Molecular Identification and Epidemiological Features of Human Adenoviruses Associated with Acute Respiratory Infections in Hospitalized Children in Southern China, 2012-2013

Background Acute respiratory infections (ARI) are the major worldwide health problem associated with high morbidity and mortality rates. Human adenovirus (HAdV) is one of the most common pathogens associated with viral ARI, and thus calls for specific diagnosis and better understanding of the epidemiology and clinical characteristics. Methods Total 4,130 children with ARI requiring hospitalization from 2012 to 2013 were retrospectively studied. Throat swab specimens were collected from each patient. Fluorescence Quantitative PCR was performed to detect adenovirus as well as other common ARI-related pathogens. The seven HAdV hypervariable regions (HVRs) of the hexon gene from fifty-seven HAdVs-positive samples collected in the seasonal peaks were sequenced. Phylogenetic analysis of HVRs was also conducted to confirm the molecular types and genetic variation. In addition, epidemiological features and co-infection with other human respiratory pathogens were investigated and analyzed. Results Of 4,130 hospitalized pediatric patients tested, the positive rates of respiratory syncytial virus (RSV), Mycoplasma pneumoniae (MP), and HAdV were 13.7%, 13.2%, and 12.0%, respectively. The HAdV positive patients accounted for 7.9%, 17.2%, 17.5% and 10.7% in age groups <1, 1–3, 3–6 and 6–14 years, respectively. Eighty-four HAdV positive children were co-infected with other respiratory pathogens (84/495, 17.0%). The most common co-infection pathogens with HAdV were MP (57.1%) and Human Bocavirus (HBoV) (16.7%). The majority of HAdV infected patients were totally recovered (96.9%, 480/495); However, four (0.8%) patients, who were previously healthy and at the age of 2 years or younger died of pneumonia. Seasonal peaks of HAdV infection occurred in the summer season of 2012 and 2013; the predominant HAdV type was HAdV-3 (70%), followed by HAdV-7 (28%). These epidemiological features were different from those in Northern China. The HAdV-55 was identified and reported for the first time in Guangzhou metropolitan area. Phylogenetic analysis indicated that all the HVR sequences of the hexon gene of HAdV-3 and -7 strains have high similarity within their individual types, and these strains were also similar to those circulating in China currently, indicating the conservation of hexon genes of both HAdV-3 and HAdV-7. Conclusions Knowledge of the epidemiological features and molecular types of HAdV, a major pathogen of pediatric ARI, as well as other co-infected respiratory pathogens circulating in Guangzhou, southern China, is vital to predict and prevent future disease outbreaks in children. This study will certainly facilitate HAdV vaccine development and treatment of HAdV infections in children.


