Human Metapneumovirus, Peru

We retrospectively studied 420 pharyngeal swab specimens collected from Peruvian and Argentinean patients with influenzalike illness in 2002 and 2003 for evidence of human metapneumovirus (HMPV). Twelve specimens (2.3%) were positive by multiple assays. Six specimens yielded HMPV isolates. Four of the 6 isolates were of the uncommon B1 genotype.

Screened specimens that gave bands within 200 bp of the expected 347-bp product were further tested with a 2step RT-PCR with F1-, F2-, and N-gene primer sets. The 2step RT-PCR was performed by using the RETROscript Kit (Ambion, Austin, TX, USA) with heat denaturation of RNA. PCR products were analyzed by gel electrophoresis. Specimens were designated RT-PCR-positive if the confirmatory N-gene primer set and at least 1 of the confirmatory F-gene primer sets yielded a band within 50 bp of the expected size (primers available from the corresponding author) (9).
Both 1-and 2-step RT-PCR procedures were adapted from previous reports (9)(10)(11)(12). With every specimen batch, a known HMPV-positive and HMPV-negative sample was tested in parallel to validate the run.
RT-PCR-positive specimens were further studied with shell-vial cell culture for viable HMPV. A shell vial containing a near confluent monolayer of LLC-MK2 cells (Diagnostic Hybrids, Inc., Athens, OH, USA) was injected with 100 µL specimen and 900 µL HMPV growth media (1× minimum essential medium with L-glutamine and Earle salts, 0.1% bovine albumin, 1× HEPES, 0.001% porcine pancreatic trypsin, 0.4505 mol/L D-glucose, 10,000 U penicillin, 10 mg streptomycin, and 50 µg amphotericin), centrifuged for 1 h at 37°C and 2,800 rpm (1,500 × g), followed by a 37°C incubation with 5% CO 2 . The cell monolayers were microscopically examined weekly for cytopathic effect (CPE) and contamination. Shell vials were incubated 3-4 weeks or until cell disruption occurred. Infected cell supernatant media were harvested each week upon cell media replacement. From an aliquot of the infected media, RNA was extracted and subsequent RT-PCR was performed by the HMPV F2-gene 1step protocol.
Sequencing was performed on the RT-PCR-positive specimens by using G univ primer set (available from the corresponding author), adapted to amplify an 800-to 1,000-bp region. Products were subsequently electrophoresed across a 1.0% agarose gel stained with ethidium bromide. RT-PCR-positive products were purified with QIAquick PCR Purification/Gel Extraction Kits (Qiagen). Strands of the amplicons were sequenced by automated sequencing with the G univ primers. Big Dye Terminator Kit v3.1 (Applied BioSystems, Foster City, CA, USA) was used in sequencing reactions. Samples were run on a 3730xl DNA Analyzer (Applied BioSystems).
Alignments of partial amino acid sequences of the HMPV G protein were generated with the ClustalW software (National Center for Biotechnology Information, Bethesda, MD, USA). Prototypic sequences of different types (A and B) and subtypes (A1, A2, B1, and B2) from the Netherlands and Canada were included in the alignments. Phylogenetic analysis was performed by the neighbor-joining method by using MEGA 2 (University of Pittsburgh, Pittsburgh, PA, USA).
Specimen laboratory results were studied for demographic and temporal predictors of RT-PCR positivity by using standard categoric data techniques. Age group cut points were selected based on age quartiles. Exact binomial 95% confidence intervals (CIs) were calculated around prevalence statistics. Similarly, 95% CIs around odds ratios were calculated by using logistic regression. Analyses were performed by using SAS software version 9.1 (SAS Institute, Inc., Cary, NC, USA).

Conclusions
We studied 420 posterior pharyngeal swab specimens collected from January 2002 to November 2003 ( Table 1). Because of differences in clinic focus, the distribution of influenzalike illness differed by site; children made up higher proportions in each country (median age 11 years, range <1-89 years, Table 2). Overall, 51% of the 302 specimens for which patient's sex was known were from male patients. Most influenzalike illness specimens were obtained during the coldest months (July through September, data not shown).
Twelve (2.9%) of 420 specimens were considered HMPV RT-PCR-positive ( Table 2). All 12 positive specimens were cultured on LLC-MK2 cells. Six of the 12 specimens grew HMPV, and none of them showed evidence of viral CPE before 7 days. The nonviability of the 6 remaining positive specimens was likely due to the 4 freeze-thaw cycles that occurred before LLC-MK2 cell culturing or possibly the degradation of HMPV RNA within the specimens during transport and storage.
All 6 of the specimens that yielded an HMPV isolate in cell culture were successfully sequenced and were used to develop a phylogenetic tree (Figure) (13). Sequencing was not attempted until ≈2 years after specimen collection. This delay in sequencing and multiple freeze-thaw cycles may explain our inability to amplify and sequence G-gene product from the other 6 positive specimens. Sequence data were compared to previously sequenced HMPV isolates, showing a high prevalence of genotype B, with 4 isolates (Peru2-2002 Our data suggest that HMPV is circulating in Peru. Consistent with results of other studies, the prevalence of DISPATCHES HMPV infection in this research was low among patients with influenzalike illness and more common among younger children (6% in children <7 years of age, Table 1) (1). In our study, HMPV was more often detected in male patients and from April to June.
Of the 12 HMPV RT-PCR-positive patients, 9 had clinical reports available for review. Three children from a small Peruvian Amazon village whose specimens were collected within 3 days of each other were infected with HMPV from the B1 subtype. Among these 3 children, the youngest (3 and 4 years of age) were the most debilitated and had the highest maximum oral temperature (39.8°C and 39.6°C). Among the remaining 6 HMPV-positive patients, 1 had pneumonia and 1 was hospitalized. These data show a higher likelihood (odds ratio 4.3, 95% CI 1.3-13.8) of detecting HMPV from patients with influenzalike illness during the Southern Hemisphere's autumn (March to June) ( Table 1).