European Journal of Epidemiology 13: 653–657, 1997. 1997 Kluwer Academic Publishers. Printed in the Netherlands.
Detection of BK polyomavirus genotypes in healthy and HIV-positive children C. Di Taranto1, V. Pietropaolo1, G.B. Orsi2, L. Jin3, L. Sinibaldi1 & A.M. Degener4 1
Institute of Microbiology, 2 Institute of Hygiene, and 4 Department of Cellular & Developmental Biology, University ‘La Sapienza’, Rome, Italy; 3 Virus Reference Division, Central Public Health Laboratory, London, UK Accepted in revised form 20 February 1997
Abstract. Urine samples from 211 community children (3–7 years age), from 33 HIV type-1 infected children and from 56 HIV-negative children were collected and analyzed for the presence of BK virus (BKV) DNA by PCR. PCR amplifications were carried out using primers specific for the BKV structural region VP1. We also investigated the distribution of BKV subtypes by a restriction fragment polymorphism assay (RFLP). We demonstrated BKV DNA in 3.8% of 211 community children with a higher prevalence of subtype I. In HIV-1 positive
children we detected BKV DNA in 2 urine samples (6%) out of 33, both belonging to subtype I. The HIV-negative cluster did not show any positivity to BKV DNA. The results confirm a more frequent primary BKV infection in children of 3–5 years of age and a higher prevalence in hospitalized children affected by HIV-1. The most relevant finding was that among both the community and HIV-1 positive children the subtype I was the most frequently detected.
Key words: BK virus, Polymerase chain reaction, Primary infection, Restriction fragment length polymorphism assay Abbreviations: BKV = BK virus; PBS = phosphate buffer saline; PCR = polymerase chain reaction; RFLP = restriction fragment length polymorphism assay Introduction The human BK polyomavirus (BKV), originally isolated from the urine of a renal allograft recipient who developed ureteric stenosis [1], has a worldwide distribution in the human population [2]. Primary infection takes place during childhood [2–4] and is usually not apparent but may be accompanied by mild respiratory illness [5]. BKV seroconversion has been linked to respiratory infections and the rapid acquisition of specific antibodies during childhood is consistent with an infection disseminated from the respiratory tract. During primary infection, viremia occurs and the virus is transported to the kidney, where it persists indefinitely [6–8]. Regarding primary infection, the portals of entry, the target cells for primary replication and all the habor cells for latent BKV genome have yet to be identified. Immunocompromising conditions result in reactivation of virus and viruria [6, 9]. In fact infection with BKV has been demonstrated by detection of virus in urine in individuals after kidney and bone marrow transplant and in pregnant women. Moreover BKV is associated with ureteral stenosis in renal allograft recipients and hemorrhagic cystitis in recipients
of bone marrow transplants [10]. Recently the brain has also been included as a selective organ for latency, since BKV genomes have been recovered from brain tumors [11, 12]. Following the first isolation of BKV, numerous variants have been characterised and these display a variation in the structure of the non-coding control region. It is not yet clear whether these sequence differences are relevant to infection or reactivation in the natural host and in the determination of cellular tropism. Also four subtypes of BKV have been characterised by nucleotide sequencing of the VP1 region [13]. These sequences confirmed that this region (aa 61–aa 83) is conserved within each subtype and is responsible for its unique antigenicity. In the present study we carried out an epidemiological survey to investigate the BKV prevalence and subtype distribution in a cluster of community children between 3–7 years age. A second cluster of hospitalized children (33 HIV-1 positive and 56 HIV1 negative) was also selected to make a comparison with the preceding one. Urine samples from children were collected and analyzed for the presence of BKV DNA by PCR amplification with primers specific for the structural region (VP1). The distribution of BKV subtypes was investi-
654 gated using a Restriction Fragment Length Polymorphism Assay (RFLP) recently developed for subtyping BKV directly from clinical specimens [14].
