Seroepidemiology and Primary Infections of the Recently Detected Merkel Cell Polyomavirus and Trichodysplasia-Spinulosa Associated Polyomavirus

Seroepidemiology and primary infections of the recently detected Merkel cell polyomavirus and trichodysplasia-spinulosa associated polyomavirus

Tingting Chen

Haartman Institute Department of Virology University of Helsinki Helsinki, Finland

ACADEMIC DISSERTATION

To be presented with the permission of the Medical Faculty of the University of Helsinki, for public examination in the Lecture Hall 2, Haartman Institute, Haartmaninkatu 3, on December 5th, 2013, at 12 noon

Helsinki 2013

SUPERVISED BY: Professor Klaus Hedman Department of Virology Haartman Institute University of Helsinki Helsinki, Finland & Docent Maria Söderlund-Venermo Department of Virology Haartman Institute University of Helsinki Helsinki, Finland

REVIEWED BY: Professor Timo Vesikari University of Tampere Medical School Vaccine Research Center Tampere, Finland & Docent Thedi Ziegler University of Turku Research Center for Child Psychiatry Institute of Clinical Medicine Turku, Finland

OPPONENT: Associate Professor Mariet Feltkamp Department of Medical Microbiology Leiden University Medical Center Leiden, The Netherlands

ISBN 978-952-10-9546-7 (paperback) ISBN 978-952-10-9547-4 (PDF)

Helsinki 2013

CONTENTS

ABSTRACT ...... 5 LIST OF ORIGINAL PUBLICATIONS ...... 7 ABBREVIATIONS ...... 8 INTRODUCTION ...... 9 REVIEW OF THE LITERATURE ...... 10 1. Overview of human polyomaviruses ...... 10 1.1 Classification and virion properties ...... 10 1.2 General aspects of HPyV life cycle ...... 11 1.3 General aspects of HPyV infection and diseases ...... 12 2. Merkel cell polyomavirus ...... 13 2.1 Merkel cell carcinoma ...... 13 2.2 Discovery of MCPyV ...... 14 2.3 MCPyV genome structure and function ...... 14 2.4 Infection and immunity ...... 17 2.5 Molecular epidemiology ...... 18 2.6 Clinical aspects on MCPyV-positive and -negative MCCs ...... 20 2.7 Association of MCPyV with pathogenesis of MCC ...... 21 3. Trichodysplasia-spinulosa associated polyomavirus ...... 23 3.1 Trichodysplasia-spinulosa disease ...... 23 3.2 Discovery of TSPyV...... 23 3.3 TSPyV genome ...... 23 3.4 Epidemiology and immunity...... 24 4. Serodiagnosis ...... 26 4.1 Antibody response in viral infection ...... 26 4.2 EIA for antibody detection ...... 27 4.3 Interpretation of results ...... 28 4.4 Biotinylation-based anti-viral IgM EIA ...... 29 AIMS OF THE STUDY ...... 30 MATERIALS AND METHODS ...... 31 1. Materials ...... 31 1.1 Patients and serum samples ...... 31 1.2 Plasmids, bacteria and cells (I, II) ...... 32 2. Methods ...... 32 2.1 Virus-like particle production (I, II) ...... 32 2.2 Biotinylation (I, II, IV) ...... 33 2.3 Immunoblotting (I, II, IV)...... 33 2.4 In-house antibody assays (I-IV) ...... 34 2.5 Cutoff determination (I-IV) ...... 35 2.6 Affinity determination (IV) ...... 36 2.7 Quantitative PCR (III) ...... 36 2.8 Statistical analysis (I-IV) ...... 36

RESULTS ...... 37 1. MCPyV and TSPyV serodiagnostics and infections ...... 37 1.1 Virus-like particles (I, II) ...... 37 1.2 Antibody responses (I-III) ...... 37 1.3 Seroprevalences (I-III) ...... 39 1.4 Asymptomatic MCPyV and TSPyV primary infections among children (III) ..... 40 2. Biotin IgM antibodies...... 42 2.1 Observation of false MCPyV IgM positivity due to biotin antibody ...... 42 2.2 Seroprevalence of biotin IgM ...... 42 2.3 The effect of biotin antibody on various IgM-antibody assays ...... 42 2.4 Characteristics of Biotin IgM ...... 43 DISCUSSION ...... 50 1. MCPyV and TSPyV serodiagnostic assays (I-III) ...... 50 1.1 Generation of MCPyV and TSPyV VLPs ...... 50 1.2. Serodiagnostic tools for MCPyV and TSPyV ...... 50 2. MCPyV and TSPyV infections (I-III) ...... 52 2.1 MCPyV and TSPyV antibody responses and seroprevalence ...... 52 2.2 Antigenic cross-reactivity between different HPyVs ...... 53 2.3 Asymptomatic MCPyV and TSPyV primary infections in childhood ...... 54 3. Biotin IgM antibodies (IV) ...... 55 3.1 The effect of biotin IgM on biotinylation-associated EIAs ...... 55 3.2 Potential origins of biotin IgM...... 56 3.3 Possible clinical biomedical effects of biotin IgM...... 56 CONCLUSIONS AND FUTURE PROSPECTIVES ...... 57 ACKNOWLEDGEMENTS ...... 59 REFERENCES ...... 60

ABSTRACT

Two newly found polyomaviruses are related to human skin diseases; Merkel cell polyomavirus (MCPyV), identified in Merkel cell carcinomas (MCC) is believed to be an etiologic factor in the pathogenesis of MCC; trichodysplasia spinulosa-associated polyomavirus (TSPyV), discovered in TS lesions, has an apparently causative role in this rare skin disease. To investigate their seroprevalence and possible clinical significance, we have developed for these two polyomaviruses comprehensive serodiagnostic methods and studied the occurrence of antibodies to these viruses, the quality of immune responses, diagnostic criteria, and possible disease associations. In addition, the thesis aims at searching for factors that cause false results in biotinylation-based antiviral IgM enzyme immunoassays (EIAs). For MCPyV and TSPyV, we developed EIAs for virus-specific IgG, IgM and IgG avidity, by using the major capsid protein VP1 expressed in insect cells as self-assembled into virus-like particles. To determine seroprevalence and seroconversion rates, virus- specific IgG antibody assays were applied to examine serum samples from adults and three groups of children less than 13 years of age. In the first two children´s groups, children underwent initial and 3-8-year follow-up sampling. In the third group, children were followed clinically from birth until 13 years of age, and from each such child consecutive serum samples were obtained. Among Finnish adults, the seroprevalence for MCPyV was 58% and for TSPyV 70%. Among Finnish children between 1-13 years of age, the seroprevalence for MCPyV were determined at 9-45%, with seroconversion rates of 16-33% during the 3-8-year follow-up, while for TSPyV the seroprevalence was 5-39%, with seroconversion rates of 30-45%. Among 144 children followed from birth up to 13 years, 31% and 27% had IgG seroconversion to MCPyV and to TSPyV, respectively. These data indicate that MCPyV and TSPyV circulate widely in the general population, and that primary infections by these two viruses occur extensively in childhood with the first exposures taking place at young age. We then investigated whether in children primary infections with these two viruses were associated with any symptoms or signs. For each virus, EIAs for antiviral IgG, IgM and IgG avidity, along with quantitative PCR were employed on consecutive serum samples from the seroconverting children in the third children´s group. At the time of IgG seroconversion, additional infection markers (IgM response and/or low-avidity IgG) were found to be present in 62% of the MCPyV seroconverters, and in 82% of the TSPyV seroconverters. Comparing clinical events during the seroconversion intervals with those during the previous and subsequent intervals, we found that with neither virus primary infection was significantly associated with symptoms or signs of any infection-related illness. These observations imply that in childhood primary infections with MCPyV or TSPyV are in general asymptomatic. To improve the quality of serodiagnostic assays, we looked for factors causing false IgM reactivities in IgM assays in which captured virus-specific IgM antibodies are detected by biotinylated viral antigens bound to horseradish peroxidase conjugated streptavidin. Employed as negative control for our MCPyV IgM EIA, one adult serum

sample exhibited IgM reactivity against the biotinylated viral antigen, but did not show reactivity against the non-biotinylated one. This observation pointed to the occurrence of anti-biotin IgM in human blood, causing false-positive results in biotinylation-based antiviral IgM assays. To investigate the prevalence of biotin-specific IgM antibodies in the general population, we developed an indirect EIA for biotin IgM, and found these antibodies to be present in 3% of adults regardless of age but rarely in children. Adverse effects of the biotin IgM were examined with a variety of biotinylation-based antiviral IgM assays. Interestingly, soluble biotin failed to bind the IgM and streptavidin/avidin simultaneously, pointing to the possibility that the IgM and streptavidin/avidin share the same binding site on biotin. The affinities of biotin IgM antibodies were determined at a range of 10-3 to 10-4 mol/L. These findings provide new information on biotinylation- based immunoassays, and open new insights into anti-vitamin immune responses.

LIST OF ORIGINAL PUBLICATIONS

The thesis is based on the following publications that are referred to in the text by their Roman numerals. The original articles are reprinted with the permission of their copyright holders.

I. Chen T, Hedman L, Mattila PS, Jartti T, Ruuskanen O, Söderlund-Venermo M, Hedman K. Serological evidence of merkel cell polyomavirus primary infections in childhood. J Clin Virol. 2011 Feb; 50(2):125-9.

II. Chen T, Mattila PS, Jartti T, Ruuskanen O, Söderlund-Venermo M, Hedman K. Seroepidemiology of the newly found trichodysplasia spinulosa-associated polyomavirus (TSV). J Infect Dis. 2011 Nov; 204(10):1523-6.

III. Chen T, Tanner L, Simell V, Hedman L, Mäkinen M, Sadeghi M, Veijola R, Hyöty H, Ilonen J, Knip M, Toppari J, Ruuskanen O, Simell O, Söderlund-Venermo M, Hedman K. Merkel cell polyomavirus and Trichodysplasia spinulosa-associated polyomavirus primary infections among children determined by comprehensive immunoassays - diagnostics and clinical pictures. Submitted.

IV. Chen T, Hedman L, Mattila PS, Jartti L, Jartti T, Ruuskanen O, Söderlund-Venermo M, Hedman K. Biotin IgM antibodies in human blood: a previously unknown factor eliciting false results in biotinylation-based immunoassays. PLoS One 2012;7(8):e42376.

ABBREVIATIONS

B19V human parvovirus B19 bio-BSA biotinylated BSA bio-EIA biotin IgM indirect EIA BKPyV or BKV BK polyomavirus BSA bovine serum albumin DIPP Diabetes Prediction and Prevention EIA enzyme immunoassay EM electron microscopy GST glutathione S-transferase HBoV human bocavirus HPyV human polyomavirus HRP horseradish peroxidase JCPyV or JCV JC polyomavirus KIPyV KI polyomavirus LPyV B lymphotropic polyomavirus LT large T antigen MCC Merkel cell carcinoma MCPyV or MCV Merkel cell polyomavirus NCCR non-coding control region OBD origin binding domain OPD orthophenylene diamine PBMCs peripheral blood mononuclear cells PBS phosphate-buffered saline PBST PBS containing 0.05% Tween 20 PML progressive multifocal leukoencephalopathy pRb retinoblastoma protein PVAN polyomavirus-associated nephropathy RCA rolling-circle amplification RT room temperature SDS-PAGE sodium dodecyl sulfate polyacrylamide gel electrophoresis Sf Spodoptera frugiperda sT small T antigen STLPyV STL polyomavirus SV40 simian vacuolating agent 40 or simian virus 40 TBS-T Tris-buffered saline with 0.1% Tween 20 TS trichodysplasia spinulosa TSPyV or TSV trichodysplasia-spinulosa associated polyomavirus VLPs virus-like particles WUPyV WU polyomavirus

INTRODUCTION

Polyomaviruses belong to the family Polyomaviridae, consisting of small non-enveloped viruses with a double-stranded, covalently closed circular DNA genome (1). The family name originates from Greek, “poly” meaning “many” and “oma”, meaning “tumor”. A number of mammalian and avian polyomaviruses have been identified, however, with each virus species demonstrating a narrow host range (1). In contrast, the polyomaviruses can experimentally transform non-permissive cells and induce tumors in their non-natural hosts (2-5). So far, twelve human polyomaviruses (HPyVs) have been found, and ten of them are discoveries of the past six years. Among these recently found HPyVs, two are found to be associated with human skin diseases: Merkel cell polyomavirus (MCPyV or MCV) with Merkel cell carcinoma (MCC) and trichodysplasia spinulosa-associated polyomavirus (TSPyV or TSV) with trichodysplasia spinulosa (TS) disease. In 2008, MCPyV was found in MCC, an uncommon and aggressive skin cancer, and subsequently the virus was shown to be present in 30%-100% of MCCs around the world (6, 7). The virus DNA is integrated into the genome of tumor cells and harbors specific mutations that eliminate viral replication capacity, but retain their oncogenicity, strongly suggesting that the virus is a pathogenetic factor in MCC (6, 8). In 2010, TSPyV was discovered in TS lesions of a heart transplant patient, with an apparent causal relation of the virus with this rare skin disease that is predominant in immunosuppression (9). Moreover, by analysis of all available TS cases worldwide, an active TSPyV infection was demonstrated to be associated with TS pathogenesis (10). As MCPyV and TSPyV are recent discoveries, information on their impact beyond MCC and TS is just beginning to accumulate. The purposes of this thesis are to develop for these two recently found HPyVs comprehensive serodiagnostic methods, and to disclose the occurrence and possible disease associations of these two HPyVs.

REVIEW OF THE LITERATURE

1. Overview of human polyomaviruses

HPyVs comprise twelve members (Table 1). JC polyomavirus (JCPyV or JCV) and BK polyomavirus (BKPyV or BKV) are the classical members known to infect humans (JC and BK stand for the initials of the patients from whom the viruses were first detected). In 1971, the former was identified in the brain tissue of a progressive multifocal leukoencephalopathy (PML) patient (11), and the latter in the urine of a nephropathic kidney transplant recipient (12). Since 2007, many additional HPyVs have been found. KI polyomavirus (KIPyV) and WU polyomavirus (WUPyV) (KI stands for Karolinska Institute and WU for Washington University) were in 2007 identified in the nasopharyngeal aspirates of children with respiratory tract infection (13, 14). MCPyV was in 2008 found in a rare but aggressive skin tumor  MCC (6). HPyV6 and HPyV7 were in 2010 identified in the skin swabs from healthy persons (15), and TSPyV in the lesion of a rare skin disease  TS (9). HPyV9, phylogenetically close to the B lymphotropic polyomavirus (LPyV) of African green monkey, was in 2011 detected in the blood and urine of asymptomatic kidney transplant recipients (16). Subsequently, the following three were discovered sequentially: HPyV10/MWPyV/MXPyV in stool samples of children with diarrhea and anal condylomas (17-19), STL polyomavirus (STLPyV) in feces of diarrheal as well as healthy children (20), and HPyV12 in organs of the digestive tract (21).

1.1 Classification and virion properties

In the past, polyomaviruses were classified as members of the genus Polyomavirus in the family Papovaviridae, which included papillomaviruses, polyomaviruses, and the simian vacuolating agent 40 (SV40). In 2000, polyomaviruses were separated into an independent family, Polyomaviridae, which contained a sole genus including different species. Since 2011, by the criteria of host range and genetic relationship, polyomaviruses have been grouped into three genera: two mammalian virus genera named Orthopolyomavirus and Wukipolyomavirus, and an avian virus genus named Avipolyomavirus (22, 23). The genus Wukipolyomavirus comprises viruses infecting humans: WUPyV, KIPyV, HPyV6 and HPyV7, among which KI polyomavirus is the type species (22). The genus Orthopolyomavirus, with Simian virus 40 serving as type species, contains human (BKPyV, JCPyV, MCPyV and TSPyV) and other mammalian polyomaviruses (22, 23). Polyomavirus particles are non-enveloped and composed of icosahedral capsids of 40- 45nm in diameter. The capsids contain 72 capsomers in a skewed lattice arrangement and one pentameric capsomer consists of five VP1 molecules associated with one molecule of either VP2 or VP3 (24, 25). The VP1, the major capsid protein, constitutes over 70% of the total content of virus particles. The VP2 and VP3 are thought to help with virus entry, and to function in the encapsidation of viral genomes in the formation of virions (22, 23).

The nonenveloped virions have a buoyant density of 1.34-1.35g/cm3 in CsCl gradients, and are resistant to ether, acid and heat treatment (1, 23, 25, 26).

Year of Names Associated diseases Source of isolation References Discovery JCPyV or JCV 1971 PML Brain 11 PVAN, BKPyV or BKV 1971 Urine 12 haemorrhagic cystitis Nasopharyngeal KIPyV 2007 None 13 tissue Nasopharyngeal WUPyV 2007 None 14 tissue MCPyV or MCV 2008 MCC MCC lesion 6 HPyV6 2010 None Skin 15 HPyV7 2010 None Skin 15 TSPyV or TSV 2010 TS TS lesion 9 HPyV9 2011 None Blood, urine, skin 16 HPyV10/MWPyV/ 2012 None Stool, wart 17,18,19 MXPyV STLPyV 2012 None Stool 20 HPyV12 2013 None Liver 21

Table 1. Human Polyomaviruses. Modified by permission from DeCaprio et al. 2013 (27).

1.2 General aspects of HPyV life cycle

Polyomaviruses can infect their natural hosts to commence a productive life cycle that results in nuclear DNA replication, assembly of the viral capsid, and release of virions by cell lysis (1). HPyVs attach to the host cell by the interaction of VP1 with receptors on the host cell membrane, and after attachment, enter the cell by endocytosis through various pathways (1). For example, JCPyV attaches to the cell through sialic acid and 5HT2A serotonin receptor, and enter the cell through the clathrin-mediated pathway, whereas BKPyV interact with sialic acid of gangliosides on the cell membrane and depends on the caveolin-mediated pathway to enter the cell (28-32). Recently, MCPyV VP1 has been shown to bind to heparan sulfate of glycosaminoglycans and sialic acids on ganglioside GT1b sequentially for infectious entry (33, 34). After entry, polyomaviruses are transported through the cytosol to the nucleus, where the early transcription occurs (1, 35). The early-gene product T antigen initiates replication of viral genome and transcription of late genes. After translation, the late-gene products VP1, VP2 and VP3 are translocated into the nucleus, and assemble with viral genomes into virions. The virions are finally released from the cell probably by cell lysis (1, 36-38). Polyomaviruses can establish in non-permissive cells a non-productive infection that results in viral early gene expression in absence of viral genome replication. The

interaction between viral T antigens and tumor suppressor proteins will lead to abnormal cell cycle progression, the result of which is oncogenic transformation (39). In permissive cells, cellular transformation by T antigens rarely occurs because the release of virions leads to cell death. However, in non-permissive cells, T antigen expression can induce oncogenic transformation (36). On the other hand, if viruses undergo mutations resulting in their disability to complete lytic replication, oncogenesis may occur (1). Among HPyVs, BKPyV and JCPyV are integrated into the chromosomal DNA by recombination, and develop tumors in their non-natural hosts (2, 3, 36). MCPyV in MCC presents the first evidence that an HPyV is integrated into the natural host genome and associated with a cancer (6).

1.3 General aspects of HPyV infection and diseases

HPyVs, as a rule, establish persistence or latency in humans after primary infection, and in some rare cases, they reactivate and lead to pathological processes (37, 40, 41). It is a common feature among HPyVs that infections with these viruses are ubiquitous in the general population (37). After a short period of viremia following the first exposure, the viruses are thought to escape elimination by the host’s immune mechanisms and establish a lifelong persistent infection in the human body (37). The sites of persistence are not clearly known, but HPyV DNA is shown in a number of human organs. JCPyV and BKPyV DNA are detected in tonsils, urogenital system, colon, lymph nodes, bone marrow, blood cells and the central nervous system (42). Among asymptomatic individuals, low- level secretion of BKPyV and JCPyV occurs in the urinary tract, suggesting that the viral genome persisting in kidney and urogenital system intermittently replicates (42-45). HPyVs can reactivate from latency, actively replicate and cause disease. During this, virions can be visualized and viral capsid proteins detected in affected tissues (37). Different from the asymptomatic low-level virus replication described above, the viral genome synthesis under reactivation is usually at a high level and associated with the formation of virions. Under immunosuppression, JCPyV can reactivate in the human brain and cause PML  a disease associated with HIV/AIDS and as side effect of immunomodulatory therapies (40, 46, 47). In the case of PML, active replication of JCPyV in oligodendrocytes destroys myelin-producing cells, resulting in brain lesions and death (47). Similarly, BKPyV can reactivate from latency in immunocompromized individuals, and cause nephropathy. In renal transplant recipients, BKPyV replication in their kidney cells causes polyomavirus-associated nephropathy (PVAN)  a leading cause of allograft failure associated with immunosuppressive therapies (48-50). In addition, BKPyV replication in the urothelium causes hemorrhagic cystitis (51). Notably, JCPyV replication is observed in the brain of almost all of PML patients, and BKPyV replication is found in kidneys of 80% of renal transplant patients with 10% progressing to PVAN (52, 53). Moreover, the viremia of JCPyV and BKPyV rarely occurs in healthy individuals after primary infection, but appears in 17% of PML patients and up to 100% of PVAN patients, suggesting that an active viral replication bypassing the control of host immunity is associated with the progression of these diseases (40, 48, 54, 55).

To date, among the recently found HPyVs, KIPyV and WUPyV are predominantly found in nasopharyngeal aspirates, bronchoalveolar lavages, and feces from pediatric patients (56-58). Though initially found in nasopharyngeal aspirates from children with respiratory infections, these two viruses have been later shown to lack a clear etiological link to respiratory disease (13, 14, 58, 59). The shedding of KIPyV and WUPyV, similar to that of JCPyV and BKPyV, increases with immunosuppression (60, 61). MCPyV, considered to participate in the development of MCC, has been detected in approximately 75% MCC tumors with clonal integration, whereas the virus is also present in the healthy skin, and other tissues (7). HPyV6 and 7, predominantly present in cutaneous swabs from healthy persons, have not been reported to be associated with any diseases (15, 62). TSPyV is mostly present in the inner-root sheath cells of hair follicles in TS lesions, and thought to play a role in the pathogenesis of TS disease (9, 10). HPyV 9 is occasionally observed in plasma, urine and skin swabs (16, 63, 64), but no linkage to any disease has been established so far. There are few reports on HPyV10, STLPyV, and HPyV12, of which the first two have been occasionally seen in stools and the latter might be related to the digestive tract (17-21). Their disease association remains unclear.

