Cord blood samples were collected during September 2013–January 2014 at Saiseikai Niigata Daini Hospital, a hospital affiliated with Niigata University in Niigata, Japan. Samples were collected from healthy newborns born at full term to women with uncomplicated pregnancies. Preterm infants (gestational age <37 weeks) were excluded because they have lower levels of maternally derived immunoglobulins than do full-term neonates (17). Also excluded were babies born to mothers with maternal conditions that could affect levels of maternal antibodies in newborns (e.g., pregnancy-induced hypertension and abnormalities in the placenta). After the umbilical cord was double-clamped and cut, an obstetrician obtained 3 mL of cord blood from the umbilical artery. The samples were centrifuged at 700 × g for 10 min at 4°C; plasma (>500 μL) was frozen at −80°C; and the samples were sent to the laboratory of the Department of Pediatrics, Niigata University, where they were stored at −80°C until analysis. The study was approved by the Ethics Committees of Niigata University and Saiseikai Niigata Daini Hospital. Informed written consent was obtained from the parents of all study participants.

Since 2012, the viral causes of fever in neonates and young infants (i.e., <3 months of age) have been routinely evaluated by use of real-time PCR for enteroviruses (18), herpes simplex virus types 1 and 2 (19), and HPeVs (20) at Niigata University Hospital and its 40 affiliated hospitals in Niigata Prefecture (a population of ≈88,000 children <5 years of age). We evaluated patients with severe diseases related to HPeV3 who were admitted to Niigata University Hospital or 1 of its 11 affiliated hospitals during August 2013–September 2014. At onset, severe disease related to HPeV3 was defined as clinical symptoms and signs indicating sepsis or sepsis-like syndrome; disease status was confirmed by a positive result for HPeV3 by PCR analysis of samples of serum or cerebrospinal fluid (CSF). Information about patients’ contacts who were ill was obtained from interviews with parents and caregivers. With informed consent from parents, blood samples were collected prospectively at 3 and 6 months of age. Frozen serum samples were sent to the laboratory at Niigata University and stored at −20°C until analysis. The study was approved by the Ethics Committee of Niigata University.

Of the 5 commercially available IVIG preparations in Japan, 3 are from plasma pools of several thousand Japanese donors: polyethylene glycol (PEG)–treated IVIG (Venoglobulin IH; Mitsubishi Tanabe Pharma, Osaka, Japan), freeze-dried PEG-treated IVIG (Glovenin-I; Takeda, Osaka), and 2 batches of sulfonated IVIG (Venilon-I; Teijin Pharma, Tokyo, Japan); Another available IVIG preparation, ion-exchange resin-treated IVIG (Gammagard; Baxter, Tokyo), is from US donors; another, pH 4–treated IVIG (Sanglopor; CSL Behring, Tokyo), is from German donors. The plasma pools were collected during 2010–2012.

LLC-MK2 cells from the kidney of a healthy adult rhesus monkey were used for the neutralization assay (3). The cells were maintained in Eagle’s minimum essential medium (Sigma-Aldrich, St. Louis, MO, USA) containing 8% fetal bovine serum, 200 μg/mL gentamicin, and 2.5 μg/mL amphotericin B at 37°C, with 5% CO₂, for 1 week before passage. HPeV1 Harris (21), HPeV3 A308/99 (8), and HPeV6 NII561–2000 (3) were cultured in LLC-MK2 cells to obtain a sufficient amount of working seed viruses. The viruses were stored at −80°C until the neutralization test. The median tissue-culture infectious dose (TCID₅₀) was calculated by using the method of Reed and Muench (22).