Materials and methods Sampling. A cluster of community children aged 3–7 years from 2 schools was selected in the city of Rome. One school was in an urban area (Nazario Sauro School), while the second one was in suburban area (Angelini School). A medical team collected the urine samples from the children at the school and by means of sealed sterile glass containers transported the samples to the laboratory. General information about health conditions of the children was collected from the school doctors. The sampling was carried out between February–May 1994 (n = 211). Urine samples were collected again after 8 months from 8 children who resulted positive to BKV after the first test. After one year urine samples were collected again from all the previous 211 children and screened for BKV DNA presence. A second cluster of hospitalized children was selected from a city hospital between May–October 1995 and included 56 children HIV-1 negative, with characteristics very similar to the community children, and 33 children who were found to be HIV-1 positive. Also for the hospitalized cluster the urine samples were collected and transported to the laboratory under the same conditions. The age range (1–12 years) of the hospitalized children was wider than the community cluster. Data about the immunodeficiency level and the drug therapy were collected from the HIV-1 positive children. Immunophenotyping. Peripheral CD3+CD4+ lymphocytes were determined in EDTA-whole blood by flow cytometry, according to the CDC guidelines [15], using a mixture of fluorochrome-conjugated antibodies (Ortho, Raritan, NJ, USA). Samples treatment. Urine samples collected were stored at –80 °C until use. One ml of sample was centrifuged at 12000 g for 2 min, sediments were washed with phosphate buffer saline and resuspended in 100 µl of distilled water. After heating at 95 °C for 5 min and spinning at 12000 g for 10 sec, supernatants were stored at –20 °C for PCR amplification. Oligonucleotide primers. One set of primers (3271/327-2) from nucleotides (nt) 1630 to 1956 were used for amplification of the 61–83 amino acid (aa) region of VP1. The primers were chosen to anneal with the invariant region flanking the subtypespecific region in the VP1 of BKV [4]. The sensitivity and specificity of the chosen primers in PCR
were estimated by Jin et al. [16]. By this method it is possible to detect 10 ag of BKV DNA. PCR amplification. PCR amplifications were performed in a GeneAmp PCR System 2400 (PerkinElmer Cetus, Emeryville, CA). Precautions recommended by Kwok [17] were followed. Twenty-five pmol of each primer were used in a 50 µl reaction mix containing 1.25 U of Taq polymerase and 200 µM of each dNTP. Ten µl of resuspended urinary sediment was enclosed in the reaction. The reactions were subjected to an initial denaturation at 94 °C for 2 min, followed by 35 cycles consisting of 1 min at 55 °C, 1 min at 72 °C and 1 min at 91 °C, and an extension cycle of 1 min at 55 °C and 4 min at 72 °C. Ten µl of each PCR product were electrophoresed on a ethidium bromide-stained 3% agarose gel. Positive (BKV DNA) and negative (pure water) controls were set up in parallel to exclude contamination and false-negative results. The expected PCR product was 327 bp in length. Restriction fragment length polymorphism assay (RFLP). In order to subtype BKV present in PCR products, the RFLP method was used [13]. Ten µl of PCR product were digested with 1–2 U of endonuclease in a total of 20 µl of the supplied buffer at 37 °C for 2 hours. A two-step approach was performed. Digestion with Alu I enabled BKV strains I and II to be differentiated from subtypes III and IV. Subsequently, the digestion patterns produced by Xmn I and Ava II respectively were used to distinguish strains of subtypes I from II and subtypes III from IV. The reaction mixture was electrophoresed on a ethidium bromide-stained 3% agarose gel. Molecular hybridization. Amplified DNA products were applied onto nitrocellulose filters by dotblotting. A threefold volume of SCC (10 X) and formaldehyde (6.15 M) were added to each sample, denatured at 65 °C for 15 min and put in ice for 5 min. A plasmid containing the entire genome of PT strain of BKV was used as template in a PCR amplication in the presence of 32P-α dATP. The probe was purified from uncorporated nucleotides by filtration through a 4 ml Sephadex G 50 column and eluted with a 250 mM Na Cl 1 mM TRIS-HCl pH 7.5 1 mM EDTA pH 8 buffer. Twenty fractions were collected. The 2 or 3 fractions corresponding to the exclusion peak were collected and added to the hybridization buffer. Hybridization was performed with 4–5 × 106 cpm/ml of hybridization buffer under high stringency condition [18]. Statistical analysis. In order to compare quantitative data non parametric tests were used (Mann-Whitney) and data were expressed as medians and ranges. For qualitative data, the Fisher’s exact test or the contingency tables were used whenever necessary.
655 Results
both under treatment with AZT and with CD4 cells/mmc counts respectively 605 and 3720. In HIV-negative children the PCR analysis did not show any positivity to BKV DNA.