2. Merkel cell polyomavirus

2.1 Merkel cell carcinoma

MCC is a highly aggressive skin cancer of neuroendocrine origin, with high rates of recurrence and metastasis (65). The tumor, named after its possible cell of origin, is thought to arise from mechanoreceptor Merkel cells in the basal layer of the epidermis (66). Some studies argued for a stem cell as the tumor origin, since some MCCs showed epithelial and sarcomatous features (67). The controversy seems to be explained by a recent finding that demonstrates mammalian Merkel cells developing from epidermal stem cells (68). MCC has incidence rates of 0.18-0.41 per 100 000 persons, occurs predominantly among Caucasians between 60 and 85 years of age, and the incidence rate is about twice as high in males than in females (67). Because of the improvement of diagnostic techniques, the age-adjusted incidence rate seems to have tripled during the past 20 years (69). The overall 5-year relative survival rate is about 60% (65), and among patients with local, nodal, and metastatic disease, the rates are 64%, 39% and 18%, respectively (70). The tumor arises predominantly in the skin of head and neck, but can also occur at extracutaneous sites including salivary glands, nasal cavity, lip, lymph nodes, vulva, vagina and esophagus (67). The clinical features of MCC are summarized as AEIOU: asymptomatic/lack of tenderness, expanding rapidly, immune suppression, older than 50 years, and ultraviolet exposed site on a person with fair skin (71). MCC has been observed most often in sun-exposed skin of Caucasians, pointing to ultraviolet exposure as a major risk factor for tumorigenesis (69, 71). In a study of 1380 psoriasis patients treated by ultraviolet phototherapy, three developed MCCs, an incidence

rate 100 times higher than expected in the general population (72). Another significant risk factor is immunosuppression. MCC occurs more frequently in severely immunosuppressed populations, such as patients with HIV/AIDS and leukemia, and after organ transplantation (73-77). Patients with AIDS have a 13-fold increased risk for MCC, compared with the general population (75). The MCC incidence is also associated with the occurrence of other cancers, such as squamous cell carcinomas, basal cell carcinomas and malignant melanoma, suggesting the existence of shared risk factors among these cancers (78).

2.2 Discovery of MCPyV

Using the digital transcriptome subtraction technique, Feng et al. in 2008 discovered a previously unknown human polyomavirus in MCC (6). This approach combines high- throughput cDNA sequencing and in silico methods to sequentially exclude human sequences from a diseased tissue sample transcriptome (79). In the case of MCPyV, cDNA libraries were generated with random RT-PCR from MCC tumors and sequenced. Known human sequences were subtracted from MCC cDNA libraries by the help of specific software, and some transcripts were detected with similarity to LPyV and BKPyV. By viral genome walking, two complete MCPyV genomes MCPyV350 and MCPyV339 were obtained. Meanwhile, MCPyV DNA was observed in eight of ten MCCs, and in six of them the viral genome was integrated into the tumor genome (6). Importantly, the patterns of integration suggest that MCPyV infection and integration occur preceding the clonal expansion of the tumor cells.

2.3 MCPyV genome structure and function

MCPyV has similar genome structures to those found in other polyomaviruses. The genome of MCPyV is approximately 5.4kb in length, and composed of three functional parts: the early and late regions with a non-coding control region (NCCR) in between (Figure 1). The early region encodes non-structural proteins (tumor antigens), which are involved in viral DNA replication and gene expression. The late region encodes structural proteins, which assemble into viral particles. The NCCR contains a single origin of DNA replication and the transcriptional control elements. The early and late regions, without overlapping in their sequences, bidirectionally generate transcripts from the opposite strands of DNA (1, 22, 80). The early antigens large T antigen (LT), small T antigen (sT) and 57kT antigen are generated from a common pre-mRNA via alternative splicing, and thus share identical N-terminus but possess different C-terminal DNA sequences (8). The agnoprotein, which exists in JCPyV and BKPyV, has not been seen in MCPyV or in any other novel HPyVs (22, 81)

Figure 1 The genome organization of MCPyV. Reproduced by permission from Feng et al. 2008 (6).

The large T antigen MCPyV LT retains conserved domains that are essential for replication of viral genome and regulation of viral gene expression. These domains include the origin-binding domain, helicase/ATPase domain, retinoblastoma protein (pRb) binding domain, and the DnaJ domain (7, 8, 82-84). LT plays an important role in initiation of viral replication. In the case of SV40, LT binds to GAGGC motifs at the viral origin and recruits cellular replication proteins to initiate viral replication (85-89). The ATPase activity of LT enhances the assembly of the LT-origin complex, and helicase activity helps unwind the double-stranded DNA (88, 90). With MCPyV, a similar function of LT has been demonstrated for viral replication. MCPyV harbors GAGGC-like pentanucleotides that serve as LT binding sites (6, 8, 82, 91). Importantly, MCPyV tumor-derived LT, without the origin-binding domain or helicase domain, reduces LT-origin assembly and fails to initiate MCPyV origin replication (8, 82). The DnaJ domain, located at the very N-terminus of LT, acts as a co-chaperone regulator, binding chaperone Hsc70 and stimulating its ATPase activity. The DnaJ-Hsc70 chaperone is able to mediate rearrangements of the initiation complex leading to the stimulation of viral DNA replication (92-95). An intact DnaJ region has been shown to be compulsory for MCPyV replication (82). The pRb family proteins form a complex with E2F transcriptional factors to repress genes that are associated with cell cycle progression, DNA replication, DNA repair, differentiation, and apoptosis (96, 97). The release of pRb from E2F can be induced by DnaJ domain-mediated hydrolysis of Hsc70 ATP, and the LT intrinsic pRb-binding LXCXE motif. The release of pRb from E2F leads to transactivation of E2F-responsive genes and thus promotes cell proliferation (88, 92, 98). Both the pRb-binding LXCXE motif and DnaJ domain are conserved in MCPyV LT (8, 84). An intact pRb-binding motif

is required for promoting the proliferation of MCC cells (99). Consequently, the targeting of LT with pRb induce survivin expression, which is critical for the survival of MCPyV- positive MCC cells (100). The tumor suppressor protein p53 binding domain, which has been found in the LT of SV40, BKPyV and JCPyV (80), is absent from MCPyV LT (8, 101). p53 is responsible for repressing abnormal cellular proliferation, stimulating apoptosis and regulating DNA repair (102, 103). The binding of LT to p53 leads to abnormal cell proliferation (104). The lack of a p53-binding site in MCPyV LT suggests factors other than p53 inactivation involved in MCPyV tumorigenesis. Bub1 is a mitotic spindle checkpoint protein kinase responsible for correct chromosome segregation (105, 106). Neither is the Bub1 binding site, another LT targeting motif conserved in most polyomaviruses, found in MCPyV (80). MCPyV LT contains a nuclear localization signal (107), and the nuclear localization of LT is observed in cell cultures as well as in the MCC tumors (108). In addition, MCPyV contains a unique region in the T antigen, which is called the “MCPyV-Unique Region”. This unique region interacts with the cytoplasmic protein Vamp6 and redistributes it to the nucleus, preventing Vamp6 function in lysosome clustering (83).

The small T antigen MCPyV sT shares with LT approximately 80 N-terminal amino acids and possesses a unique C-terminus (8). Compared to LT, larger quantities of sT are detectable in MCC tumors (109). With SV40, LT is a potent in vitro transforming oncoprotein, whereas with MCPyV, sT functions in the cellular transformation (82, 109). MCPyV sT expression is essential for cell transformation but not compulsory for tumor cell growth and survival (109, 110). The sT induces transformation not through interacting with phosphatase 2A, but rather by promoting hyperphosphorylation of 4E-BP1 (109). The 4E-BP1 is a downstream target of the mTOR signaling pathway, which is critical for tumor cell survival in many types of genetic cancers (111).

57kT The MCPyV early transcript encodes an additional protein of Mr57000, named 57kT. This protein, identical to LT except for a deletion in the C-terminal region, contains a pRb- binding site but eliminates the origin-binding domain and the DNA helicase/ATPase domain (6, 83). The biological function of 57kT is unclear, but the protein is thought to be an analog of the SV40 17 kT antigen, which appears to function in the process of cellular proliferation (112, 113).

Capsid proteins The late region of HPyV genome, such as SV40, encodes structural proteins required for viral assembly, including a major capsid protein VP1 and two minor proteins VP2 and VP3. These proteins are generated from alternatively spliced mRNAs. The latter two are overlapping, with the entire VP3 sequence located at the C-terminus of VP2. The translation of VP3 initiates at a highly conserved motif within VP2 open reading frame (6,

114). However, phylogenetic analysis has shown that MCPyV lacks the conserved VP3 N- terminal motif, indicating the absence of a functional VP3 protein in this virus (115, 116). Correspondingly, in MCPyV-infected cells, VP3 is neither detectable in cell lysate nor present in MCPyV native virions (116). It has been demonstrated with SV40 that the viral VP2 and VP3 are transported into the nucleus depending on VP1, and probably act in trafficking of incoming virus to the nucleus in the early phase of infection (98, 117, 118). In the case of MCPyV, the viral VP2 is shown to be compulsory for efficient MCPyV infectious entry into certain cell types (116).

MicroRNA Viral-encoded microRNA has been detected for MCPyV (119, 120). A microRNA MCPyV-miR-M1-5p is expressed at low levels in 50% of MCPyV-positive MCCs (119). MCPyV microRNA is antisense to early transcripts and can negatively regulate LT expression late in infection (119). This downregulation of LT levels may be involved in immune evasion of the virus, compared to the function of SV40 microRNA (121).

2.4 Infection and immunity

MCPyV infection is ubiquitous in the general population, as MCPyV specific antibodies are detected in a large proportion of healthy persons. The seroprevalence of MCPyV is 60- 80% among adults (122-126). In contrast to the normal population in MCC patients with MCPyV-positive tumors the seroprevalence of MCPyV is up to 100% and these patients also have higher antibody titers than healthy individuals (115, 126-128). One possible explanation is that the immunosuppressed status of MCC patients allows reactivation of MCPyV, leading to a stage of viremia preceding tumor development. The other is that MCC patients have been challenged with a high virus burden to stimulate antibody responses. It has been shown that the antibody levels correlate with the presence of viral DNA in the skin irrespective of disease, and that MCC patients have high MCPyV viral load in their unaffected skin (129, 130). However, VP1 antibody levels are not associated with tumor viral loads, but are associated with a better progression-free survival in MCC patients (128). In a subset of MCC patients, antibodies against LT are detected (131). The levels of LT antibodies correlate strongly with the presence of MCPyV DNA and the expression of T antigens in MCC tumor cells (131). Interestingly, elevated LT antibody levels are associated with progression of the disease, suggesting that LT antibody cannot protect MCC patients against cancer progression, but may serve as a clinically useful indicator of disease status (131). With MCPyV, cellular immunity apparently plays an important role in the control of the virus and of MCC formation. Tumor infiltrating lymphocytes are present in MCPyV- positive tumors (132-134), and high counts of intratumoral T cells contribute to favorable survival in MCC patients (132, 133). When implanted into immunodeficient mice, MCC tumors fail to grow unless T cells in the tumor grafts are depleted, suggesting that tumor- specific T cells are essential in controlling tumor growth (135). Circulating T cells and tumor-infiltrating T cells can recognize specific peptides from MCPyV LT antigen and

capsid protein, and circulating T-cell responses are detectable regardless of MCC (134). MCPyV-specific T-cells against VLPs are found in blood from healthy persons. Compared to MCPyV-seronegative individuals, MCPyV-seropositive subjects show significantly stronger IFN-γ responses, suggesting that T-cells play a critical role in the control of MCPyV in the general population (136). Obviously, as host immune functions are important in the progression of MCC, boosting host immune response might be a novel therapeutic or prophylactic option for this cancer.

2.5 Molecular epidemiology

MCPyV, since its discovery, has been shown in MCCs worldwide. Overall, MCPyV is detectable in 75% of MCCs (7), with integration into human genome and tumor-specific mutations (6, 137-140). Using a more sensitive detection method, a recent study implies that the rate of MCPyV detected in MCC must be higher than those reported (141). MCPyV is found in MCC tumors irrespective of the sites of origin (142), and as shown with MCC patients in their unaffected skin, nasal swabs, urine samples, and peripheral blood mononuclear cells (PBMCs) (137, 139). Among these latter samples, only PBMCs have been shown with integrated forms of MCPyV, probably due to circulation of metastatic cells (139). Nevertheless, in MCC patients, compared to non-MCC tissues, MCC tumors generally show higher MCPyV prevalence and higher viral load (137, 143- 145). MCPyV is shed predominantly in the epidermis of normal skin (15, 130, 137, 143, 145). In one study, MCPyV DNA was found in cutaneous swabs of all participants (130). In another study, MCPyV genomes of wild type (without mutation and integration) were recovered from 40% of skin swabs of healthy adults (15). The viral load does not differ in the skin biopsy samples from MCC and non-MCC individuals (137), but are of higher level in cutaneous swabs than in skin biopsies from MCC patients (130). Besides in skin, MCPyV has been detected in various tissues, including respiratory tract and its secretions (146-148), tonsillar biopsies (148), lung tissues (143, 144, 149), oral mucosa (150), saliva (130, 143), anogenital samples (143, 145), digestive tract (143, 151), lymphoid tissue (61, 152), urine (153, 154) and blood (144, 148, 154-156) (Table 2). The level of MCPyV DNA is higher in the digestive system than in the respiratory and genitourinary systems (143). HIV patients have MCPyV detected in normal tissues more frequently than healthy persons have (157, 158). However, renal transplant patients show MCPyV shedding in urine at low level regardless of immunosuppression (159). Notably, viral mutation and integration have not been reported so far in any type of non-tumor tissues of individuals without MCC.

MCPyV Study Sample materials DNA Schowalter et al. (15) Forehead skin swabs of 35 volunteers 40% Foulongne et al. (130) Cutaneous swabs of 25 healthy volunteers 100% Loyo et al. (143) 21 non-MCC skin samples 48% Wieland et al.(145) 34 healthly skins from immunocompetent patients 24% Wieland et al. (145) 14 plucked eyebrow hairs from 14 HIV+ patients 50% 526 respiratory tract samples from patients with Bialasiewicz et al. (146) 1.3% upper or lower respiratory tract symptoms Goh et al. (147) 635 nasopharyngeal aspirate samples 4.2% 140 nasopharyngeal aspirate samples from Kantola et al. (148) 2.1% wheezing children Kantola et al. (148) 229 tonsillar tissues from asymptomatic children 3.5% Loyo et al. (143) 43 samples of lung 26% Fukumoto et al. (144) 17 pneumonia samples 18% 66 transbronchial biopsies from lung transplant reci Bergallo et al. (149) 33.8% pients Foulongne et al. (130) buccal mucosa swabs of 42 pateints 4.8% 120 mucosal samples (anal, penile, oral) from HIV+ Wielande et al. (145) 31% men Loyo et al. (143) 10 saliva from patients without cancer 100% Loyo et al. (143) 10 samples of bladder 60% 27 anogenital samples from an immunodeficient Wielande et al. (145) 74% patient Loyo et al. (143) 37 samples of liver tissue 54% Loyo et al. (143) 50 samples of colon tissue 18% Pancaldi et al. (155) 60 buffy coats of healthy blood donors 22% 840 sera from immunocompromised and Kantola et al. (148) 0.1% immunocompetent individuals Fukumoto et al. (144) 23 serum samples from HIV patients 39% 621 serum samples from 394 hospitalized elderly Sadeghi et al. (156) 10% individuals Shuda et al. (108) 21 PBMC from HIV+ patients 14% 29 normal lymph nodes from patients with Toracchio et al. (152) 10.3% metastatic carcinoma or melanoma 55 lymphoid tissue samples (lymph node and Sharp et al. (61) 1.8% spleen) from immunocompetent individuals

Table 2. MCPyV DNA prevalence among non-tumor samples.

MCPyV has been detected in a number of tumor types other than MCC, particularly in skin cancers. MCPyV is detectable in 15-35% of squamous cell carcinomas, which is the

second most frequently diagnosed skin cancer and associated with immunosuppression (150, 160-168). In one study, MCPyV has been identified in 52% of squamous cell carcinomas from immunocompromised and in 25% of squamous cell carcinomas from immunocompetent individuals (160). MCPyV is also present in basal cell carcinomas (162, 165), adenocarcinomas (163, 164), large-cell carcinomas (163), pleomorphic carcinomas (163), lymphomas (152, 169-171), colon cancers (151), non-small cell lung carcinomas (172), and Kaposi's sarcomas (173). Viral DNA integration and mutation were reported in a few cases, such as squamous cell carcinomas and adenocarcinomas (163), but the detection rates seem incidental. However, in one study, six of 19 MCPyV-positive chronic lymphocytic leukemia cases showed a truncating MCPyV large T antigen deletion, suggesting a potential role of MCPyV in a subset of chronic lymphocytic leukemias (174).

2.6 Clinical aspects on MCPyV-positive and -negative MCCs

A majority of MCCs (approximately 75%) have MCPyV involved in their pathogenesis, whereas those tumors negative for the virus, must stem from other etiological pathways. Accordingly, a number of MCC research groups have investigated the clinical differences between the tumors with and without MCPyV. Compared to virus-negative tumors, as a general conclusion, MCPyV-positive MCCs are more frequently present on the extremities such as limbs, and less often on the trunk (175-177). Several studies have demonstrated that MCPyV-positive tumors are less frequently present in patients with metastases at diagnosis (139, 175, 176), though this conclusion was not reached in a subsequent study (177). In a small-size study of 26 MCC cases, a Japanese group observed more irregular nuclei and more abundant cytoplasm in virus-negative than in virus-positive tumors (178). Recently, this research group further proved this observation with additional 43 MCC samples (179). However, with 144 MCC cases originating from Australia and Germany, Schrama et al. did not observe significant differences between MCPyV-positive and -negative tumors concerning histological type, microanatomic compartment, mitotic rate and expression of p16, p53, RB1, and Ki67 (177). The reports on the effect of MCPyV status on MCC outcome seem contradictory. Some investigators found patients with MCPyV-positive tumors associated with better survival than those with MCPyV-negative ones (175, 176, 180, 181), whereas others did not observe such association (168, 177, 182). This discrepancy might be due to different composition of study materials, as MCPyV-positive MCCs are likely composed of two subgroups of tumors with distinct clinical features. Together with MCPyV-negative tumors, MCCs can be divided into three distinct groups based on the MCPyV status, viral load and T-antigen expression, i.e. MCC with high, low and no viral detection (168, 175, 180, 183, 184). MCC tumors with high MCPyV detection show correlations between viral load, pRb expression, LT expression, absence of p53 expression, and better clinical outcome (180, 183, 184). However, MCC tumors with low or no MCPyV detection are predominant in p53 expression but lack LT and pRb expression, suggesting a different route involved in the MCC development (180, 184, 185). Tuoze et al. postulated that tumors without or with MCPyV at a low viral burden, developing from a similar etiology, compose a group of MCCs with a progressive

phenotype, while MCCs with a high MCPyV burden, presence of LT expression, and high level of MCPyV antibodies, compose the other group with a less aggressive phenotype (128). Recently, a study identified a new type of MCPyV-positive MCC, in which the cell growth is T-antigen independent. This finding, too, supports that alternative pathways other than the T antigen-based may be associated in the formation of certain MCPyV- positive MCC tumors (186). Moreover, this study suggests that some MCC tumors may only need MCPyV for tumor initiation and lose the virus during tumor progression, comprising the MCPyV-negative MCCs with a progressive phenotype (186).

2.7 Association of MCPyV with pathogenesis of MCC

The criteria proposed by zur Hausen for defining a causal role for an infection in cancer are (187): "(i) Epidemiological plausibility and evidence that a virus infection represents a risk factor for the development of a specific tumor. (ii) Regular presence and persistence of the nucleic acid of the respective agent in cells of the specific tumor. (iii) Stimulation of cell proliferation upon transfection of the respective genome or parts therefrom in corresponding tissue culture cells. (iv) Demonstration that the induction of proliferation and the malignant phenotype of specific tumor cells depend on functions exerted by the persisting nucleic acid of the respective agent." A number of evidences support an association between MCPyV and the pathogenesis of MCC. (i) MCPyV has been detected in a large proportion of MCCs regardless of the geographical distribution, and the tumor-associated viral genome is integrated into the host cells by a monoclonal pattern (6, 139, 140). (ii) The metastatic tumors reveal the same monoclonal pattern of integration as the primary tumor, suggesting that viral integration is an early event in tumorigenesis preceding tumor cell expansion and tumor progression (6). (iii) The integrated MCPyV harbors specific mutations in the LT gene resulting in truncated LT expression, while MCPyV genomes found in non-MCC tissues are wild-type (6, 139, 154, 188). (iv) MCPyV LT contains motifs critical for virus-positive tumor cell replication and survival in the case of viral integration (8, 99, 108, 110, 184, 189, 190). (v) MCPyV sT functions in cell transformation (109). However, epidemiological investigations show that MCPyV presents as a constituent of normal skin flora and infection with this virus is widespread in the general population. To answer how this ubiquitous viral infection eventually causes a rare skin cancer, virologists have proposed a model, wherein a series of rare mutagenic events associated with loss of immune control are involved in MCC formation (7, 191, 192). As shown in Figure 2 (191), MCPyV infection occurs asymptomatically in the early childhood (126, 193, 194). UV radiation or other environmental factors may induce integration of MCPyV into the host cell genome (6). If the wild-type MCPyV T antigen induces DNA to replicate from the origin of replication of the integrated virus, the host-cell DNA strands break, which would lead to cell death. On the other hand, if a second mutation is induced to generate a truncated T antigen without replication capacity but with oncogenicity (195), the nascent tumor cell survives (8). The persistent expression of T antigen drives the pathogenesis of MCC (99, 109, 189). In addition, loss of immune control through ageing,

AIDS or immunosuppressive therapy, allows for an increase of MCPyV burden in the skin as well as T-antigen expression (115, 126, 129, 131).

