Enteric viruses, such as noroviruses (NoVs), enteroviruses, and adenoviruses, have been a major research target of environmental virology. Development and improvement of methods to detect enteric viruses in environmental water, especially virus concentration methods and molecular detection techniques, have enabled us to accumulate qualitative and quantitative data on occurrence and behavior of enteric viruses in the environments. NoVs are considered to be an important agent in causing waterborne gastroenteritis. Although no routine in vitro infectivity assay system for human NoVs is available, discovery of murine noroviruses (MNVs) that can be cultivated in routine cell culture system have facilitated our understanding on environmental stability of NoVs. That is, recent progress in environmental virology has been elucidating the behavior of enteric viruses including human NoVs in water environments and water treatment systems. On the other hand, emerging and re-emerging viral diseases for humans have been reported. In 1997, highly pathogenic avian influenza A (HPAI) of the H5N1 subtype viruses transmitted from birds to humans and caused the deaths of 6 out of 18 infected persons in Hong Kong. This was the first incidence of transmission of avian influenza virus to humans with fatal outcome. Further outbreaks of HPAI H5N1 viruses had started since 2003 in Hong Kong, Vietnam, Indonesia, and Thailand. The HPAI H5N1 viruses have then spread to other Asian countries (including Japan), the Middle East, Southeast Europe, Central Europe, and Africa. Therefore, human infection with HPAI H5N1 virus is of a great public health concern in the world, and the H5N1 avian influenza is one of the important emerging diseases for humans. As represented by the H5N1 avian influenza, a novel virus with unknown biological characteristics from non-human source can emerge, and thus, environmental virologists should target broad range of pathogenic viruses to understand their behavior in the environments and control risks.

On the basis of the above background, the present study was performed to obtain systematic and comprehensive knowledge to manage infection risks caused by waterborne viruses. The major objectives of the present study were (1) to investigate the prevalence, seasonality, and genetic diversity of enteric viruses, namely NoVs, sapoviruses (SaVs), and Aichi viruses (AiVs), in water environments and (2) to characterize infection risks of highly pathogenic avian influenza viruses to humans through water.

Chapters 3 and 4 focus on gastroenteritis viruses, namely NoVs, SaVs, and AiVs. Circulation of the viruses between contaminated environmental water and human populations is a key issue in understanding their epidemiology and health risks for humans. Since wastewater contains viruses shed from all populations regardless of their health status, monitoring of viruses in wastewater and urban river water receiving effluents from multiple wastewater treatment plants (WWTPs) could be an appropriate approach for determining the actual prevalence and molecular epidemiology of gastroenteritis viruses in catchment areas rather than clinical studies. In Chapter 3, prevalence and genetic diversity of NoVs, SaVs, and AiVs in wastewater and river water was investigated during a 1-year period. Presence of viral genomes in the water samples were examined with RT-PCR assays, and the strains were further characterized based on the nucleotide sequences of the PCR amplicons. GI NoV strains were more frequently detected than GII NoV strains in both wastewater and river water samples. This result disagrees with the epidemiological data obtained from the Infectious Disease Surveillance Center, National Institute of Infectious Diseases, Japan, where GII strains were much more frequently detected than GI in feces of hospitalized patients, suggesting that GI strains are more widely spread among humans than previously appreciated. Moreover, GIV NoV strains were successfully identified in both wastewater and river water samples by utilizing the newly developed semi-nested RT-PCR assay specific for GIV. Interestingly, it was demonstrated that genetically diverse GIV NoV strains, including those of a newly identified genetic cluster, were circulating in Japan. The newly developed assay can be a powerful tool to detect and characterize GIV strains in water environments. SaV strains were also successfully identified in both wastewater and river water. A nested RT-PCR assay utilizing a novel primer (1245Rfwd), rather than any previously reported assays, more efficiently amplified SaV genomes in the samples. The newly developed RT-PCR assay is useful for identifying SaV strains in environmental samples. Another highly important results obtained in this chapter is that Aichi viruses showed the highest prevalence in both wastewater (23/24; 96%) and river water (36/60; 60%) among the viruses tested, suggesting that AiV are prevalent in aquatic environments of Japan. The results described in this chapter demonstrates that genetically diverse NoV, SaV, and AiV strains are circulating between human populations and water environments. These results provide novel findings contributing our understanding of the prevalence and genetic diversity of the viruses in water environments. Gastroenteritis viruses in water environments may reflect, more accurately, the actual prevalence and molecular epidemiology in the human population rather than reported cases which represent a small portion of total cases.