Introduction
Adenoviruses are non-enveloped, double stranded DNA viruses that vary in size from 70 to 100 nm [1]. Up to now, at least 68 genotypes of human adenovirus (HAdV) have been identified [2] and classified into 7 species from A to G [3] based on serology, whole-genome sequencing, and phylogenetic analyses [4]. HAdVs can cause a wide range of illnesses, such as acute respiratory infections (ARIs), gastroenteritis, conjunctivitis, cystitis, and meningoencephalitis. They are usually responsible for 5-7% of respiratory illnesses in infants and children [5]. ARI is one of the most common causes of morbidity and mortality in children. HAdV-3, -4, and -7 are the major HAdV types associated with ARI in children and adults in the world [6][7][8][9][10][11][12][13][14]. The re-emergent genotype HAdV-55 was recently reported in children with ARIs in Beijing and Shaanxi Province, China [15]. However, information on the epidemiological and clinical features of HAdV circulating in hospitalized children is limited in China.
The purpose of this retrospective study was to determine the prevalence, epidemiology as well as the types of HAdVs circulating among hospitalized children with ARI in Guangzhou, Southern China during 2012-2013. All the specimens were collected in Guangzhou Women and Children's Medical Center (GWCMC), which has 1,358 beds, receives over 3,000,000 pediatric outpatient person-times and admits 57,000 inpatients each year from Guangzhou as well as other cities in Southern China.
In addition, a previous investigation found that some patients infected by HAdVs were also co-infected with other viral pathogens [16], leading to severe clinical consequences in hospitalized patients. Thus, co-infections with other respiratory viruses were also investigated in this study.
From January 1, 2012 to December 31, 2013, 4,130 hospitalized pediatric patients (younger than 14 years old) with symptoms of ARI (at least two of the following symptoms: cough, pharyngeal discomfort, nasal obstruction, snivel, sneeze, sore throat, dyspnea, and fever) or diagnosed as pneumonia as assessed by means of chest radiography, were included in this study at GWCMC. Chest radiography was conducted according to the clinical situation of the patients, and pneumonia was classified as an acute illness with ICD-10 (international classification of diseases-10). The Patients with the following conditions were excluded from our study: HIV infection; leukemia; receiving immunosuppressive agents; chemotherapy; known or suspected active tuberculosis.
Clinical characteristics of the patients were retrospectively analyzed. Throat swabs were collected in 2.5 ml of viral transport medium and were delivered to the Central Laboratory of GWCMC, which was used for further respiratory pathogen detection. This project was approved by the Ethics Committee of the GWCMC and was carried out in accordance with the principles expressed in the Declaration of Helsinki. Data records and collected clinical specimens are de-identified and completely anonymous.
Sequencing, molecular typing and phylogenetic analysis of the hypervariable regions of HAdV hexon gene The seven hypervariable regions (HVRs) of HAdV hexon gene [19][20] from positive specimens at the seasonal peak were PCR amplified, purified and directly sequenced. The HVRs of the hexon gene are responsible for the viral serum neutralization and contain serotype-specific residues [19][20]. Primers HVRF and HVRR were used in the PCR reaction to amplify the 1.6-kbp fragment of the seven HVRs, as described earlier [6]. The amplicon was submitted to Invitrogen (Guangzhou) for sequencing. Additional sequencing primer was also used (Hex1F: 5'-GCCAGAGCCTCAAGTTGGA-3'). The sequencing reaction was carried out by using an ABI Prism BigDye Terminator v3.1 Cycle Sequencing Ready Reaction kit with AmpliTaq DNA polymerase on an ABI 3730 DNA sequencer (Applied Biosystems). DNA sequencing reads were assembled into a single contig using the SEQMAN software from the Lasergene package (DNAStar; Madison, WI).

Statistical analysis
The data were evaluated for statistical significance with Chi-square test and Fisher's exact test where appropriate. All tests were two-tailed and the value of p<0.05 was considered to represent a statistically significant difference.

Demographic data of the children admitted with acute respiratory infections
Total 4,130 samples were collected and analyzed from 2012 to 2013. It showed that 80.8% of the ARI pediatric patients were at age groups 0-1 and 1-3 years ( Table 1). The HAdV-positive patients accounted for 7.9%, 17.2%, 17.5%, and 10.7% for age groups <1, 1-3, 3-6, and 6-14 years, respectively (Table 1). Although patients at age groups 1-3 and 3-6 had higher HAdV infection rates, but there was no statistical significance difference among different age groups. The total HAdV-positive rate is 12.0% (495/4,130) for the whole age group younger than 14. The male and female HAdV-positive rates were 11.9% and 12.0%, respectively. No significant gender difference was found (p = 0.793).
The 495 HAdV-positive patients with ARI included 347 male and 148 female, and the median age was 21 months, ranging from 1 day to 13 years old. The clinical diagnoses included pneumonia, bronchitis, acute asthmatic bronchitis, and Tonsillitis ( Table 4). The most common symptoms were fever (98.2%), cough (97.4%), and Dyspnea (72.1%); 208 patients (42%) had moist rales. Radiographic evidence of pneumonia was defined as the presence of consolidation (10.1%), alveolar or interstitial infiltrate (82.2%), and pleural effusion (1.8%). Thirty-two children (6.5%) diagnosed as pneumonia and with unstable vital signs were admitted into the Intensive Care Unit (ICU), 24 of which required respiratory support with mechanical ventilation. The ratio of male to female of these patients was 3:1, and the median age is 11 months, ranging from 54 days to 12 years old. Four hundred eighty HAdV-infected patients were totally recovered (96.9%,). However, four (0.8%) patients, who were previously healthy and aged 2 years and 1 month, 1 year and 3 months, 1 year and 1 month, and 8 months, respectively, died. Three of them were tested HAdV-positive only and the other one was co-infected with HBoV. They all suffered from high fever (>40°C) and rapid exacerbation of lower respiratory tract infection. All four cases required mechanical ventilation. Two of them died due to respiratory failure, and the other two died due to multiple organs dysfunction syndrome (MODS) and shock.