Figure 2 The molecular evolution of MCPyV in the pathogenesis of MCC. Reproduced with permission from Moore & Chang 2010 (191).

Notably, T antigen as oncoprotein plays a key role in MCC pathogenesis. MCPyV LT expressed in MCC tumors is uniformly truncated, retaining the pRb binding or LXCXE motif but eliminating the DNA binding and helicase domains (6, 8, 192, 196). In MCC cell lines and xeno-transplantation models, knockdown of this viral LT leads to growth inhibition while restoration of LT expression leads to rescued cell growth, indicating that LT expression is a critical factor in the formation of MCPyV-positive MCC tumors (99, 189). Moreover, an intact pRb-binding site is required for LT to rescue cell growth (99). In a recent study, expression of LT C-terminal 100 residues showed inhibitory activities in growth of several cell types, suggesting that the deletion of the C terminus of LT may also result in the loss of a growth-inhibitory function intrinsic to this region (101). The sT is UV-inducible, and capable of transforming human cells by targeting 4E-BP1 (109).

3. Trichodysplasia-spinulosa associated polyomavirus

3.1 Trichodysplasia-spinulosa disease

TS is a rare skin disease only observed in severely immunocompromised patients, such as solid organ transplant recipients and patients with leukemia and lymphoma (10, 197, 198). The disease is characterized by the appearance of follicular-based papules and keratin spikes on the skin, sometimes along with alopecia of the eyebrows and eyelashes (197, 199). The lesions appear most often in the face, especially the nose, eyebrows and auricles, but also in other parts of the body (200). The histological features include abnormal hair follicle maturation with the follicles distended and enlarged, and high production of inner root sheath cells with excessive amounts of trichohyalin (9, 197, 199). In 1999, Haycox and co-workers first introduced the term "trichodysplasia spinulosa" to describe this skin disorder in a renal/pancreas transplant patient with immunosuppressive regimen (199). They also proposed a viral origin for TS, because by electron microscopy (EM) honeycomb-arranged viral particles with papovavirus morphology were seen in the affected follicle clusters (199). Subsequent reports confirmed the presence of polyomavirus-like particles, and the entity was termed "virus-associated trichodysplasia spinulosa" (201-206).

3.2 Discovery of TSPyV

In 2010, using rolling-circle amplification (RCA), van der Meijden et al. identified in TS lesions a previously unknown human polyomavirus  designated TSPyV or TSV (9). The technique employed the DNA polymerase of bacteriophage phi29, an enzyme with strand displacement and proofreading activities. With random primers, the polymerase can efficiently amplify circular DNA molecules from complex biological samples (207). For TSPyV identification, the DNA extracts of the TS lesions were subjected to RCA, followed by restriction enzyme digestion. Based on the sizes of digestion products, the amplicon was estimated to be about 5000 bp in length, a size typical for the genome of known polyomaviruses. Through sequencing of the RCA fragments, the entire TSPyV genome was obtained (9). With a close phylogenetic relation to Bornean orangutan polyomavirus, TSPyV is categorized into the genus Orthopolyomavirus (9, 22, 23).

3.3 TSPyV genome

TSPyV has a double stranded DNA genome of 5232 base pairs (Figure 3) (208). Based on an analysis of the putative viral sequence, the components commonly found in polyomaviruses have also been revealed in the TSPyV genome  two early genes (LT and sT) and three late genes (VP1, VP2 and VP3) with NCCR in between. The early and late genes reside on the opposite strands, with an organization similar to that of other polyomaviruses. The putative LT-binding sites, origin and TATA box are present in NCCR. The LT sequence is characteristic, including DnaJ-domain, pRb-binding motif, the

origin-binding domain, ATPase/helicase domain, and a nuclear localization signal. The putative TSPyV sT contains PP2A-binding motifs (9). The potential functions of these motifs and domains are similar to those described for MCPyV.

Figure 3 The genome organization of TSPyV. Reproduced with permission from Kazem et al. 2013 (208).

3.4 Epidemiology and immunity

The presence of TSPyV in TS disease has been confirmed by a number of case reports after the discovery of TSPyV. Extracted from TS lesions, the TSPyV sequences were amplified by PCR, and viral particles observed by EM (197, 209). Notably, by analyzing all available TS cases worldwide, Kazem et al. showed the relation of TSPyV infection with TS disease (10). In their study, TSPyV DNA was detectable in the lesional samples of all TS cases, whereas only in 2% of forehead skin swabs from healthy controls (10). The viral load in the TS lesions is 1,000,000-fold higher than that in the forehead skin of TS patients, and 10,000 – 100,000-fold higher than that in asymptomatic controls, suggesting that a high rate of TSPyV replication occurs in TS-affected skin (10). Moreover, TSPyV VP1 was present in all lesional TS samples, and restricted to the nuclei of the inner root sheath cells that over-expressed trichohyalin. However, the VP1 was absent from the non-lesional samples (10), indicating that an active TSPyV infection is involved in the development of TS. Furthermore, in several studies, TS patients´ conditions improved, when treated with anti-viral drugs, for instance cidofovir, which can selectively inhibit viral DNA polymerase and thus suppress viral replication (9, 201) . So far, TSPyV has been searched in various non-TS samples including skin swabs, plucked eyebrows, nasopharyngeal swabs, feces, urine, blood, and cerebrospinal fluid (208). However, the viral DNA is only present in less than 5% of the samples from asymptomatic individuals, regardless of whether they are immunosuppressed or immunocompetent (208). TSPyV DNA has been shown in 4% of plucked eyebrows, 2%

forehead skin swabs, 3% nasopharyngeal swabs, 3% fecal and 1% urine samples (Table 3) (9, 10, 16, 208, 210). Viral DNA has, however, not been detected in blood and cerebrospinal fluid (16, 156, 208). In the case report by Fischer et al., TSPyV DNA was found in a renal allograft biopsy specimen of a TS patient (211). In two independent studies investigating TSPyV in cancers, TSPyV DNA was absent from various skin cancers such as pilomatricomas, basal cell carcinomas and squamous cell carcinomas (165, 212). In contrast, a seroepidemiological study showed TSPyV infection to be ubiquitous in the general population (208). In addition, T-cell immunity against TSPyV has been studied in healthy persons. Interestingly, using a VP1-VLP antigen, the TSPyV T-cell responses did not differ between virus-seropositive and seronegative subjects (213).

TSPyV Study Sample materials DNA van der Meijden et al. Plucked eyebrows from 69 immunosuppressed 4% (9) renal transplant patients Plucked eyebrows from 81 renal transplant Kazem et al. (208) 2% patients Kazem et al. (10) Forehead skin swabs of 249 healthy individuals 2% Siebrasse et al. (210) Nasopharyngeal swab from 32 transplant patients 3% Siebrasse et al. (210) Fecal samples from 32 transplant patients 3% 160 fecal samples from pediatric patients with Kazem et al.(208) 1% gastroenteritis 38 fecal samples from pediatric patients with Scuda et al. (16) 0% gastroenteritis 179 urine samples from immunocompromised Scuda et al. (16) 1% patients 17 urine samples from immunocompetent Kazem et al. (208) 0% individuals 88 serum samples from immunocompromised Scuda et al. (16) 0% patients 621 serum samples from 394 hospitalized elderly Sadeghi et al. (156) 0% individuals 19 serum samples from immunocompetent Kazem et al. (208) 0% individuals Scola et al. (165) 193 samples of various skin diseases 0% Kanitakis et al. (212) 10 pilomatricoma samples 0%

Table 3. TSPyV DNA prevalence among non-TS samples. Reproduced with permission from Kazem et al. 2013 (208).

4. Serodiagnosis

Serodiagnosis is the process that employs serological testing for diagnosis of diseases. The serodiagnostic tools entail the detection and quantification of a specific antibody or antigen usually in blood serum samples or in other body fluids such as cerebrospinal fluid and saliva. For infectious diseases e.g. caused by viruses, serodiagnostic methods can be used to indicate an acute primary infection, to distinguish stages of infection (recent or reactivated versus past), or in some cases to assess the efficacy of vaccination or antiviral therapy. The classical methods for serological testing include precipitation reaction, complement-fixation, hemagglutination inhibition, immunofluorescence, radioimmunoassay, Western blot, and enzyme immunoassay (EIA) (214, 215). The following chapters focus on the application of antibody detection in the serodiagnosis of viral infections.

4.1 Antibody response in viral infection

When the host encounters a viral infection, a series of immune responses must in order to defend the host against the invading pathogen. For instance, B-cells produce IgM and IgG antibodies, which can neutralize the pathogen and serve as markers for the progression of infection. Figure 4 shows the classical pattern of antibody response to virus infections. Upon primary exposure to the pathogen, a lag phase precedes the appearance of antibodies. In the course of primary infection, the first isotype of antibody that appears is IgM, the level of which increases rapidly to a plateau and thereafter declines to undetectable levels in the following weeks or months. Following IgM, IgG appears due to isotype switching, and most often persists in lifelong circulation (216). Once the host encounters the same pathogen again, a secondary immune response occurs, sometimes resulting in a detectable boost of IgG level, however, as a rule in the absence of IgM (214, 217). Therefore, during an acute primary viral infection, it is possible to observe a seroconversion of both virus- specific IgM and IgG antibodies. Notably, one essential feature of the immune response against infection is that the avidity of antibodies increases from low to high because of the processes of somatic mutation and selection (218-220). Due to pathogen clearance, B-cell clones of higher avidity can compete out those of lower avidity, resulting in progressive production of higher avidity IgG antibodies (218, 221). Thus, individuals with past immunity usually present with high-avidity IgG antibodies already during the acute phase of the reinfection, whereas patients undergoing a more recent infection show predominantly low-avidity IgG (214).

Figure 4 Typical evolution of antibody response following viral infection.

4.2 EIA for antibody detection

Because of its capacity of large-scale screening, ease of automation, and high sensitivity, EIA is one of the most common serological methods in clinical diagnostic laboratories (214, 215). The method involves an enzyme conjugated with an antibody or antigen reacting with a substrate to produce color changes, therefore allowing quantitative measurement. In serological testing for virus-specific antibodies, three types of EIAs are widely utilized: indirect EIA, IgM-capture EIA, and competitive EIA (222). In an indirect EIA (Figure 5A), specific antibody in human serum binds to viral antigen immobilized on a solid phase. The specific antibody is detected by an enzyme- labeled anti-human immunoglobulin, of which the enzyme reacts with the substrate to generate a color. Depending on the isotype of immunoglobulin, either IgM or IgG can be detected. However, because IgG not only accounts for the majority of immunoglobulins in human serum, but also shows higher affinity than does IgM, the prevailing of virus- specific IgG may interfere with the detection of virus-specific IgM. On the other hand, if rheumatoid factor, an IgM-class autoantibody against IgG, is present, a false positive IgM result may occur. Thus, IgM-capture EIA is preferred for detection of IgM antibodies in clinical samples. In this format (Figure 5B), a repertoire of IgM antibodies are captured out of human serum by an anti-human IgM antibody coated on the solid phase. The specific IgM is detected by binding to the corresponding antigen preparation. Competitive EIA, used to distinguish false positivity, is widely employed when high specificity is required. This assay employs a labeled antibody to compete for binding to immobilized antigen with antibody in the clinical sample (Figure 5C). Alternatively, soluble antigen is used to compete with immobilized antigen for binding to the antibody in the serum (214, 215, 222).

Figure 5 EIA formats.

Testing for IgG avidity may be accomplished by a modified indirect EIA aiming at the distinction between low- and high-avidity bindings. In practice, the reactivity of the serum in the well treated with a protein-denaturing agent such as urea that will disrupt the antigen-antibody binding, is compared with the reactivity of a well without such treatment. If the patient serum has low-avidity IgG, it will be susceptible to the urea treatment. In contrast, the high-avidity IgG is more resistant (223).

4.3 Interpretation of results

As mentioned, a primary infection can be determined by seroconversion of virus-specific IgG or by presence of specific IgM. Usually, a fourfold or more rise in IgG antibody concentration between acute and convalescent samples is considered evidence for an acute or recent infection. Increasingly, to account for issues such as IgM persistence for a long period or short presence, an avidity assay may serve as a more accurate way to determine the time of infection. Detection of high-avidity IgG indicates exposure to the virus in the past, whereas detection of low-avidity IgG indicates a recent infection. Nevertheless, it should be taken into account that the patients´ clinical status may affect the results. The antibody production may be impaired among the immunosuppressed patients or the

elderly, leading to a false negative diagnosis. Transfer of maternal antibodies and cross- reactive antibodies derived from a recent infection by related pathogens, may lead to false information. Overall, essential characteristics of assay performance, referred to as sensitivity and specificity, play important roles in result interpretation. The term sensitivity refers to the proportion of patients with an infection that test positive, and specificity refers to the proportion of individuals without an infection who test negative. With antiviral IgM assays, the specificity is critically important, as the presence of pathogen-specific IgM antibody is considered a marker of ongoing or recent infection and thus affects clinical management (214, 215, 222).

4.4 Biotinylation-based anti-viral IgM EIA

Biotinylation is the process of covalently linking biotin to macromolecules, such as nucleic acid and protein (224, 225). Biotin is water soluble, and due to its small size, can attach to proteins or nucleic acids without changing their natural function (224, 225). The interaction between biotin and avidin/streptavidin shows extremely high affinity and specificity. Once biotin binds to avidin/streptavidin, it is rarely released by extremes in pH, temperature and denaturing reagents (224). Since a number of labeling reagents and methods have been developed for biotin, the technique of biotinylation is widely applied in biomedical assays (224-228). The biotinylation-based anti-viral IgM EIA is essentially an IgM-capture EIA using as antigen biotinylated viral proteins detected by conjugated streptavidin (Figure 4A of Study IV). Our lab has employed this method in detection of anti-viral IgM for human parvovirus B19 (B19V), parvovirus 4 and human bocaviruses (HBoVs) (229-231).

AIMS OF THE STUDY

1. To generate MCPyV and TSPyV virus-like particles (VLPs) in insect cells via the baculovirus expression system

2. To develop for comprehensive serodiagnostic methods of MCPyV and TSPyV infections VLP-EIA-based assays for measurement of the respective IgM, IgG and IgG avidity

3. To determine the MCPyV and TSPyV seroprevalences among Finnish adults and children

4. To determine age at and frequency of MCPyV and TSPyV primary infections among Finnish children

5. To determine the clinical pictures of MCPyV and TSPyV primary infections in childhood

6. To assess the occurrence of anti-biotin IgM antibodies in the general population

7. To assess whether the anti-biotin IgM antibodies exert any adverse effects on biotinylation-based anti-viral IgM assays, and the degree of adverse effects on different assays

MATERIALS AND METHODS

1. Materials

1.1 Patients and serum samples

Children´s group 1 (Thesis studies: I, II, IV) From September 2000 to May 2002, children aged 3 months to 15 years with acute respiratory wheezing were admitted to the Department of Pediatrics of Turku University Hospital (Turku, Finland). Acute and convalescent phase serum samples were collected at the time of hospital admission and 2 weeks later. From a number of children follow-up samples were obtained 5-8 years later (232, 233). Study I, for MCPyV, included altogether 217 children, of which 72 underwent follow-up sampling. Study II, for TSPyV, enclosed 65 children with both paired acute-phase and follow-up serum samples available. Study IV, for biotin IgM, included 366 serum samples obtained from these children.

Children´s group 2 (I, II, IV) Between March 2001 and December 2002, children aged 1-4 years with recurrent or persistent otitis media were randomly selected to undergo adenoidectomy in a clinical study trial (234), and the initial serum samples were obtained. An average of three years later, these children underwent follow-up sampling. Study I comprised from the clinical trial a proportion of 158 children, 86 of which underwent adenoidectomy. Except one child with adenoidectomy excluded, Study II and Study IV included the same study subjects as Study I.

Children´s group 3 (III) Since 1994, children have been recruited in the population-based Diabetes Prediction and Prevention (DIPP) study for investigation of preclinical events in development of type-1 diabetes (235, 236). The children were invited for follow-up visits at 3- to 12-month intervals, from birth until young adolescence. At each visit, blood samples were collected, and the child´s clinical and physiological data were recorded by a study nurse via parental interviewing. Study III comprised from this cohort 144 children, who were born during 1995-2004 in whom type-1 diabetes did not develop by the end of 2004. These 144 children were called for an average of 13 visits (median 12, range 8-27), at a mean interval of 3.65 months until two years of age, and thereafter of 6.35 months interval.

Adult group (II, IV) This group comprised 115 university students and 34 staff members of our institute. A single serum sample was obtained from each volunteer of normal health. All these volunteers were included in Study II, and 144 in Study IV.

Senior group (IV): Senior citizen in-patients of Turku City Hospital participated in the study “Virel” which aims to investigate the molecular virus etiology in respiratory diseases in the elderly and its links to illness severity and long-term outcome. A total of 468 sera from 316 senior citizens (mean age 82 years, range 70-100) were analyzed in Study IV.

Ethical aspects (I-IV) The study protocols were approved by the Ethics Committees of Turku University Hospital, Helsinki University Hospital or the Hospital District of Southwest Finland. Written informed consent was obtained from all participants. If a senior participant was unsure about the consent, a written consent was obtained from the next of kin. For participating children, consent was obtained from the corresponding parents or guardians.

1.2 Plasmids, bacteria and cells (I, II)

The genes coding for MCPyV VP1 and TSPyV VP1 were chemically synthesized (GenScript, USA), and cloned into pUC57 vectors. For improved expression, the codon usage was optimized for that of Spodoptera frugiperda (Sf). For MCPyV, the synthesized VP1 was based on MCPyV339 (accession number EU375804), and for TSPyV, on the reported TSPyV DNA sequence (accession number GU989205). In the Bac-to-Bac System (Invitrogen), pFastBacDual vector is used to express two proteins simultaneously, and DH10Bac competent cells are used for production of recombinant bacmids. DH10Bac contains an AcMNPV parent bacmid that can recombine with a donor plasmid, e.g., with pFastBacDual. Sf9 insect cells were used for propagation of the recombinant baculovirus and production of recombinant proteins.

2. Methods

2.1 Virus-like particle production (I, II)

VLPs were generated through Bac-to-Bac protein expression system in accordance with the manufacturer´s (Invitrogen) instructions. The VP1 coding sequence was cloned into pFastBacDual transfer vector downstream of the polyhedrin promoter and the sequence was confirmed. The VP1-containing vector was transformed into DH10Bac to generate recombinant baculovirus genomic DNA (bacmid), followed by alkaline-lysis isolation.

The recombinant bacmids were transfected into Sf9 cells with cellfectin (Invitrogen). The recombinant baculovirus was passaged for two or three times. Sf9 cells were infected with recombinant virus at a high multiplicity of infection and grown in Insect Express medium (Lonza, Basel, Switzerland) for 4-5 days at 27°C. The infected cells were harvested by centrifugation at 1000g for 5 min, washed once with phosphate-buffered saline (PBS), suspended in lysis buffer 1 (20mM Tris-HCl, pH 7.5; 150mM NaCl; 1% Triton) and kept on ice for 10 min, followed by centrifugation at 10,000g for 10 min. Then, the pellet was resuspended in lysis buffer 2 (10mM Tris-HCl, pH 8.0; 500mM NaCl; 1mM CaCl2; 1mM MgCl2; 0.01% Triton) supplemented with a protease inhibitor cocktail (complete EDTA-free; Roche). After sonication on ice, the lysate was collected by centrifugation at high speed, 12,000 -13,000 rpm, for 3 -10 min, and applied to CsCl gradient ultracentrifugation. Two types of CsCl gradient were employed. In Study I, 1 ml of lysate preparation was loaded on the top of 11 ml of 28% CsCl, and centrifuged in a Beckman SW40Ti rotor at 24,200 rpm at 4°C for 48h. The fractions containing visible bands were collected. In Study II, 3.5 ml preparation was loaded on a 1.5-ml of a 25% (wt/vol) sucrose shelf (in 20mM Tris, pH 7.5), and centrifuged in a Beckman SW55Ti rotor at 103,246×g at 4°C for 2h. The pellets containing partially purified VLPs were resuspended in 400µl 20mM Tris (pH 7.5), and loaded on top of a 1.23 to 1.36 g/cm3 CsCl gradient. After ultracentrifugation in a Beckman SW55Ti rotor at 103,246×g at 4°C for 15h, the upper 2ml of the gradient was removed, and 250µl fractions were collected. The VLP-containing fractions were determined by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE), and the presence of VLPs was confirmed by EM. The fractions were dialyzed against PBS three times over night. The VLPs in PBS were supplemented with NaN3 (final concentration of 0.05%) and stored at 4 °C . The protein concentration was measured by BCA Protein Assay kit (Pierce).

2.2 Biotinylation (I, II, IV)

Biotinylation of protein were performed by using the EZ-LinkSulfo-NHS-LC- Biotinylation kit in accordance to the manufacturer´s (Pierce) instructions. The target protein  VLPs in Studies I and II, and bovine serum albumin (BSA) (Sigma-Aldrich, A7906) in Study IV  in PBS (approximately 1mg/ml) were incubated with 30-fold molar excess of biotin at room temperature (RT) for 1 hour. The biotinylated products were dialyzed against PBS three times over night.

2.3 Immunoblotting (I, II, IV)

In Studies I and II, dot blotting was employed. The nitrocellulose membranes (Whatman Protran BA 85) spotted with 1-2µL of protein were blocked in 5% BSA in Tris buffered saline with 0.1% Tween 20 (TBS-T) for 30 min at RT. Serum samples diluted in 0.1% BSA/TBS-T were put onto the membranes for 30 min. The secondary antibody, horseradish peroxidase (HRP)-conjugated anti-human IgG (DakoCytomation, Denmark)

diluted 1:1000 in 5% BSA/TBS-T was added to the membrane and kept for 30 min. The bound antibody was visualized with 3,3´-diaminobenzidine and H2O2. To detect biotinylated VLPs in Study I, the blocked membranes were incubated with HRP- conjugated streptavidin (Dako, Glostrup, Denmark) at 1:10,000 in 5% BSA/TBS-T and visualized as above. In Study IV, Western blot was employed. Biotinylated BSA (bio-BSA) or BSA loaded at 1µg/lane was separated by SDS-PAGE. After transferred onto nitrocellulose membranes, the membrane was blocked in 5% BSA in TBST for 30min at RT. The serum samples diluted in 5% BSA/TBST were applied on the blots at 4°C overnight. The secondary antibody, HRP-conjugated rabbit anti-human IgM (DakoCytomation, Glostrup, Denmark) diluted 1:20,000 in 5% BSA in TBST was applied for 1 hour at RT. The blots were visualized using the SuperSignal West Dura Extended Substrate Detection kit (Pierce, Rockford, IL.) and exposed on X-ray films (Fuji Super RX).