All samples (i.e., cord blood samples, serum samples from patients infected with HPeV3, and IVIG preparations) were subjected to neutralization testing. Serum samples for real-time PCR were used to measure NATs at disease onset. A total of 25 μL of 100 TCID₅₀ viruses and 25 μL of serially diluted serum samples or IVIG preparation (started at 1:4 and then serially diluted twice until 1:2,048 in cord blood samples and IVIG preparations and until 1:512 in HPeV3-infected patients) were mixed in a 96-well plate and incubated at 37°C for 1 h. Then, 100 μL of suspended LLC-MK2 cells (10⁶/mL) was added to the wells. Cell controls and virus back titration were performed. The appearance of a cytopathic effect (CPE) was evaluated daily by light microscopy from day 3 and determined at day 10, when CPE was positive in control wells containing 100 TCID₅₀ and was confirmed by virus back titration (3). CPE >50% was considered positive. The test was performed in triplicate when the IVIG preparations were evaluated.

Real-time PCR was performed on serum or CSF samples collected from patients during the acute phase of the disease (<2 days after disease onset). CSF samples were centrifuged at 700 × g for 6 min at 4°C, and the supernatants were used in the analysis. Viral RNA was extracted from the serum and CSF samples by using a QIAamp MinElute Virus Spin Kit (QIAGEN, Valencia, CA, USA), according to the manufacturer’s instructions. Real-time reverse transcription PCR (RT-PCR) was performed by using a PrimeScript One Step RT-PCR Kit (TaKaRa, Tokyo, Japan) and the Thermal Cycler Dice Real Time System II (TaKaRa) with virus-specific primers and a TaqMan probe that targeted the conserved 5′ untranslated region (20). The thermocycling settings were 42°C for 5 min for cDNA synthesis; 95°C for 3 min followed by 45 cycles at 95°C for 5 s for denaturation; and 60°C for 40 s for annealing and extension.

A seminested RT-PCR assay with modifications was used to amplify the viral protein (VP) 1 region (23). After viral RNA was converted to cDNA by using SuperScript VILO MasterMix (Invitrogen, Carlsbad, CA, USA), an HPeV-VP1S forward primer (5′-GGD ARR MTK GGD VAW GAY GC-3′) and HPeV-VP1AS2 reverse primer (5′-TCY ARY TGR TAY ACA YKS TCT CC-3′) pair was used in the first PCR; and HPeV-VP1AS (5′-CCA TAR TGY TTR TAR AAA CC-3′) was used as reverse primer in the second PCR. These primer sequences were determined by using the International Union of Pure and Applied Chemistry nucleotide ambiguity codes. The 20-μL PCR mixture contained 2 μL of cDNA, 0.5 μM of each primer, and 10 μL of iQ Supermix (Bio-Rad Laboratories, Hercules, CA, USA). The cycling conditions were 95°C for 3 min, followed by 30 cycles at 95°C for 15 s, 42°C for 30 s, and 72°C for 1 min, terminating in a final extension at 72°C for 10 min. Amplicon size was 830 bp. HPeV-positive samples were typed by sequencing the VP1 PCR product. When the amount of the second PCR product was insufficient to sequence the VP1 region, a third PCR was performed with the same primers used in the second PCR. The PCR products were purified by using Illustra ExoProStar (GE Healthcare, Tokyo, Japan) and sequenced with a BigDye Terminator v3.1 Cycle Sequencing Kit (Applied Biosystems, Foster City, CA, USA) on an automated sequencer (Applied Biosystems 3130xl Genetic Analyzer). Each type was determined by comparing the nucleotide sequence of the VP1 region with available HPeV sequences from the DNA data banks of DDBJ, EMBL, and GenBank.

Serum-derived cDNA was used for VP1 sequencing. When the serum sample was unsuccessful in amplifying the VP1 region, CSF or stool samples collected during the acute phase of disease were used. Stool samples were diluted with 1× phosphate buffered saline to 10% wt/vol suspensions, followed by vortex and centrifugation at 6,250 × g for 20 min at 4°C. The supernatants were filtered with a Millex-GV Filter Unit (EMD Millipore, Billerica, MA, USA) with a pore size of 0.22 μm. After sample preparation, stool samples underwent RNA extraction and cDNA synthesis reactions as described for previous samples.

