table 5 shows Picornavirus genera and enterovirus species associated with disease in humans. While most enterovirus infections are asymptomatic and mild febrile illness (e.g., respiratory illness and febrile rash illness) has been recognized as the most common symptomatic manifestation of infection, EVs are also associated with severe neurological illness, such as aseptic meningitis and encepha­litis. Aseptic meningitis seems to be the most common severe clinical manifestation associated with enterovirus infection in children and adults, especially in developed countries (Pallansch et al., 2013; Stellrecht et al., 2010; Tapparel et al., 2013). PVs are perhaps the most studied and characterized of the EVs, especially given the clinical consequences of infection which can infect motor neurons of the anterior horn of the spinal cord and lead to paralytic poliomyelitis in humans (Rhoades et al., 2011). Infants infected with Coxsackievirus (CV) have been shown to be extremely susceptible to myocarditis, meningitis and encephalitis with a subsequent mortality rate as high as 10% (Rhoades et al., 2011). EV-71 is a major public health issue across Asia and of increasing concern globally, causing HFMD with potential neurological complications as previously discussed (Rhoades et al., 2011). Echoviruses are highly infectious and preferentially target infants and young children. These viruses can cause a mild, nonspecific illness similar to that of CVs. Other clinical syndromes associated with NPEV infections include acute hemorrhagic conjunctivitis caused by multiple EV serotypes such as coxsackievirus type A24 variant (CVA24), enterovirus type 70 (EV-70), and E-71; coxsackieviral myocarditis (i.e., inflammation of the myocardium) caused by coxsackievirus group B (CVB); enteroviral respiratory infections caused by Coxsackievirus group A (CVA) and CVB serotypes as well as echovirus types 6, 9, 11, 16, 17, 22, and 25; acute gastroenteritis caused by echovirus types 1, 7, 11, 13, 14, 30 and 33. Consequently, no enteroviral disease is uniquely associated with any specific serotype, and no serotype is uniquely associated with any one disease (Pallansch et al., 2013; Tapparel et al., 2013).

1.2 Taxonomic Classification

1.2.1 Physical description of the agent

EVs, like other members of the family Picornaviridae, are non-enveloped icosahedral particles (i.e., virions) ~ 30 nm in diameter. Virions consist of a protein shell (capsid) surrounding the naked single-stranded positive (message)-sense RNA genome (Racaniello, 2013; Stellrecht et al., 2010). EVs are distinguished from other picornaviruses on the basis of physical properties, including buoyant density in cesium chloride and acid stability (Pallansch et al., 2013). EVs and cardioviruses have a buoyant density of 1.34 g/mL, while parechoviruses (HPeVs) and rhinoviruses (HRVs) have an intermediate value of 1.36 g/mL and 1.40 g/mL, respectively and such differences lie in the permeability of the viral capsid to cesium (Racaniello, 2013). EVs (except HRVs), hepatoviruses, and HPeVs are acid stable and retain infectivity at pH values of 3.0 and lower. In contrast, HRVs are labile at pH values of less than 6.0. Differences in pH stability influence the sites of replication of the virus, for instance, HRVs replicate in the respiratory tract and need not be acid stable while cardioviruses, EVs, hepatoviruses, and HPeVs pass through the stomach to gain access to the intestine and, therefore, must be resistant to low pH (Racaniello, 2013).

EVs are relatively resistant to many common laboratory disinfectants, including 70% ethanol, isopropanol, dilute Lysol, and quaternary ammonium compounds. The virus is insensitive to lipid solvents, including ether and chloroform, and it is stable in many nonionic detergents at ambient temperature. Formaldehyde, glutaraldehyde, strong acid, sodium hypochlorite, and free residual chlorine inactivate EVs (Pallansch et al., 2013; Stellrecht et al., 2010)). Chlorine disinfection is a critical step for inactivation of viruses in water and wastewater treatment processes Ideal conditions for virus inactivation are high chlorine residual, long contact time, high water temperature, low pH, low turbidity, and an absence of interfering substances (DEP, 2011)(see Section III).

In the environment, EV virions are relatively thermostable, but less so than hepatitis A virus that is distinguished from other picornaviruses by its high thermal stability (Anderson and Counihan, 2011; Pallansch et al., 2013). EVs, like other picornaviruses, have evolved thermal stability (stable at 42°C or up to 50°C in the presence of sulfhydryl reducing agents and magnesium cations) and pH stability (pH 3-9) in the extracellular environment, which is the result of the assembly of virus particles of all picornaviruses (Gerba et al., 2007; Hurst and Adcock, 2000; Hurst and Murphy, 1996; National Research Council, 2004; Racaniello, 2013). Indeed, the biophysical properties of picornaviruses such as their small-size (30 nm), genome type (ssRNA), and non-enveloped protein capsid structure of the virion are known to play an important role on the mechanisms of virus survival and transport in the environment (Bosch et al., 2006; Fong and Lipp, 2005; Gerba et al., 2013; National Research Council, 2004; Sobsey and Meschke, 2003; Wigginton and Kohn, 2012; Xagoraraki et al., 2014). In particular, the electrostatic and hydrophobic interactions involving functional groups of the viral capsid (i.e., charged amino acids residues) are essential for capsid assembly, capsid stability, and virus particle adsorption to solid surfaces (e.g., suspended solids, sediments and environmental surfaces or fomites) (Mateu, 2011; Michen and Graule, 2010). Studies have suggested that picornaviral capsids are also stabilized by interactions with the RNA genome (Pallansch et al., 2013). Additional studies conducted with PV strains (serotype 1 virulent Mahoney and serotype 1 attenuated Sabin) have indicated that viral binding to bacteria or bacterial polysaccharides offers a selective advantage by enhancing environmental stability of the virions (Robinson et al., 2014).

The capsids of EVs and other picornaviruses are composed of four viral structural proteins (i.e., VP1, VP2, VP3, and VP4) arranged in 60 repeating protomeric units (protomers) that confer an icosahedral shape to the virion. The shell is formed by VP1 to VP3, and VP4 lies on its inner surface. VP1, VP2, and VP3 comprise the surface of the virion with a similar topology forming an eight-stranded, antiparallel β-barrel core structure. Packing of the β-barrel domains is strengthened by a network of protein-protein contacts on the interior of the capsid, particularly around the fivefold axes of symmetry. This network, which is formed by the N-terminal extensions of VP1 to VP3 and VP4, is essential for the stability of the virion (Racaniello, 2013). Similarly, a network formed by the N-termini on the interior of the capsid contributes significantly to the stability of the virions. Furthermore, studies have suggested that picornaviral capsids are stabilized by interactions with the ssRNA genome. The positive strand ssRNA genome of EVs and all picornaviruses serves as a template for both viral protein translation and RNA replication. In fact, the viral ssRNA is infectious because it is translated on entry into the cell to produce all the viral proteins required for viral replication (Pallansch et al., 2013; Stellrecht et al., 2010).