2.4 In-house antibody assays (I-IV)

Virus-specific IgG EIAs (I-IV) The biotinylated VLPs, serving as antigen, were diluted in PBS containing 0.05% Tween 20 (PBST) to a final concentration of 60ng/100µl. Then, 100 µL of antigen preparation were added to each well of streptavidin-coated plates (Thermo Scientific) followed by incubation for 60 min at RT in a rocking (400 rpm) EIA incubator. To minimize nonspecific background, the antigen-coated plates were post-blocked with sample diluents (three times, 10 min each). The serum samples diluted 1:200 were applied (100µl/well) for 60 min as the above. After washes with PBST (three times, 5 min each), HRP-conjugated rabbit anti-human IgG diluted 1:2,000 in the sample diluents, was added to each well (100µl/well), and incubated for 60 min as the above. After washes with PBST (four times, 5 min each), orthophenylene diamine (OPD) and H2O2 were added and incubated at RT. After 15min, 0.5 M H2SO4 was added to stop the reaction, and the absorbances at 492 nm were recorded.

Virus-specific IgM EIAs (III, IV) With virus-specific IgM EIA, a µ-capture format was employed. A polyclonal goat anti- human IgM antibody (Cappel/ICN Biomedicals) diluted 1:1200 in 50mM carbonate buffer, pH 9.6 (Sigma). Preparations were added (100µl/well) to each well of a microtiter plate and incubated at RT overnight. On the next day, the plate was rinsed with PBST (five times), followed by incubation with 150 µl of 3% BSA in PBS for 30 min at 37 °C . After a rinse, the pre-coated plate can be stored at 4 °C for up to one month. The human serum samples diluted 1:200 in PBST were applied at 100µl/well to the pre-coated plate and incubated for 60 min at RT in the EIA incubator. After rinses with PBST (five times), the plates were applied with the preparation of detection antigen and incubated for 45 min at 37 °C. The biotinylated MCPyV and TSPyV VLPs were prepared in 0.5% BSA in PBST at a concentration of 10ng/100µl. After rinses with PBST (five times), the plates were

treated (100µl/well) with HRP-conjugated streptavidin (Dako, Glostrup, Denmark) diluted 1:12,000 in 0.5% BSA in PBST for 45 min at 37°C, followed by OPD and H2O2 for 15 min at 37°C.

IgG avidity EIAs (III) The IgG-avidity EIA was based on the IgG EIA described above. However, two additional steps were included: (i) From the serum two dilution series in PBST were made at fourfold steps  one from 1:50 to 1:3200 and the other from 1:200 to 1:12,800. (ii) After antibody binding, the microwells of the former series were washed with urea in PBST, whereas, those of the latter series were washed with PBST alone. Thus, two dilution curves and IgG-end-point titers were obtained. Their ratio was calculated with curve-fitting software, representing IgG avidity (237).

Biotin IgM indirect EIA (IV) The Bio-BSA prepared in PBS at a concentration of 200ng/100µl was added (100µl/well) to microtiter wells (Nunc-Immuno™ Modules), and incubated for 2.5 hours at 37°C. After rinses with PBST (five times), the plates were blocked (150µl/well) in 5% BSA in PBS for 30 min at 37°C. After a rinse with PBST, the pre-coated plate can be stored at 4 °C for one day. The serum samples diluted 1:100 in PBST were applied (100µl/well) for 60 min at RT in the EIA incubator. The secondary antibody, HRP-conjugated rabbit anti-human IgM diluted 1:1,000 in 3% BSA in PBST, was added (100µl/well), and incubated for 60 min as above. After washes, OPD and H2O2 were added RT, and the reaction was stopped after 25 min.

2.5 Cutoff determination (I-IV)

For virus-specific IgG EIA (I, II), cutoff values were determined as follows: (i) the absorbance values from lowest to highest were placed in a dot plot, yielding a polynomial fit curve; (ii) a provisional cutoff was determined by visual inspection of the fitting curve; (iii) values below the provisional cutoff were used to calculate their mean and SD; (iv) the exact positions of the cutoff were determined by competition of serum samples spanning an absorbance range of 0.05 to 0.500. Two cutoffs were set for each assay. Samples with absorbance values exceeding the higher cutoff were considered virus-specific IgG positive, and samples with values below the lower cutoff were considered negative. With the samples of absorbance values between the two cutoffs, competition assays were performed to determine unequivocal positivity or negativity. In Study I, a specific MCPyV-IgG signal was obtained by subtraction of the antigen- free background in the MCPyV EIA. The higher and lower cutoffs were 0.120 and 0.150. In Study III, subtraction of the antigen-free background was omitted, and the two cutoffs were adjusted to 0.120 and 0.210. With TSPyV IgG EIA, the cutoffs were 0.100 and 0.240. For the virus-specific IgM EIA (III), the cutoff was calculated from 40 serum samples of adults with serological evidence for past infection with the virus (Methods in Study III).

The cutoffs for MCPyV IgM EIA were 0.207 and 0.260, and the cutoffs for TSPyV were 0.194 and 0.240. For Bio-EIA (IV), the cutoffs were determined in accordance with the WB results (Methods in Study IV).

2.6 Affinity determination (IV)

Based on the Biotin IgM indirect EIA, a competition assay was set up to measure biotin IgM affinity. The serum samples were incubated with increasing concentrations of soluble biotin (Sigma-Aldrich, B4501) over night. On the next day, the mixtures were applied onto Bio-EIA plates and the amount of antibody bound to the solid phase was measured. The absorbance values were plotted to yield a fitting curve by a nonlinear regression program (GraphPad Prism). The affinity was then calculated from the curve.

2.7 Quantitative PCR (III)

The DNA in 30µL of serum was extracted by phenol-chloroform, precipitated by sodium acetate and ethanol, eluted in 30µL of 10mmol/L Tris-Cl buffer (pH 8.0) and stored at - 20°C. An amount of 5µL was applied in quantitative PCR (qPCR). The MCPyV LT gene- based qPCR and TSPyV LT gene-based qPCR were performed as described (156) with a Stratagene Mx3005p instrument (Agilent Technologies, Santa Clara, CA, USA).

2.8 Statistical analysis (I-IV)

The seroprevalence and seroconversion rate differences between males and females were compared by logistic regression (I, II). Absorbance values of two groups were compared by a non parametric Mann-Whitney test with the tied P-value (I). Liddell´s exact test was employed to compare infection-related symptoms in each child during the seroconverted interval with symptoms during the previous sample interval and the subsequent interval (III). Wilcoxon rank-sum test, Spearman Correlation and Logistic regression, were employed to analyze whether the children experienced illness during the seroconversion period associated with background factors of the child, which were detailed in Study III. Absorbance values of more than two groups were analyzed using one-way ANOVA for overall comparison, followed by Tukey-Kramer HSD test for pairwise comparison (IV).

RESULTS

1. MCPyV and TSPyV serodiagnostics and infections

1.1 Virus-like particles (I, II)

For generation of VLPs of MCPyV and TSPyV, the synthetic viral VP1 genes were cloned into a recombinant baculovirus and expressed in Sf9 cells. By SDS-PAGE, a major protein band was seen in the whole cell lysates, and in the CsCl gradient-purified material. When fractionally purified by the two-step method, the VLPs of both MCPyV and TSPyV were barely seen in the cytosolic lysates, whereas they were highly enriched in the nuclear lysates (Fig. 1 of Study I and Figure 6). By EM, the purified VLPs had an icosahedral shape (Fig. 2 of Study I and Figure S1A of Study II).

Figure 6 SDS-PAGE of TSPyV VP1 in cell lysates and VLPs after purification and enrichment of TSPyV VP1 in nuclear lysate. Sf9 cells were harvested 3 or 4 days postinfection and processed by two-step lysis. As a control, GST was expressed in Sf9 cells by the same procedure. Whole: the whole cell lysate. Lys1: the supernatant of first lysis (cytosolic lysates). Lys 2: the supernatant of second lysis (nuclear lysates). Purified: Purified TSPyV VLPs after CsCl ultracentrifugation, arrow: TSPyV VP1; arrow head: Glutathione S-transferase (GST).

1.2 Antibody responses (I-III)

MCPyV and TSPyV IgG EIAs were set up by using VLPs as antigen, to detect antibody responses against the capsid protein VP1. The IgG EIA reactivities among each group are shown in Figure 7A. Of 1749 serum samples tested by MCPyV IgG EIA (Figure 7B), 66% were shown with absorbance values below 0.1 OD units, 20% with OD between 0.1 and 1.0, 9% with OD between 1.0 and 2.0, and 4% with OD exceeding 2.5. Of 1316 serum samples tested respectively by TSPyV IgG EIA, 65%, 9%, 13% and 13% were shown.

Figure 7 (A) Distributions of MCPyV and TSPyV seroreactivities. GR_CH1.1: the initial serum samples obtained from the children´s group 1. GR_CH1.2: the 5-8 year follow-up serum samples obtained from the children´s group 1. GR_CH2.1: the initial serum samples obtained from the children´s group 2. GR_CH2.2: the 3-year follow-up serum samples obtained from the children´s group 2. GR_CH3: the serum samples obtained from the children´s group 3. GR_AD: the serum samples obtained from the adult group. Line indicates higher cutoff in the corresponding assay. (B) The bar chart with percentage of 1749 serum samples at the various ODs in MCPyV EIA and 1316 serum samples in TSPyV EIA.

In Studies I and II, immunoreactivities of MCPyV and TSPyV VLPs with human IgG were examined by dot-blotting. The EIA-seropositive samples with high to low absorbance values showed corresponding reactivity with the native VLPs, but little or no reactivity with the same VLPs that had been pre-denatured (Fig. 3 of Study I and Figure 2 of Study II). To assess the degree of antigenic cross-reactivity between TSPyV and MCPyV VLPs, sera that were EIA-seropositive for both TSPyV and MCPyV were applied in competition assays. In TSPyV EIA, the seroreactivities decreased to undetectable level when the serum was preincubated with unbiotinylated TSPyV VLPs, but not with MCPyV VLPs. Conversely, their reactivities in MCPyV EIA declined to undetectable level by MCPyV

VLPs, but not by TSPyV VLPs. The antibody reactivities of adult sera showed no correlation between TSPyV and MCPyV VLP EIAs (Figure S2 of Study II). In Study III, for each seroconverting child, the antibody responses including IgG, IgM and IgG avidity, of the consecutive serum samples are shown in a line graph (Supplement 1 of Results). The MCPyV IgM and IgG antibodies occurred in five patterns among the 45 MCPyV seroconverting children (Supplement 2 of Results and Figure 2 of Study III): IgG seroconversion without IgM detectable in 28 (62%), or with IgM transiently present in 11 (24%), IgM detectable before IgG in 4 (9%), transient IgM re-occurring after reversion of the initial IgM in 2 (4%), and persistence of IgM till the end of follow-up in 1 (2%). Among 39 TSPyV seroconverting children, four patterns were seen (Supplement 2 of Results and Figure 2 of Study III): IgG seroconversion without IgM detectable in 9 (23%), or with IgM transiently present in 28 (72%), transient IgM re-occurring after the initial IgM in 1 (3%), and persistence of IgM till the end of follow-up in 1 (3%). From each seroconverter who had at least two follow-up visits after seroconversion, three serum samples  one taken at the IgG-seroconversion visit, one at the subsequent visit, and one at the last visit, were analyzed for IgG avidity (Supplement 2 of Results and Supplementary Figure of Study III). Among 33 such MCPyV seroconverters, 22 showed maturation of IgG avidity from low or borderline to high during the follow-up, 7 showed consistent high avidity, and 4 showed consistent low avidity. Of 27 such TSPyV seroconverters, 14, 12 and 1 showed the respective avidity patterns. The sampling intervals are shown in Supplementary Figure of Study III.

1.3 Seroprevalences (I-III)

Among the children´s group 1 (Table 4), 217 children were studied for MCPyV IgG. The MCPyV IgG seropositivities increased from 4% among children less than 1 year of age to 35% at 4-13 years. Among the children who underwent follow-up sampling 5-8 years later at 6-11 years of age, 37% were seropositive for MCPyV IgG. Of note, among the 65 children who were in the first samples seronegative, 22 (33 %) seroconverted for MCPyV during the 5-8-year follow-up. Of the same children´s group, 65 were studied for TSPyV IgG. Among the children below 1 year of age, 10% were TSPyV IgG positive, and at 1-3 years of age, 7% were. After a follow-up of 5-8 years, of initially TSPyV-seronegative children 45% experienced a seroconversion. The seropositivity increased to 48% among children at 6-11 years of age. Among the children´s group 2 (Table 4), 158 children were studied for MCPyV IgG. At the initial sampling (at 1-3 years of age), 4% were seropositive, while at 3-year follow- up, 18% were (at 4-7 years of age). Of the initially seronegative children, 16% seroconverted for MCPyV during an average of 3 years. Among the same children group, 157 children were studied for TSPyV IgG. At the initial sampling, 5% were seropositive, while at the follow-up, 34% had IgG antibodies in their sera. Of the initially TSPyV seronegative children, 30% seroconverted. Among the children´s group 3 (Table 4 and Figure 1 of Study III), the MCPyV seroprevalence increased from 12% at 1-2 years of age to 45% at 6-13 years of age, while the TSPyV seroprevalence increased from 3% to 39%. Of 144 children, 45 showed IgG

seroconversion by MCPyV and 39 by TSPyV. Of these seroconverters, four children showed IgG seroconversion with MCPyV during infancy, and one with TSPyV, while one child showed MCPyV IgG, IgM and low avidity of IgG in the first sample at infancy. In Study III, the respective maternal IgG for MCPyV and TSPyV was seen in 22% (10/45) of MCPyV seroconverters and 31% (12/39) of TSPyV seroconverters at sampling ages of 0.23-0.62 year, and these maternal IgGs disappeared by 0.49-1.07 year.

Children´s group 1 Children´s group 2 Children´s group 3

Age MCPyV TSPyV Age MCPyV TSPyV Age MCPyV TSPyV 2/51 2/20 <1y (4%) (10%) 17/144 5/144 1-2y (12%) (3%) 7/79 3/45 6/158 8/157 26/144 8/144 1-4y 1-3y 2-3y (9%) (7%) (4%) (5%) (18%) (6%) 31/142 18/142 3-4y (22%) (13%) 25/111 21/111 4-5y 9/26 28/158 53/157 (22%) (19%) 4-13y 4-7y (35%) (18%) (34%) 18/75 21/75 5-6y (24%) (28%) 26/71 31/65 23/51 20/51 6-11y 6-13y (37%) (48%) (45%) (39%)

Table 4. Seroprevalence among the children´s groups.

Among the adult group, of 115 volunteers who were studied for MCPyV IgG, 53% were seropositive, whereas of 149 volunteers studied for TSPyV IgG, 70% were seropositive.

1.4 Asymptomatic MCPyV and TSPyV primary infections among children (III)

Of 45 MCPyV seroconverting children, 62% showed serological markers of primary infection, other than IgG seroconversion: 33% showed the presence of IgM antibodies and 51% showed low-avidity of IgG (Table 5). Of 39 TSPyV seroconverting children, 82% showed these acute-phase markers: 77% IgM antibodies and 33% low-avidity IgG. Co- presence of IgM antibodies and low-avidity of IgG were seen in 22% of the MCPyV seroconverters, and in 28% of the TSPyV seroconverters. MCPyV viremias were found neither in the IgG seroconversion samples, nor in the previous or subsequent samples. TSPyV viremias were observed at low quantity in the IgG seroconversion samples of two children at 4.5 and 3.0 years of age.

No. (%) MCV Virologic findings Mean Median Range children 15 (33) IgM antibodies, abs 0.394 0.285 0.213–1.080

0 (0) qPCR positive 23 (51) low IgG avidity, % 10.1 10.3 1.9–15.0

10 (22) IgM together with low IgG avidity

No. (%) TSV Virologic findings Mean Median Range children 30 (77) IgM antibodies, abs 0.447 0.405 0.197–0.851 2 (5) qPCR positive Low positivity Ct 35.4 and 39.6 13 (33) low IgG avidity, % 8.8 7.9 1.1–15.0 11 (28) IgM together with low IgG avidity 2 (5) IgM together with qPCR positive

Table 5. Serologic findings among MCPyV and TSPyV seroconverters.

For determination of any possible clinical correlates during MCPyV or TSPyV primary infections, all infection-related symptoms and illnesses during the seroconversion interval were compared with those during the previous or the subsequent interval for each IgG-seroconverted child. Symptoms and signs of infection-related illness during the seroconversion interval were reported in 73% MCPyV seroconverters and in 82% TSPyV seroconverters. Neither the overall symptoms nor any individual symptom significantly differed from those of the previous and subsequent intervals (Table of Study III).

2. Biotin IgM antibodies

2.1 Observation of false MCPyV IgM positivity due to biotin antibody

In the in-house MCPyV IgM µ-capture EIA, high reactivity was seen in one serum sample originating from our staff (designated #m22), which was also positive for MCPyV IgG. Another serum sample obtained from the same individual one year earlier was found positive for MCPyV IgG, but negative in the MCPyV IgM assay. In dot blotting for the MCPyV IgM antibodies, the sample #m22 was reactive with biotinylated but not with non-biotinylated MCPyV VLPs (Figure 8). Similarly, in Western blot, the sample was reactive with biotinylated but not with nonbiotinylated BSA, and furthermore in a competition assay the reactivity was reduced dose-dependently by increasing amounts of soluble biotin (Figure 8). These observations demonstrated in this human serum the occurrence of anti-biotin IgM antibodies, which caused false positivity in a biotinylation- based antiviral IgM assay.

Figure 8 In dot blotting, serum sample #m22 reactive with biotinylated antigens but not with the non-biotinylated one. In Western blot, the IgM reactivity was reduced when the serum was co-incubated with 0µg/ml, 10µg/ml and 100µg/ml of competition soluble biotin. First on the right side is the blot treated with the protein stain Ponceau S red.

2.2 Seroprevalence of biotin IgM

A bio-EIA using biotinylated BSA as antigen was then developed, to determine the occurrence of biotin IgM antibodies in the general population. The EIA OD values of 612 serum samples from 459 adults are shown in Figure 1 of Study IV. In accordance with Western blot results (Figure 2 of Study IV), in this bio-EIA the cutoff for IgM positivity was set at 0.3 OD units. Among 144 adults, 4 (2.8%) had biotin IgM, and among 316 senior citizens 10 (3.2%) were biotin IgM-positive. The donor #22 who was initially positive lost his IgM during follow-up of six months, without any detectable anti-biotin IgG. Among 678 samples from 349 children aged 1-12 years, only one was positive. Overall, the seroprevalence of biotin IgM among adults was 3.0%, and among children 0.3%.

2.3 The effect of biotin antibody on various IgM-antibody assays

We then assessed the frequency of adverse-effects of the biotin IgM on various biotinylation-based antiviral IgM assays of µ-capture format, including four inhouse virus

IgM EIAs (MCPyV, TSPyV, HBoV1, B19V) and one commercial kit (B19V EIA of Biotrin). The results are shown in Figure 3 of Study IV and Table 6. Of note, all the four samples with Bio-EIA reactivities above OD 3, showed high reactivities in all these five assays. Among 20 serum samples positive for biotin IgM, the numbers of respective samples that were positive, negative, and at borderline range in each antiviral IgM assay were 17, 3, 0 in inhouse MCPyV EIA, 7, 11, 2 in TSPyV, 4, 16, 0 in HBoV, 7, 12, 1 in B19V (inhouse), and 1, 18, 1 in B19 Biotrin EIA (Table 6).

2.4 Characteristics of Biotin IgM

Bio-IgM positive and negative samples were applied in a µ-capture EIA (Figure 4 of Study IV), and tested with one of three different antigens: biotinylated VLPs, biotinylated BSA and soluble biotin. Bio-IgM-positive samples showed positive IgM results with biotinylated VLPs at a minimum amount of 5ng/well, with biotinylated BSA at a minimum amount of 10µg/well, while yielded negative results with soluble biotin irrespective of amount applied. In competition assays in a µ-capture IgM EIA (Figure 5 of Study IV), the reactivities of two Bio-IgM-positive sera were blocked dose-dependently by increasing concentrations of soluble biotin, but not by non-biotinylated MCPyV VLPs. Conversely, with two MCPyV-IgM-positive sera, the reactivities were blocked by MCPyV VLPs, but not by biotin. The binding of bio-IgM to biotinylated BSA was inhibited by biotin, but not by vitamin B5, with IC50s ranging from 2.1×10-3 to 1.7×10-4 mol/L.

Inhouse IgM EIAs Commercial Bio-IgM Subject # B19 biotrin EIA MCV TSV Boca B19 EIA

107.1 0.312 0.205 0.190 0.021 0.070 0.086

107.2 0.379 0.203 0.185 0.021 0.130 0.093

191.1 0.525 0.269 0.092 0.053 0.105 0.115

191.2 0.438 0.202 0.069 0.042 0.095 0.118

199 0.420 0.378 0.045 0.024 0.096 0.081

250.1 0.323 0.437 0.338 0.049 0.280 0.111

250.2 0.360 0.480 0.295 0.064 0.269 0.085

304 0.429 0.344 0.327 0.033 0.103 0.080

312 0.434 0.674 0.156 0.030 0.151 0.102

343 0.345 0.315 0.141 0.043 0.309 0.101

400 0.401 0.429 0.118 0.036 0.074 0.080

54 0.529 0.284 0.209 0.029 0.066 0.105

61 0.512 0.525 0.108 0.021 0.070 0.094

13.1 1.004 0.420 0.141 0.013 0.171 0.079

13.2 0.887 0.427 0.148 0.027 0.174 0.078

28 1.118 0.547 0.227 0.072 0.146 0.171 > OD 3.0 08PR2372 3.570 1.224 0.621 0.345 0.633 0.316

08PR2450 3.503 0.822 0.520 0.205 0.415 0.217

06PR384 3.445 1.766 0.328 0.212 0.861 0.434

88 3.299 1.846 0.843 0.222 1.289 0.172

Table 6. Effects of biotin IgM on various anti-viral IgM assays. Bold: positive in the corresponding IgM EIA. Underline: at the borderline zone of the IgM EIA.

