Global Distribution of Human Protoparvoviruses – Volume 24, Number 7—July 2018 – Emerging Infectious Diseases journal – CDC
Global Distribution of Human Protoparvoviruses – Volume 24, Number 7—July 2018 – Emerging Infectious Diseases journal – CDC
Global Distribution of Protoparvoviruses Download the PDF of the podcast transcript (66KB) Dr. Eric Delwart, an investigator at the Blood Systems Research Institute, discusses the global distribution of protoparvoviruses. Running time = 27:46 Read the associated article in the July 2018 issue of the EID Journal
Global Distribution of Human Protoparvoviruses — E. Väisänen et al.
Volume 24, Number 7—July 2018 Research Global Distribution of Human Protoparvoviruses On This Page Materials and Methods Results Discussion Cite This Article Figures Figure 1 Figure 2 Tables Table 1 Table 2 Table 3 Podcast Listen to audio/Podcast Downloads Article RIS [TXT – 2 KB] Altmetric Metric Details Elina Väisänen, Ushanandini Mohanraj, Paula M. Kinnunen 1 , Pikka Jokelainen, Haider Al-Hello, Ali M. Barakat, Mohammadreza Sadeghi 2 , Farid A. Jalilian, Amir Majlesi, Moses Masika, Dufton Mwaengo, Omu Anzala, Eric Delwart, Olli Vapalahti, Klaus Hedman, and Maria Söderlund-Venermo Author affiliations: University of Helsinki, Helsinki, Finland (E. Väisänen, U. Mohanraj, P.M. Kinnunen, P. Jokelainen, O. Vapalahti, K. Hedman, M. Söderlund-Venermo) ; Statens Serum Institut, Copenhagen, Denmark (P. Jokelainen) ; Estonian University of Life Sciences, Tartu, Estonia (P. Jokelainen) ; National Institute for Health and Welfare, Helsinki (H. Al-Hello); Al-Hussein Teaching Hospital, Thi-Qar Governorate, Iraq (A.M. Barakat) ; Blood Systems Research Institute, San Francisco, California, USA (M. Sadeghi, E. Delwart) ; University of California, San Francisco (M. Sadeghi, E. Delwart) ; Hamadan University of Medical Sciences, Hamadan, Iran (F.A. Jalilian, A. Majlesi) ; University of Nairobi, Nairobi, Kenya (M. Masika, D. Mwaengo, O. Anzala) ; Helsinki University Hospital, Helsinki (O. Vapalahti, K. Hedman) Cite This Article Abstract Development of next-generation sequencing and metagenomics has revolutionized detection of novel viruses. Among these viruses are 3 human protoparvoviruses: bufavirus, tusavirus, and cutavirus. These viruses have been detected in feces of children with diarrhea. In addition, cutavirus has been detected in skin biopsy specimens of cutaneous T-cell lymphoma patients in France and in 1 melanoma patient in Denmark. We studied seroprevalences of IgG against bufavirus, tusavirus, and cutavirus in various populations (n = 840), and found a striking geographic difference in prevalence of bufavirus IgG. Although prevalence was low in adult populations in Finland (1.9%) and the United States (3.6%), bufavirus IgG was highly prevalent in populations in Iraq (84.8%), Iran (56.1%), and Kenya (72.3%). Conversely, cutavirus IgG showed evenly low prevalences (0%–5.6%) in all cohorts, and tusavirus IgG was not detected. These results provide new insights on the global distribution and endemic areas of protoparvoviruses. Parvoviruses are small, nonenveloped, single-stranded DNA viruses that infect a wide variety of animals ranging from insects and shrimp to birds and mammals. Human parvoviruses belong to 4 genera: Erythroparvovirus , Bocaparvovirus , Tetraparvovirus , and Dependoparvovirus ( 1 ). The recently described bufavirus, tusavirus, and cutavirus are the first members of the genus Protoparvovirus found in humans. All 3 viruses were identified by next-generation sequencing and metagenomics in feces of children with diarrhea: bufavirus from Burkina Faso in 2012, tusavirus from Tunisia in 2014, and cutavirus from Brazil and Botswana in 2016 ( 2 – 4 ). In addition, cutavirus was detected by in silico analysis of existing next-generation sequencing libraries and by PCR of malignant skin tissues of patients in France with cutaneous T-cell lymphoma ( 4 ). To date, 3 genotypes of bufavirus have been detected, and bufavirus DNA has been detected in 1 nasal swab specimen of a child in Finland and in 0.1 by using a competition assay, as described ( 12 , 27 ). In the competition assay, serum antibodies were blocked separately with 3 unbiotinylated antigens in solution: the same (homologous) antigen as in the EIA, the heterologous antigen of the phylogenetically closest protoparvovirus, and the heterologous antigen of a more distant protoparvovirus, before repeating the EIAs. A sample was considered IgG positive when full homologous blocking but no (or partial) heterologous