High Prevalence of Human Metapneumovirus Infection in Young Children and Genetic Heterogeneity of the Viral Isolates

RNA of the newly identified human metapneumovirus (HMPV) wasdetected in nasopharyngeal aspirates of 11 of 63 (17.5%) youngchildren with respiratory tract disease. Markers of infectioncaused by another member of the Pneumovirinae subfamily of thefamily Paramyxoviridae, respiratory syncytial virus (RSV), wereidentified in 15 of these patients (23.8%). Three patients weresimultaneously infected with HMPV and RSV. Studies of the clinicalcharacteristics of HMPV-infected children did not reveal anydifference between HMPV-infected patients and a control populationof RSV-infected patients with regard to disease severity, butthe duration of symptoms was significantly shorter for HMPV-infectedpatients. Phylogenetic analysis of the amplified viral genomefragments confirmed the existence and simultaneous circulationwithin one epidemic season of HMPV isolates belonging to twogenetic lineages.

INTRODUCTION

Top

Introduction

Recently, a new infectious agent, human metapneumovirus (HMPV),was isolated from nasopharyngeal aspirates of young childrenwith respiratory tract illness from The Netherlands (7). HMPV,together with avian pneumovirus serotypes A, B, C, and D, formthe Metapneumovirus genus of the Pneumovirinae subfamily ofthe family Paramyxoviridae (7, 8). Information on the biologyof HMPV, as well as on the prevalence and clinical significanceof HMPV infection, is scarce. Preliminary data, however, demonstratedthat HMPV can cause severe respiratory disease in children youngerthan age 5 years (2, 3, 4, 6, 7). These observations suggestthat HMPV can play an important role in the infectious pathologyof humans and indicate the necessity of more detailed studyof this agent and infections caused by this agent. Here, wereport on the prevalence and clinical characteristics of HMPVinfection in children with respiratory disease from Germanyand on the genetic heterogeneity of the HMPV isolates identified.

MATERIALS AND METHODS

Top

Materials and Methods

Specimens. Nasopharyngeal aspirates were collected from 63 hospitalizedpatients younger than age 2 years. These patients representedall children in this age group admitted to the Essen UniversityHospital with upper or lower respiratory tract infection betweenJanuary and May 2002. Most of these patients presented withwheezing and/or signs of respiratory distress (Table 1). Theclinical data for the HMPV-infected patients were compared tothose for a control group of children from the same populationinfected with respiratory syncytial virus (RSV).

View this table:
[in this window]
[in a new window]
TABLE 1. Clinical data for the study population

Reverse transcription (RT)-PCR for HMPV RNA. RNA was extracted from nasopharyngeal aspirates with the RNeasykit (QIAGEN) and reverse transcribed and amplified with twoprimer sets. Information on the first set, which included twoprimers whose sequences were derived from the L gene and whichgenerated a DNA fragment of 171 bp, was kindly provided by vanden Hoogen et al. (7). The second set was designed in the presentstudy. It included four primers whose sequences were derivedfrom the nucleocapsid (N) gene. Primer 750as (5'-TGCTTTGCTGCCTGTAGATGATGAG)was used to generate the cDNA. Reverse-transcribed cDNA wassubjected to the first round of a 30-cycle PCR with primers111s (5'-AGAGTCTCAGTACACAATMAAAAGAG) and 750as. The second roundof the 30-cycle PCR was run with primers 114s (5'-AGAGTCTCAGTACACAATMAAAAGRGATG)and 442as (5'- GCCATTGTTTTYCTTGCYTC). Each cycle included denaturationat 94°C for 1 min, annealing at 60°C for 1 min, andextension at 72°C for 2 min. The PCR products were separatedby agarose gel electrophoresis and visualized by ethidium bromidestaining.

Detection of RSV. Infection caused by RSV was diagnosed by direct immunofluorescence(BioMerieux), according to the instructions of the manufacturer.

Sequencing and phylogenetic analysis. The amplified DNA was gel purified with the QIAquick Gel Extractionkit (QIAGEN), cloned in the pTOPO TA cloning vector (Invitrogen),and subjected to sequencing in both directions (Dye TerminatorDNA sequencing kit; Perkin-Elmer). Newly obtained sequenceswere compared with HMPV sequences taken from the GenBank databaseby phylogenetic analysis with the programs DNADIST and NEIGHBORfrom the package PHYLIP (version 3.5c; J. Felsenstein, Departmentof Genetics, University of Washington, Seattle, 1993). The significanceof the grouping was evaluated by bootstrap analysis (100 replicates)with the program SEQBOOT.

Nucleotide sequence accession numbers. The HMPV sequences obtained in the present study were depositedin GenBank under accession numbers AF543686 to AF543690.

RESULTS

Top

Results

Nasopharyngeal aspirate samples from 63 patients with respiratoryillness were tested for the presence of HMPV RNA in parallelwith two primer sets. Five positive samples from three patientswere detected with the set used in the original study of HMPV(7). All five of these samples plus nine samples from an additionaleight patients were positive with the second set of primers.Cloning and sequencing of the amplified fragments confirmedthe specificity of the PCR for all positive samples. Thus, thenested RT-PCR based on the primer set designed in the courseof the present study appears to be more sensitive. Overall,HMPV RNA was found in 11 of 65 patients (17.5%). Five patientswere positive for HMPV RNA after the first round of PCR, andan additional six patients were positive for HMPV RNA afterthe second round of PCR. The frequency of detection of the RSVantigen by direct immunofluorescence in the nasopharyngeal aspiratesfrom the same patients was slightly higher (15 of 63 samples;23.8%). Three patients were simultaneously found to be positivefor markers of HMPV infection and RSV infection. One patientbecame positive for RSV antigen during hospitalization and afterthe disappearance of HMPV RNA.