1.2.2 Taxonomy and diversity in the Picornaviridae

Enterovirus corresponds to the genus within the Picornaviridae family. This family belongs to the order Picornavirales and comprises 12 genera, which all contain viruses that infect vertebrates (i.e., Apthovirus, Avihepatovirus, Cardiovirus, Enterovirus, Erbovirus, Hepatovirus, Kobuvirus, Parechovirus, Sapelovirus, Senecavirus, Tremovirus, Teschovirus) (Racaniello, 2013). Picornavirus genera and enterovirus serotypes infecting humans are presented in Table 5.

Members of the Echovirus genus originally known as echovirus 22 and 23 have been reclassified within the genus Parechovirus based on distinctive biological and molecular properties. To date, 16 genotypes of human parechoviruses (HPeVs) have been identified (Harvala and Simmonds, 2009). Likewise, the former Rhinovirus genus has been reclassified within the Enterovirus genus based on its genome organization and particle structure (Tapparel et al., 2013).

With respect to EVs, the current taxonomic classification scheme (Virus Taxonomy: 2014 Release (International Committee on Taxonomy of Viruses) http://www.ictvonline.org/taxonomyReleases.asp last accessed Feb. 2016) based on genome organization and biological properties of viruses plus high-throughput EV genome sequencing and bioinformatics analysis, divides human EVs into four species: Enterovirus A, B, C, and D. The classification scheme keeps traditional names for individual serotypes (i.e., coxsackievirus, echovirus, enterovirus, poliovirus). The following species designation criteria based on molecular and biological features have been established: HEV species share amino acid percent identities from 60% to greater than 70% within specific genomic coding regions, share a limited range of host cell receptors and a limited natural host range; have a G + C genome base composition which varies by no more than 2.5%; and share a significant degree of compatibility in proteolytic processing, replication, encapsidation, and genetic recombination.

Approximately 107 serotypes/genotypes of NPEVs have been identified based on a threshold of 25% nucleotide divergence in the VP1 coding region (Pallansch et al., 2013). The serotype of an unknown isolate is inferred by comparison of the VP1 sequence with a database containing VP1 sequences for the prototype and variant strains of all human EV serotypes according to the following suggested guidelines:

1. A partial or complete VP1 nucleotide sequence identity of 75% (>85% amino acid) identity between a clinical EV isolate and serotype prototype strain is used to establish the serotype of the isolate, on the provision that the second highest score is <70%
2. A best-match nucleotide sequence identity of <70% may indicate that the isolate represents an unknown serotype,
3. A sequence identity between 70% and 75% indicates that further characterization is required before the isolate can be firmly identified.

VP1 nucleotide sequence comparison can be used to rapidly determine whether viruses isolated during an outbreak are epidemiologically related (Manor et al., 2007; Pallansch et al., 2013; Shulman et al., 2012; Shulman et al., 2015).

New genera of the Picornaviridae family (i.e., Cardiovirus also referred to as Saffold cardiovirus or SAFV, Cosavirus , a common stool-associated picornavirus, and Salivirus, a stool Aichi-like virus) have been identified in fecal specimens and in sewage using conventional and highly-sensitive genome sequencing technologies (Greninger et al., 2009; Jones et al., 2007; Kapoor et al., 2008; Tapparel et al., 2013). Although these viruses have been associated with disease in humans, they do not appear to be a major cause of human illness; therefore, their clinical significance remains to be fully elucidated.

It is known that EV activity in populations can be either sporadic or epidemic, and that certain EV types may be associated with both sporadic and epidemic disease occurrences. The patterns of virus shedding and routes of transmission for EVs are consistent with few exceptions. The largest amount and duration of virus shedding occurs upon primary infection with a given serotype and most primary infections occur during childhood. Consequently, young children are possibly the most important transmitters of EV within households (Pallansch et al., 2013).

1.3 Transmission

1.3.1 Routes of transmission

Transmission of human EVs to susceptible individuals occurs via fecal – oral, oral – oral and nasopharyngeal routes (respiratory droplets), and via fomites. Most EVs enter primarily by the fecal – oral route and replicate in the gastrointestinal tract, including the oropharyngeal and intestinal mucosa. Following replication, EVs may cause both localized and systemic infections, affecting many organ systems in patients of all ages. Some EVs have a restricted respiratory tropism (e.g. HRVs) while others can invade the CNS and induce neurological disease (Pallansch et al., 2013; Rhoades et al., 2011; Tapparel et al., 2013). It is well documented that the fecal – oral pathway of EV transmission predominates in areas with poor sanitary conditions, while respiratory transmission predominates in more developed areas. Consequently, the relative importance of the different modes of transmission probably varies with the particular EV and environmental settings. HRVs are traditionally associated with upper respiratory tract infection, otitis media, and sinusitis, and therefore are predominantly transmitted from person to person via direct contact or through a fomite or small or large particle aerosol (Jacobs et al., 2013).

Evidence of more specific modes of enterovirus transmission by the fecal – oral route via environmental contamination (e.g., contaminated food [shellfish] and water sources) are described in detail in specific sections of this chapter. Since EVs are enteric viruses, water-related transmission remains of primary importance for indirect human enterovirus transmission.

1.3.2 Reservoirs and fecal shedding

Contaminated water and some food sources may serve as reservoirs of the environmentally-stable EV virions, but humans are thought to make up the only important natural reservoir for replication of human EVs (Gerba et al., 2013; Pallansch et al., 2013). Most EVs, excluding Enterovirus 70 (EV70) coxsackievirus A24 (CVA24) are abundant in stool specimens of infected individuals and levels of excretion of PVs are known to reach maximal amounts of 10⁶ infectious virons per gram of feces (Dowdle et al., 2002; Hovi et al., 2007). Less information is available for levels of excretion of other enterovirus genotypes; however, the consensus from previous and current references indicates that humans excrete between 10⁶ and 10⁷ EVs per gram of feces (Feachem et al., 1983b; Gerba, 2000; Melnick and Rennick, 1980). The duration of intestinal virus shedding and thus the spreading (i.e., transmission) of a particular enterovirus may vary among the different genotypes. For example, longer periods of excretion have been documented for PVs and CVs (Pallansch et al., 2013). Additional studies have indicated that EV excretion may persist for up to 11 weeks after an EV infection (Chung et al., 2001). Furthermore, long-term PV excretion ranging from 10 months to 28 years has been documented for individuals with immunodeficiencies (Dunn et al., 2015; Kew et al., 1995).

1.4 Population and Individual Control Measures

1.4.1 Poliomyelitis vaccination and eradication update

Following the eradication of smallpox, paralytic poliomyelitis (i.e., AFP) caused by wild polioviruses (WPV, serotypes 1, 2 and 3) became a major focus of public health activity. Most childhood EV and PV infections are asymptomatic. Surprisingly, on average only about 1 in 200 to 1000 poliovirus infections in fully susceptible individuals results in paralytic disease and drops to 1 in a few millions for vaccinated individuals. The most common symptomatic disease caused by PVs is a mild febrile illness with or without gastrointestinal signs (i.e., abortive poliomyelitis) that occurs in 4% to 8% of individuals. Less frequently infection may lead to non-paralytic illness characterized by the typical features of viral meningitis. (Pallansch et al., 2013). Person-to-person WPV transmission can occur in the absence of the relatively rare symptomatic infections especially in highly vaccinated populations (Shulman et al., 2014).