blocking occurred, as described ( 12 ). Statistical Analysis We performed statistical analysis by using 2 × 2 tables and test statistics (mid p-exact value) in OpenEpi software ( https://www.OpenEpi.com ). A 2-tailed p value 0.5 for bufavirus 2 or cutavirus, a reaction was always observed with the other antigen. In samples showing weak reactivity (OD 0.1–0.5) in cutavirus or bufavirus 2 IgG EIAs, both cross-reactivities and single-specific reactivities were observed. In the competition assay, the specific reactivity of bufavirus or cutavirus was blocked completely only by homologous antigen, whereas cross-reactive reactions were blocked by homologous and heterologous antigens. In several instances, heterologous antigen slightly reduced the specific EIA reactivity; however, this reduction was much less than that caused by homologous antigen. In cohorts from the Middle East and Africa that showed high prevalences of bufavirus IgG, some cross-reactivity was also observed among the 3 bufavirus genotypes, mostly for samples with higher ODs. However, the competition assay used for all positive samples could distinguish genotype-specific reactivity for correct interpretation. Tusavirus IgG did not cross-react with bufavirus IgGs or cutavirus IgG. Bufavirus IgG in Adults Bufavirus IgG was rare among veterinarians from Finland and blood donors from the United States: only 1.9% of the veterinarians and 3.6% of the blood donors had bufavirus IgG ( Table 2 ). In these cohorts, each bufavirus IgG–positive person had antibodies against only 1 bufavirus genotype. No indications of specific animal contact being associated with bufavirus seropositivity were found when we compared background information for bufavirus IgG–positive and bufavirus IgG–negative veterinarians. For blood donors from the United States, all 3 bufavirus IgG–positive samples were from Mississippi. However, all samples from Arizona were negative for bufavirus IgG. The most commonly detected genotype was bufavirus 1 in Finland and bufavirus 3 in the United States ( Table 2 ). In striking contrast to adults from the United States and Finland, including our previous results for students and staff members from Finland ( 12 ), bufavirus IgG was common in Iraq, Iran, and Kenya, for which 84.8%, 56.1%, and 72.3%, respectively, of adult populations had IgG against > 1 bufavirus genotypes ( Table 2 ). In the Middle East, bufavirus 1 was the most common type, whereas in Kenya, bufavirus 3 was the predominant genotype ( Table 2 ). Bufavirus 2 was the second most prevalent bufavirus in all 3 high-prevalence countries. In Iraq, we found that 30 (30.3%) of 99 persons had antibodies against 2 bufavirus genotypes, and 7 (7.0%) of 99 persons had antibodies against all 3 bufavirus genotypes. In adults from Kenya, we found similar prevalences: 30 (25.2%) of 119 persons had antibodies against 2 bufavirus genotypes, and 6 (5.0%) of 119 persons had antibodies against all 3 bufavirus genotypes. However, in Iran, we found that double or triple prevalences were lower: 14 (13.1%) of 107 persons had antibodies against 2 bufavirus genotypes, and 2 (1.9%) of 107 persons had antibodies against all 3 bufavirus genotypes. In Kenya, HIV-positive patients had a similar bufavirus IgG prevalence as the rest of the cohort: 78.9% (30/38) in HIV-positive adults (mean age 46.3 years, range 27–85 years) vs. 69.1% (56/81) in HIV-negative adults (mean age 41.9 years, range 18–88 years) (p = 0.275). When we compared only the 38 HIV-positive persons and 60 HIV-negative persons within the same age range (27–85 years), bufavirus seroprevalences were even more similar: 79% for HIV-positive persons and 75% for HIV-negative persons (p = 0.669). However, possible undiagnosed cases of infection with HIV and unequal numbers could affect the accuracy of this comparison. When the adult cohorts were analyzed more closely and persons were divided by age into equal-sized groups of 40 years of age, we found that for younger adults in Iran, bufavirus seroprevalence was lower than that for older adults (21/54 [38.9%] vs. 39/53 [73.6%]; p = 0.0003). However, a similar distinction was not observed for adults from Iraq or Kenya ( Table 3 ). For veterinarians from Finland and adults from the United States, we found that younger adults also had a lower bufavirus seroprevalence, albeit without statistical power, because of the low overall prevalence of bufavirus IgG in these countries. When we divided the cohorts into persons 30 years of age, a similar trend was also observed