DISCUSSION

1. MCPyV and TSPyV serodiagnostic assays (I-III)

1.1 Generation of MCPyV and TSPyV VLPs

For use as antigens in antibody assays, we generated for MCPyV and TSPyV the corresponding VLPs by expression of virus capsid protein VP1 in insect cells. Recombinant MCPyV and TSPyV VLPs have been reported in several independent studies, either by a similar methodology as ours or by co-expression of MCPyV VP1 and VP2 in eukaryotic cells (123-126, 129). The fact that MCPyV and TSPyV VP1 self- assembled into VLPs without requirement of the minor proteins VP2 and VP3 is analogous to those reported with the classical polyomaviruses  SV40, BKPyV and JCPyV (25, 26, 238-241). It is thought that VP2 and VP3 are required in encapsidation of DNA to form infectious virions, but not compulsory for formation or stability of viral capsids (117). The nuclear localization signal of VP1 of JCPyV and BKPyV has been identified. JCPyV or BKPyV VP1 are expressed in the cytoplasm of insect cells and transported to the nucleus for VLP assembly (26, 238). We purified the VLPs of MCPyV and TSPyV by two-step lysis to separate the cytosolic and nuclear fractions, and found the VLPs to be highly enriched in the nuclear lysates. Thus, these two new viruses are assembled in the nuclei, similar to the classical HPyVs. Indeed, the basic sequence (MAPKRKASSTCK) at the N terminus of MCPyV VP1 and the one (MAPKRKGEGCA) at TSPyV VP1, resemble the nuclear localization signals in VP1 of JCPyV (MAPTKRKGERKD) or BKPyV (MAPTKRKGECP) (26, 238). Evidently, in both MCPyV and TSPyV, VP1 alone contains sufficient structural information for viral capsid formation. The conformation dependence of antibody reactivity has been reported with BKPyV and JCPyV (26, 238, 242). We assessed the nature of capsid epitopes of MCPyV and TSPyV, and found them, likewise, to be conformational rather than linear. A similar conclusion with MCPyV was reached by peptide library screening and Western blotting (126). Of note, studying BKPyV and JCPyV epitopes with monoclonal antibodies, Randhawa et al. showed that antibodies reacting with intact VLPs recognized conformational epitopes located on exposed surfaces of VLPs, whereas antibodies reactive with disrupted VLPs recognize linear epitopes buried inside (243).

1.2. Serodiagnostic tools for MCPyV and TSPyV

In the Studies I and II, EIAs were developed to detect antibodies for the two recently found HPyVs. For SV40, JCPyV, and BKPyV, a variety of antibody assays have been set up, such as immunoblotting, hemagglutination inhibition assay, EIA, neutralization assay, complement fixation, and immunofluorescence assay (244). However, for the newly detected HPyVs, the serological assays developed thus far have been EIA-format or

Luminex-based (122, 127, 239, 245, 246). The antigens have been either VP1 capsomers or VLPs (122-126, 245). However, since self-assembled VLPs morphologically resemble native viral particles and thus retain the conformational epitopes, VLP-based EIAs are likely to be more sensitive than the VP1-capsomers based (193, 238, 242). Neutralization assay, identifying protective antibodies, have not been widely applied for the newly found HPyVs, apparently due to the lack of permissive cell lines for infection by these viruses. For MCPyV, an alternative format of neutralization assay has been set up with a reporter- gene-enclosed pseudovirus (115). Other than viral capsid protein VP1, HPyV LT has also been investigated as serodiagnostic antigen (131, 242, 247). Compared to anti-VP1 responses, the antibody responses against BKPyV LT have been found to occur at lower frequency and at lower level, and to be associated with more extensive BKPyV replication after primary exposure (242). With MCPyV, anti-LT antibodies are shown in 40% of MCC patients but only in 0.9% of healthy controls (131). These findings suggest that anti- LT antibody is not an appropriate indicator for an early exposure of HPyVs, but may serve as serological markers for extensive viral replication, such as viral reactivation. For diagnosis of primary infections by MCPyV and TSPyV, in addition to assays for viral IgG, we set up EIAs for viral IgM and IgG avidity. As noted, both presence of IgM and low avidity of IgG are serological markers for an acute infection. In Study III, we investigated the occurrences of these two markers post-IgG seroconversion among children followed up prospectively for MCPyV or TSPyV, and observed these two in more than half of IgG-seroconversion samples for each virus. Of the two markers, presence of IgM was more frequent than presence of low-avidity IgG among TSPyV seroconverters, whereas presence of low-avidity IgG was frequent among MCPyV seroconveters. This might suggest some difference in immunoglobulin class shift or affinity maturation for these new HPyVs. Regarding viremia, MCPyV DNA was absent from all the samples at or flanking the IgG seroconversions, while TSPyV DNA was seen at low copy number in two seroconverters. In Tolstov´s study on adult seroconverters, MCPyV viremia was neither found (194). These observations suggest that MCPyV or TSPyV may not be viremic or that viremia is of short duration. Thus, serodiagnosis is likely to be the approach of choice in diagnosis of primary infection both for MCPyV and TSPyV. The cutoffs, used as criteria for positivity and negativity, crucially affect the sensitivity and specificity of diagnostic assays. In general, serodiagnostic cutoffs should be determined using subjects with or without defined infection (222). In the absence of such standards, we determined the cutoffs via statistical calculation of potential negative/positive results obtained through visual observation of EIA results from a large number of samples. To limit false results, we employed a broad borderline zone between the higher and lower cutoffs, and samples yielding results in this zone were further examined with VLP competition.

2. MCPyV and TSPyV infections (I-III)

2.1 MCPyV and TSPyV antibody responses and seroprevalence

We found among Finnish adults MCPyV VP1 seroprevalence of 57% and TSPyV VP1 seroprevalence of 70%, which is in line with the seroprevalence of the classical HPyVs among European and US populations. The seroprevalence for the other known HPyVs are 80-100% for BKPyV, 40-70% for JCPyV, 70-100% for WUPyV, 55-90% for KIPyV, 70- 80% for HPyV6, 35-60% for HPyV7, 30- 50% for HPyV9, and 20% for HPyV12 (43, 122, 123, 246, 248-252). For MCPyV, seroprevalence of 50-95% have been reported in adults of different geographical regions (115, 122-127, 194), while for TSPyV, seroprevalence of 70% and of 76% were reported in Dutch and Italian cohorts, respectively (123, 245). The high seroprevalence of these two recently detected HPyVs among asymptomatic adults indicate that infections with MCPyV or TSPyV are common in the general population. In three different Finnish pediatric study cohorts, MCPyV seroprevalence was 9-45% and TSPyV was 7-48% among children at 1-13 years of age, consistent with prior and subsequent reports on the presence of MCPyV and TSPyV IgG in childhood. MCPyV seroprevalences among children as reported by Kean et al. were 20.5% at 1-5 years, by Tolstov et al. 43% at 2-5 years, by Viscidi et al. 45% under 10 years, and by Nicol et al. 58% under 10 years (122, 123, 125, 126). The TSPyV seroprevalences in childhood as reported by van der Meijden et al. were 41% under 9 years of age, and by Nicol et al., 31% at 1-4 year and 70% at 5-9 years (123, 245). Concerning pediatric infections by other HPyVs, the JCPyV seroprevalence reaches approximately 25% in young adolescence whereas BKPyV seroprevalence reaches up to 90% among children under 10 years of age (43, 248). The rates of seropositivity for both WUPyV and KIPyV increase rapidly among children of 4-6 years, reaching 87 % and 93%, respectively (250). In Study III, the average seroconversion age for MCPyV was slightly lower than for TSPyV, whereas the seroprevalence among adults of TSPyV was higher than of MCPyV, suggesting that the transmission routes of these two viruses might differ. A feco-oral route has been proposed for MCPyV, effected via skin contact or saliva (15, 143, 153, 253, 254). A quite recent study demonstrated that MCPyV infection is acquired through close contact, especially between siblings and between mothers and their children (255). Furthermore, the number of children experienced MCPyV infection during infancy is higher than that of TSPyV, suggesting that MCPyV shedding from maternal skin may affect infection. We conducted follow-up studies in these three pediatric populations, in order to get a clearer idea about time and frequency of primary infections with MCPyV and TSPyV. With MCPyV, 33% of children seronegative at < 3 years of age were shown to seroconvert during 5-8 years, 16% of children seronegative at 1-4 years to seroconvert during an average of 3 years, and 31% of children followed from birth up to 13 years of age to seroconvert during the follow-up. With TSPyV, among these three cohorts the corresponding seroconversion rates were 45%, 30% and 27%. In addition to high seroprevalence, these observations on seroconversion further provide evidence that the primary exposures to these two newly found HPyVs occur extensively in early childhood.

BKPyV antibody titers and seroprevalences decrease with increasing age, while those of JCPyV remain relatively unchanged or may increase (43, 248). For both MCPyV and TSPyV, we observed in a substantial proportion of seroconverting children in Study III the occurrence of maternal antibodies. These maternal antibodies mostly declined to undetectable level by the end of the first year of age. In contrast, the MCPyV and TSPyV IgG, acquired through infection, can persist long. From all of the seroconverting children, IgG antibodies were detectable until the end of follow-up. Interestingly, we observed in one fourth of the MCPyV seroconverters a further increase in antibody level after 0.8-3.0 years. This booster of IgG response is apparently due to antigenic re-stimulation, as MCPyV DNA is widely shed in human skin (15, 129). To date, the antigens used in MCPyV serological assays are derived from two MCPyV isolates  MCPyV339 and MCPyV350 (Figure 9), the former of which is employed in most of MCPyV seroepidemiological studies including ours (115, 125, 126, 129). Though the known MCPyV isolates share as high as 99% of the amino acid sequences in their VP1 regions, Kean et al. found in the general population a lower seroprevelence with MCPyV350 than with MCPyV339. They also observed as many as 164 sera to be IgG- positive for VP1 of isolate 350 but not of isolate 339, whereas 560 sera were positive for isolate 339 but not 350, suggesting that between these two variants certain epitopes recognized by human sera are different (122). Carter et al. found that among MCC patients the seroprevalence and seroreactivity with MCPyV339 were significantly higher than with MCPyV350, implying that MCPyV339 may contain certain MCPyV VP1 epitopes not present in isolate 350 (122, 127). Thus, our studies do not reflect the precise MCPyV seroprevalence in Finnish population. It is plausible to expect some individuals harboring antibodies to certain epitopes of MCPyV isolates other than 339. Sequence alignment shows MCPyV339 and MCPyV350 differing in the VP1 regions by four amino acids, two of which are shown by mutagenic analysis to be critical for the epitopes present in MCPyV339 (127). Since BKPyV has four antigenic types, it will be worthwhile to know whether MCPyV isolates can be grouped into different serotypes (256, 257).

2.2 Antigenic cross-reactivity between different HPyVs

MCPyV and TSPyV VP1 have an overall amino acid similarity of approximately 50% to each other (6, 9, 258). In our competition assays, the VLPs of these two viruses did not compete with each other, demonstrating that in humans the majority of TSPyV and MCPyV antibodies are, at least in regards to the antibodies measured by these assays, species specific. A number of studies, either using as antigens VLPs or GST recombinant VP1 protein, have shown that antibodies against BKPyV and JCPyV do not cross-react with MCPyV (122, 124-126). Van der Meijden et al. showed that sera positive for TSPyV VP1 antibodies did not cross-react with MCPyV VP1 and BKPyV VP1 (245). Among the 12 HPyVs, the VP1 amino acid similarity ranges from 25% between BKPyV and HPyV7 to 78% between BKPyV and JCPyV (Figure 9) (258). However, seroepidemiological studies indicate cross-reactivity unlikely exists between antibodies against the HPyVs VP1 (258). On the other hand, as the VP1 antibodies in human sera only recognize conformational epitopes, cross-reactivity might exist in the linear epitopes between

HPyVs with high VP1 homogeneity. Of note, in Western blot, hyperimmune rabbit serum against GST fused VP1 protein of BKPyV showed strong cross-reactivity with the VP1 of JCPyV, suggesting linear epitopes shared by these two viruses (250). The minor capsid proteins VP2 and VP3, as lining the interiors of HPyV particles, are thought to induce antibodies less frequently than VP1. Interestingly, Kantola et al. found that some human sera showed IgG reactivity with KIPyV and WUPyV VP2 and VP3 in Western blot. Moreover, they found cross-reactivity between these two viruses with regard to IgG for denatured VP2 and VP3 (259).

Figure 9 Phylogenetic analysis of HPyVs VP1 protein sequences. Neighbour-joining tree was generated using online ClustalW2 programme (http://www.ebi.ac.uk/Tools/msa/clustalw2/). Thirteen HPyVs sequences were obtained from Genbank. These included JCPyV (NC_001699), BKPyV (NC_001538), KIPyV (NC_009238), WUPyV (NC_009539), MCPyV 339 (EU375804), MCPyV 350 (EU375803), HPyV6 (NC_014406), HPyV7 (NC_014407), TSPyV (NC_014361), HPyV9 (HQ696595), MWPyV (JQ898292), STLPyV (NC_020106), and HPyV12 (JX308829)

2.3 Asymptomatic MCPyV and TSPyV primary infections in childhood

In Study III, we examined the clinical correlates of primary infections with MCPyV and TSPyV among the children followed over several years. With neither MCPyV nor TSPyV was the occurrence of symptoms or signs significantly associated with the time of seroconversion, suggesting that primary exposures to these viruses in childhood are asymptomatic. Similarly, Tolstov et al. neither found clinical associations of MCPyV primary infection among male adults (194). Upper respiratory tract infections, as expected, were the most frequent infections in Study III. However, their occurrence was not associated with MCPyV or TSPyV primary infections. In Study I, no MCPyV

seroconversions were found in children at the time of acute wheezing. Although MCPyV DNA has been found in respiratory tract samples (143, 146, 148), our observations rule against the involvement of MCPyV in respiratory tract infection. A similar conclusion has been reached with WUPyV and KIPyV, also discovered in respiratory samples (13, 14, 59, 260). So far, there has been no firm evidence for an etiological role of any of the newly discovered polyomaviruses in respiratory illness. The prototypic HPyVs, BKPyV and JCPyV are known to establish a persistent, seemingly asymptomatic infection in immunocompetent individuals (261, 262). Although BKPyV seroconversion has been reported in a small number of children with clinical signs of acute upper respiratory tract infection, so far no solid evidence exists to link BKPyV to the disease (263). Different from the prior study which investigated BKPyV seroconversion in children with certain disease, our study investigated whether symptoms and signs of any infection-related illness in children were associated with viral seroconversion. We found primary infection with MCPyV and TSPyV in children to be asymptomatic. BKPyV and JCPyV can reactivate and lead to serious diseases such as PVAN and PML (46, 48). As the MCPyV and TSPyV-related diseases occur under immunosuppression, it is tempting to speculate that these viruses might persist long after primary infection and reactivate to pathogenicity. MCPyV is considered a member of our normal skin flora. Apparently, the virus, after encountering mutation and integration, can give rise to this rare cancer (7, 192). Previous studies showed that an active TSPyV infection was associated with TS pathogenesis (9, 10). However, our observation here rules against a connection between TSPyV primary infection and disease, and raise the possibility that the disease is due to TSPyV reactivation. Interestingly, exanthema, a common sign of skin infection, appeared in seven children (15.9%) at the MCPyV seroconversion interval but only in one child in the adjacent periods (2.2 % before, 2.3 % after), though the rate did not reach statistical significance, possibly due to small size of cohort. Since MCPyV and TSPyV are of skin tropism, it is possible that some skin-related signs beyond our interview were present in our cohort. In addition, infection-related symptoms might appear in geographic regions outside Finland. In the future, it could be worthwhile to investigate dermatological manifestations in larger- size cohorts as well as in various geographic locations.

3. Biotin IgM antibodies (IV)

3.1 The effect of biotin IgM on biotinylation-associated EIAs

For detecting anti-viral IgM antibodies, our laboratory, as many others, is using and also setting up µ-capture EIAs employing as antigen biotinylated VLPs detected by HRP- conjugated streptavidin. With MCPyV IgM EIA, we employed as negative controls serum samples obtained from individuals with past immunity against MCPyV, and found one sample exhibiting strong reactivity in the assay. Interestingly, this sample showed IgM reactivity with the biotinylated MCPyV VLP but not with the corresponding

unbiotinylated VLP, suggesting that IgM antibodies in human blood cause false results in biotinylation-based antiviral IgM assays. The adverse effects of the biotin IgM on several antiviral IgM assays were then assessed. As expected, serum samples with high levels of biotin IgM affected adversely all five IgM assays of µ-capture format, whereas samples with low levels exhibited low interference with those assays. Furthermore, among the IgM assays for different viruses, the effects of biotin IgM differed and MCPyV IgM EIA was the most susceptible. The degree of adverse effects seemingly correlates to the conformation of viral particles or their degree of biotinylation. Therefore, two ways are proposed to limit the adverse effect. One is to use appropriate amount of biotin to label antigens, and the other is to employ a borderline zone upon setting up the cutoffs according to the false reactivity derived from biotin IgM.

3.2 Potential origins of biotin IgM

The origin of biotin IgM is an intriguing question. Existing reports on occurrence of human antibodies to vitamins are scarce. Vitamin D IgG antibodies were present at a low prevalence in patients with autoimmune diseases (264). For the biotin IgM we found a seroprevalence of 3% among Finnish adults. This prevalence is quite close to those of some known autoantibodies among the general population, e.g. 1-3% of rheumatoid factor (265-267). It is possible that the biotin IgM antibodies are related to autoimmunity. Secondly, in the human blood, biotin only accounts for half the total avidin-binding substances, and biotin metabolites account for the rest (268). It is possible that some imbalance in biotin metabolism will induce an immune response to these biotin metabolites, resulting in the generation of biotin antibody. Thirdly, in Study IV the affinity levels of biotin IgM ranged from 10-3-10-4 mol/L, close to that of many polyreactive antibodies generated without known antigenic stimulation (269-272). Polyreactive antibodies generally belong to the IgM class and target endogenous host antigens (269- 272). As biotin can interact with host proteins, biotin IgM might be a group of polyreactive antibodies induced by the biotin-host complexes (273-275).

3.3 Possible clinical biomedical effects of biotin IgM

Biotin acting as a coenzyme is essential for in most organisms, and catalyzes critical reactions in the intermediary metabolism of gluconeogenesis, fatty acid synthesis, and amino acid catabolism (274). Biotin deficiency can lead to a number of clinical abnormalities, such as hair loss, growth retardation, neurological disorders and vulnerability to infections (276, 277). The avidin in raw eggs binds biotin and prevents its intestinal absorption, so biotin deficiency can ensue in persons excessively consuming raw eggs (278-280). After intestinal absorption, biotin circulates in the human blood for uptake by cells (273). We showed that biotin IgM and streptavidin/avidin share the same binding site on biotin. It is possible that biotin IgM in blood will function as avidin in the intestinal tract, preventing biotin uptake by tissues and cells. This assumption indeed deserves further investigation.

CONCLUSIONS AND FUTURE PROSPECTIVES

This thesis includes studies on the two recently detected human polyomaviruses MCPyV and TSPyV, with respect to (i) development of serodiagnostic assays, (ii) determination of their seroepidemiology among Finnish cohorts, and (iii) identification of possible clinical pictures associated with primary infections in childhood. In addition, the thesis aims at improving the quality of antiviral IgM assays, and identifies and characterize endogenous biotin IgM antibodies circulating in our blood. In the thesis, MCPyV and TSPyV VP1 proteins were readily expressed in the baculovirus system and self-assembled into virus like particles. Using VLPs as antigen, we developed for MCPyV and TSPyV comprehensive diagnostic assays including EIAs for IgG, IgM and IgG avidity, for estimation of MCPyV and TSPyV primary infections. We found with the two viruses high seroprevalence rates among adults, and high rates of primary infections among children, implying that MCPyV and TSPyV circulate widely among the general population, in particular during early childhood. In a longitudinal study, comparing the seroconversion intervals with the previous and subsequent intervals for clinical events, we found the primary infections by neither virus to be significantly associated with symptoms or signs of any infection-related illness, indicating that these pathogenic polyomaviruses circulate silently among constitutionally healthy children. With MCPyV and TSPyV, future studies may aim at addressing the following issues. Firstly, MCPyV and TSPyV apparently are capable to enter latency and later reactivate. Thus, the persistence sites and patterns, reactivation mechanisms, as well as reactivation- related diseases other than MCC and TS, are calling for further investigation. Secondly, these two viruses have been shown to be associated with human skin diseases  MCPyV with MCC and TSPyV with TS. Therefore, it is worthwhile to know whether these viruses can serve as markers in diagnosis of their related diseases or as targets for treatment. Thirdly, since at least two MCPyV variants have been found, it will be interesting to know with MCPyV and TSPyV whether any other variants exist, and whether the pathogenicity differs among variants. In addition, since ten HPyVs have been discovered by novel detection methods during the past six years, one could expect more such viruses that silently circulate in the general population, to be discovered in the future. Presence of IgM antibodies to biotin in some human sera has been accidentally found during thorough evaluation of the described serodiagnostic assays. These antibodies adversely affect a variety of biotinylation-based antiviral IgM assays, causing false- positive results for viral IgM. The prevalence of biotin IgM was 3% among adults, close to the prevalence of the rheumatoid factor autoimmune antibody. However, the affinity of biotin IgM was at a range close to that of polyreactive antibodies. The biotin IgM and streptavidin were found to share the same binding site on biotin. These findings provide new insights into biotinylation-based immunoassays and new information on the immune recognition of vitamins. Thus, several aspects deserve further investigations. (i) The origin of biotin IgM. These antibodies may originate from an autoimmune process or may represent a group of polyreactive antibodies. (ii) The biomedical effects of biotin IgM. Biotin IgM might

inhibit the uptake of biotin from circulation, or might keep the balance of biotin with its metabolites in blood. Occurrence of biotin IgM might prevail in certain risk groups, e.g. patients in context of biotin deficiency. (iii) The side effects of biotin IgM on serological assays. As biotinylation is widely employed in serological assays, in addition to antiviral IgM assays these antibodies might affect a variety of biotinylation-based immunoassays.