In Chapter 4, TaqMan-based real-time RT-PCR assays for rapid detection, quantification, and genotyping of AiVs have been established. The assays were subsequently applied for quantitative analysis of occurrence and behavior of AiVs at two WWTPs (A- and B-WWTP) during a 1-year period. AiVs were detected in all influent as well as effluent samples of both WWTPs, demonstrating that they are continuously circulating between human populations and water environments. A trend observed in both WWTPs was that the concentrations of AiVs in influent increased in winter season, and were comparably lower in summer-autumn season. The removal ratios of AiVs at A-WWTP and B-WWTP were 2.41±0.42 log10 (n=12) and 2.96±0.40 log10 (n=12), respectively. It should be noted here that AiVs can potentially be an appropriate indicator of viral contamination in the environments because of their high prevalence in the environments and structural as well as genetic similarity with some of other important enteric viruses. AiVs are small round viruses possessing a single-stranded RNA genome and are members of the family Picornaviridae that includes enteroviruses and hepatitis A virus. These latter two viruses are listed in the latest U.S. Environmental Protect Agency's Contaminant Candidate List (CCL3), which identifies emerging contaminants of aquatic environments that may pose a public health risk. This is the first study providing protocols for quantification and genotyping of AiVs by real-time RT-PCR assays and describing quantitative data on the occurrence of AiVs in wastewater over a 1-year period. Behavior of AiVs in water environments had been unknown, since there have been only a few studies reporting detection of AiVs in environmental samples. This study provides useful knowledge toward understanding their occurrence and fate in water environments.

Chapters 5, 6, and 7 involve experimental and analytical studies on influenza A viruses. It is likely that avian influenza viruses with feces or other secretions from both symptomatic and asymptomatic waterfowl will be released into water environments where the birds gather. Although avian influenza viruses can persist for extended periods of time in water, the occurrence of the viruses in environmental water remains unknown. Since virus concentration is an essential step to detect viruses at low levels in water, Chapter 5 was designed to determine recovery yields of influenza A viruses in water by conventional virus concentration methods with both plaque assay and real-time RT-PCR assay. The results demonstrated that infectious influenza viruses were not efficiently recovered by Mg- or Al-methods because of pH sensitivity of influenza viruses. A novel method using an HA electronegative filter and surfactant-based eluting solution (pH 7.9) was judged to be an appropriate concentration method that can efficiently recover viable influenza A virus particles from water to obtain a recovery rate of more than 30%.

Open bodies of water, including drinking water sources, can be contaminated by infected waterfowl. The oral ingestion or aspiration of water containing influenza A virus could be a possible mode of transmission to humans, although no evidence has been reported. Despite growing concerns about the public health threat posed by influenza A viruses, there has been limited knowledge on efficacy of disinfectants in inactivating influenza A viruses in water. In Chapter 6, inactivation kinetics of influenza A viruses by disinfectants, namely chlorination, chloramination, and UV irradiation, were investigated. For the purpose of drinking water production, the U.S. Environmental Protection Agency requires free chlorine CT values of 6 and 8mg-min/L to achieve enteric virus inactivation of 3 and 4 log10, respectively. According to the results obtained in this study, these CT values would be more than sufficient to inactivate influenza A viruses in water. Inactivation ratio of influenza A viruses by each disinfectant was compared with those of enteric viruses, which demonstrated that influenza A viruses are less resistant to the disinfectants than most of enteric viruses. These results demonstrated that water disinfection process could achieve appreciable inactivation of influenza A viruses when a proper pre-treatment process (e.g. coagulation, sedimentation and rapid sand filtration) is performed to remove the chlorine demand and suspended solid. The information on inactivation of influenza A viruses described in this study is useful for developing risk management procedures regarding the role of water in the transmission of the viruses to humans and poultry.