Molecular types of HAdVs detected in hospitalized children with ARI in summers of 2012-2013
The 195 HAdV-positive samples during July to Spetember of 2012-2013 were collected. The viral DNA were extracted and the HVRs of HAdV hexon gene were amplified by PCR. Among these specimens, 57 were both amplified and sequenced successfuly for HVRs. The molecular types of these samples were determined by the Blastn of the sequences in GenBank. In 2012, 71.4% were identified as type 3 and 28.6% were type 7. In 2013, three types of HAdVs were identified: HAdV-3, -7 and -55. HAdV-3 account for 69.8% (30/43), followed by HAdV-7 (12, 27.9%) (Fig  2A). One case of HAdV-55 in 2013 was identified and reported for the first time in Guangzhou metropolitan area, after the first and second reports of this type in Shaanxi Province (2006) and Beijing (2012), China [15,23] (Fig 2A). In summers of 2012-2013, the most prevalent HAdV type was HAdV-3, which accounted for 70% of the total cases (40/57); HAdV-7 accounted for 28% (16/57) (Fig 2B). No significant difference was found in clinical manifestations and laboratory findings when compared between the two HAdV types in these 56 cases (Table 5).

Phylogenetic analysis of the hexon genes of HAdVs from hospitalized cases in Guangzhou, Southern China
The HVR sequences of the hexon genes from these clinical strains isolated from hospitalized children have been deposited in GenBank under the accession numbers KR090744-KR090819.
The phylogenetic tree was built based on the alignment of the nucleotide sequences of the seven HVRs with the sequences from relevant prototypes and circulating strains in China. Both the neighbor-joining and maximum likelihood trees provided the same HAdV clusters and similarly supported topology (Fig 3 and S1 Fig). This further confirmed the types of these isolates. HAdV-3 (n = 40) was the most prevalent during summer of 2012-2013. These HVRs formed a subclade with another China HAdV-3 isolate Guangzhou01 circulating in 2004 [19], but they were distinguished from the HAdV-3 prototype strain GB, all of which formed HAdV-3 clade. HAdV-7 was the second prevalent strain during the summers of 2012-2013 (n = 16). All the HVRs formed a subclade with another recently re-emergent HAdV-7 strain DG01 in 2011 [24], indicating the hexon genes of HAdV-7 circulating in China were much conserved. However, they were discriminated from HAdV-7 prototype Gomen. The HVR sequence of HAdV-55 was very similar with the recently identified HAdV-55 strain BJ01 (Beijing; 2012) (bootstrap value 99) [15].