ACKNOWLEDGEMENTS

This study was carried out at the Department of Virology, Haartman Institute, University of Helsinki, during the years 2008-2013. I thank the former Head of the Department, Professor Antti Vaheri, and the current Head of the Department, Professor Kalle Saksela for the excellent working environment. I would like to express the deepest appreciation to my supervisors, Professor Klaus Hedman and Docent Maria Söderlund-Venermo, for your excellent guidance, caring, patience, and providing me with an excellent atmosphere for doing research. Your wisdom, knowledge and commitment to the highest standards always inspired and motivated me. Thank you for allowing me the opportunity to learn from you and to carry out my PhD project. Without your guidance and persistent help, this dissertation would not have been possible. I would like to acknowledge the Reviewers of this thesis, Professor Timo Vesikari and Docent Thedi Ziegler. I am thankful for your insightful comments on the manuscript that made this thesis better, and for completing the review in a quite tight time schedule. I thank all the collaborators for your contribution to this work: Petri S. Mattila, Tuomas Jarrti, Olli Ruuskanen, Laura Jartti, Laura Tanner, Ville Simell, Marjaana Mäkinen, Riitta Veijola, Heikki Hyöty, Jorma Ilonen, Mikael Knip, Jorma Toppari, and Olli Simell for providing invaluable clinical materials. I would like to express my special gratitude to all the former and present members of our group: Lea, Päivi, Elina, Kalle, Laura, Reza, Arun, Anne, Mira, Claudia, Mari, Xuemeng, Yilin, Mafe, Visa and Benedict. Lea is especially thanked for teaching me ELISA and helping me doing numerous ELISAs in the third paper. Reza is particularly thanked for your excellent PCR work. I warmly thank all my Chinese friends here (too many to list here but you know who you are). The festival parties, game nights, advices as well as general help, and friendship are all greatly appreciated. I want to express my deepest appreciation to my dear parents. Thank you for your continuous love, care, and support. Finally, I must express my very profound gratitude to my husband Wei for providing me with unfailing support and continuous encouragement throughout my years of study. This study was supported by the Helsinki University Central Hospital Research and Education Fund, the Sigrid Jusélius Foundation, the Medical Society of Finland, the Academy of Finland, the Research Funds of the University of Helsinki, the Orion-Farmos Research Foundation, the Paulo Foundation, the Biomedicum Helsinki Foundation, The foundation for research on Viral diseases, and Kliinisen Kemian Foundation.

Helsinki, November 2013

REFERENCES

1. Imperiale M,J., Major E,O. Polyomaviruses. In: Knipe D,M., Howley P,M., editors. Fields virology. Fifth edition ed. Lippincott, Williams & Wilkins; 2007. p. 2263-98. 2. Mandl CW, Frisque RJ. Characterization of cells transformed by the human polyomavirus JC virus. J Gen Virol. 1986;67 ( Pt 8):1733-9. 3. Chenciner N, Meneguzzi G, Corallini A, Grossi MP, Grassi P, Barbanti-Brodano G, et al. Integrated and free viral DNA in hamster tumors induced by BK virus. Proc Natl Acad Sci U S A. 1980;77(2):975-9. 4. Hirai K, Lehman J, Defendi V. Integration of simian virus 40 deoxyribonucleic acid into the deoxyribonucleic acid of primary infected chinese hamster cells. J Virol. 1971;8(5):708-15. 5. Khalili K, Croul S, Del Valle L, Krynska B, Gordon J. Oncogenic potential of human neurotropic virus: Laboratory and clinical observations. Isr Med Assoc J. 2001;3(3):210-5. 6. Feng H, Shuda M, Chang Y, Moore PS. Clonal integration of a polyomavirus in human merkel cell carcinoma. Science. 2008;319(5866):1096-100. 7. Arora R, Chang Y, Moore PS. MCV and merkel cell carcinoma: A molecular success story. Curr Opin Virol. 2012;2(4):489-98. 8. Shuda M, Feng H, Kwun HJ, Rosen ST, Gjoerup O, Moore PS, et al. T antigen mutations are a human tumor-specific signature for merkel cell polyomavirus. Proc Natl Acad Sci U S A. 2008;105(42):16272-7. 9. van der Meijden E, Janssens RW, Lauber C, Bouwes Bavinck JN, Gorbalenya AE, Feltkamp MC. Discovery of a new human polyomavirus associated with trichodysplasia spinulosa in an immunocompromized patient. PLoS Pathog. 2010;6(7):e1001024. 10. Kazem S, van der Meijden E, Kooijman S, Rosenberg AS, Hughey LC, Browning JC, et al. Trichodysplasia spinulosa is characterized by active polyomavirus infection. J Clin Virol. 2012;53(3):225-30. 11. Padgett BL, Walker DL, ZuRhein GM, Eckroade RJ, Dessel BH. Cultivation of papova-like virus from human brain with progressive multifocal leucoencephalopathy. Lancet. 1971;1(7712):1257-60. 12. Gardner SD, Field AM, Coleman DV, Hulme B. New human papovavirus (B.K.) isolated from urine after renal transplantation. Lancet. 1971;1(7712):1253-7. 13. Allander T, Andreasson K, Gupta S, Bjerkner A, Bogdanovic G, Persson MA, et al. Identification of a third human polyomavirus. J Virol. 2007;81(8):4130-6. 14. Gaynor AM, Nissen MD, Whiley DM, Mackay IM, Lambert SB, Wu G, et al. Identification of a novel polyomavirus from patients with acute respiratory tract infections. PLoS Pathog. 2007;3(5):e64. 15. Schowalter RM, Pastrana DV, Pumphrey KA, Moyer AL, Buck CB. Merkel cell polyomavirus and two previously unknown polyomaviruses are chronically shed from human skin. Cell Host Microbe. 2010;7(6):509-15. 16. Scuda N, Hofmann J, Calvignac-Spencer S, Ruprecht K, Liman P, Kuhn J, et al. A novel human polyomavirus closely related to the african green monkey-derived lymphotropic polyomavirus. J Virol. 2011;85(9):4586-90. 17. Siebrasse EA, Reyes A, Lim ES, Zhao G, Mkakosya RS, Manary MJ, et al. Identification of MW polyomavirus, a novel polyomavirus in human stool. J Virol. 2012;86(19):10321-6. 18. Buck CB, Phan GQ, Raiji MT, Murphy PM, McDermott DH, McBride AA. Complete genome sequence of a tenth human polyomavirus. J Virol. 2012;86(19):10887-12. 19. Yu G, Greninger AL, Isa P, Phan TG, Martinez MA, de la Luz Sanchez M, et al. Discovery of a novel polyomavirus in acute diarrheal samples from children. PLoS One. 2012;7(11):e49449.

20. Lim ES, Reyes A, Antonio M, Saha D, Ikumapayi UN, Adeyemi M, et al. Discovery of STL polyomavirus, a polyomavirus of ancestral recombinant origin that encodes a unique T antigen by alternative splicing. Virology. 2013;436(2):295-303. 21. Korup S, Rietscher J, Calvignac-Spencer S, Trusch F, Hofmann J, Moens U, et al. Identification of a novel human polyomavirus in organs of the gastrointestinal tract. PLoS One. 2013;8(3):e58021. 22. Johne R, Buck CB, Allander T, Atwood WJ, Garcea RL, Imperiale MJ, et al. Taxonomical developments in the family polyomaviridae. Arch Virol. 2011;156(9):1627-34. 23. King A, Adams M, Carstens E, Lefkowitz E. Virus taxonomy. ninth report of the international committee on taxonomy of viruses. London, San Diego: Elsevier Academic Press; 2011. 279- 290. 24. Liddington RC, Yan Y, Moulai J, Sahli R, Benjamin TL, Harrison SC. Structure of simian virus 40 at 3.8-A resolution. Nature. 1991;354(6351):278-84. 25. Hale AD, Bartkeviciute D, Dargeviciute A, Jin L, Knowles W, Staniulis J, et al. Expression and antigenic characterization of the major capsid proteins of human polyomaviruses BK and JC in saccharomyces cerevisiae. J Virol Methods. 2002;104(1):93-8. 26. Chang D, Fung CY, Ou WC, Chao PC, Li SY, Wang M, et al. Self-assembly of the JC virus major capsid protein, VP1, expressed in insect cells. J Gen Virol. 1997;78:1435-9. 27. DeCaprio JA, Garcea RL. A cornucopia of human polyomaviruses. Nat Rev Microbiol. 2013;11(4):264-76. 28. Elphick GF, Querbes W, Jordan JA, Gee GV, Eash S, Manley K, et al. The human polyomavirus, JCV, uses serotonin receptors to infect cells. Science. 2004;306(5700):1380-3. 29. Low JA, Magnuson B, Tsai B, Imperiale MJ. Identification of gangliosides GD1b and GT1b as receptors for BK virus. J Virol. 2006;80(3):1361-6. 30. Pho MT, Ashok A, Atwood WJ. JC virus enters human glial cells by clathrin-dependent receptor-mediated endocytosis. J Virol. 2000;74(5):2288-92. 31. Eash S, Querbes W, Atwood WJ. Infection of vero cells by BK virus is dependent on caveolae. J Virol. 2004;78(21):11583-90. 32. Neu U, Stehle T, Atwood WJ. The polyomaviridae: Contributions of virus structure to our understanding of virus receptors and infectious entry. Virology. 2009;384(2):389-99. 33. Erickson KD, Garcea RL, Tsai B. Ganglioside GT1b is a putative host cell receptor for the merkel cell polyomavirus. J Virol. 2009;83(19):10275-9. 34. Schowalter RM, Pastrana DV, Buck CB. Glycosaminoglycans and sialylated glycans sequentially facilitate merkel cell polyomavirus infectious entry. PLoS Pathog. 2011;7(7):e1002161. 35. Qu Q, Sawa H, Suzuki T, Semba S, Henmi C, Okada Y, et al. Nuclear entry mechanism of the human polyomavirus JC virus-like particle: Role of importins and the nuclear pore complex. J Biol Chem. 2004;279(26):27735-42. 36. White MK, Khalili K. Polyomaviruses and human cancer: Molecular mechanisms underlying patterns of tumorigenesis. Virology. 2004;324(1):1-16. 37. White MK, Gordon J, Khalili K. The rapidly expanding family of human polyomaviruses: Recent developments in understanding their life cycle and role in human pathology. PLoS Pathog. 2013;9(3):e1003206. 38. Norkin LC, Ouellette J. Cell killing by simian virus 40: Variation in the pattern of lysosomal enzyme release, cellular enzyme release, and cell death during productive infection of normal and simian virus 40-transformed simian cell lines. J Virol. 1976;18(1):48-57. 39. White MK, Khalili K. Expression of JC virus regulatory proteins in human cancer: Potential mechanisms for tumourigenesis. Eur J Cancer. 2005;41(16):2537-48. 40. Ferenczy MW, Marshall LJ, Nelson CD, Atwood WJ, Nath A, Khalili K, et al. Molecular biology, epidemiology, and pathogenesis of progressive multifocal leukoencephalopathy, the

JC virus-induced demyelinating disease of the human brain. Clin Microbiol Rev. 2012;25(3):471-506. 41. Jiang M, Abend JR, Johnson SF, Imperiale MJ. The role of polyomaviruses in human disease. Virology. 2009;384(2):266-73. 42. Doerries K. Human polyomavirus JC and BK persistent infection. Adv Exp Med Biol. 2006;577:102-16. 43. Egli A, Infanti L, Dumoulin A, Buser A, Samaridis J, Stebler C, et al. Prevalence of polyomavirus BK and JC infection and replication in 400 healthy blood donors. J Infect Dis. 2009;199(6):837-46. 44. Kling CL, Wright AT, Katz SE, McClure GB, Gardner JS, Williams JT, et al. Dynamics of urinary polyomavirus shedding in healthy adult women. J Med Virol. 2012;84(9):1459-63. 45. McClure GB, Gardner JS, Williams JT, Copeland CM, Sylvester SK, Garcea RL, et al. Dynamics of pregnancy-associated polyomavirus urinary excretion: A prospective longitudinal study. J Med Virol. 2012;84(8):1312-22. 46. White MK, Khalili K. Pathogenesis of progressive multifocal leukoencephalopathy--revisited. J Infect Dis. 2011;203(5):578-86. 47. Berger JR. Progressive multifocal leukoencephalopathy in acquired immunodeficiency syndrome: Explaining the high incidence and disproportionate frequency of the illness relative to other immunosuppressive conditions. J Neurovirol. 2003;9 Suppl 1:38-41. 48. Nickeleit V, Singh HK, Mihatsch MJ. Polyomavirus nephropathy: Morphology, pathophysiology, and clinical management. Curr Opin Nephrol Hypertens. 2003;12(6):599- 605. 49. Moens U, Van Ghelue M. Polymorphism in the genome of non-passaged human polyomavirus BK: Implications for cell tropism and the pathological role of the virus. Virology. 2005;331(2):209-31. 50. Nickeleit V, Mihatsch MJ. Polyomavirus nephropathy in native kidneys and renal allografts: An update on an escalating threat. Transpl Int. 2006;19(12):960-73. 51. Dropulic LK, Jones RJ. Polyomavirus BK infection in blood and marrow transplant recipients. Bone Marrow Transplant. 2008;41(1):11-8. 52. Moens U, Ludvigsen M, Van Ghelue M. Human polyomaviruses in skin diseases. Patholog Res Int. 2011;2011:123491. 53. Hirsch HH, Knowles W, Dickenmann M, Passweg J, Klimkait T, Mihatsch MJ, et al. Prospective study of polyomavirus type BK replication and nephropathy in renal-transplant recipients. N Engl J Med. 2002;347(7):488-96. 54. Viscidi RP, Khanna N, Tan CS, Li X, Jacobson L, Clifford DB, et al. JC virus antibody and viremia as predictors of progressive multifocal leukoencephalopathy in human immunodeficiency virus-1-infected individuals. Clin Infect Dis. 2011;53(7):711-5. 55. Nickeleit V, Klimkait T, Binet IF, Dalquen P, Del Zenero V, Thiel G, et al. Testing for polyomavirus type BK DNA in plasma to identify renal-allograft recipients with viral nephropathy. N Engl J Med. 2000;342(18):1309-15. 56. Bialasiewicz S, Whiley DM, Lambert SB, Nissen MD, Sloots TP. Detection of BK, JC, WU, or KI polyomaviruses in faecal, urine, blood, cerebrospinal fluid and respiratory samples. J Clin Virol. 2009;45(3):249-54. 57. Bialasiewicz S, Whiley DM, Lambert SB, Jacob K, Bletchly C, Wang D, et al. Presence of the newly discovered human polyomaviruses KI and WU in australian patients with acute respiratory tract infection. J Clin Virol. 2008;41(2):63-8. 58. Babakir-Mina M, Ciccozzi M, Perno CF, Ciotti M. The novel KI, WU, MC polyomaviruses: Possible human pathogens? New Microbiol. 2011;34(1):1-8.

59. Norja P, Ubillos I, Templeton K, Simmonds P. No evidence for an association between infections with WU and KI polyomaviruses and respiratory disease. J Clin Virol. 2007;40(4):307-11. 60. Knowles WA, Pillay D, Johnson MA, Hand JF, Brown DW. Prevalence of long-term BK and JC excretion in HIV-infected adults and lack of correlation with serological markers. J Med Virol. 1999;59(4):474-9. 61. Sharp CP, Norja P, Anthony I, Bell JE, Simmonds P. Reactivation and mutation of newly discovered WU, KI, and merkel cell carcinoma polyomaviruses in immunosuppressed individuals. J Infect Dis. 2009;199(3):398-404. 62. Duncavage EJ, Pfeifer JD. Human polyomaviruses 6 and 7 are not detectable in merkel cell polyomavirus-negative merkel cell carcinoma. J Cutan Pathol. 2011;38(10):790-6. 63. Sauvage V, Foulongne V, Cheval J, Ar Gouilh M, Pariente K, Dereure O, et al. Human polyomavirus related to african green monkey lymphotropic polyomavirus. Emerg Infect Dis. 2011;17(8):1364-70. 64. Foulongne V, Sauvage V, Hebert C, Dereure O, Cheval J, Gouilh MA, et al. Human skin microbiota: High diversity of DNA viruses identified on the human skin by high throughput sequencing. PLoS One. 2012;7(6):e38499. 65. Allen PJ, Bowne WB, Jaques DP, Brennan MF, Busam K, Coit DG. Merkel cell carcinoma: Prognosis and treatment of patients from a single institution. J Clin Oncol. 2005;23(10):2300- 9. 66. Rywlin AM. Malignant merkel-cell tumor is a more accurate description than trabecular carcinoma. Am J Dermatopathol. 1982;4(6):513-5. 67. Albores-Saavedra J, Batich K, Chable-Montero F, Sagy N, Schwartz AM, Henson DE. Merkel cell carcinoma demographics, morphology, and survival based on 3870 cases: A population based study. J Cutan Pathol. 2010;37(1):20-7. 68. Van Keymeulen A, Mascre G, Youseff KK, Harel I, Michaux C, De Geest N, et al. Epidermal progenitors give rise to merkel cells during embryonic development and adult homeostasis. J Cell Biol. 2009;187(1):91-100. 69. Hodgson NC. Merkel cell carcinoma: Changing incidence trends. J Surg Oncol. 2005;89(1):1- 4. 70. Lemos BD, Storer BE, Iyer JG, Phillips JL, Bichakjian CK, Fang LC, et al. Pathologic nodal evaluation improves prognostic accuracy in merkel cell carcinoma: Analysis of 5823 cases as the basis of the first consensus staging system. J Am Acad Dermatol. 2010;63(5):751-61. 71. Heath M, Jaimes N, Lemos B, Mostaghimi A, Wang LC, Penas PF, et al. Clinical characteristics of merkel cell carcinoma at diagnosis in 195 patients: The AEIOU features. J Am Acad Dermatol. 2008;58(3):375-81. 72. Lunder EJ, Stern RS. Merkel-cell carcinomas in patients treated with methoxsalen and ultraviolet A radiation. N Engl J Med. 1998;339(17):1247-8. 73. Zwald FO, Brown M. Skin cancer in solid organ transplant recipients: Advances in therapy and management: Part I. epidemiology of skin cancer in solid organ transplant recipients. J Am Acad Dermatol. 2011;65(2):253-61. 74. Gooptu C, Woollons A, Ross J, Price M, Wojnarowska F, Morris PJ, et al. Merkel cell carcinoma arising after therapeutic immunosuppression. Br J Dermatol. 1997;137(4):637-41. 75. Engels EA, Frisch M, Goedert JJ, Biggar RJ, Miller RW. Merkel cell carcinoma and HIV infection. Lancet. 2002;359(9305):497-8. 76. Kanitakis J, Euvrard S, Chouvet B, Butnaru AC, Claudy A. Merkel cell carcinoma in organ- transplant recipients: Report of two cases with unusual histological features and literature review. J Cutan Pathol. 2006;33(10):686-94.

77. Kurnatowska I, Zawiasa A, Narbutt J, Wagrowska-Danielewicz M, Stempien M, Nowicki M. Merkel cell carcinoma in a kidney transplant patient: Case report and update on management. Ann Transplant. 2010;15(3):66-70. 78. Kaae J, Hansen AV, Biggar RJ, Boyd HA, Moore PS, Wohlfahrt J, et al. Merkel cell carcinoma: Incidence, mortality, and risk of other cancers. J Natl Cancer Inst. 2010;102(11):793-801. 79. Feng H, Taylor JL, Benos PV, Newton R, Waddell K, Lucas SB, et al. Human transcriptome subtraction by using short sequence tags to search for tumor viruses in conjunctival carcinoma. J Virol. 2007;81(20):11332-40. 80. Van Ghelue M, Khan MT, Ehlers B, Moens U. Genome analysis of the new human polyomaviruses. Rev Med Virol. 2012;22(6):354-77. 81. Khalili K, White MK, Sawa H, Nagashima K, Safak M. The agnoprotein of polyomaviruses: A multifunctional auxiliary protein. J Cell Physiol. 2005;204(1):1-7. 82. Kwun HJ, Guastafierro A, Shuda M, Meinke G, Bohm A, Moore PS, et al. The minimum replication origin of merkel cell polyomavirus has a unique large T-antigen loading architecture and requires small T-antigen expression for optimal replication. J Virol. 2009;83(23):12118-28. 83. Feng H, Kwun HJ, Liu X, Gjoerup O, Stolz DB, Chang Y, et al. Cellular and viral factors regulating merkel cell polyomavirus replication. PLoS One. 2011;6(7):e22468. 84. Liu X, Hein J, Richardson SC, Basse PH, Toptan T, Moore PS, et al. Merkel cell polyomavirus large T antigen disrupts lysosome clustering by translocating human Vam6p from the cytoplasm to the nucleus. J Biol Chem. 2011;286(19):17079-90. 85. Meinke G, Bullock PA, Bohm A. Crystal structure of the simian virus 40 large T-antigen origin-binding domain. J Virol. 2006;80(9):4304-12. 86. Meinke G, Phelan P, Moine S, Bochkareva E, Bochkarev A, Bullock PA, et al. The crystal structure of the SV40 T-antigen origin binding domain in complex with DNA. PLoS Biol. 2007;5(2):e23. 87. Meinke G, Phelan PJ, Fradet-Turcotte A, Bohm A, Archambault J, Bullock PA. Structure- based analysis of the interaction between the simian virus 40 T-antigen origin binding domain and single-stranded DNA. J Virol. 2011;85(2):818-27. 88. Simmons DT. SV40 large T antigen functions in DNA replication and transformation. Adv Virus Res. 2000;55:75-134. 89. Valle M, Chen XS, Donate LE, Fanning E, Carazo JM. Structural basis for the cooperative assembly of large T antigen on the origin of replication. J Mol Biol. 2006;357(4):1295-305. 90. Li D, Zhao R, Lilyestrom W, Gai D, Zhang R, DeCaprio JA, et al. Structure of the replicative helicase of the oncoprotein SV40 large tumour antigen. Nature. 2003;423(6939):512-8. 91. Harrison CJ, Meinke G, Kwun HJ, Rogalin H, Phelan PJ, Bullock PA, et al. Asymmetric assembly of merkel cell polyomavirus large T-antigen origin binding domains at the viral origin. J Mol Biol. 2011;409(4):529-42. 92. Srinivasan A, McClellan AJ, Vartikar J, Marks I, Cantalupo P, Li Y, et al. The amino- terminal transforming region of simian virus 40 large T and small t antigens functions as a J domain. Mol Cell Biol. 1997;17(8):4761-73. 93. Kelley WL, Georgopoulos C. The T/t common exon of simian virus 40, JC, and BK polyomavirus T antigens can functionally replace the J-domain of the escherichia coli DnaJ molecular chaperone. Proc Natl Acad Sci U S A. 1997;94(8):3679-84. 94. Sullivan CS, Pipas JM. T antigens of simian virus 40: Molecular chaperones for viral replication and tumorigenesis. Microbiol Mol Biol Rev. 2002;66(2):179-202. 95. Campbell KS, Mullane KP, Aksoy IA, Stubdal H, Zalvide J, Pipas JM, et al. DnaJ/hsp40 chaperone domain of SV40 large T antigen promotes efficient viral DNA replication. Genes Dev. 1997;11(9):1098-110.

