A total of 30 HPeV strains, including the reference strains Harris (HPeV-1, ATCC VR-52), and Williamson (HPeV-2, ATCC VR-53), as well as 28 clinical HPeV isolates were used in this study (Table). Clinical specimens of various origins, such as nasopharyngeal aspirates (NPA), throat swabs, stools, cerebrospinal fluid (CSF), and endotracheal secretions (Table A1), were first added to different continuous cell lines, including human lung adenocarcinoma (A-549), human rhabdosarcoma (RD), transformed human kidney (293), human colon adenocarcinoma (HT-29), human laryngeal carcinoma (Hep-2), human foreskin fibroblast, mink lung, African green monkey kidney (Vero), Madin Darby canine kidney , and rhesus monkey kidney (LLC-MK2) cells. The viral cultures were incubated for 3 weeks at 37°C in a 5% CO₂ atmosphere. Viral isolates with cytopathic effects (CPE) suggestive of HEV or HPeV were further analyzed by neutralization assays with Lim and Benyesh-Melnick antiserum pools A-H (National Institutes of Health, Bethesda, MD, USA) and specific antisera for HPeV-1 and -2 (MA Bioproducts, Walkersville, MD, USA).

HPeV genomic RNA was isolated from the supernatant of infected cell cultures with the QIAamp Viral RNA kit (Qiagen, Mississauga, Ontario, Canada). cDNA was prepared by using the HPeVUniv5´ primer selected from the C-terminal region of the capsid VP0 encoding gene: 5´-GCT GAC CTA TGY ATC CCC TAT GT-3´ (nt 1358–1379, GenBank accession no. AJ998818) and the SuperScript II reverse transcriptase (Gibco BRL, Burlington, Ontario, Canada). Viral cDNA was then amplified by PCR by using the Pfu Turbo Polymerase (Stratagene, La Jolla, CA, USA) and primers HPeVUniv5´ and HPeVUniv3´ (selected from the N-terminal region of the capsid VP1 encoding gene): 5´-GTG AAC CCC AYG AAT TTT GGA A-3´ (nt 2351–2330, accession no. AJ998818). After a primary denaturation step at 95°C for 5 min, 35 cycles of denaturation at 94°C for 50 s, annealing at 58°C for 50 s, and extension at 72°C for 2 min were performed, followed by a final extension step at 72°C for 7 min. PCR products were analyzed by electrophoresis on a 1.2 % agarose gel and the expected amplicons of ≈1,000 bp were purified by using the GenElute PCR clean-up kit (Sigma-Aldrich, St. Louis, MO, USA). Nucleotide sequences of PCR products were determined with PCR primers by using an automated DNA sequencer (ABI Prism 377A; Perkin-Elmer Applied Biosystems, Foster City, CA, USA). The nucleotide sequences were analyzed, and the deduced amino acid sequences were aligned by using ClustalW (available from http://www.infobiogen.fr/). The resulted alignment was analyzed by the neighbor-joining method with the MEGA2 (Biodesign Institute, Arizona State University, Phoenix, AZ, USA) program for the construction of the phylogenetic tree.

Demographic and clinical informations were retrospectively collected from clinical charts of hospitalized patients for whom a positive culture for HPeV was obtained in our laboratory in the past 20 years. The Mann-Whitney rank sum test was used to compare the mean age of patients infected with HPeV-1 versus HPeV-3; the Fisher exact test was used to compare HPeV-1 versus the 2 other HPeV types regarding the rate of underlying diseases and the nature of initial diagnosis.

Among the different cell lines used, HT-29 was found to be the most suitable for efficient isolation of HPeV-1 and -2. Both grew efficiently in this cell line with an evident CPE noted after a mean incubation time of 3.8 days (range 1–8 days). A presumptive distinction between HPeV- and HEV-induced CPE in this cell line was possible. Parechovirus-infected cells were large, regularly shaped spheres, whereas enterovirus-infected cells were rather small with an irregular shape (Figure 1A). On the other hand, HPeV-3 isolates only initially grew on LLC-MK2 cells after a mean incubation time of 16 days (range 14–17 days). On passage, however, these viruses grew rapidly (in ≈1–3 days) in the same cell line (Figure 1B) and in Vero cells. All tested HPeV-1 and only one third of HPeV-2 (Can82047-01) isolates were neutralized by specific antisera for HPEV-1 and -2 (Table).