In 1988 the World Health Assembly adopted the Global Polio Eradication Initiative (GPEI) to eradicate all three serotypes of poliovirus. This effort has involved both health professionals and more than ten million volunteers in all countries (WHO, 2014). Success of any eradication program depends on an effective intervention strategy and sensitive and easily performed surveillance systems able to detect the pathogen. For polio, intervention was based on the availability of an oral polio vaccine (OPV) consisting of live attenuated poliovirus strains and an inactivated polio vaccine (IPV) containing inactivated neurovirulent vaccine strains. Each vaccine type has advantage and disadvantages (see reviews by (Plotkin and Vidor, 2008; Shulman, 2012; Sutter et al., 2008). Both OPV and IPV were trivalent to raise antibodies against the three non-cross-reacting poliovirus serotypes. The gold standard for PV surveillance is based on investigation of the cause of all cases of AFP in children ≤15 years of old to determine whether or not the AFP was caused by poliovirus (Pallansch et al., 2013). Poliovirus is considered to be absent when the number of AFP cases investigated is 1 per 100,000 children ≤15 of age (the rate for all non-PV causes of AFP) and none were due to PV. A supplementary or alternative approach is to conduct environmental surveillance (ENVS) of sewage and other wastewaters, to test whether PVs are present in samples collected from these sources. The advantage of ENVS is that it can detect PV circulation in the absence of clinical cases since virus is excreted into sewage during asymptomatic infections as well as symptomatic infections. Moreover, results can be semi quantitative as well as qualitative (Hovi et al., 2012; Hovi et al., 2001; Lodder et al., 2012; Manor et al., 2014).

A review of publications between 1935 and 1995 on excretion of polioviruses by Alexander and associates (Alexander et al., 1997) indicated that in most infections of naıve children, wild polioviruses were excreted for 3 – 4 weeks with a mean rate of 45% at 28 days, and 25% of the cases were still excreting during the sixth week. In contrast, fewer than 20% excreted vaccine strains after 5 weeks. Excretion of polio ranged from a few days to several months (Minor, 1992). The highest probability of detecting poliovirus positive stool samples was reported to be at 14 days after the onset of paralysis (Alexander et al., 1997; Dowdle and Birmingham, 1997) and is the basis of stool sample collection for diagnosis of polio by AFP surveillance (WHO, 2004). The Global Polio Laboratory Network of laboratories accredited by the WHO provides the surveillance arm necessary for successful eradication. Standard laboratory procedures were adopted for AFP) surveillance (WHO, 2004).

The GPEI has made significant advances towards eradication of poliomyelitis. To date, nearly 80% of the world's population lives in areas certified as polio-free. The last reported infection with WPV2 was reported in 1999 (MMWR, 2001) and type two WPV was declared eradicated by the WHO on September 20, 2015 http://www.polioeradication.org/mediaroom/newsstories/Global-eradication-of-wild-poliovirus-type-2-declared/tabid/526/news/1289/Default.aspx. This declaration prepared the way for the global withdrawal of type 2 PV from trivalent OPV by April 2016, and the subsequent destruction of all materials that contain type 2 OPV by July 2016 (WHO, 2015b) in all but a few specially designated laboratories. No circulating WPV3 has been reported since November 2012 (Kew et al., 2014). Countries where transmission of wild poliovirus has not yet been interrupted for greater than three years are Afghanistan and Pakistan. Significantly, no polio cases have been reported for more than one year in Nigeria. The international spread of WPV1 was declared a public health emergency of international concern (PHEIC) in May 2014 during the 4^th IHR Emergency Committee meeting regarding the international spread of wild poliovirus http://www.who.int/mediacentre/news/statements/2015/polio-27-february-2015/en/ since continued circulation of WPV1 in endemic countries risked reintroduction of PV1 into polio-free areas of the world, for example by travelers, refugees from Africa and the Middle East.

Evidence of global interruption of transmission of WPV has been based primarily on AFP surveillance. ENVS and AFP surveillance will be critically important during the end stage of eradication and post eradication for documenting eradication and ensuring very early detection of any reemergence. ENVS will be especially important for populations with high IPV vaccine coverage, since the ratio of AFP cases to asymptomatic infections would be orders of magnitude lower than for under-vaccinated populations (Hovi et al., 2012). (The latest information on global eradication can be viewed at http://www.polioeradication.org/). Improving methodologies and expanded use of environmental surveillance have become one of the goals of the WHO, the Rotary International and the Bill & Melinda Gates Foundation.

Live attenuated polio vaccine can be transmitted from person-to-person, a trait considered advantageous when WPV infections were endemic, but which is now problematic as the eradication endgame approaches (Dowdle et al., 2003). Specifically, like wild poliovirus, the OPV genome evolves during circulation. Some of the OPV progeny may revert to neurovirulence. AFP from reverted vaccine strains (e.g., vaccine associated paralytic poliomyelitis (VAPP)) occurs very rarely (1 per 500,000 to 1,000,000 primary vaccinations of naive individuals) and in rarely, these neurovirulent vaccine-derived strains (VDPVs) have circulated and caused outbreaks of poliomyelitis in poorly vaccinated populations (Kew et al., 2005). As a consequence there is a major effort to develop new, improved, and safer live polio vaccine stocks promoted by the WHO, Rotary and the Bill and Melinda Gates association. Three different approaches have been funded to reduce the risk for vaccine-derived poliovirus causing outbreaks of poliomyelitis and are in advanced stages of completion prior to clinical testing. These approaches include genetically engineering PV to de-optimize codon usage in the capsid region; increasing poliovirus polymerase fidelity and reducing recombination capacity; and relocating the cis-acting replication element (cre) to attenuate or minimize recombination. Alternative approaches include use of noninfectious DNA, immunogenic peptides and antiviral drugs. All will need more testing against PV.

Development of a general anti-EV vaccine appears to be very difficult due to significant EV genetic variability. However efforts have been initiated to develop a vaccine against specific EVs most notably EV-71 (Liang and Wang, 2014; Lu, 2014; Zhu et al., 2014).

