Novel Human Adenovirus Strain, Bangladesh

Yuki Matsushima, Hideaki Shimizu, Atsuko Kano, Etsuko Nakajima, Yoko Ishimaru, Shuvra Kanti Dey, Yuki Watanabe, Fuyuka Adachi, Keiichiro Suzuki, Kohnosuke Mitani, Tsuguto Fujimoto, Tung Gia Phan, and Hiroshi Ushijima

Author affiliations: Kawasaki City Institute of Public Health, Kanagawa, Japan (Y. Matsushima, H. Shimizu, A. Kano, E. Nakajima, Y. Ishimaru); Nihon University School of Medicine, Tokyo, Japan (K. Dey, H. Ushijima); Saitama Medical School, Saitama, Japan (Y. Watanabe, F. Adachi, K. Suzuki, K. Mitani); National Institute of Infectious Diseases, Tokyo (T. Fujimoto); Blood Systems Research Institute, San Francisco, California, USA (T.G. Phan)

Abstract

We report a novel human adenovirus D (HAdV-65) isolated from feces of 4 children in Bangladesh who had acute gastroenteritis. Corresponding genes of HAdV-65 were related to a hexon gene of HAdV-10, penton base genes of HAdV-37 and HAdV-58, and a fiber gene of HAdV-9. This novel virus may be a serious threat to public health.

Human adenoviruses (HAdVs) are common pathogens that cause several diseases, such as pneumonia, acute gastroenteritis, and epidemic keratoconjunctivitis (1). HAdV infection is also associated with a serious adenovirus syndrome in immunocompromised patients after stem cell transplantation (2). Acute gastroenteritis causes illness and death in humans worldwide. Illness is associated with infection of enteric viruses, including rotavirus, astrovirus, norovirus, sapovirus, and adenovirus.
HAdVs are divided into 7 species (HAdV-A–G) on the basis of DNA genome homology. Most acute gastroenteritis related to HAdVs is caused by HAdV-F species (HAdV-40 and HAdV-41 (3,4). Recently, we detected HAdV-D in feces of children with diarrhea in Bangladesh (5). Other HAdV-D strains have also been associated with diarrhea in Kenya and Brazil (6,7).
We report a novel HAdV-D (HAdV-65) strain detected in feces of 4 children with acute gastroenteritis during October 2004–March 2005 in Bangladesh (5) and results of hexon, penton base, and fiber gene sequence analyses. We also report the full genome sequence of this virus, whose corresponding genes are closely related to the hexon gene of HAdV-10, penton base genes of HAdV-37 and HAdV-58, and the fiber gene of HAdV-9.

The Study

Cloned virus (3 plaque purifications) was propagated in an A549 cell line, which was maintained in minimal essential medium supplemented with 1% fetal bovine serum (Cansera International Inc., Toronto, Ontario, Canada). Cultures were observed for 3–4 weeks for a cytopathic effect. After a cytopathic effect was observed, cell lysates were centrifuged at 1,430 × g for 20 min at 4°C. Supernatants were centrifuged at 72,000 × g for 3 h at 4°C. Pellets were resuspended in sterile water and treated with 10 µL (20 mg/mL) of proteinase K. DNA was extracted by using the phenol:chloroform:isoamyl alcohol (25:24:1) method (Invitrogen, Carlsbad, CA, USA) and precipitated with isopropyl alcohol.
PCR was performed in a total volume of 50 μL containing 20 pmol/μL of each primer, 2.5 mmol/L of dNTP, 1.25 units of GXL DNA polymerase (Takara, Shiga, Japan), and 5 μL of DNA template. After PCR products were purified by using the MinElute PCR Purification Kit (QIAGEN, Hilden, Germany), cycle sequencing was conducted by using the Genome Lab DTCS Quick Start Kit (Beckman Coulter Inc., Fullerton, CA, USA).
The complete genome of HAdV-65 was sequenced by using the primer walking method. The 5′ terminus of full-length DNA was phosphorylated with 20 units of T4 polynucleotide kinase and 0.5 μL of 100 mmol/L ATP, and ligated to a blunt EcoRI-NotI-BamHI adaptor (1 pmol/μL) for 3 h at 8°C by using the DNA Ligation Mighty Mix Kit (Takara), PCRs with primer pairs containing adaptor sequences were conducted as described (8). DNA sequences were assembled by using the CEQ 2000XL DNA Analysis System version 4.3.9 (Beckman Coulter Inc.).
The genome of HAdV-65 was 35,172 bp. It had a GC content of 56.9%, and the inverted terminal repeat sequence of this virus was 150 bp. Phylogenetic trees were generated by using the maximum-likelihood method with MEGA5 (www.megasoftware.net