All HPeV clinical isolates as well as the 2 HPeV-1 and -2 reference strains were efficiently amplified by our RT-PCR test, generating an amplicon of the expected size (≈1,000 bp). No PCR products were generated from enterovirus (coxsackievirus A4 and coxsackievirus B4) and rhinovirus (n = 2) clinical isolates. Conversely, our HPeV isolates were negative when tested by our HEV RT-PCR assay, which is based on the amplification of the 5´ noncoding region. Sequencing of HPeV PCR products confirmed the presence of the VP3 gene in our HPeV-1 (n = 20), HPeV-2 (n = 3) and HPeV-3 (n = 5) clinical isolates. Sequence analysis showed that HPeV-1 clinical isolates had 75.0% to 85.1% nt and 89.2 to 97.5% amino acid identities with the prototype Harris strain (accession no. S45208) (Table). HPeV-2 isolates had nucleotide and amino acid identities of 67.8% to 81.1% and 72.5% to 96.4%, respectively, with the prototype Williamson strain (accession no. AJ005695). Of note, Can95219-03 and Can95224-03 HPeV-2 isolates had the same sequence and were more related to the Connecticut CT-80-6760 HPeV-2 strain (accession no. AF055846) with which they had nucleotide and amino acid identities of 85.1% and 95.8%, respectively. HPeV-3 isolates had nucleotide and amino acid identities of 95.0% to 95.5% and 95.7% to 96.4%, respectively, with the Japanese A308/99 HPeV-3 isolate (accession no. AB084913).

Phylogenetic analysis confirmed the identities of HPeV isolates. The HPeV-1 isolate Can81805-01 was closely related to the reference Harris strain (lineage II), whereas the remaining 19 isolates from this study, as well as 3 other Japanese HPeV-1 isolates (A1087-99, A942-99, and A10987-00), were found to form a distinct cluster (lineage I) (Figure 2). The HPeV-2 isolate Can82047-01 was related to the reference Williamson strain (lineage I), whereas the other 2 HPeV-2 isolates, Can95219-01 and Can95224-01 HPeV-2, were somewhat related to the Connecticut CT80-6760 strain (lineage II). Finally, all HPeV-3 isolates formed a separate cluster that also included other Japanese HPeV-3 isolates (A308-99, A628-99, A317-99, and A354-99).

The alignment of the predicted amino acid sequences of the partial VP0–VP3 region of Canadian HPeV-1, -2 and -3 isolates with the reference strains showed differences in the VP0/VP3 cleavage site, which was N/N in Can952219-03 and Can95224-03 (HPeV-2), N/S in Connecticut CT80-6760, T/A in Williamson and Can82047-01 (HPeV-2), N/A in Harris and all Canadian HPeV-1 isolates, and N/G in all HPeV-3 isolates (Figure 3). The length of the N-terminal extension of the VP3 protein, which is a molecular feature of parechoviruses, was also shown to differ from 27 residues in HPeV-1 of lineage I to 34 residues in HPeV-2 of lineage II. HPeV-1 of lineage II and HPeV-2 of lineage I had an extension of 28 residues, whereas an extension of 30 residues was seen in all HPeV-3 isolates (Figure 3).

Twenty-eight HPeV isolates, recovered in the Quebec City area between 1985 and 2004, were available for this study (Table). This collection included 20 (71.4%) HPeV-1, 3 (10.7%) HPeV-2, and 5 (17.8%) HPeV-3. The viral isolates were recovered predominantly from NPAs (n = 16), but also from throat swab specimens (n = 2), endotracheal secretions (n = 2), stool samples (n = 7), and CSF (n = 1) (Table A1). In addition, for 2 patients, a second concomitant HPeV isolate was recovered from another site (urine in patient 7 and NPA from patient 20). Complete clinical findings were obtained for 23 (82.1%) of the 28 patients, whereas partial information was available for the remaining 5 patients.

Twenty-seven (96.4%) of the 28 patients were children <4 years (mean 11.1 months, median 7.0 months, range 1 week to 4 years). Mean age was 14.6, 6.3, and 0.7 months for HPeV-1–, HPeV-2–, and HPeV-3–infected children, respectively (p<0.001 for comparison of HPeV-1 vs HPeV-3). The 78-year-old adult excreted HPeV-1 in her endotracheal secretions. The female-to-male ratio was 1.2 (54.2% vs. 45.8%). Seventy-five percent of all HPeV isolates were recovered during the typical enterovirus season, i.e., during summer (28.6%) and fall (46.4%). The epidemiology of the newly described HPeV-3 was similar to that of other HPeVs, with 80% of this viral serotype isolated during the summer-fall period.