1.4.2 Hygiene measures

Wastewater treatment (WWT) includes biological treatment (e.g., conventional activated sludge, waste stabilization ponds, trickling filters), physicochemical treatment (e.g., sedimentation, media or membrane filtration), conventional oxidation (e.g., chlorine and ozone), and advanced treatment technologies (e.g., membrane bioreactors, microfiltration, ultrafiltration, constructed wetlands, chemical coagulation, and disinfection with ultraviolet irradiation and ozone. WWT can also include steps that remove other hazardous contaminants of urban wastes, e.g., toxic chemicals, heavy metals, pharmaceuticals. A by-product of domestic wastewater or sewage treatment is sewage sludge, a semi-solid slushy mass, deposit, or sediment produced by sedimentation during water and sewage treatment processes. This sludge contains enteric pathogens and needs to be properly treated for recycling or disposal (discussed below in Section III). Multiple levels of wastewater treatment (preliminary, primary, secondary, and tertiary or advanced) and combination of their corresponding treatment technologies are required to consistently reduce, remove and/or inactivate the vast numbers of viruses present in all the wastes sent into the sewer from drains and toilets (Costan-Longares et al., 2008; Francy et al., 2012; Haramoto and Otagiri, 2013; Harwood et al., 2005; Hegazy et al., 2013; Katayama et al., 2008; Kitajima et al., 2014a; Kitajima et al., 2014b; Kitajima et al., 2015; La Rosa et al., 2010; Locas et al., 2010; Ng et al., 2012; Ottoson et al., 2006a; Rock et al., 2015; Simmons et al., 2011; Simmons and Xagoraraki, 2011a; Symonds et al., 2014). The greater the exposure of the public to these contaminated sources, the greater the need for treatment to reduce all viral pathogens and hazardous contaminants present in these wastes (Gerba et al., 2013; Rock et al., 2015; Xagoraraki et al., 2014).

Person-to-person EV transmission can also be interrupted by physical hygienic means. These include directly stripping EVs from the surfaces of foods, hands, and other surfaces by adequate washing, disinfecting materials and surfaces used for food production, heating the food to inactivate virus, and using uncontaminated or treated water. Unfortunately, facilities that hygienically separate human excreta from human contact are unavailable for approximately 2.4 billion people or one third of the current global population (Graham and Polizzotto, 2013; WHO and UNICEF, 2015).

2.0 Environmental Occurrence and Persistence

Numerous methods have been used to evaluate the occurrence and persistence of EVs and other human picornaviruses in environmental waters. 2.1 of this Section details the extensive research that has been accumulated over the past few years. The current state of knowledge about the occurrence and persistence of EVs and other human picornaviruses in different environmental sources is documented in 2.2 and 2.3 of this Section. See Reviews by Faechem et al. (Feachem et al., 1980; Feachem et al., 1983a) for a review of literature before 1983.

2.1 Detection Methods

The selection of an appropriate filtration method for the primary concentration of viruses is crucial for successful virus detection (Rajal et al., 2007). Interestingly, the efficiency of virus recovery by these methods depends largely on the viral target and less on water types or method used (Cashdollar and Wymer, 2013). To date, no single efficient method to recover multiple viral pathogens from environmental waters has been developed. Adsorption of viral particles to charged microporous filters (electronegative and electropositive membranes and cartridges) followed by elution or desorption of the virus by a pH adjusted solution is perhaps the most common approach used in water analysis applications (Fong and Lipp, 2005; Fout et al., 1996; Haramoto et al., 2005; Katayama et al., 2002; Wyn-Jones and Sellwood, 1998; Wyn-Jones, 2007).

Ultrafiltration, i.e. tangential flow filtration (TFF) and hollow-fiber ultrafiltration, based on size exclusion or entrapment of viral particles to a filter matrix, has emerged as the most promising methods for virus concentration from large volumes of water (Gibson and Schwab, 2011; Hill et al., 2005; Olszewski et al., 2005; Polaczyk et al., 2008; Rhodes et al., 2011). Furthermore, these methods have been optimized for the simultaneous concentration of multiple microorganisms (e.g., enteric bacteria, protozoa, and viruses) from diverse water matrixes, including source and finished drinking water, tap water, surface water, and reclaimed water (Gibson and Schwab, 2011; Hill et al., 2007; Hill et al., 2005; Liu et al., 2012; Morales-Morales et al., 2003; Polaczyk et al., 2008). In addition, a new type of filtration media composed of highly electropositive nanoalumina (AlOOH) fibers (~2 nm in diameter by ~250 nm in length) and microglass is currently available for concentrating EVs and other enteric viruses from large volumes of water (Gibbons et al., 2010; Ikner et al., 2011; Karim et al., 2009; Li et al., 2010). NanoCeram cartridge filters combine the benefits of standard pleated filters and nanotechnology (Karim et al., 2009; Li et al., 2010).

Cell culture (i.e., primary cell cultures, cell strains, and continuous cell lines) has been widely used for cultivation and isolation of viruses from environmental samples, especially viruses that are amenable to culture (e.g., EVs, adenoviruses, reoviruses, and rotaviruses) (Fout et al., 1996; Gerba et al., 2013). In fact, virus isolation in cell culture has long served as the “gold standard” for virus detection (Leland and Ginocchio, 2007; WHO, 2004, 2015a). However, EVs behave differently on different types of cultured cells and therefore the use of multiple appropriate cell lines is essential to increase yield and to improve the detection assay (Chonmaitree et al., 1988; Prim et al., 2013). Table 6 describes the susceptibilities of different cell lines commonly used for isolation of EV and HPeV. Cytopathic effect (i.e., morphological changes in individual cells or groups of cells induced by virus infection and easily recognizable under a light microscope) is the simplest and most widely used criterion for enterovirus infectivity; however not all EVs, cause a cytopathic effect, and in these cases other methods must suffice (Condit, 2013; Gerba et al., 2013). Quantitative measures of infectious enterovirus in environmental samples have been performed by the plaque assay or various endpoint methods that measure virus infectivity but not the number of virus particles in a preparation, many of which may be non-infectious (Condit, 2013).

Numerous molecular biology-based detection methods for EVs in environmental concentrates have been developed during the last decades as an alternative to cell-culture method. Molecular biological detection of enteroviral genomic RNA by conventional reverse transcription – polymerase chain reaction (RT-PCR) and by reverse transcription quantitative real-time PCR (qRT-PCR) has been widely used for direct detection and identification of human EVs in environmental water concentrates. Genus-specific primers against the highly conserved region of the 5’ untranslated region (5’UTR) from all enterovirus genomes are broadly used for generic detection of a few copies of enteroviral RNA. Furthermore, the enterovirus 5’ UTR PCR assay, also called pan-EV PCR, allows the detection of enterovirus serotypes that do not readily grow in cell culture standardly used in most laboratories and additional fastidious serotypes that that have been unculturable in all cells tested to date.

Highly specific identification of enterovirus serotypes can be achieved by RT-PCR and direct nucleotide sequencing of all or a portion of the genomic region that encodes capsid proteins especially VP1 (Oberste et al., 1999; Oberste et al., 2003; Kroneman et al., 2011; Nix et al., 2006; Oberste et al., 2006; Pallansch et al., 2013). These sequences can be submitted to an online tool at http://www.rivm.nl/mpf/enterovirus/typingtool/introduction that returns the genotype by comparison with a regularly updated sequence reference set of human Picornaviridae genera and enterovirus species and (sero)types (Kroneman et al., 2011).