All statistical analyses were performed by using SPSS Statistics 22.0 (IBM SPSS, Chicago, IL, USA). Geometric mean titers (GMTs) and seropositivity rates to HPeV1, 3, and 6 were compared by using the Kruskal-Wallis test and χ² test, respectively. In the calculations, an antibody titer <1:4 was regarded as 1, and a titer >1:2,048 was regarded as 2,048. A 2-tailed p<0.05 was used to indicate statistical significance.

During the study period, 175 cord blood samples were collected. Median gestation age was 39.7 (range 37.1–41.9) weeks, and median maternal age was 32 (range 16–44) years. GMTs of antibodies to HPeV1, 3, and 6 were 52.0 (95% CI 40.5–66.8), 33.9 (95% CI 25.4–45.3), and 48.9 (95% CI 35.7–66.9), respectively. GMTs showed no significant differences among the 3 genotypes (p = 0.17) (Figure 1).

During the study period, 46 patients had a diagnosis of severe diseases related to HPeV, determined by PCR analysis of serum or CSF samples. NATs were measured in serum samples from 45 of these patients (1 patient was excluded because only a CSF sample was available on admission). Most (42/45 [93%]) were enrolled in the study during the epidemic of HPeV3 infection in Niigata in 2014. HPeV3 infection was identified in most (44/45 [98%]) HPeV-positive samples by sequencing the VP1 region of the virus by using cDNA derived from samples of serum (n = 40), CSF (n = 3), or stool cDNA (n = 1). The VP1 region could not be successfully amplified in 1 sample, so HPeV3 infection was diagnosed on the basis of a positive PCR result and a >4-fold increase in NATs to HPeV3.

The 45 patients with HPeV3-related diseases had a median age of 1 month (range 4 days–3 months, 21 days); 43 (96%) were <3 months of age, and 28 (62%) were male. For 25 (56%) patients, contact with a sick family member was identified. Clinical diagnoses at time of admission were sepsis (n = 37), sepsis-like illness (n = 7), and encephalitis with septic shock (n = 1).

All 45 serum samples at disease onset were available for neutralization testing. Most (42/45 [93%]) patients had a NAT of <1:4 to HPeV3 (HPeV3 negative). Three patients had low NATs to HPeV3: 1:4 in 1 (2%) patient and 1:16 in 2 (5%) patients.

Follow-up NATs were measured in the 34 (76%) patients whose parents agreed to participate in the follow-up study. During ongoing follow-up, 33 patients were evaluated at age 3 months and 19 patients at age 6 months. NATs to HPeV3 were >1:512 in all patients at age 3 months and remained >1:512 at age 6 months (Figure 2).

Because the maximum antibody titer at disease onset was 1:16 in patients infected with HPeV3 and animal studies have shown high protection against enteroviral infection at titers of 1:32 (24), we chose a titer of >1:32 to indicate the level of neutralizing antibody necessary to prevent infection. At this level, seropositivity rates in cord blood samples for HPeV1, 3, and 6 were 65%, 61%, and 71%, respectively, and did not significantly differ among the 3 genotypes (p = 0.12) (Table 1). When the data were stratified by maternal age, the seropositivity rate for HPeV3 declined as maternal age increased (Table 1). When mothers were divided into age groups of 16–24 years (n = 11), 25–34 years (n = 107), and 35–44 years (n = 57), the group of youngest mothers had significantly higher GMTs (336.5, 95% CI 176.1–642.9) than those for older mothers (31.9 [95% CI 22.0–46.4] for mothers ages 25–34 and 24.4 [95% CI 15.4–38.6] for mothers ages 35–44; p<0.001). In contrast, more cord blood samples had a titer <1:4 for HPeV3 (n = 29, 17%) than for HPeV1 (n = 6, 3%), and seropositivity rates for these 2 groups differed significantly when a lower cutoff value (1:4) was considered (83% vs. 97%, respectively; p<0.001).

All IVIG preparations we examined contained high NATs to all HPeVs (Table 2). NAT was ≥1:1,024 to HPeV1 and HPeV6 and ≥1:512 to HPeV3. No differences in antibody levels were observed among IVIG preparations originating from Japan, the United States, and Germany.