Recently, highly pathogenic avian influenza A (HPAI) of the H5N1 subtype viruses have infected humans and caused severe diseases in many countries. Quantitative microbial risk assessment (QMRA) framework can be a powerful tool to understand how to control pandemics mediated by environmental reservoirs or human-to-human transmission (e.g. calculating risk of infection due to a low dose). The objectives of Chapter 7 were to estimate risks of waterborne and airborne infections, and to develop a time-dependent dose-response model, based on the QMRA framework. Waterborne infection risks of the HPAI H5N1 virus quantitatively evaluated assuming that a single HPAI H5N1 virus-infected duck shed the virus in the river water used for drinking water production and recreational purposes. The results of the Monte Carlo simulation demonstrated that the median probability of infection associated with swimming in the contaminated river was 9.4×10・11 [infection/person/swim]. The median probability of infection associated with consumption of tap water was less than 10・13 [infection/person/year] when the virus reduction efficiency at the drinking water treatment plant was 4 log10. Furthermore, a basic model to evaluate a household, airborne, secondary transmission risk was constructed. An essential step in the QMRA process is dose-response assessment; however, to date, none of the prior studies investigated the dose-response relationship to describe HPAI H5N1 virus infection in humans. In this study, the time-dose-response model to describe mortality of mice exposed to an HPAI H5N1 virus, describing the mortality over time and represents experimental responses accurately, has been successfully developed and validated. This is the first study describing a time-dependent dose-response model for HPAI H5N1 virus. These models developed in the present study will be useful to evaluate the risks of HPAI H5N1 virus infection under various exposure scenarios and to estimate the mortality of HPAI H5N1 virus, which may depends on time post exposure, for preparation of a future influenza pandemic caused by this lethal virus.

Environmental virology consists of four major elements: (1) development of concentration and detection methods of viruses in water, (2) investigation of the occurrence and behavior of viruses in water environments, (3) characterization of inactivation of viruses by disinfectants used for water treatment and water distribution processes, and (4) risk assessment. A lot of knowledge on environmental and medical virology has been accumulated since several decades ago. The present study provides novel knowledge on the prevalence, seasonality, and genetic diversity of enteric viruses, namely NoVs, sapoviruses (SaVs), and Aichi viruses (AiVs), in water environments, and on infection risks of highly pathogenic avian influenza viruses to humans through water. This dissertation summarizes previously known and newly obtained knowledge that greatly contributes to evaluate and reduce infection risks of pathogenic viruses in water.

本論文は、Molecular Epidemiological Analysis of Pathogenic Viruses in Water Environments and Risk Assessment (水環境中における病原ウイルスの分子疫学的解析および感染リスク評価)と題する。近年世界規模での大きな社会問題となってきているノロウイルスや鳥インフルエンザウイルスを含む幅広い病原ウイルスを対象として、水環境中におけるウイルスの遺伝的多様性とヒトへの感染リスクについて研究したものである。本論文は以下の8章で構成されている。

第1章では、水環境中の病原ウイルスの遺伝子解析および高病原性鳥インフルエンザウイルスの感染リスク評価を実施することの意義ならびに本論文の目的と構成を示している。

第2章では、水系感染性の病原ウイルスについての既存の知見をまとめている。病原ウイルスについての概論、ヒトへの健康被害と集団感染報告事例、消毒処理感受性や環境中における生残性、水中のウイルスの検出法と検出事例、ヒトへの感染リスクの定量的評価などに関する既存の研究事例を整理して示している。

第3章では、下水および河川水中におけるノロウイルス、サポウイルスおよびアイチウイルスの存在状況ならびに遺伝的多様性を調査している。具体的には、水試料中のウイルス遺伝子をNested RT-PCR法により増幅し、PCR産物をクローニング後、塩基配列を決定することで遺伝子型を同定するとともに遺伝的多様性を解析している。ここで、GIVノロウイルス、サポウイルスについては、遺伝子増幅のための新たなプライマーを開発している。ノロウイルスについては、胃腸炎感染者から検出されるウイルス株はGII株が大部分(90%)を占めるが、本研究の調査では水試料からGI株のほうが高頻度に検出されたことを報告しており、急性胃腸炎事例として報告されないGI感染者が流域に多数存在していた可能性を示している。さらに、GIVノロウイルスの遺伝子増幅のための新たなNested RT-PCRを開発し水試料からのウイルス検出に適用することで、下水および河川水中におけるGIVノロウイルスの存在状況および遺伝的多様性に関する新たな知見を得ることにも成功している。サポウイルスについても新たなNested RT-PCRを構築し、水環境中のサポウイルスの遺伝的多様性に関する新たな知見を得ている。アイチウイルスに関しては、これまで考えられてきた以上に遺伝的に多様なウイルス株がヒトの間で広く流行していることを示す知見を示している。以上の結果より、水環境中のウイルス遺伝子を網羅的に解析することにより流域におけるウイルス感染症の真の流行状況像を把握することが可能であるとの結論を導いている。