Integrated cell culture and PCR (i.e., ICC-PCR) developed in the late 1980s still provides a reliable method for practical analysis and direct monitoring of environmental samples for EVs and other viral pathogens recovered from environmental samples (Amdiouni et al., 2012; Chapron et al., 2000; Choo and Kim, 2006; Lee et al., 2005; Reynolds, 2004; Reynolds et al., 1996). More recently, these techniques have been used in combination with viral metagenomics approaches for cell culture evaluation of the virologically quality of wastewater (Aw et al., 2014).

2.2 Data on the Occurrence in the Environment

Human EVs have been recovered worldwide from surface waters including coastal waters, rivers, streams, and lakes, from wastewaters, ground waters and finished drinking water (Boehm et al., 2003; Borchardt et al., 2007; Borchardt et al., 2012; Chapron et al., 2000; Donaldson et al., 2002; Fong and Lipp, 2005; Fout et al., 2003; Fuhrman et al., 2005; Gantzer et al., 1997; Griffin et al., 2003; Hamza et al., 2009; Hot et al., 2003; Jung et al., 2011; Pusch et al., 2005; Springthorpe and Sattar, 2007; Ye et al., 2012). EVs along with other culturable viruses (adenovirus, rotavirus, and reovirus) have been extensively monitored for regulatory purposes in the US in order to derive baseline data on the occurrence of viral agents in surface and ground waters (Fout et al., 1996). Due to their ubiquitous excretion in feces and frequent detection in untreated domestic wastewaters EVs are the best known viruses and commonly associated with water quality (Grabow, 2007; Rodriguez et al., 2012). The detection of EVs in sources of drinking water and recreational water bodies has been broadly accepted as a marker of a possible failure of the sanitation systems and as an indicator of the potential role of water in disease outbreaks (Hovi et al., 2007). Moreover, since the natural host for all human EVs is humans, the detection and quantification of amplifiable enterovirus genetic material in environmental waters has been proposed and used as a water quality assessment tool for tracking sources of human fecal pollution (Boehm et al., 2003; Fong et al., 2005; Wong et al., 2012).

2.2.1 Occurrence in sewage

The concentration and types of EVs and other picornaviruses in municipal waste waters (i.e., sewage) in different geographic locations, including highly polluted surface waters in areas where wastewater treatment is either inadequate or nonexistent, are shown in Table 7.

Members of the genus Enterovirus are universally distributed among human populations and therefore are always found in sewage waters worldwide (Gerba et al., 2013; Pallansch et al., 2013).

Quantitative estimation of human EVs and other picornaviruses (i.e., viral load) in domestic wastewaters (i.e., primary source of human EVs and enteric picornaviruses to the environment) from different regions may vary due to differences and limitations of the methods used for recovery and detection of viruses in water, differences in the viral load that may exist in human populations for each enteric virus, and daily plus seasonal fluctuations of viruses in human sewage. Nevertheless, the assessment of the total and infectious viral load in untreated and treated domestic wastewaters as well as the identification of virus types are critical to understand the fate of viral pollutants in the environment and to estimate the potential health risks from waterborne viral pathogens from discharges to waters used for recreation, shellfish farming or as sources of drinking water (de Roda Husman and Bartram, 2007; Gerba et al., 2013; Gibson, 2014; Rao et al., 1986; Symonds et al., 2009). Quantitative information of the virus concentration in various types of water, including quantitative data of the inactivation and removal of pathogenic viruses by water and wastewater treatment processes, are essential parameters for the quantitative microbial risk assessment (QMRA) process, the new concept for evaluating microbial safety of drinking water and wastewater (Haas et al., 1993; Haas et al., 1999; Hijnen et al., 2006; Mena, 2007; Soller, 2006).

Screening of waste waters for human viruses in multiple geographical areas has revealed the presence of newly described genera of the Picornaviridae family, i.e., Cosavirus, Cardiovirus, Salivirus, Kobuvirus (Alcala et al., 2010; Blinkova et al., 2009; Haramoto and Otagiri, 2013; Kitajima et al., 2014b; Kitajima et al., 2015). More recent, deep-sequencing analysis of viral metagenomes from sewage and sewage sludge (also known as sewage viromes) has indicated that these novel human picornaviruses can be found in most cases at relatively higher abundance and occurrence than members of the Enterovirus genus (Aw et al., 2014; Bibby and Peccia, 2013; Cantalupo et al., 2011; Ng et al., 2012). Taken together, these studies have demonstrated the importance of targeted molecular assays and unbiased high-throughput sequencing (i.e., metagenomics viral pathogen detection technologies) to the study of virus diversity in waste waters and to monitor shedding of known and newly characterized viral human pathogens on a population level scale.

2.2.2 Occurrence in sludge

Numerous studies conducted during the last two decades have evaluated the occurrence of EVs in biosolids and the efficiency of sludge treatment on the reduction of these and other enteric viruses (Goddard et al., 1981; Goyal et al., 1984; Guzman et al., 2007; Monpoeho et al., 2001; Pourcher et al., 2005; Schwartzbrod and Mathieu, 1986; Sidhu and Toze, 2009; Soares et al., 1994; Straub et al., 1994; USEPA, 1988). Despite the extreme heterogeneity of sludge samples and variations in sampling processing methods, these studies demonstrated the occurrence of infectious EVs and other enteric viruses in treated biosolids.

Soares et al. (Soares et al., 1994) evaluated the occurrence of EVs in undigested and anaerobically digested sludge. For undigested sludge the concentration of EVs varied between 4.36MPN/g (most probable number per gram dry weight) to 7.00x10² MPN/g with geometric average of 3.29x10¹ MPN/g ± 0.0044. For anaerobically digested sludge the concentration varied from <0.0062 MPN/g to 2.52x10² MPN/g with a geometric average of 1.61 MPN/g ± 0.02. Removal of EVs during sludge treatment ranged from 0.5 log₁₀ to >3 log₁₀ with a geometric mean 1 log₁₀. The study revealed that despite sludge treatment the concentration of EVs in anaerobically digested sludge could exceed 10² MPN/g and therefore additional treatment was required in order to reduce the concentration of EVs to levels established for land application in the study area.

