Molecular Typing of Human Adenovirus Infection among Hospitalized Patients with Respiratory Tract Infections in Guangzhou, China

Background Human Adenoviruses (HAdVs) cause a wide array of illnesses in all age groups. They particularly cause frequent morbidity among children. In China, human adenovirus types 3, 4, 7, 11, 14, 21, and 55 have caused at least seven outbreaks since 2000. However, limited studies are available regarding the epidemiological patterns and diversity of HAdVs types among hospitalized patients with respiratory tract infections (RTIs). Methods To understand the epidemiology and type distribution of HAdV infections associated with RTIs in China, nasal swab (NS) clinical samples were collected from 4129 patients in a Guangzhou hospital between August 2017 and October 2019. PCR, sequencing, and phylogenetic analysis were performed on these specimens to identify HAdV virus types. Results HAdV was detected in 99 (2.4%) of the 4,129 NS specimens, with the highest HAdV prevalence (6.3%) found in children between the ages of 5 and 10 years. Among HAdV-positive specimens, the most prevalent genotypes identied were HAdV-B3 (55.6%) and HAdV-B7 (25.3%). The most common symptoms in the HAdV-infected patients were fever (100%), cough (80.8%) and rhinorrhea (71.8%). HAdV infections were detected throughout the year with a relatively higher prevalence in summer. Conclusion All ages suffer adenovirus infections, but young children are at greatest risk. These study data demonstrate that at least three species of HAdVs (species B, C, and E) are circulating in Guangzhou City, China. As antiviral therapies and type-specic vaccines become available, such epidemiological data will be useful in guiding therapy and public health interventions.

Province in 2011 [13]. Two outbreaks of acute respiratory diseases caused by HAdV-7 were detected in military training camps, one in Hubei Province and another in Shaanxi Province between 2012 and 2013 [7]. HAdV-B3 was also frequently reported as the common cause of epidemic ARI outbreaks in China [14].
However, limited data are available regarding the epidemiological and clinical features of HAdV in hospitalized patients. This study we sought to determine the prevalence and types of HAdVs among hospitalized patients with RTIs in Guangzhou, China. Such epidemiological data are useful to health professionals regarding decisions in employing appropriate therapy and adopting effective prevention strategies for adenovirus control.

Materials And Methods
Case De nition A total of 4,129 nasal swab (NS) specimens were previously collected from patients with RTIs at the Second A liated Hospital of Guangzhou University of Chinese Medicine between August 2017 and October 2019. These patients were recruited for participation in the hospital's study if they met the following inclusion criteria: (i) had a fever (temperature ≥ 37.3°C, measured at the hospital) with the apparent respiratory symptom(s); (ii) did not have the bacterial infection (white blood cell count > 12x109/L, or PCT > 0.5ng/L). Individuals who met the inclusion criteria were consented (parental consent required for adolescents under the age of 18) and invited to complete a brief questionnaire about their demographics and symptoms.

Subject Sample Collection
Each participant permitted the collection of one NS specimen (the swab stayed in the nose for 15 seconds), which was placed in viral transport media (COPAN Diagnostics Inc, Italy), stored in the icebox and then transported to the hospital's laboratory. Specimens were preserved at -80°C until further processing.
At the hospital's laboratory, all clinical NS specimens were rst screened with the commercially available Immuno uorescence assay (respiratory viral panel 1 screening and identi cation kit, Light DiagnosticTM, Chemicon International Inc, Temecula, USA) targeting adenovirus, in uenza A and B virus, parain uenza 1-3, and respiratory syncytial virus (RSV). A total of 117 HAdV-positive specimens were detected. Later these HAdV-positive specimens were sent to the Duke Kunshan University (DKU) One Health Research Laboratory for further study.

Detection of HAdV with Real-Time PCR
At the DKU One Health Research Laboratory, DNA was extracted from HAdV-positive specimens and eluted in 60µl of elution buffer using a QIAamp MinElute Virus Spin Kit (Cat. No.57704, Qiagen Inc., Hilden, Germany). Extracted DNA was then tested for adenovirus by real-time reverse transcriptasepolymerase chain reaction (qPCR) assays using a BioRad SsoAdvanced Universal Probes Supermix (Bio-Rad Laboratories, Richmond, CA) as previously described [15] on a Mic qPCR Cycler (BioMolecular Systems, EI Cajon, CA). Positive and negative template controls were included in each qPCR run. The qPCR cycling was run as follows: 90°C for 2 min, 95°C for 3 min, followed by 40 cycles of 95°C for 15s, and 60°C for 30s. Samples with quanti cation cycles (Cq) values < 38 were considered positive for adenovirus.

