Serotyping of adenoviruses on conjunctival scrapings by PCR and sequence analysis

Abstract

To detect and identify adenovirus (Ad), we investigated hypervariable regions (HVRs) of Ad by using a combination of PCR and direct sequencing (PCR-sequence) method. Primers for nested PCR to amplify the conserved region in the hexon protein containing HVRs were designed based on hexon gene sequences derived from GenBank. These two primer sets amplified a DNA fragment of 7 HVRs from 16 prototypes of Ad, which were divided into five subgenera, including seven serotypes that are the predominant causative agents of acute conjunctivitis in Japan, and from 31 recent conjunctival scraping specimens from patients with adenoviral conjunctivitis. HVR DNA sequences were determined by means of universal sequence primers. Analysis of the predicted amino acid homology of HVRs among Ad prototypes suggested three regions, HVR4, -5, and -7, to be candidates for the neutralization epitopes. The clinical serotype of specimens was determined by the PCR-sequence method with reference to these three HVRs. The serotype determined according to this method was identical to that obtained by culture isolation and the neutralization test (NT) in all scraping samples, whereas the results of this method did not match PCR and restriction fragment length polymorphism (PCR-RFLP) analysis in five samples. It took only three days to detect Ad and to identify the serotype, in contrast to culture isolation-NT, which took at least 2 weeks. These findings indicate that our newly developed PCR-sequence method is applicable for the detection and serotyping of human Ads.

FIG. 1

Ad genome. Gene map of the regions surrounding the Ad hexon protein showing the seven HVRs, the primers, and the regions sequenced.

FIG. 2

Electrophoresis of PCR-amplified products. Agarose gel electrophoresis shows the specific band of ca. 2.8 kbp obtained in the first PCR (a) and of ca. 1.8 kbp obtained in the second PCR (b) from DNA samples of all 16 prototypes. Numbers above the lanes correspond to the Ad serotypes, and lane M contains molecular weight standards (pHY).

FIG. 3

Comparison of predicted amino acid sequences of seven hexon HVRs among 16 prototypes. The sequences of the region of loop 1 and loop 2 have been aligned to obtain maximal homology. Deduced amino acid sequences taken from previous reports were as follows (accession number): Ad1 (X67709) (25), Ad2 (J 01917) (17), Ad5 (X76550) (19), Ad6 (X67710) (25), Ad3 (X76549) (27), Ad7 (X76551) (26), Ad8 (X74663), Ad19 (X98359), Ad37 (X98360), Ad4 (X84646) (27), Ad40 (X51782) (35), and Ad41 (X51783) (34).