Monpoeho et al. (2001) evaluated several viral extraction techniques for simultaneous detection of EV genomes (number of copies per 10 g of dry matter) and infectious particles (most-probable-number cytopathogenic units [MPNCU]/10 g of dry matter) in different types of sludge samples. The concentration of EVs detected varied considerably from one sample to another. For primary sludge the concentrations varied from <0.3 to 2.24x10² MPNCU/g and from <40 to 3.84x10⁴ genome copies/g (dry weight). The mean concentrations detected in primary sludge were 45.7 MPNCU/g and 1.37 x10⁴ genome copies/g. The equivalent values were 2.9 MPNCU/g and 9.36x10² copies/g in activated sludge, 0.9 MPNCU/g and 1.06x10³ copies/g in thickened sludge, and 0.7 MPNCU/g and 4.8x10³ copies/g in digested sludge. According to these results, from primary sludge to activated sludge the concentrations of virus decreased by a factor of 16 in terms of infectivity and by a factor of 15 in terms of genomes. From activated sludge to thickened sludge viruses decreased by a factor of 3 in terms of infectivity but remained unchanged in terms of genomes. Finally, from thickened sludge to digested sludge they decreased slightly, by a factor of 1.3, in terms of infectivity and increased by a factor of 4.5 in terms of genomes. The study provided a rapid sludge testing protocol capable of determining concentrations of viral genomes and proportion of infectious virus particles in sludge samples.

Wong et al. (2010) investigated the occurrence and the quantitative levels of human EVs and other enteric viruses (human adenovirus, NoVGGI, NovGGII ,and hepatitis A virus) in class B mesophilic anaerobically digested (MAD) biosolid samples using molecular (qPCR) and cell culture methods. In this study, the levels of human adenoviruses were significantly higher than the levels of other enteric viruses and they were also detected more frequently than other enteric viruses. The average levels of EVs and adenoviruses in the MAD biosolids were 1.9x10⁴ copies/g and 5.0x10⁵ copies/g, respectively. The average level of human NoVs corresponded to 5x10⁴ copies/g (NV GGI) and 1.5x10⁵ copies/g (NV GGII). The ICC-PCR assays demonstrated the occurrence of infectious adenoviruses and EVs in the biosolid samples thereby indicating the limited inactivation effectiveness of MAD treatment as revealed in previous studies (Monpoeho et al., 2004; Soares et al., 1994). This study also highlighted the importance of monitoring for additional human pathogenic viruses (adenoviruses, NoVs) considering their frequent occurrence and levels reported in biosolids. Indeed, due to the large potential pathogen diversity in sewage sludge and biosolids, continuous surveillance of biosolids for the culturable and infective content of a broad diversity of pathogens may be required to understand potential risks (Viau et al., 2011).

Reduction/inactivation of EVs in biosolids is discussed in Part 3.1.8 of Section 3.0 Monitoring the occurrence of infectious EVs in biosolids is required to assess pathogen disinfection efficacy and compliance with pathogen land-application standards. Furthermore, quantitative data from pathogen monitoring can be used to assess the infection risk associated with biosolids pathogen exposure. Class-B-treatment processes must yield virus reduction levels (i.e., log reduction value) established for each biosolids disinfection process (i.e., 0.5 – 2 logs) (USEPA, 1999). The European limit value for virus concentration in biosolids is 3 MPNCU (for most probable cytopathic units) per 10 g of dry matter. More recently, QMRA of biosolids exposure based on new pathogen data has demonstrated that the tradition of monitoring pathogen quality by Salmonella spp. and concentration of EVs underestimates the infection risk of pathogens contained in the biosolids (to the public), and that a rigorous biosolids pathogen treatment process is the most efficient method of reducing pathogen exposure and infection risk (Viau et al., 2011). Indeed, the use of improved risk assessment methods has been recommended to supplement technological approaches to establishing regulatory criteria for pathogens in biosolids (National Research Council, 2002). In the United Kingdom, the use of sewage sludge as a soil enhancer and fertilizer on agricultural land remains the environmentally favored option, with around 80% being applied to agricultural land. In Australia and New Zealand strict state and national guidelines specify the way in which specific biosolids can be used.

2.2.3 Occurrence in drinking water

EVs were detected in treated drinking water from supplies that had been treated and disinfected according to international specifications for the production of safe drinking water in South Africa (Ehlers et al., 2005; Vivier et al., 2004). The frequency of detection ranged from 11 to 18.7% and CV-B was the EV predominantly found in drinking water samples. In Egypt, two conventional drinking water treatment plants differing in treatment capacity were investigated for the occurrence of EVs and other enteric viruses. EVs were found in finished water samples and the infective EV count ranged between 1.6 and 33.3 plaque forming unit (PFU)/L (Ali et al., 2004). These viruses were also detected in the distribution system as a result of inadequate water pipe connections. Similarly, EVs were detected in finished water samples collected from seven conventional water treatment plants in the Montreal, Quebec, area that ranged in capacity from 2.4 to 26.4 millions of gallons per day and produced finished water essentially free of indicator bacteria. The frequency of detection of cytopathogenic viruses expressed as MPNCU/L was 7% (11 of 155) and the average density was 0.0006 MPNCU/L with the highest virus density measured corresponding to 0.02 MPNCU/L (Payment et al., 1985b). All viruses isolated were EVs (types 1, 2, and 3 PVs, CVB3, CVB4, CVB5, Echovirus 7, and untyped picornaviruses). EVs were also detected in 47.8% of tap water samples collected from different locations in three urban areas in Seoul, Korea (Lee and Kim, 2002). The level of infectious EVs was always below 10 MPN per 1000 Lbut the frequency of detection was considerably high compared to previous studies.

2.2.4. Occurrence in coastal waters

Human EVs have been found in varying concentrations in marine coastal waters as demonstrated by numerous research studies conducted from early 2000 around the world (Table 8). Previous studies from 1970s – 1980s on viruses in coastal waters focused on EVs due to the ability to detect these viruses using standard-cell culture monolayer methods. The level of EVs reported per liter of sample analyzed ranged from 0.007 to 2.6 PFU; 0.05 to 16 TCID₅₀; 0.05 to 6.5 MPNCU as summarized by Bosch et al. (2005). Recent investigations have introduced PCR detection and quantification of enterovirus genomes by qPCR followed by molecular characterization of the EV types including identification of other human picornaviruses by amplicon sequencing. Types 1, 2 and 3 PVs; coxsackievirus, echovirus, EVs and other human picornaviruses such as HPeVs have been isolated from marine coastal waters. The concentrations reported in the different studies included in Table 8 vary as a result of the level of sewage treatment provided and the methods used to concentrate and detect the viruses.

Solid-associated virions in marine waters are protected from inactivation by environmental stressors (e.g., sunlight irradiation, increasing temperatures, and microbial enzymatic degradation), can be transported long distances after discharge in surficial sediment pore water, and particle-bound virions in sediments can be easily resuspended by mild turbulence and water movements including tidal currents, patterns of water circulation, prevailing winds, storm action, dredging, boating and in coastal and shelf environments (Bosch et al., 2006; Dupuy et al., 2014; Gerba, 2007; Hurst and Murphy, 1996; Pianetti et al., 2007; Rao et al., 1986).

2.2.4.1 Regulatory considerations for coastal waters

Marine coastal waters have received a great deal of attention most likely due to the public health concerns associated with the contamination of shellfish growing beds and the potential exposure pathway to enteric virus infection risk from coastal recreation, e.g. swimming, water skiing, scuba diving, and surfing (Bosch et al., 2006; Dyble et al., 2008; Gerba, 2006; Griffin et al., 2003; Stewart et al., 2008).