HMPV-positive samples were detected in materials collected betweenJanuary (beginning of the study) and April 2002. Although RSV-positivesamples were identified in May, no HMPV-positive samples wereidentified in May.

No obvious difference in the frequencies of HMPV RNA detectionin different age groups of patients was noted. Serial aspiratesamples were available from three HMPV-infected patients. Inone of these patients, HMPV RNA was detected for at least 20days.

Comparative analysis of the clinical characteristics of HMPV-positivepatients and children infected with RSV showed that both groupswere similar with respect to age and sex. An underlying respiratorydisorder was present in one patient in both groups (larynx stenosisfor the patient in the HMPV group, bronchopulmonary dysplasiafor the patient in the RSV group). Most patients had bronchiolitisalone or bronchiolitis in combination with pneumonia (Table1). There were no significant differences with regard to theseverity of symptoms between the two groups, although pneumoniatended to be more frequent in RSV-infected patients (5 of 12patients among the RSV-infected children versus 1 of 8 patientsamong the HMPV-infected children [P = 0.17 by Fisher's exacttest]). Only one patient in the RSV-infected group requiredintubation and mechanical ventilation. The duration of symptomswas variable between patients, but was significantly shorterin the HMPV-infected group (P < 0.04 by the Wilcoxon test).

Alignment of the nucleotide sequences of the N-gene fragmentsof the HMPV isolates identified showed heterogeneity distributedevenly throughout the whole fragment sequenced. Phylogeneticanalysis of the HMPV sequences corresponding to the 613-bp fragmentsobtained after the first round of PCR from five patients revealedtwo major clusters (Fig. 1A). The levels of similarity betweenthe sequences within each cluster varied from 93 to 99%, andthe level of similarity between the sequences forming thesetwo clusters was about 86%. Smaller fragments (71 bp) of thesefive sequences were additionally compared with the correspondingsequences of HMPV variants isolated in The Netherlands (1).A tree with a similar topography, namely, a segregation of theHMPV sequences into two major groups, was observed upon analysiswith those sequences (Fig. 1B).

View larger version (15K):
[in this window]
[in a new window]
FIG. 1. Phylogenetic analysis of HMPV isolates. Comparison of the sequences of 613-bp (A) or 71-bp (B) fragments of the N gene. Nld, sequences of the HMPV variants isolated in The Netherlands (7); APV-C, sequence of the N-gene fragment of avian pneumovirus serotype C (GenBank accession no. AF176590).

DISCUSSION

Top

Discussion

This paper reports on the frequency of HMPV infection in a nonpreselectedgroup of young children attending a hospital with symptoms ofa respiratory tract disease. Most of the previous publicationsreferred to the detection of HMPV in special groups of patientswho were found to be negative for the markers of infectionscaused by other respiratory viruses (2-7). Taking into accountour observations on the occurrence of double infections causedby both HMPV and RSV, one might presume that such a preselectioncould lead to an underestimation of the true incidence of HMPVinfection. In our study, HMPV RNA was identified in 11 of 63patients (17.5%). This frequency is comparable to that of RSVinfection in the same group (23.8%). One should note, however,that such a comparison should be considered carefully becausethe cases of HMPV and RSV infections in this study were identifiedby absolutely different diagnostic techniques (RT-PCR and immunofluorescenceassay, respectively). It is also possible that the RT-PCR protocolsused in this and other studies are not optimal for the detectionof all HMPV variants.

Our results demonstrating that the spectrum of clinical symptomsin HMPV-infected children is similar to that in patients withrespiratory disease caused by RSV confirm the previous observations(7). Taken together with the data on the high incidence of HMPVinfection, these results suggest that for young children, HMPVcould be a pathogen as important as RSV. Preliminary data fromthe most recent reports (1, 5) indicate that this may also holdtrue for adults, especially those with impaired immunity. Eventhough disease severity was not significantly different betweenHMPV- and RSV-infected children, the duration of symptoms requiringtreatment appeared to be shorter in HMPV-infected patients.A larger sample size will be needed to confirm these preliminaryfindings.

The data from the present study further extend the observationson the evident heterogeneity of the HMPV genome (5, 7). At themoment, there is some difficulty with the detailed analysisof the HMPV sequences available, as the genome fragments sequenced(most of which were rather short) were generated with primersderived from different parts of the viral genome. Nevertheless,comparison of the sequences of the relatively short fragments(71 bp) of the N gene of HMPV isolates from the present studywith the sequences available in the GenBank revealed the existenceof two major clusters (Fig. 1B). Similar bifurcation into twomajor branches occurred upon comparison of a smaller numberof sequences of longer fragments (613 bp) of the N gene (Fig.1A). These results, together with the data reported earlierfor both the N gene and several other viral genome regions (5,7), provide evidence for the existence and simultaneous circulationof HMPV isolates belonging to two evolutionary lineages. Theheterogeneity of the HMPV genome evidently may pose a diagnosticproblem. One cannot exclude the existence of other variantsof HMPV which are not detected by the sets of primers in use.Further studies of the heterogeneity of the HMPV genome aretherefore needed to clarify this crucial issue and to optimizediagnostic procedures to detect this important pathogen of humans.

ACKNOWLEDGMENTS

We are grateful to Yvonne Pietsch for skillful technical assistance.

FOOTNOTES

* Corresponding author. Mailing address: Institute of Virology, Essen University Hospital, Hufelandstr. 55, D-45122 Essen, Germany. Phone: 49 201 7233533. Fax: 49 201 7235929. E-mail: sergei.viazov{at}uni-essen.de.

REFERENCES

Top