for adults in Kenya (14/35 [56.0%] vs. 70/92 [76.1%]; p = 0.0595). This trend was not observed for persons in Iraq. Bufavirus IgG in Children in Kenya Figure 2 . Seroprevalence of bufavirus in Kenya, by age. Several persons (mostly adults) had IgG against > 1 bufavirus genotypes; such persons are counted as 1 person in the bufavirus column. Differences in… In Kenya, the bufavirus IgG prevalence in children was significantly lower than that in adults (20.6% in children <18 years of age vs. 72.3% in adults; p<0.0001), but we observed similar proportions of bufavirus genotypes and a predominance of bufavirus 3 in both adults and children. ( Table 2 ). When we divided the cohort of children into those <5 years of age and those 5–17 years of age, the prevalence of bufavirus IgG by age increased from 12% to 28.1% ( Figure 2 ). Cutavirus and Tusavirus IgG in Adults and Children The prevalence of cutavirus IgG was generally low for all groups, ranging from 1.0% in Iraq to 5.6% in Iran, and cutavirus IgG was not detected in adults in the United States ( Table 2 ). In veterinarians in Finland, cutavirus IgG (4.9%) was more common than bufavirus IgG (1.9%) (p = 0.032). Two adults from Kenya and 1 veterinarian from Finland had both cutavirus IgG and bufavirus 2 IgG in their samples, which showed that these 2 antigenically similar viruses can infect the same person and elicit specific immune responses against each virus. However, for 6 patients (1 in Finland, 2 in Iraq, and 3 in Kenya), we could not determine whether the reactivity detected was specific for bufavirus 2, cutavirus, or both. These results were not included in the final prevalence calculations ( Table 2 ). Tusavirus IgG was not detected in any cohort ( Table 2 ). Top Discussion During the current decade, several new parvoviruses have been detected, mostly because of the development of NGS methods. Bufavirus, tusavirus, and cutavirus are the newest of these viruses detected in human samples ( 2 – 4 ). Bufavirus has been associated with gastroenteritis, and cutavirus is being studied for its relationship to skin cancers ( 4 , 8 – 10 , 12 , 24 ). However, studies that attempted to detect bufavirus, tusavirus, and cutavirus DNA in any sample type or virus antibodies in serum samples have been infrequent ( 13 ). We found high (50%–85%) seroprevalences of bufavirus IgG in cohorts from the Middle East and Africa, which indicated that bufavirus infections are endemic to these areas. The observed low (1.9%) seroprevalence in veterinarians in Finland is consistent with our previous results for staff members and medical students born in Finland (3.1%) ( 12 ). The seroprevalence of bufavirus in the United States was similar to that in Finland, although the major genotype was different. In contrast to the diverse epidemiology of antibodies against bufavirus, antibodies against cutavirus appeared globally and were much more evenly distributed and showed a low prevalence. These results provide new insights on the global distribution and identify areas to which protoparvoviruses are endemic. Because the difference in seroprevalence between persons born in Finland and staff born in Asia in our previous study could also be caused by more frequent animal contacts for 5 persons from Asia ( 12 ), we included veterinarians in this current study. However, no specific animal contacts for veterinarians from Finland were associated with bufavirus IgG or cutavirus IgG seropositivity. Although species jumps have occurred within protoparvoviruses ( 28 ), this result is consistent with the general rule of host-order specificity of parvoviruses ( 1 ). No animal contact information was available for persons from the Middle East, Kenya, or the United States. The age group results from Kenya, which showed continuously increasing seroprevalences of all 3 bufavirus genotypes, also showed that bufaviruses infect persons of all ages. The lower seroprevalence in children 1 bufavirus genotypes; such persons are counted as 1 person in the bufavirus column…. Tables Table 1 . Characteristics of cohorts used in study of global distribution of human protoparvoviruses Table 2 . Seroprevalence of IgG against protoparvoviruses in different population cohorts Table 3 . Seroprevalence of bufavirus IgG in adult cohorts, by age, in study of global distribution of human protoparvoviruses Top Cite This Article
DOI: 10.3201/eid2407.172128 1 Current affiliation: Finnish Food Safety Authority Evira, Helsinki, Finland. 2 Current affiliation: University of Turku, Turku, Finland.