Discussion
HAdV is one of the major causative pathogens of severe acute respiratory disease in human beings. Although most of respiratory infections caused by HAdVs are self-limiting, fatal infections also occur in children and adults [19,[25][26][27]. In this retrospective study, HAdV was the third most frequent pathogen which causes 12% children hospitalized due to ARI (495/4,130). Surprisingly, more than 6.5% of HAdV-positive patients with ARI needed intensive care, and 4.8% needed respiratory support with mechanical ventilation. Four children died of pneumonia due to HAdV infection, three of which had only HAdV infection. Thus, pediatrician must take HAdV infection into account when they have pediatric patients with ARI.  The previously reported HAdV infection rates among pediatric patients varied from 4.5% to 25% in different countries [18,[28][29][30][31]. For example, the HAdV infection rate in North-East Brazil is 25% [28], 17.9% in Italy [30]. In Northern China, the HAdV infections rate was 10.4%-20.1% [9,31], which is consistent with our result in Guangzhou, Southern China (12.0%). However, the type distributions of HAdV in Southern and Northern China are completely different. In Southern China, most of HAdV-positive cases were caused by HAdV-3 (70% in this study) [6], followed by HAdV-7 (28%), while HAdV-7 dominated in Northern China (46.2%) [9]. However, another recent report found that HAdV-3 was the most predomi-  [32]. In their report, the molecular types were determined by nest PCR rather than sequencing; both outpatient and inpatient children were included, not just hospitalized children. These factors may account for the type distribution difference between the two reports in Northern China.
HAdV-55 was re-emergent in 2006 in China and named as type 55 in 2009 [33][34]. Subsequently it was re-emergent in Beijing in 2012 [35][36]. In this study, HAdV-55 was identified and reported for the first time in Guangzhou, Southern China in August 2013. Given that the high morbidity of severe pneumonia caused by HAdV-55, continuous monitoring of HAdVs types in children and adults are critically necessary. In this study, all the patients were inpatients and only HAdVs from the cases in seasonal peaks were molecular typed. These factors may account for only three types identified in this study.
A previous epidemiological study of HAdV infection in Guangzhou, China showed a tendency of gradual increase in infection rate in children, from 5% in 2006-2008 [29] to 9% in 2009-2012 [18]. Our study also indicates that HAdV infection in children increased to 9% (172/1867) in 2012 and 14% (323/2263) in 2013. For the pediatricians, more attention should be paid to this increasing HAdV infection rate during the past 7 years. This study also revealed the incidence and pattern of seasonal prevalence of HAdV associated with hospitalized children in Southern China. Although HAdV infection can occur during the whole year, the peak season for HAdV infection in this study was summer, which was different from the previous reports in Northern China and other countries or regions. For examples, in Northern China, Mexico and Taiwan, the peak seasons were in winter and spring [9] [37] [38]. However, in Tennessee, USA [39] and Brazil [28], HAdV infection was observed all the year without a clear seasonality. The difference of seasonal distribution between Southern and Northern China remains unknown. Our hypothesis is that HAdV-3 predominant in Southern China might be more resistant to higher temperature than HAdV-7, which probably leads to more HAdV-3 cases in Southern China in summer. More experiments to compare the biological characteristics difference between HAdV-3 and HAdV-7 may confirm this hypothesis.
The co-infections with HAdV were also investigated in this study. In recent US study, multiple pathogens were detected in 26% of the hospitalized children [39]. In our study, 17.0% (84/ 495) of the HAdV positive children were co-infected with other pathogens detected from their nasopharyngeal swabs. Interesting, the recent HAdV epidemiology study showed that there was very high co-infection rate between HAdV and other pathogens (90.4%) in Northern Bootstrapped, neighbor-joining trees with 1,000 replicates were constructed using the MEGA 5.1.0 software (http://www.megasoftware.net) and by applying default parameters, with a maximum-composite-likelihood method. Bootstrap numbers shown at the nodes indicate the percentages of 1,000 replications producing the clade. A bootstrap value of 80 indicates robustness and confidence in the branching. The scale bar indicates units of nucleotide substitutions per site. doi:10.1371/journal.pone.0155412.g003 China [9]. In our 495 HAdV-positive children, the most common co-infection pathogens with HAdV were MP and HBoV (48/84, 57.1% and 14/84, 16.7%, respectively), which was slightly different from that in Northern China [9], where RSV, RHV and HBoV were the most common co-infection pathogens with HAdV. The association between co-infection and severe illness is still unknown. The co-infection pathogens in the hosts might interact indirectly or directly with each other, but the interaction mechanism is poorly understood. Given the large proportion and diversity of the co-infection pathogens, further study is needed.
However, this study has some limitations. First, considering vast number of studied samples and the relatively limited resources, we performed the sequence analysis and molecular typing only from those samples in seasonal peaks, but not from the whole year. Therefore, the molecular typing of all HAdV positive samples in this study is not available. The cases in seasonal peaks might not reflect the difference in clinical manifestations and laboratory findings between the HAdV-3 and HAdV-7 in the whole year. Second, the sensitivity of current realtime fluorescent PCR kit for some pathogens (e.g., rhinovirus) is not optimal. It could not identify all the subtypes of rhinovirus, which might lead to the underestimation of rhinovirus infection rate.
In conclusion, this study reported the epidemiology among the HAdV-infected pediatric hospitalized patients with ARI in Guangzhou, Southern China in 2012-2013. HAdV was detected in 495 out of 4,130 (12.0%) children with ARI. The most prevalent HAdV type was HAdV-3 (70%) and HAdV-7 (28%) in summer. HAdV-55 was reported for the first time in Guangzhou in this study. Seasonal peaks of HAdV infections in Guangzhou occurred in summer, not in winter or spring. Co-infections with other respiratory viruses were common (17%). The clinical observations indicate that HAdVs is a major agent of ARI and HAdV infections can lead to severe clinical consequences, including death. Additional studies are needed to determine the role of HAdV infection in the severity of disease and their co-infections with other respiratory pathogens. More attention should be paid to the difference in circulating types and seasonal peaks of HAdVs between Southern and Northern China. Therefore, extensive and continuous surveillance for HAdV molecular epidemiology is necessary and HAdV vaccine should be developed to prevent the infection in children.