• The European Union regulations concerning the quality of bathing waters, European Directive 76/160/ECC, stipulates monitoring for viruses, with a standard of 0 EVs per 10 liters.

Recent studies are addressing the fate and distribution of human EVs and other human picornaviruses in tropical coastal marine environments in Latin America where research studies had been limited and disposal of untreated wastewater into the coastal ocean is a common practice (Betancourt et al., 2015). Overall, these studies have highlighted several relevant findings related to the occurrence and public health significance of human EVs in the coastal environment: (i) EVs are suitable and valuable indicators of human fecal pollution in coastal waters, although they are not ideal viral indicators for all fecally excreted viruses in all situations and geographic locations; (ii) EV detection by RT-PCR/RT-qPCR is a highly sensitive and easy-to-use tool for rapid assessment of marine water quality and fecal contamination provided that adequate sample process controls are included - from concentration to genome amplification – to accurately detect and quantify EV genomes; (iii) the inclusion of ICC-RT-PCR in studies of human EVs in coastal waters serves as a further processing step for enhancing the sensitivity of the EV detection assay with multiple advantages including the capacity to reveal the circulation of specific EV serotypes and other human picornaviruses (e.g. Aichi virus and Parechovirus) in sewage-polluted coastal waters, i.e. ENVS; (iv) molecular typing methods based on RT-PCR and nucleotide sequencing of portions of the VP1 gene (i.e., direct genotyping identification) can be used to obtain location-specific data on enterovirus serotypes in coastal waters, which is of practical value in epidemiological surveillance and in outbreak investigation as previously mentioned; and (v) most frequently EV concentrations in marine waters are not correlated with concentrations of bacterial indicators of sewage pollution as previously documented, although the simultaneous occurrence of EVs and fecal indicator bacteria in water quality monitoring can be used in most geographical settings as a reasonable indication of recent fecal contamination and therefore of chronic pollution problems in coastal waters (Betancourt et al., 2015; Boehm et al., 2003; Connell et al., 2012; Donaldson et al., 2002; Fuhrman et al., 2005; Gersberg et al., 2006; Gregory et al., 2006; Kitajima and Gerba, 2015; Lipp et al., 2001b; Lipp et al., 2001c; Moce-Llivina et al., 2005; Wetz et al., 2004; Wyer et al., 1995).

2.2.4.2 Epidemiological evidence for marine recreational disease

Epidemiological evidence of EV-associated infections transmitted by swimming in contaminated seawater has been limited, most likely due to variability and inefficiency of analytical techniques and lack of epidemiological studies that establish a link between EV illnesses and marine water exposure. Begier et al. (Begier et al., 2008) reported the first outbreak of primarily EV illnesses among travelers to Mexico that likely occurred as a result of swimming in sewage-contaminated seawater. All sea swimming exposures were associated with increased risk of EV infection, and only the time spent on swimming was significantly associated with increased risk of EV infection. QMRAs can be used to evaluate the likelihood that adverse health effects will occur following exposure to coastal recreational waters in which human viruses are present. For these purposes, effective investigations and documentation of EV outbreaks, quantitative data of EVs and thorough consideration of the sources of their infections are necessary, otherwise such assessments will remain difficult to conduct or, when conducted, will be quite uncertain in their outcomes (Parkin et al., 2003). Most commonly, fecal indicator bacteria have been used as predictors of acute gastroenteritis or gastrointestinal illness – the main health outcome – among swimmers at wastewater-impacted recreational beaches and in exploratory quantitative microbial risk assessments (Arnold et al., 2013; Wade et al., 2010). Acute gastroenteritis is a generic term used to refer to gastrointestinal illness involving diarrhea, nausea, or vomiting, which may or may not also be combined with abdominal pain, abdominal cramps, or systemic symptoms, such as fever (Parashar and Glass, 2012; Roy et al., 2006). It is also the common definition given to waterborne diseases (Annon, 1999). However, EVs cause a wide spectrum of acute disease as previously mentioned. Therefore, the use of QMRAs when investigating recreational exposure assessments to enteric viruses must be based on quantitative detection of specific viruses and not on fecal indicator bacteria, as fecal indicator bacteria-based QMRAs frequently underestimate virus occurrence and associated health risks (Symonds and Breibart, 2015).

2.2.5 Occurrence in shellfish

EV contamination has been reported in areas routinely impacted by sewage with a frequency of detection of 19% that can increase up to 45% in highly contaminated sites (Le Guyader et al., 2000). Another study reported EV detection in shellfish tissue from shellfish collected from six of nine beach (66%) sites tested. In this case study, the frequency of detection of EV was higher in shellfish than in water samples from corresponding locations (Connell et al., 2012). EVs have been detected in high quality harvesting areas in Portugal with 35% of the batches analyzed showing signs of enteroviral contamination, 33% showing hepatitis A virus contamination, and 37% showing signs of NoV contamination (Mesquita et al., 2011). Due to bioaccumulation of human enteropathogenic viruses (e.g., EVs and NoVs) by indigenous filter-feeding bivalve mollusks these organisms have been suggested as potential bio-sentinels of human waste in marine coastal waters (Asahina et al., 2009). No PVs were found in shellfish during ENVS for PV in shellfish in the Netherlands downstream from an accidental discharge of high amounts of virulent poliovirus from a vaccine production facility in Brussels (Duizer, 2015; ECDC, 2014).

Although several studies have demonstrated EV contamination in shellfish harvested from polluted and even relatively clean sites, these viruses have not been directly associated with shellfish or seafood vectored infection (Lees, 2000).