第4章では、Real-time RT-PCRによるアイチウイルスの核酸定量法を開発し、下水中におけるアイチウイルスの挙動の定量的解析を実施している。開発した核酸定量法は、アイチウイルスの迅速検出、定量およびGenotype識別が可能であり、アイチウイルスの疫学と環境中での挙動を解明するための有用なツールであるとしている。また、この方法を国内二ヶ所の下水処理場の下水中からのアイチウイルス検出に適用し、下水中におけるアイチウイルス濃度の年間変動を調査するとともに、下水処理場における除去率を算出してその他のウイルスや指標微生物の挙動と比較している。アイチウイルスは下水流入水および放流水から年間を通じて検出され、下水処理による除去率は他のウイルスと同程度であったことから、新たなウイルス指標として有望であるとの結論を導いている。

第5章では、水中のA型インフルエンザウイルスの濃縮法を開発している。まず、水中の腸管系ウイルスの濃縮を目的として開発され、様々な水試料からのウイルス検出に適用されてきた既存の3種類の濃縮法(Mg法、Al法、1MDS法)によるA型インフルエンザウイルスの濃縮回収率を測定したところ、これら既存の濃縮法では感染性のA型インフルエンザウイルス粒子を効率良く回収できなかったことを示している。そこで、感染性のA型インフルエンザウイルス粒子の回収を目的として、界面活性剤系の誘出液を使用した濃縮法を新たに開発している。開発した濃縮法(Mg-Tween法)を使用することで、実用可能なレベル(30%以上)の感染性ウイルス回収率が得られたとしている。

第6章では、水の消毒処理(塩素、モノクロラミンおよび紫外線)によるA型インフルエンザウイルスの不活化特性を解析している。本研究でのA型インフルエンザウイルスの不活化実験結果と既に報告されている腸管系ウイルスの実験結果とを比較し、A型インフルエンザウイルスは、いずれの消毒処理についても大部分の腸管系ウイルスよりも速やかに不活化することを示している。我が国では、水道法により給水栓末端で0.1mg/L以上の遊離塩素濃度を確保することが義務付けられているが、この濃度の遊離塩素で消毒した場合にはA型インフルエンザウイルスは速やかに不活化することから、仮に水道原水がA型インフルエンザウイルスにより汚染されたとしても上水道で施されている通常の塩素消毒により十分な不活化が期待できると結論付けている。

第7章では、高病原性鳥インフルエンザウイルスの定量的感染リスク評価モデルを開発している。具体的には、まず一羽の感染カモ糞便に汚染された河川水を感染源とした水系感染リスクを定量的に評価している。不確実性を含む原単位はその分布を推定した上でモンテカルロ法による計算を実施した結果、河川での遊泳に伴う感染確率の中央値は9.4×10・11 [infection/person/swim]、浄水場でのウイルス除去率を4 log10とした場合の水道水飲用に伴う感染確率の中央値は 10・13 [infection/person/year]未満であることを示している。更に、空気を介した感染リスク評価モデルも開発している。また、ヒトが高病原性鳥インフルエンザウイルスに感染した場合には感染後経時的に致死率が上昇するが、この反応を表現するために従来型Dose-Responseモデルに時間項(感染後死亡までの時間)を組み込み、新たなTime-Dose-Responseモデルを開発することに成功している。

第8章は総括であり、本論文の結論および今後の展望について整理して示している。

本論文では、水環境中におけるノロウイルス、アイチウイルスおよびサポウイルスの挙動と遺伝的多様性に関する数多くの新たな知見を得ているとともに、インフルエンザウイルスの濃縮法の開発、不活化特性の解析および感染リスク評価モデルの開発を行っている。このように、本論文は、水中病原ウイルスの感染リスク管理に資する新たな知見をまとめたものであり、都市環境工学の学術分野に大いに貢献する成果である。

よって本論文は博士(工学)の学位請求論文として合格であると認められる。