96. Chen HZ, Tsai SY, Leone G. Emerging roles of E2Fs in cancer: An exit from cell cycle control. Nat Rev Cancer. 2009;9(11):785-97. 97. van den Heuvel S, Dyson NJ. Conserved functions of the pRB and E2F families. Nat Rev Mol Cell Biol. 2008;9(9):713-24. 98. DeCaprio JA, Ludlow JW, Figge J, Shew JY, Huang CM, Lee WH, et al. SV40 large tumor antigen forms a specific complex with the product of the retinoblastoma susceptibility gene. Cell. 1988;54(2):275-83. 99. Houben R, Adam C, Baeurle A, Hesbacher S, Grimm J, Angermeyer S, et al. An intact retinoblastoma protein-binding site in merkel cell polyomavirus large T antigen is required for promoting growth of merkel cell carcinoma cells. Int J Cancer. 2012;130(4):847-56. 100. Arora R, Shuda M, Guastafierro A, Feng H, Toptan T, Tolstov Y, et al. Survivin is a therapeutic target in merkel cell carcinoma. Sci Transl Med. 2012;4(133):133ra56. 101. Cheng J, Rozenblatt-Rosen O, Paulson KG, Nghiem P, Decaprio JA. Merkel cell polyomavirus large T antigen has growth-promoting and inhibitory activities. J Virol. 2013;87(11):6118-26. 102. Vogelstein B, Lane D, Levine AJ. Surfing the p53 network. Nature. 2000;408(6810):307-10. 103. Vousden KH, Prives C. Blinded by the light: The growing complexity of p53. Cell. 2009;137(3):413-31. 104. Levine AJ. The common mechanisms of transformation by the small DNA tumor viruses: The inactivation of tumor suppressor gene products: P53. Virology. 2009;384(2):285-93. 105. Bolanos-Garcia VM, Blundell TL. BUB1 and BUBR1: Multifaceted kinases of the cell cycle. Trends Biochem Sci. 2011;36(3):141-50. 106. Elowe S. Bub1 and BubR1: At the interface between chromosome attachment and the spindle checkpoint. Mol Cell Biol. 2011;31(15):3085-93. 107. Nakamura T, Sato Y, Watanabe D, Ito H, Shimonohara N, Tsuji T, et al. Nuclear localization of merkel cell polyomavirus large T antigen in merkel cell carcinoma. Virology. 2010;398(2):273-9. 108. Shuda M, Arora R, Kwun HJ, Feng H, Sarid R, Fernandez-Figueras MT, et al. Human merkel cell polyomavirus infection I. MCV T antigen expression in merkel cell carcinoma, lymphoid tissues and lymphoid tumors. Int J Cancer. 2009;125(6):1243-9. 109. Shuda M, Kwun HJ, Feng H, Chang Y, Moore PS. Human merkel cell polyomavirus small T antigen is an oncoprotein targeting the 4E-BP1 translation regulator. J Clin Invest. 2011;121(9):3623-34. 110. Angermeyer S, Hesbacher S, Becker JC, Schrama D, Houben R. Merkel cell polyomavirus- positive merkel cell carcinoma cells do not require expression of the viral small T antigen. J Invest Dermatol. 2013; 133(8):2059-64. 111. Buchkovich NJ, Yu Y, Zampieri CA, Alwine JC. The TORrid affairs of viruses: Effects of mammalian DNA viruses on the PI3K-akt-mTOR signalling pathway. Nat Rev Microbiol. 2008;6(4):266-75. 112. Zerrahn J, Knippschild U, Winkler T, Deppert W. Independent expression of the transforming amino-terminal domain of SV40 large I antigen from an alternatively spliced third SV40 early mRNA. EMBO J. 1993;12(12):4739-46. 113. Comerford SA, Schultz N, Hinnant EA, Klapproth S, Hammer RE. Comparative analysis of SV40 17kT and LT function in vivo demonstrates that LT's C-terminus re-programs hepatic gene expression and is necessary for tumorigenesis in the liver. Oncogenesis. 2012;1:e28. 114. Sedman SA, Mertz JE. Mechanisms of synthesis of virion proteins from the functionally bigenic late mRNAs of simian virus 40. J Virol. 1988;62(3):954-61. 115. Pastrana DV, Tolstov YL, Becker JC, Moore PS, Chang Y, Buck CB. Quantitation of human seroresponsiveness to merkel cell polyomavirus. PLoS Pathog. 2009;5(9):e1000578.

116. Schowalter RM, Buck CB. The merkel cell polyomavirus minor capsid protein. PLoS Pathog. 2013;9(8):e1003558. 117. Forstova J, Krauzewicz N, Wallace S, Street AJ, Dilworth SM, Beard S, et al. Cooperation of structural proteins during late events in the life cycle of polyomavirus. J Virol. 1993;67(3):1405-13. 118. Kasamatsu H, Nehorayan A. Vp1 affects intracellular localization of Vp3 polypeptide during simian virus 40 infection. Proc Natl Acad Sci U S A. 1979;76(6):2808-12. 119. Lee S, Paulson KG, Murchison EP, Afanasiev OK, Alkan C, Leonard JH, et al. Identification and validation of a novel mature microRNA encoded by the merkel cell polyomavirus in human merkel cell carcinomas. J Clin Virol. 2011; 52(3):272-5 120. Seo GJ, Chen CJ, Sullivan CS. Merkel cell polyomavirus encodes a microRNA with the ability to autoregulate viral gene expression. Virology. 2009;383(2):183-7. 121. Sullivan CS, Grundhoff AT, Tevethia S, Pipas JM, Ganem D. SV40-encoded microRNAs regulate viral gene expression and reduce susceptibility to cytotoxic T cells. Nature. 2005;435(7042):682-6. 122. Kean JM, Rao S, Wang M, Garcea RL. Seroepidemiology of human polyomaviruses. PLoS Pathog. 2009;5(3):e1000363. 123. Nicol JT, Robinot R, Carpentier A, Carandina G, Mazzoni E, Tognon M, et al. Age-specific seroprevalence of merkel cell polyomavirus, human polyomaviruses 6, 7 and 9 and trichodysplasia spinulosa-associated polyomavirus. Clin Vaccine Immunol. 2013;20(3):363-8. 124. Touze A, Gaitan J, Arnold F, Cazal R, Fleury MJ, Combelas N, et al. Generation of merkel cell polyomavirus (MCV)-like particles and their application to detection of MCV antibodies. J Clin Microbiol. 2010;48(5):1767-70. 125. Viscidi RP, Rollison DE, Sondak VK, Silver B, Messina JL, Giuliano AR, et al. Age-specific seroprevalence of merkel cell polyomavirus, BK virus, and JC virus. Clin Vaccine Immunol. 2011;18(10):1737-43. 126. Tolstov YL, Pastrana DV, Feng H, Becker JC, Jenkins FJ, Moschos S, et al. Human merkel cell polyomavirus infection II. MCV is a common human infection that can be detected by conformational capsid epitope immunoassays. Int J Cancer. 2009;125(6):1250-6. 127. Carter JJ, Paulson KG, Wipf GC, Miranda D, Madeleine MM, Johnson LG, et al. Association of merkel cell polyomavirus-specific antibodies with merkel cell carcinoma. J Natl Cancer Inst. 2009;101(21):1510-22. 128. Touze A, Le Bidre E, Laude H, Fleury MJ, Cazal R, Arnold F, et al. High levels of antibodies against merkel cell polyomavirus identify a subset of patients with merkel cell carcinoma with better clinical outcome. J Clin Oncol. 2011;29(12):1612-9. 129. Faust H, Pastrana DV, Buck CB, Dillner J, Ekstrom J. Antibodies to merkel cell polyomavirus correlate to presence of viral DNA in the skin. J Infect Dis. 2011;203(8):1096- 100. 130. Foulongne V, Kluger N, Dereure O, Mercier G, Moles JP, Guillot B, et al. Merkel cell polyomavirus in cutaneous swabs. Emerg Infect Dis. 2010;16(4):685-7. 131. Paulson KG, Carter JJ, Johnson LG, Cahill KW, Iyer JG, Schrama D, et al. Antibodies to merkel cell polyomavirus T antigen oncoproteins reflect tumor burden in merkel cell carcinoma patients. Cancer Res. 2010;70(21):8388-97. 132. Paulson KG, Iyer JG, Tegeder AR, Thibodeau R, Schelter J, Koba S, et al. Transcriptome- wide studies of merkel cell carcinoma and validation of intratumoral CD8+ lymphocyte invasion as an independent predictor of survival. J Clin Oncol. 2011;29(12):1539-46. 133. Sihto H, Joensuu H. Tumor-infiltrating lymphocytes and outcome in merkel cell carcinoma, a virus-associated cancer. Oncoimmunology. 2012;1(8):1420-1.

134. Iyer JG, Afanasiev OK, McClurkan C, Paulson K, Nagase K, Jing L, et al. Merkel cell polyomavirus-specific CD8(+) and CD4(+) T-cell responses identified in merkel cell carcinomas and blood. Clin Cancer Res. 2011;17(21):6671-80. 135. Dowlatshahi M, Huang V, Gehad AE, Jiang Y, Calarese A, Teague JE, et al. Tumor-specific T cells in human merkel cell carcinomas: A possible role for tregs and T-cell exhaustion in reducing T-cell responses. J Invest Dermatol. 2013; 133(7):1879-89. 136. Kumar A, Chen T, Pakkanen S, Kantele A, Soderlund-Venermo M, Hedman K, et al. T- helper cell-mediated proliferation and cytokine responses against recombinant merkel cell polyomavirus-like particles. PLoS One. 2011;6(10):e25751. 137. Foulongne V, Dereure O, Kluger N, Moles JP, Guillot B, Segondy M. Merkel cell polyomavirus DNA detection in lesional and nonlesional skin from patients with merkel cell carcinoma or other skin diseases. Br J Dermatol. 2010;162(1):59-63. 138. Kassem A, Schopflin A, Diaz C, Weyers W, Stickeler E, Werner M, et al. Frequent detection of merkel cell polyomavirus in human merkel cell carcinomas and identification of a unique deletion in the VP1 gene. Cancer Res. 2008;68(13):5009-13. 139. Laude HC, Jonchere B, Maubec E, Carlotti A, Marinho E, Couturaud B, et al. Distinct merkel cell polyomavirus molecular features in tumour and non tumour specimens from patients with merkel cell carcinoma. PLoS Pathog. 2010;6(8):e1001076. 140. Martel-Jantin C, Filippone C, Cassar O, Peter M, Tomasic G, Vielh P, et al. Genetic variability and integration of merkel cell polyomavirus in merkel cell carcinoma. Virology. 2012;426(2):134-42. 141. Rodig SJ, Cheng J, Wardzala J, Dorosario A, Scanlon JJ, Laga AC, et al. Improved detection suggests all merkel cell carcinomas harbor merkel polyomavirus. J Clin Invest. 2012; 122(12):4645-53. 142. de Biase D, Ragazzi M, Asioli S, Eusebi V. Extracutaneous merkel cell carcinomas harbor polyomavirus DNA. Hum Pathol. 2012;43(7):980-5. 143. Loyo M, Guerrero-Preston R, Brait M, Hoque MO, Chuang A, Kim MS, et al. Quantitative detection of merkel cell virus in human tissues and possible mode of transmission. Int J Cancer. 2010;126(12):2991-6. 144. Fukumoto H, Sato Y, Hasegawa H, Katano H. Frequent detection of merkel cell polyomavirus DNA in sera of HIV-1-positive patients. Virol J. 2013;10:84,422X-10-84. 145. Wieland U, Mauch C, Kreuter A, Krieg T, Pfister H. Merkel cell polyomavirus DNA in persons without merkel cell carcinoma. Emerg Infect Dis. 2009;15(9):1496-8. 146. Bialasiewicz S, Lambert SB, Whiley DM, Nissen MD, Sloots TP. Merkel cell polyomavirus DNA in respiratory specimens from children and adults. Emerg Infect Dis. 2009;15(3):492-4. 147. Goh S, Lindau C, Tiveljung-Lindell A, Allander T. Merkel cell polyomavirus in respiratory tract secretions. Emerg Infect Dis. 2009;15(3):489-91. 148. Kantola K, Sadeghi M, Lahtinen A, Koskenvuo M, Aaltonen LM, Mottonen M, et al. Merkel cell polyomavirus DNA in tumor-free tonsillar tissues and upper respiratory tract samples: Implications for respiratory transmission and latency. J Clin Virol. 2009;45(4):292-5. 149. Bergallo M, Costa C, Terlizzi ME, Astegiano S, Curtoni A, Solidoro P, et al. Quantitative detection of the new polyomaviruses KI, WU and merkel cell virus in transbronchial biopsies from lung transplant recipients. J Clin Pathol. 2010;63(8):722-5. 150. Dworkin AM, Tseng SY, Allain DC, Iwenofu OH, Peters SB, Toland AE. Merkel cell polyomavirus in cutaneous squamous cell carcinoma of immunocompetent individuals. J Invest Dermatol. 2009;129(12):2868-74. 151. Campello C, Comar M, D'Agaro P, Minicozzi A, Rodella L, Poli A. A molecular case-control study of the merkel cell polyomavirus in colon cancer. J Med Virol. 2011;83(4):721-4.

152. Toracchio S, Foyle A, Sroller V, Reed JA, Wu J, Kozinetz CA, et al. Lymphotropism of merkel cell polyomavirus infection, nova scotia, canada. Emerg Infect Dis. 2010;16(11):1702- 9. 153. Bofill-Mas S, Rodriguez-Manzano J, Calgua B, Carratala A, Girones R. Newly described human polyomaviruses merkel cell, KI and WU are present in urban sewage and may represent potential environmental contaminants. Virol J. 2010;7:141,422X-7-141. 154. Mertz KD, Junt T, Schmid M, Pfaltz M, Kempf W. Inflammatory monocytes are a reservoir for merkel cell polyomavirus. J Invest Dermatol. 2010;130(4):1146-51. 155. Pancaldi C, Corazzari V, Maniero S, Mazzoni E, Comar M, Martini F, et al. Merkel cell polyomavirus DNA sequences in the buffy coats of healthy blood donors. Blood. 2011;117(26):7099-101. 156. Sadeghi M, Aronen M, Chen T, Jartti L, Jartti T, Ruuskanen O, et al. Merkel cell polyomavirus and trichodysplasia spinulosa-associated polyomavirus DNAs and antibodies in blood among the elderly. BMC Infect Dis. 2012;12(1):383. 157. Wieland U, Kreuter A. Merkel cell polyomavirus infection and merkel cell carcinoma in HIV- positive individuals. Curr Opin Oncol. 2011;23(5):488-93. 158. Wieland U, Silling S, Scola N, Potthoff A, Gambichler T, Brockmeyer NH, et al. Merkel cell polyomavirus infection in HIV-positive men. Arch Dermatol. 2011;147(4):401-6. 159. Husseiny MI, Anastasi B, Singer J, Lacey SF. A comparative study of merkel cell, BK and JC polyomavirus infections in renal transplant recipients and healthy subjects. J Clin Virol. 2010;49(2):137-40. 160. Kassem A, Technau K, Kurz AK, Pantulu D, Loning M, Kayser G, et al. Merkel cell polyomavirus sequences are frequently detected in nonmelanoma skin cancer of immunosuppressed patients. Int J Cancer. 2009;125(2):356-61. 161. Jung HS, Choi YL, Choi JS, Roh JH, Pyon JK, Woo KJ, et al. Detection of merkel cell polyomavirus in merkel cell carcinomas and small cell carcinomas by PCR and immunohistochemistry. Histol Histopathol. 2011;26(10):1231-41. 162. Mertz KD, Paasinen A, Arnold A, Baumann M, Offner F, Willi N, et al. Merkel cell polyomavirus large T antigen is detected in rare cases of nonmelanoma skin cancer. J Cutan Pathol. 2013; 40(6):543-9. 163. Hashida Y, Imajoh M, Nemoto Y, Kamioka M, Taniguchi A, Taguchi T, et al. Detection of merkel cell polyomavirus with a tumour-specific signature in non-small cell lung cancer. Br J Cancer. 2013;108(3):629-37. 164. Imajoh M, Hashida Y, Nemoto Y, Oguri H, Maeda N, Furihata M, et al. Detection of merkel cell polyomavirus in cervical squamous cell carcinomas and adenocarcinomas from japanese patients. Virol J. 2012;9:154,422X-9-154. 165. Scola N, Wieland U, Silling S, Altmeyer P, Stucker M, Kreuter A. Prevalence of human polyomaviruses in common and rare types of non-merkel cell carcinoma skin cancer. Br J Dermatol. 2012;167(6):1315-20. 166. Rollison DE, Giuliano AR, Messina JL, Fenske NA, Cherpelis BS, Sondak VK, et al. Case- control study of merkel cell polyomavirus infection and cutaneous squamous cell carcinoma. Cancer Epidemiol Biomarkers Prev. 2012;21(1):74-81. 167. Kaibuchi-Noda K, Yokota K, Matsumoto T, Sawada M, Sakakibara A, Kono M, et al. Detection of merkel cell polyomavirus in cutaneous squamous cell carcinoma before occurrence of merkel cell carcinoma. J Am Acad Dermatol. 2011;65(5):e152-4. 168. Garneski KM, Warcola AH, Feng Q, Kiviat NB, Leonard JH, Nghiem P. Merkel cell polyomavirus is more frequently present in north american than australian merkel cell carcinoma tumors. J Invest Dermatol. 2009;129(1):246-8. 169. Cimino PJ,Jr, Bahler DW, Duncavage EJ. Detection of merkel cell polyomavirus in chronic lymphocytic leukemia T-cells. Exp Mol Pathol. 2013;94(1):40-4.

170. Teman CJ, Tripp SR, Perkins SL, Duncavage EJ. Merkel cell polyomavirus (MCPyV) in chronic lymphocytic leukemia/small lymphocytic lymphoma. Leuk Res. 2011;35(5):689-92. 171. Haugg AM, Speel EJ, Pantulu ND, Pallasch C, Kurz AK, Kvasnicka HM, et al. Fluorescence in situ hybridization confirms the presence of merkel cell polyomavirus in chronic lymphocytic leukemia cells. Blood. 2011;117(21):5776-7. 172. Lasithiotaki I, Antoniou KM, Derdas SP, Sarchianaki E, Symvoulakis EK, Psaraki A, et al. The presence of merkel cell polyomavirus is associated with deregulated expression of BRAF and bcl-2 genes in non-small cell lung cancer. Int J Cancer. 2013;133(3):604-11. 173. Katano H, Ito H, Suzuki Y, Nakamura T, Sato Y, Tsuji T, et al. Detection of merkel cell polyomavirus in merkel cell carcinoma and kaposi's sarcoma. J Med Virol. 2009;81(11):1951- 8. 174. Pantulu ND, Pallasch CP, Kurz AK, Kassem A, Frenzel L, Sodenkamp S, et al. Detection of a novel truncating merkel cell polyomavirus large T antigen deletion in chronic lymphocytic leukemia cells. Blood. 2010;116(24):5280-4. 175. Sihto H, Kukko H, Koljonen V, Sankila R, Bohling T, Joensuu H. Clinical factors associated with merkel cell polyomavirus infection in merkel cell carcinoma. J Natl Cancer Inst. 2009;101(13):938-45. 176. Andres C, Belloni B, Puchta U, Sander CA, Flaig MJ. Re: Clinical factors associated with merkel cell polyomavirus infection in merkel cell carcinoma. J Natl Cancer Inst. 2009;101(23):1655-6. 177. Schrama D, Peitsch WK, Zapatka M, Kneitz H, Houben R, Eib S, et al. Merkel cell polyomavirus status is not associated with clinical course of merkel cell carcinoma. J Invest Dermatol. 2011;131(8):1631-8. 178. Kuwamoto S, Higaki H, Kanai K, Iwasaki T, Sano H, Nagata K, et al. Association of merkel cell polyomavirus infection with morphologic differences in merkel cell carcinoma. Hum Pathol. 2011;42(5):632-40. 179. Iwasaki T, Matsushita M, Kuwamoto S, Kato M, Murakami I, Higaki-Mori H, et al. Usefulness of significant morphologic characteristics in distinguishing between merkel cell polyomavirus-positive and merkel cell polyomavirus-negative merkel cell carcinomas. Hum Pathol. 2013;44(9):1912-7. 180. Bhatia K, Goedert JJ, Modali R, Preiss L, Ayers LW. Immunological detection of viral large T antigen identifies a subset of merkel cell carcinoma tumors with higher viral abundance and better clinical outcome. Int J Cancer. 2010;127(6):1493-6. 181. Nardi V, Song Y, Santamaria-Barria JA, Cosper AK, Lam Q, Faber AC, et al. Activation of PI3K signaling in merkel cell carcinoma. Clin Cancer Res. 2012;18(5):1227-36. 182. Handschel J, Muller D, Depprich RA, Ommerborn MA, Kubler NR, Naujoks C, et al. The new polyomavirus (MCPyV) does not affect the clinical course in MCCs. Int J Oral Maxillofac Surg. 2010;39(11):1086-90. 183. Bhatia K, Goedert JJ, Modali R, Preiss L, Ayers LW. Merkel cell carcinoma subgroups by merkel cell polyomavirus DNA relative abundance and oncogene expression. Int J Cancer. 2010;126(9):2240-6. 184. Sihto H, Kukko H, Koljonen V, Sankila R, Bohling T, Joensuu H. Merkel cell polyomavirus infection, large T antigen, retinoblastoma protein and outcome in merkel cell carcinoma. Clin Cancer Res. 2011;17(14):4806-13. 185. Higaki-Mori H, Kuwamoto S, Iwasaki T, Kato M, Murakami I, Nagata K, et al. Association of merkel cell polyomavirus infection with clinicopathological differences in merkel cell carcinoma. Hum Pathol. 2012;43(12):2282-91. 186. Houben R, Grimm J, Willmes C, Weinkam R, Becker JC, Schrama D. Merkel cell carcinoma and merkel cell polyomavirus: Evidence for hit-and-run oncogenesis. J Invest Dermatol. 2012;132(1):254-6.