Among patients for whom information was available, 33.3% (8/24) had an underlying disease, including 2 children with acute lymphoblastic leukemia (ALL) (Table A1). Patients with underlying diseases represented 47.0% (8/17) of those who excreted HPeV-1 compared to none of the 7 HPeV-2–or HPeV-3–infected patients (p = 0.18 for comparison of HPeV-1 vs other HPeVs). Twenty-four (96.0%) of the 25 patients with available information were hospitalized (mean number of days 11.4, median 3.0, range 1–129). However, if the 3 patients with nosocomial HPeVs are excluded, the mean duration of hospitalization was 3.7 days; this finding was similar for the different HPeV types (mean of 3.3, 3.0, and 5.0 days for HPeV-1, -2 and -3, respectively). Only 3 (12.0%) of 25 patients were admitted to the intensive care unit. These 3 patients were infected with HPeV-1, and 2 of them (patients 6 and 26), a 4-year old boy with ALL and a 78-year-old woman, died of respiratory failure. A copathogen was found in 5 (20.8%) of 24 HPeV-infected patients, including 3 patients with human respiratory syncytial virus (HRSV), 1 with adenovirus, and 1 with human parainfluenza virus (HPIV) type 3.

Among the 24 children with available information, bronchiolitis, pneumonitis, or both, were reported as the final diagnosis in 50.0% of cases, whereas acute otitis media, sinusitis, and conjunctivitis were secondary diagnoses in 37.5%, 8.3%, and 4.2% of cases, respectively (Table A1). In addition, the final diagnosis was reported as viremia/sepsis in 29.2% of children, whereas enteritis, de novo convulsions, and shock were found in 29.2%, 4.2%, and 4.2% of patients, respectively. Final diagnosis was unknown for 3 (11.1%) of 27 children (all 3 had positive stool viral cultures). All children (7/7) with HPeV-3 and HPeV-2 infections were admitted with a diagnosis of viremia or sepsis, and all received broad-spectrum antimicrobial drugs until results of bacterial cultures (blood, CSF, urine), whereas none (0/17) of the HPeV-1-infected children had such an initial diagnosis (p = 0.001 for comparison of HPeV-1 vs. other HPeVs). Conversely, bronchiolitis was found in 8 (47.0%) of 17 children with HPeV-1 infections and in none of the patients (0/7) infected with the other 2 serotypes (p = 0.18 for comparison of HPeV-1 vs other HPeVs). However, another viral pathogen (HRSV or HPIV-3) was found in addition to HPeV-1 in 4 (50%) of 8 children with bronchiolitis.

A few cases are particularly worth noting. As previously mentioned, patient 6 was the only child who died after HPeV infection in our cohort. At the time of his HPeV-1 infection, the 4-year-old boy was receiving consolidation chemotherapy for ALL. He was admitted with high fever (40°C), dry cough, and abdominal pain. Over a course of 4 to 5 days, the following conditions developed: bilateral pneumonitis, which required mechanical ventilation, hepatitis, and septic shock complicated by renal insufficiency, for which dialysis was necessary. He died of multiorgan failure while receiving treatment with amphotericin B, acyclovir, meropenem, and vancomycin; no autopsy was performed. Apart from isolation of HPeV-1 from a throat swab, no other microorganisms (bacteria, fungi, and viruses) were recovered from blood, urine, and endotracheal secretions. Also of interest are the clinical signs and symptoms of 1-month-old twins infected by HPeV-2, who were admitted to the hospital on the same day with high fever and irritability. In 1 case (patient 20), HPeV-2 was recovered from CSF and NPA samples, whereas the virus was only isolated in NPA from the other twin (patient 21). The viruses from the 2 children had identical VP3 gene sequences. Both infants received broad-spectrum antimicrobial drugs (ampicillin plus cefotaxime) for 3 days for suspected sepsis, and both had rapid clinical improvement. Patient 26 was the only adult infected with HPeV (type 1) in our study. This 78-year-old woman was admitted to the hospital for several fractures caused by a car accident. One month after admission, bilateral pneumonia developed. She eventually died of respiratory failure 43 days after admission. An endotracheal culture was positive for HPeV-1 12 days before her death; however, results of bacterial cultures were unavailable because part of her clinical chart was destroyed a few years after her death in 1987. No autopsy report was available.