2.3 Persistence

The persistence and survival of human enteric viruses, including EVs and other picornaviruses, in the extracellular environment (e.g., water, soil, food, aerosols, and environmental surfaces or fomites), the combination of factors that influence survival (i.e., temperature and seasonal variation, pH, salinity, relative humidity, sunlight irradiation and UV light penetration, wet weather flows, viral aggregation, viral adsorption to suspended solids or surfaces, microbial biome and microbial predation), the risks for transmission of EV infections through environmental exposures (water, foods, aerosols, and fomites), and prevention strategies to reduce this risk have been extensively documented by peer-reviewed scientific research and reviewed in detail by several authors (Bertrand et al., 2012; Bosch et al., 2006; de Roda Husman et al., 2009; Gerba, 2006, 2007; Gerba et al., 2013; John and Rose, 2005; Rzezutka and Cook, 2004; Vasickova and Kovarcik, 2013; Vasickova et al., 2010; Xagoraraki et al., 2014; Yates and Yates, 1988) EV virions can survive outside the body for months under favorable environmental conditions (Bosch et al., 2006; Dowdle and Birmingham, 1997; Mattle et al., 2011; National Research Council, 2004; Pallansch et al., 2013; Templeton et al., 2004; Templeton et al., 2008; Young and Sharp, 1977). These conditions include neutral pH, moisture, low temperature: 4 – 10°C), and association with particles in untreated or partly treated waste waters that protects against inactivation by natural or artificial processes, including enzymatic (i.e., proteolytic bacterial enzymes) or UV degradation and chemical disinfection (discussed further in Section 3.2). The biophysical properties of picornaviruses such as their small-size (30 nm), genome type (ssRNA), and non-enveloped capsid structure of the virions are known to play an important role on the mechanisms of virus survival and transport in the environment (Bosch et al., 2006; Fong and Lipp, 2005; Gerba et al., 2013; National Research Council, 2004; Sobsey and Meschke, 2003; Wigginton and Kohn, 2012; Xagoraraki et al., 2014). Processing conditions for picornavirus capsid assembly enable thermal stability (stable at 42°C or up to 50°C in presence of sulfhydryl reducing agents and magnesium cations) and pH stability (pH 3 – 9) in the extracellular environment (Gerba, 2007; Hurst and Adcock, 2000; Hurst and Murphy, 1996; National Research Council, 2004; Racaniello, 2013). Capsid protein interaction with genomic RNA genome adds stabilization (Pallansch et al., 2013). Electrostatic and hydrophobic interactions involving functional groups of the viral capsid (i.e., charged amino acids residues) are essential for capsid assembly, capsid stability, and virus particle adsorption to solid surfaces (e.g., suspended solids, sediments and environmental surfaces or fomites) (Mateu, 2011; Michen and Graule, 2010). Binding of serotype 1 polio vaccine strains (Mahoney and Sabin 1) to bacteria or bacterial polysaccharides offers a selective advantage by enhancing capsid stability in the environment (Robinson et al., 2014).

2.3.1 Persistence in aquatic environments.

Persistence of EVs in aquatic environments can be inferred from initial concentrations and T90 and T99 inactivation times, the times required to reduce the number of virus by 1log₁₀ and 2log₁₀, respectively. T90 and T99 parameters have been derived from different studies under different experimental conditions (light intensity, type of water, turbidity, and pH) and different methods to determine EV inactivation rates (cell culture versus molecular methods). As a result, wide variations in inactivation times for the same enterovirus or other members within the Picornaviridae family have been reported in different studies. Several examples will be provided.

The T90 and T99 decay rates for infectivity and viral RNA genome detection of poliovirus strain Lsc-1 was 2 and 4 days, respectively, in unfiltered natural seawater at 30°C, while it was 2.2 and 4.4 days, respectively at 22°C in unfiltered natural seawater (Wetz et al., 2004). Longer time periods (lower rates of virus inactivation) were documented for filtered seawater and distilled water indicating that much of the virus decay was caused by factors biological in nature (i.e., microbial activity). Little difference was noted in the decay rates of the virus due to temperatures. In a different experiment with the same poliovirus strain comparing sterile and non-sterile marine waters under different environmental conditions (e.g., Temperature 22 – 27°C; Salinity 8 – 35%; Turbidity <1 – 4 nephelometric turbidity units, polluted versus non-polluted water, sunlight vs darkness), the T90 values in marine waters kept in the dark ranged between 1 and 1.1 days whereas only 0.4 days were required for a 1log₁₀ reduction of the virus in marine waters and the inactivation rate in the polluted canal water was 3x faster than in non-polluted marine waters with sunlight (Johnson et al., 1997).

PV remained infectious for 90 days at 10°C in wastewater and groundwater and persisted for more than a year in mineral water stored at 4°C (Biziagos et al., 1988).

Differences in T90 inactivation rates for EVs in fresh water depends on the water temperature: 1 to 1.5 days at 12 - 20°C (O’Brien and Newman 1977 as cited in (Carter, 2005)) while inactivation by 6.5 to 7 log₁₀ units for CV-B3, Echovirus-7 and PV1 required 8 weeks at 22°C, and a 4 to 5 log₁₀ inactivation of the same viruses at 4°C required 12 weeks (Hurst, 1988a).

2.3.2 Persistence in groundwater

Persistence of EVs in groundwater can be inferred from temperature-based inactivation studies (John and Rose, 2005; Schijven and Hassanizadeh, 2000). As above, variations in inactivation times for the same enterovirus or other members within the Picornaviridae family have been reported: The mean inactivation rate of PV was to 0.02 log₁₀ per day (range 0.005 – 0.05 log ₁₀ per day ) between 0 and 10°C and 0.4 log₁₀ per day (range 0.006 – 1.4 log₁₀ per day) at 26– 30°C; the mean inactivation rate for echovirus was 0.1 log₁₀ per day (range 0.05 – 0.2 log₁₀ per day) at temperatures between 11 and 15°C, and 0.2 log₁₀ per day-1 (range 0.06 – 0.6 log₁₀ per day) at temperatures between 21 and 25°C; and the mean inactivation rate for coxsackieviruses was 0.06 log₁₀ per day-1 (range 0.002 – 0.2 log₁₀ per day) at temperatures between 0 and 20°C and 0.1 log₁₀ per day at temperatures between 25 and 30°C.

2.3.3 Persistence In biofilms

Incorporation of enteroviruses into biofilms of drinking water distribution systems and into natural waste water biofilms protects the virus genome from degradation and to a certain extent, prevents inactivation (Skraber et al., 2009) and Vanden Bosschose (cited in (Skraber et al., 2005)). Persistent EVs may be present in biofilm sloughs in water distribution systems, and persistence of EVs in wastewater biofilms may extend the temporal dispersal of viral pollution in water environments after periods of prevalence in wastewaters.

2.3.4 Persistence In soil

Studies conducted during the early 1980s found that EV persistence in soils was influenced by temperature, soil moisture content, degree of virus adsorption to soil, soil levels of resin-extractable phosphorus, exchangeable aluminum and the pH (Hurst et al., 1980). Table 10 shows the data on sorption behavior of enteroviruses with various types of sorbents. Variations in experimental conditions in the various studies make it difficult to draw general conclusions. Variations in experimental conditions in the various studies make it difficult to draw general conclusions. However, in general, the mechanisms of virus adsorption to solid surfaces depend on the chemical composition of the liquid phase and the nature of the solid surface. Key variables influencing sorption are pH and ionic strength of the solution, presence of compounds competing for sorption sites, functional groups, and isoelectric points of virus and sorbent. EVs can persist in soils for prolonged periods of time, especially at lower temperatures (4 – 8°C) than at higher temperatures (20 – 37°C) (Rzezutka and Carducci, 2013; Vasickova and Kovarcik, 2013) and under in anaerobic conditions (Yates, 2002). Depending on the type of virus and sorbent, sorption is dominated by electrostatic interactions, van der Waals forces, hydrophobic effects, and covalent binding to surfaces (Yates, 2002). EV adsorption to clays protect the virions from inactivation by increasing stability of the viral capsid, by adsorption of enzymes and other inactivating substances, and prevention of aggregate formation (Gerba, 2007).