187. zur Hausen H. Viruses in human cancers. Eur J Cancer. 1999;35(14):1878-85. 188. Sastre-Garau X, Peter M, Avril MF, Laude H, Couturier J, Rozenberg F, et al. Merkel cell carcinoma of the skin: Pathological and molecular evidence for a causative role of MCV in oncogenesis. J Pathol. 2009;218(1):48-56. 189. Houben R, Shuda M, Weinkam R, Schrama D, Feng H, Chang Y, et al. Merkel cell polyomavirus-infected merkel cell carcinoma cells require expression of viral T antigens. J Virol. 2010;84(14):7064-72. 190. Erovic BM, Al Habeeb A, Harris L, Goldstein DP, Ghazarian D, Irish JC. Significant overexpression of the merkel cell polyomavirus (MCPyV) large T antigen in merkel cell carcinoma. Head Neck. 2013;35(2):184-9. 191. Moore PS, Chang Y. Why do viruses cause cancer? highlights of the first century of human tumour virology. Nat Rev Cancer. 2010;10(12):878-89. 192. Bhatia S, Afanasiev O, Nghiem P. Immunobiology of merkel cell carcinoma: Implications for immunotherapy of a polyomavirus-associated cancer. Curr Oncol Rep. 2011;13(6):488-97. 193. Chen T, Hedman L, Mattila PS, Jartti T, Ruuskanen O, Soderlund-Venermo M, et al. Serological evidence of merkel cell polyomavirus primary infections in childhood. J Clin Virol. 2011;50(2):125-9. 194. Tolstov YL, Knauer A, Chen JG, Kensler TW, Kingsley LA, Moore PS, et al. Asymptomatic primary merkel cell polyomavirus infection among adults. Emerg Infect Dis. 2011;17(8):1371-80. 195. Mogha A, Fautrel A, Mouchet N, Guo N, Corre S, Adamski H, et al. Merkel cell polyomavirus small T antigen mRNA level is increased following in vivo UV-radiation. PLoS One. 2010;5(7):e11423. 196. Schmitt M, Wieland U, Kreuter A, Pawlita M. C-terminal deletions of merkel cell polyomavirus large T-antigen, a highly specific surrogate marker for virally induced malignancy. Int J Cancer. 2012;131(12):2863-8 197. Matthews MR, Wang RC, Reddick RL, Saldivar VA, Browning JC. Viral-associated trichodysplasia spinulosa: A case with electron microscopic and molecular detection of the trichodysplasia spinulosa-associated human polyomavirus. J Cutan Pathol. 2011;38(5):420-31. 198. Schwieger-Briel A, Balma-Mena A, Ngan B, Dipchand A, Pope E. Trichodysplasia spinulosa- -a rare complication in immunosuppressed patients. Pediatr Dermatol. 2010;27(5):509-13. 199. Haycox CL, Kim S, Fleckman P, Smith LT, Piepkorn M, Sundberg JP, et al. Trichodysplasia spinulosa--a newly described folliculocentric viral infection in an immunocompromised host. J Investig Dermatol Symp Proc. 1999;4(3):268-71. 200. Osswald SS, Kulick KB, Tomaszewski MM, Sperling LC. Viral-associated trichodysplasia in a patient with lymphoma: A case report and review. J Cutan Pathol. 2007;34(9):721-5. 201. Benoit T, Bacelieri R, Morrell DS, Metcalf J. Viral-associated trichodysplasia of immunosuppression: Report of a pediatric patient with response to oral valganciclovir. Arch Dermatol. 2010;146(8):871-4. 202. Blake BP, Marathe KS, Mohr MR, Jones N, Novosel T. Viral-associated trichodysplasia of immunosuppression in a renal transplant patient. J Drugs Dermatol. 2011;10(4):422-4. 203. Lee JS, Frederiksen P, Kossard S. Progressive trichodysplasia spinulosa in a patient with chronic lymphocytic leukaemia in remission. Australas J Dermatol. 2008;49(1):57-60. 204. Sadler GM, Halbert AR, Smith N, Rogers M. Trichodysplasia spinulosa associated with chemotherapy for acute lymphocytic leukaemia. Australas J Dermatol. 2007;48(2):110-4. 205. Sperling LC, Tomaszewski MM, Thomas DA. Viral-associated trichodysplasia in patients who are immunocompromised. J Am Acad Dermatol. 2004;50(2):318-22. 206. Wyatt AJ, Sachs DL, Shia J, Delgado R, Busam KJ. Virus-associated trichodysplasia spinulosa. Am J Surg Pathol. 2005;29(2):241-6.

207. Johne R, Muller H, Rector A, van Ranst M, Stevens H. Rolling-circle amplification of viral DNA genomes using phi29 polymerase. Trends Microbiol. 2009;17(5):205-11. 208. Kazem S, van der Meijden E, Feltkamp MC. The trichodysplasia spinulosa-associated polyomavirus: Virological background and clinical implications. APMIS. 2013;121(8):770-82 209. Elaba Z, Hughey L, Isayeva T, Weeks B, Solovan C, Solovastru L, et al. Ultrastructural and molecular confirmation of the trichodysplasia spinulosa-associated polyomavirus in biopsies of patients with trichodysplasia spinulosa. J Cutan Pathol. 2012;39(11):1004-9. 210. Siebrasse EA, Bauer I, Holtz LR, Le BM, Lassa-Claxton S, Canter C, et al. Human polyomaviruses in children undergoing transplantation, united states, 2008-2010. Emerg Infect Dis. 2012;18(10):1676-9. 211. Fischer MK, Kao GF, Nguyen HP, Drachenberg CB, Rady PL, Tyring SK, et al. Specific detection of trichodysplasia spinulosa-associated polyomavirus DNA in skin and renal allograft tissues in a patient with trichodysplasia spinulosa. Arch Dermatol. 2012;148(6):726- 33. 212. Kanitakis J, Kazem S, Van Der Meijden E, Feltkamp M. Absence of the trichodysplasia spinulosa-associated polyomavirus in human pilomatricomas. Eur J Dermatol. 2011;21(3):453-4. 213. Kumar A, Kantele A, Jarvinen T, Chen T, Kavola H, Sadeghi M, et al. Trichodysplasia spinulosa-associated polyomavirus (TSV) and merkel cell polyomavirus: Correlation between humoral and cellular immunity stronger with TSV. PLoS One. 2012;7(9):e45773. 214. Schmitz J. Serologic testing for infectious diseases. In: O’Gorman M, Donnenberg A, editors. Hand book of human immunology. Second edition ed. CRC Press, Taylor & Francis Group, LLC; 2008. p. 421-40. 215. Jeffery K, Aarons E. Diagnostic approaches. In: Zuckerman AJ, Banatvala JE, Schoub BD, Griffiths PD, Mortimer P, editors. Principles and practice of clinical virology. 6th ed. John Wiley & Sons Ltd; 2009. p. 1-28. 216. Stavnezer J. Immunoglobulin class switching. Curr Opin Immunol. 1996;8(2):199-205. 217. Kindt TJ, Goldsby RA, Osborne BA, Kuby J. B-cell generation,activation, and differentiation. In: Kuby Immunology. 6th ed. W. H. Freeman; 2007. p. 247-75. 218. Ziegner M, Steinhauser G, Berek C. Development of antibody diversity in single germinal centers: Selective expansion of high-affinity variants. Eur J Immunol. 1994;24(10):2393-400. 219. Dal Porto JM, Haberman AM, Shlomchik MJ, Kelsoe G. Antigen drives very low affinity B cells to become plasmacytes and enter germinal centers. J Immunol. 1998;161(10):5373-81. 220. Jacob J, Miller C, Kelsoe G. In situ studies of the antigen-driven somatic hypermutation of immunoglobulin genes. Immunol Cell Biol. 1992;70:145-52. 221. Liu YJ, Joshua DE, Williams GT, Smith CA, Gordon J, MacLennan IC. Mechanism of antigen-driven selection in germinal centres. Nature. 1989;342(6252):929-31. 222. Crowther JR. The ELISA guidebook. 2ed ed. Crowther JR, editor. Humana Press; 2009. 223. Hedman K, Lappalainen M, Seppaia I, Makela O. Recent primary toxoplasma infection indicated by a low avidity of specific IgG. J Infect Dis. 1989;159(4):736-40. 224. Bratthauer GL. The avidin-biotin complex (ABC) method and other avidin-biotin binding methods. Methods Mol Biol. 2010;588:257-70. 225. Diamandis EP, Christopoulos TK. The biotin-(strept)avidin system: Principles and applications in biotechnology. Clin Chem. 1991;37(5):625-36. 226. Barat B, Wu AM. Metabolic biotinylation of recombinant antibody by biotin ligase retained in the endoplasmic reticulum. Biomol Eng. 2007;24(3):283-91. 227. Hofmann K, Wood SW, Brinton CC, Montibeller JA, Finn FM. Iminobiotin affinity columns and their application to retrieval of streptavidin. Proc Natl Acad Sci U S A. 1980;77(8):4666- 8.

228. Sutton A, Vann WF, Karpas AB, Stein KE, Schneerson R. An avidin-biotin based ELISA for quantitation of antibody to bacterial polysaccharides. J Immunol Methods. 1985;82(2):215-24. 229. Kaikkonen L, Lankinen H, Harjunpaa I, Hokynar K, Soderlund-Venermo M, Oker-Blom C, et al. Acute-phase-specific heptapeptide epitope for diagnosis of parvovirus B19 infection. J Clin Microbiol. 1999;37(12):3952-6. 230. Lahtinen A, Kivela P, Hedman L, Kumar A, Kantele A, Lappalainen M, et al. Serodiagnosis of primary infections with human parvovirus 4, finland. Emerg Infect Dis. 2011;17(1):79-82. 231. Soderlund-Venermo M, Lahtinen A, Jartti T, Hedman L, Kemppainen K, Lehtinen P, et al. Clinical assessment and improved diagnosis of bocavirus-induced wheezing in children, finland. Emerg Infect Dis. 2009;15(9):1423-30. 232. Allander T, Jartti T, Gupta S, Niesters HG, Lehtinen P, Osterback R, et al. Human bocavirus and acute wheezing in children. Clin Infect Dis. 2007;44(7):904-10. 233. Jartti T, Lehtinen P, Vuorinen T, Osterback R, van den Hoogen B, Osterhaus AD, et al. Respiratory picornaviruses and respiratory syncytial virus as causative agents of acute expiratory wheezing in children. Emerg Infect Dis. 2004;10(6):1095-101. 234. Hammaren-Malmi S, Saxen H, Tarkkanen J, Mattila PS. Adenoidectomy does not significantly reduce the incidence of otitis media in conjunction with the insertion of tympanostomy tubes in children who are younger than 4 years: A randomized trial. Pediatrics. 2005;116(1):185-9. 235. Nejentsev S, Sjoroos M, Soukka T, Knip M, Simell O, Lovgren T, et al. Population-based genetic screening for the estimation of type 1 diabetes mellitus risk in finland: Selective genotyping of markers in the HLA-DQB1, HLA-DQA1 and HLA-DRB1 loci. Diabet Med. 1999;16(12):985-92. 236. Kupila A, Keskinen P, Simell T, Erkkila S, Arvilommi P, Korhonen S, et al. Genetic risk determines the emergence of diabetes-associated autoantibodies in young children. Diabetes. 2002;51(3):646-51. 237. Korhonen MH, Brunstein J, Haario H, Katnikov A, Rescaldani R, Hedman K. A new method with general diagnostic utility for the calculation of immunoglobulin G avidity. Clin Diagn Lab Immunol. 1999;6(5):725-8. 238. Li TC, Takeda N, Kato K, Nilsson J, Xing L, Haag L, et al. Characterization of self- assembled virus-like particles of human polyomavirus BK generated by recombinant baculoviruses. Virology. 2003;311(1):115-24. 239. Viscidi RP, Rollison DE, Viscidi E, Clayman B, Rubalcaba E, Daniel R, et al. Serological cross-reactivities between antibodies to simian virus 40, BK virus, and JC virus assessed by virus-like-particle-based enzyme immunoassays. Clin Diagn Lab Immunol. 2003;10(2):278- 85. 240. Goldmann C, Petry H, Frye S, Ast O, Ebitsch S, Jentsch KD, et al. Molecular cloning and expression of major structural protein VP1 of the human polyomavirus JC virus: Formation of virus-like particles useful for immunological and therapeutic studies. J Virol. 1999;73(5):4465-9. 241. Kosukegawa A, Arisaka F, Takayama M, Yajima H, Kaidow A, Handa H. Purification and characterization of virus-like particles and pentamers produced by the expression of SV40 capsid proteins in insect cells. Biochim Biophys Acta. 1996;1290(1):37-45. 242. Bodaghi S, Comoli P, Bosch R, Azzi A, Gosert R, Leuenberger D, et al. Antibody responses to recombinant polyomavirus BK large T and VP1 proteins in young kidney transplant patients. J Clin Microbiol. 2009;47(8):2577-85. 243. Randhawa P, Viscidi R, Carter JJ, Galloway DA, Culp TD, Huang C, et al. Identification of species-specific and cross-reactive epitopes in human polyomavirus capsids using monoclonal antibodies. J Gen Virol. 2009;90:634-9.

244. Knowles WA. Discovery and epidemiology of the human polyomaviruses BK virus (BKV) and JC virus (JCV). Adv Exp Med Biol. 2006;577:19-45. 245. van der Meijden E, Kazem S, Burgers MM, Janssens R, Bouwes Bavinck JN, de Melker H, et al. Seroprevalence of trichodysplasia spinulosa-associated polyomavirus. Emerg Infect Dis. 2011;17(8):1355-63. 246. Trusch F, Klein M, Finsterbusch T, Kuhn J, Hofmann J, Ehlers B. Seroprevalence of human polyomavirus 9 and cross-reactivity to african green monkey-derived lymphotropic polyomavirus. J Gen Virol. 2012;93:698-705. 247. Bredholt G, Olaussen E, Moens U, Rekvig OP. Linked production of antibodies to mammalian DNA and to human polyomavirus large T antigen: Footprints of a common molecular and cellular process? Arthritis Rheum. 1999;42(12):2583-92. 248. Knowles WA, Pipkin P, Andrews N, Vyse A, Minor P, Brown DW, et al. Population-based study of antibody to the human polyomaviruses BKV and JCV and the simian polyomavirus SV40. J Med Virol. 2003;71(1):115-23. 249. Stolt A, Sasnauskas K, Koskela P, Lehtinen M, Dillner J. Seroepidemiology of the human polyomaviruses. J Gen Virol. 2003;84(:1499-504. 250. Nguyen NL, Le BM, Wang D. Serologic evidence of frequent human infection with WU and KI polyomaviruses. Emerg Infect Dis. 2009;15(8):1199-205. 251. Neske F, Prifert C, Scheiner B, Ewald M, Schubert J, Opitz A, et al. High prevalence of antibodies against polyomavirus WU, polyomavirus KI, and human bocavirus in german blood donors. BMC Infect Dis. 2010;10:215,2334-10-215. 252. Nicol JT, Touze A, Robinot R, Arnold F, Mazzoni E, Tognon M, et al. Seroprevalence and cross-reactivity of human polyomavirus 9. Emerg Infect Dis. 2012;18(8):1329-32. 253. Baez CF, Guimaraes MA, Martins RA, Zalona AC, Cossatis JJ, Zalis MG, et al. Detection of merkel cell polyomavirus in oral samples of renal transplant recipients without merkel cell carcinoma. J Med Virol. 2013; 85(11):2016-9. 254. Foulongne V, Courgnaud V, Champeau W, Segondy M. Detection of merkel cell polyomavirus on environmental surfaces. J Med Virol. 2011;83(8):1435-9. 255. Martel-Jantin C, Pedergnana V, Nicol JT, Leblond V, Tregouet DA, Tortevoye P, et al. Merkel cell polyomavirus infection occurs during early childhood and is transmitted between siblings. J Clin Virol. 2013; 58(1):288-91 256. Knowles WA, Gibson PE, Gardner SD. Serological typing scheme for BK-like isolates of human polyomavirus. J Med Virol. 1989;28(2):118-23. 257. Jin L, Gibson PE, Knowles WA, Clewley JP. BK virus antigenic variants: Sequence analysis within the capsid VP1 epitope. J Med Virol. 1993;39(1):50-6. 258. Moens U, Van Ghelue M, Song X, Ehlers B. Serological cross-reactivity between human polyomaviruses. Rev Med Virol. 2013; 23(4):250-64 259. Kantola K, Sadeghi M, Ewald MJ, Weissbrich B, Allander T, Lindau C, et al. Expression and serological characterization of polyomavirus WUPyV and KIPyV structural proteins. Viral Immunol. 2010;23(4):385-93. 260. Abedi Kiasari B, Vallely PJ, Corless CE, Al-Hammadi M, Klapper PE. Age-related pattern of KI and WU polyomavirus infection. J Clin Virol. 2008;43(1):123-5. 261. Heritage J, Chesters PM, McCance DJ. The persistence of papovavirus BK DNA sequences in normal human renal tissue. J Med Virol. 1981;8(2):143-50. 262. Chesters PM, Heritage J, McCance DJ. Persistence of DNA sequences of BK virus and JC virus in normal human tissues and in diseased tissues. J Infect Dis. 1983;147(4):676-84. 263. Goudsmit J, Wertheim-van Dillen P, van Strien A, van der Noordaa J. The role of BK virus in acute respiratory tract disease and the presence of BKV DNA in tonsils. J Med Virol. 1982;10(2):91-9.

264. Carvalho JF, Blank M, Kiss E, Tarr T, Amital H, Shoenfeld Y. Anti-vitamin D, vitamin D in SLE: Preliminary results. Ann N Y Acad Sci. 2007;1109:550-7. 265. Ailus K, Melamies L, Tuomi T, Palosuo T, Aho K. Measuring rheumatoid factor in nonrheumatoid subjects: Immunoturbidimetric assay, latex slide test, and enzyme-linked immunosorbent assay compared. Clin Chem. 1991;37(10):1766-9. 266. Korpilahde T, Heliovaara M, Kaipiainen-Seppanen O, Knekt P, Aho K. Regional differences in finland in the prevalence of rheumatoid factor in the presence and absence of arthritis. Ann Rheum Dis. 2003;62(4):353-5. 267. Shim CN, Hwang JW, Lee J, Koh EM, Cha HS, Ahn JK. Prevalence of rheumatoid factor and parameters associated with rheumatoid factor positivity in korean health screening subjects and subjects with hepatitis B surface antigen. Mod Rheumatol. 2012;22(6):885-91. 268. Mock DM, Lankford GL, Mock NI. Biotin accounts for only half of the total avidin-binding substances in human serum. J Nutr. 1995;125(4):941-6. 269. Binder CJ. Natural IgM antibodies against oxidation-specific epitopes. J Clin Immunol. 2010;30 Suppl 1:S56-60. 270. Burastero SE, Casali P, Wilder RL, Notkins AL. Monoreactive high affinity and polyreactive low affinity rheumatoid factors are produced by CD5+ B cells from patients with rheumatoid arthritis. J Exp Med. 1988;168(6):1979-92. 271. Chu Q, Ludtke JJ, Subbotin VM, Blockhin A, Sokoloff AV. The acquisition of narrow binding specificity by polyspecific natural IgM antibodies in a semi-physiological environment. Mol Immunol. 2008;45(5):1501-13. 272. Notkins AL. Polyreactivity of antibody molecules. Trends Immunol. 2004 Apr;25(4):174-9. 273. McMahon R. Biotin in metabolism and molecular biology. Annu Rev Nutr. 2002;22:221-39. 274. Zempleni J, Mock DM. Biotin biochemistry and human requirements. J Nutr Biochem. 1999;10(3):128-38. 275. Dakshinamurti K, Chauhan J. Nonavidin biotin-binding proteins. Methods Enzymol. 1990;184:93-102. 276. Balnave D. Clinical symptoms of biotin deficiency in animals. Am J Clin Nutr. 1977;30(9):1408-13. 277. Wolf B. Clinical issues and frequent questions about biotinidase deficiency. Mol Genet Metab. 2010;100(1):6-13. 278. Sydenstricker VP, Singal SA, Briggs AP, Devaughn NM, Isbell H. Preliminary observations on "egg white injury" in man and its cure with a biotin concentrate. Science. 1942;95(2459):176-7. 279. Peters JM. A separation of the direct toxic effects of dietary raw egg white powder from its action in producing biotin deficiency. Br J Nutr. 1967;21(4):801-9. 280. Peters JM, Boyd EM. Toxic effects from a rancid diet containing large amounts of raw egg- white powder. Food Cosmet Toxicol. 1969;7(3):197-207.