Two hundred and eleven urine samples were collected from the community children cluster and analyzed for the presence of BKV DNA by PCR amplification with primers specific for the structural region (VP1). A total of 8 urine samples (3.8%) were positive to PCR (Table 1). All the PCR products (positive and negative for BKV) were also employed in a molecular hybridization test. The results obtained confirmed the PCR results, without any additional positive sample. The distribution of BKV subtypes was investigated using a Restriction Fragment Length Polymorphism Assay (RFLP): subtype I (5 samples) was the most frequent to be detected, followed by subtype III (2) and subtype IV (1) (Table 1). We divided the community cluster in two age groups: 3–5 and 6–7 years age. Seven positive urine samples (5.2%) were collected from children 3–5 years age group. Only one sample (1.3%) was positive in 6–7 years age group. Urine specimens were collected from the 8 positive children and tested again after 8 months; in all cases PCR did not reveal any presence of BKV DNA. As to the results obtained by repeating the sampling on 211 children after one year, it can be said that in total they were similar for both the percentage of positivity and the prevalence of subtype I (data not reported). A second cluster of hospitalized children (33 HIV1 positive and 56 HIV-negative children) was selected from an ENT ward of the same hospital with characteristics very similar to the community children. In HIV-1 positive children BKV DNA was detected in 2 urine samples (6%) out of 33. The distribution of the subtypes was investigated using RFLP assay and both the isolates were found to belong to subtype I. In order to assess the effect of increasing immunosuppression, the HIV infected children were subdivided into three groups according to CD4 level as is shown in Table 2. The two positive children were both 1 year old,
Discussion As already mentioned in the Introduction, BKV infection is present everywhere in the world. The primary infection appears in childhood, but it is difficult to recognize due to very few or no clinical symptoms. Usually data regarding primary infection were obtained by means of serological studies, depending on the difficult viral isolation. PCR permits the identification of the viral genomic DNA present in urine, determining the viral spreading period. Consequently PCR is a valid method to demonstrate the presence of viral DNA in body fluids and to ascertain the viral circulation. In this context the determination of BKV DNA was performed by PCR using primers specific for the VP1 region. Other studies were carried out collecting samples from hospitalized or immunocompromised groups of the population. The aim of this study was to investigate the epidemiological prevalence of BKV primary infection in children of 3–7 years age. Also the diffusion of the various subtypes was evaluated. The community cluster was relatively large and included only healthy children. Overall BKV DNA prevalence in urine samples was 3.8% and appears lower than that observed in other countries. This is probably due to the fact that those studies were carried out exclusively with hospitalized children [14]. The 3–5 year age group presented a higher prevalence (5.2%) by comparison with the 6–7 age group (1.3%). These data, in accordance with other studies [14, 17], suggest that the BKV primary infection occurs mainly in the first years of life of the child. This is supported by the results achieved in the hospitalized non-HIV cluster, where no BKV DNA specimen was detected. These results could be explained by the following considerations: (1) the higher age range in hospitalized HIV-negative
Table 1. BKV diffusion in community children
Age
Number of of children
Number of positive
Percentage of positive
Subtype I
II
III
IV
3–5 6–7
134 077
7 1
5.2* 1.3*
4 1
– –
2 –
1 –
Total
211
8
3.8
5
–
2
1
* p = 0.001.
656 Table 2. BKV diffusion in HIV-positive children
References
CD4 cells
Number of positive
Number of negative
Total
> 500 201–500 < 200 ND*
2 – – –
06 10 13 02
08 10 13 02
Total
2
31
33
* ND: no data.
children (6–12 years old), by comparison with that of school children (3–7 years old); (2) also the HIVnegative cluster was selected from an ENT ward with children who did not present any respiratory infectious disease. However the two hospital clusters presented the same age range. For this reason it was possible to carry out comparisons between the two groups. Hospitalized children affected by HIV-1 presented the highest prevalence 2/33 (6%), in spite of their age range (1–12 years old) which was wider than that of the community cluster (3–7 years old). It must be pointed out, however, that both HIV-1/BKV positive children were 1 year old and therefore a phenomenon of reactivation would be excluded. Subtype I was the most frequently detected among both the community and HIV-1 positive children. This suggests that subtype I is probably the most frequently responsible for the primary BKV infection in Rome. These data are in contrast with our results achieved in HIV positive adults where subtypes II and IV were more frequent. In conclusion the low BKV infection prevalence in school children, the similar rate during a period of two year sampling and the community characteristics of the cluster which has been studied suggest that in our area the BKV infection diffusion in children is endemic rather than epidemic. There is the need for further study to ascertain why these age groups (children and adults) present different subtype diffusions. Various hypotheses should be looked at: (1) reinfection by different subtypes; (2) the possibility for mixed infections which could make possible genomic viral recombinations; (3) the high genetic variability which has been demonstrated during BK viral replication in vivo and in vitro.
