Sterile cotton swabs were used to collect tears and eye discharges from 15 AHC patients. Each swab was collected in 1 mL viral transport medium (1× Hank's balanced salt solution containing 5% fetal bovine serum, penicillin [100 U/mL], and streptomycin [100 μg/mL], pH 7.4) in 5-mL cryovials. All clinical samples were kept on dried ice or at –70°C and transported to the enterovirus laboratory at Fiocruz, Rio de Janeiro.

Confluent HEp-2c and RD cells cultured in minimum essential medium supplemented with 2% of fetal bovine serum and antimicrobial drugs were used for viral isolation. Cell cultures were spread with 0.2 mL viral transport medium, incubated at 37°C, and observed for 7 days for cytopathic effect (CPE). One additional blind passage was performed if CPE was not observed during the first passage. Only HEp2c cells were capable of viral isolation, with characteristic CPE observed in 9 (53%) of 15 specimens.

RNA was extracted from 250 μL virus-infected culture supernatant by using Trizol LS (Invitrogen, Carlsbad, CA, USA), and the complementary DNA was synthesized with Oligo(dT)15 (Invitrogen) by using SuperScriptII reverse transcriptase (Invitrogen). Enterovirus group–specific reverse transcription–polymerase chain reaction (RT-PCR) was performed by using a primer pair (222/292) that amplifies an ≈350-bp fragment within the VP1 gene, as described (10). RT-PCR products were analyzed by electrophoresis in 1% agarose gels containing 0.5 μg/mL ethidium bromide. Products were further gel purified by using QIAquick Gel Extraction Kit (Qiagen, Hilden, Germany) and quantified by comparison, in 1% agarose gel, with Low DNA Mass Ladder (Invitrogen). Cycle-sequencing reactions were performed by using the ABI BigDye Terminator Cycle Sequencing Ready Reaction (PE Applied-Biosystems, Foster City, CA, USA) in a GeneAmp thermocycler. VP1 sequences from our isolates were compared to those available at GenBank by the BLAST software (11) to determine viral identity and serotype and aligned with sequences obtained from the database by using ClustalW (12). To identify respective divergence and infer the genetic relationship among the isolates, we used the neighbor-joining reconstruction method included in the Mega 3 software (13).

Figure

Figure. Phylogenetic analysis of CA24 strains isolated during the acute hemorrhagic conjunctivitis outbreaks in Rio de Janeiro in 2003 and 2004. Sequences of CA24 isolated from previous outbreaks in Korea in 2002...

All isolates were identified as coxsackievirus A24v. The nucleotide sequence identities among CA24v isolates of the 2004 outbreak in Rio de Janeiro and CA24v isolates from a preceding AHC epidemic in Brazil in 2003 varied from 98.5% to 100%. The fact that several isolates, recovered 1 year apart from different AHC episodes, shared 100% identity could be partially explained by the lower resolution of the sequence interval used for the analysis (270 bp or ≈3.6% of the genome) when compared with the entire VP1 (≈900 nt or ≈12% of the enterovirus genomes), which is regularly used for poliovirus and enterovirus phylogenetic analysis. The sequence window can be widened to cover the complete VP1 or the complete genome if higher resolution is needed. Phylogenetic analyses based on the small genome segment generated by primers 222/292 provided sufficient information to identify the relationship among the CA24v isolates and the probable origin of AHC strains in circulation in Brazil. Both 2003 and 2004 isolates were closely related to the CA24v isolates (97.5%–99.0%) that caused AHC epidemics in South Korea and French Guiana in 2002 and 2003 (Figure).

During the first half of 2003, CA24v was also the etiologic agent responsible for a large outbreak of AHC in several states of Brazil, including Rio de Janeiro (unpub. data). The present outbreak (2004) also occurred during the end of summer and beginning of fall.

Sequencing analysis of a relatively small fragment in the VP1 is a simple and rapid method for enterovirus typing (10). Furthermore, comparisons of these sequences were useful to infer phylogenetic relationships among CA24v isolates. The CA24v responsible for the 2004 outbreak in Rio de Janeiro had >97% identity with the CA24v isolated in Korea and French Guiana from AHC outbreaks in 2002 and 2003, which suggests a direct route of disease dissemination among these countries. However, more data are required to support these findings. The source of the outbreak and factors that influenced its spread remain unclear.