Acknowledgments This work was supported by contributions of Project AIDS No. 9305-29 (1995, Italy) and Istituto PasteurFondazione Cenci Bolognetti.
01. Gardner SD, Field AM, Coleman DV, Hulme B. New human papovavirus (BK) isolated from urine after renal transplantation. Lancet 1971(i): 1253–1257. 02. Brown P, Tsai T, Gajdusek C. Seroepidemiology of human papovaviruses. Discovery of virgin populations and some unusual patterns of antibody prevalence among remote peoples of the world. Am J Epidemiol 1975; 102: 331–340. 03. Flaegstad T, Traavik T, Kristiansen BE. Age-dependent prevalence of BK virus IgG and IgM antibodies measured by enzyme-linked immunosorbent assays. (ELISA) J Hyg 1986; 96: 523–528. 04. Sundsfjord A, Flaegstad T, Flo R, Spein AR, Pedersen M, Permin H, Julsrud J, Traavik T. BK and JC viruses in human immunodeficiency virus type I-infected persons: Prevalence, excretion, viremia and viral regulatory regions. J Infect Dis 1994; 169: 485–490. 05. Goudsmit J, Wertheim-van Dillen P, van Strien A, vander 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: 91–99. 06. Arthur RR, Shah KV. Occurrence and significance of papovaviruses BK and JC in urine. Prog Med Virol 1989; 36: 42–61. 07. Dorries K, ter Meulen V. Progressive multifocal leucoencephalophaty: Detection of papovavirus JC in kidney tissue. J Med Virol 1983; 11: 307–317. 08. Markowitz RB, Thompson HC, Mueller JF, Cohen JA, Dynan WS. Incidence of BK virus and JC virus viruria in human immunodeficiency virus-infected and -uninfected subjects. J Infect Dis 1993; 167: 13–20. 09. Vallbracht A, Lohler J, Gossmann J, Gluck T, Petersen D, Gerth HJ, Gencic M, Dorries K. Disseminated BK type polyomavirus infection in an AIDS patient associated with central nervous system disease. Am J Pathol 1993; 143: 29–39. 10. Arthur RR, Shah KV, Charache P, Saral R. BK and JC virus infections in recipients of bone marrow transplants. J Infect Dis 1988; 158: 563–569. 11. Dorries K, Loeber G, Meixensberger J. Association of polyomaviruses JC, SV40 and BK with human brain tumors. Virology 1987; 160: 268–270. 12. Elsner C, Dorries K. Evidence of human polyomavirus BK and JC infection in normal brain tissue. Virology 1992; 1991: 72–80. 13. Jin L. Rapid genomic typing of BK virus directly from clinical specimens. Mol Cell Probes 1993; 7: 331–334. 14. Jin L, Pietropaolo V, Booth JC, Ward KH, Brown DWG. Prevalence and subtypes distribution of BK virus in healthy people and immunocompromised patients detected by PCR-Restriction Enzyme Analysis. Clin Diagn Virol 1995; 3: 285–295. 15. Centers for Disease Control. Guidelines for the performance of CD4+ cells determinations in patients with human immunodeficiency virus infection. MMWR 1992; 41: 1–17. 16. Jin L, Gibson PE, Booth JC, Clewley JP. Genomic typing of BK virus in clinical specimens by direct sequencing of polimerase chain reaction products. J Med Virol 1993; 41: 11–17. 17. Kwok S, Higuchi R. Avoiding false positive with PCR. Nature 1989; 339: 237–238.
657 18. Sambrook J, Fritsch EF, Maniatis T. Molecular cloning. New York: CSH Laboratory Press, 1989: 153–173. 19. McCance DJ, Gardner SD. Papovaviruses: Papillomaviruses and polyomaviruses. In: Zuckerman AJ, Banatvala J, Pattison JR (eds), Principles and practice of clinical virology. New York: John Wiley and Sons, 1987: 479–506.
Address for correspondence: Pietropaolo Valeria, Institute of Microbiology, University ‘La Sapienza’, P.le Aldo Moro 5, I-00185 Rome, Italy Phone: +39 6 49914627; Fax: +39 6 49914626 E-mail:
[email protected]