Sequence and Phylogenetic Analysis
Hexon gene nucleotide sequences from the specimens collected in this study were aligned with multiple reference strains available in GenBank (Supplementary Table 1 The 99 HAdV-positive pateints identi ed in this study included 66 (63.67%) males and 33 (33.33%) females. Their median age was 4 years, range 2 months to 63 years. The HAdV-positive patients were distributed across age groups: Age groups <5, 5-10, 10-19, and ≥19 years, accounted for 3.8%, 6.3%, 0.6%, 0.16% of positives, respectively (Table 1). A signi cant difference in the number of HAdV-positive patients among the different age groups was observed with the highest prevalence of HAdV infections being found among children who were between 5 and 10 years (6.3%), followed by those under ve years (3.8%). There was no signi cant difference in HAdV-positivity observed between males and females.

Seasonal Distribution of the HAdV Infections
In China, 12 months were classi ed into four seasons, including Spring ( Among the 117 immuno uorescence-positive HAdV specimens, 103 (88.0%) specimens were detected as HAdV-positive with qPCR at DKU One Health Research Laboratory.

Sequence and Phylogenetic Analysis
To further analyze the HAdV genotype, the hexon gene from 103 HAdV positive specimens was ampli ed by conventional PCR. Ninety-nine (96.1%) specimens were successfully genotyped. Eight out of 99 sequences selected as representative sequences (see details in supplementary material) were further aligned with other reference strains (partial Hexon gene) ( Figure 2). The phylogenetic analysis indicated that 84 patients belonged to species B, 7 patients belonged to species C, and 8 patients belonged to species E (Table 3). Within the HAdV-B species, type 3, 7, and 55 were identi ed. It is worth noting that all the HAdV-B strains identi ed in this study were highly identical to strains detected in Shenzhen and Jiangxi City, suggesting the genetic conservation of some HAdV-B viruses in different areas of China. Additionally, HAdV-B3 and HAdV-B7 were the most prevalent types in this study. Within the HAdV-C species, types 1, 2 and 5 were identi ed. All the HAdV-C species generated from this study seem to be more closely related to HAdV-C viruses circulated in neighboring countries. For example, the HAdV-C1 strain in this study was nearly identical to the strain CAU230/AdV/KOR/2016 found in South Korea. The HAdV-C5 strain identi ed in this study was identical to a Japanese HAdV virus. Study ndings indicated that Species B, C, and E (at least seven subtypes) circulated simultaneously in Guangzhou in 2017-2019.

Clinical Characteristics of HAdV Genotypes
HAdV genotypes differed in their characteristics ( Table 3). The most common clinical manifestations of all HAdV genotypes were fever (100%), cough (50 -100%), and rhinorrhea (50-100%). The comparison among the seven HAdV types revealed that HAdV-B7 caused more severe diarrhea than other HAdV types. The majority (94.6%) of pneumonia patients were found to be associated with HAdV-B species (including type 3, 7, and 55). Unexpectedly, three out of four gastroenteritis patients were found in patients who were infected with HAdV-B7.

Discussion
A recent multicenter, prospective registry study found that adenovirus was the third-leading cause of viral infection among community-acquired pneumonia (CAP) patients in China, after in uenza viruses and respiratory syncytial virus [17]. An increasing number of HAdV outbreaks have been reported in China in recent years [7,12,14,18]. But because there is no national surveillance system for HAdV in China and there is often no difference in symptom clustering between viral and bacterial infections, the diagnosis of HAdV is often underestimated [2].
The present study recorded the epidemiological distribution of circulating HAdV strains among hospitalized patients with respiratory tract infections (RTIs) between 2017 and 2019 in Guangzhou City, China. In this study, the overall positive rate of HAdV was 2.4%, which is consistent with the positive rate (2.0%-6.1%) found in hospitalized patients with acute viral respiratory infection in other recent reports [3,[19][20][21]. However, two previous studies conducted in the Northern part of China showed a higher HAdV prevalence (10.4%-20.1%) than our study [22,23]. These ndings demonstrate that HAdV prevalence may differ by geographic locations. Such differences in HAdV prevalence could be in uenced by a number of factors, including sample type, small sample size, sampling period, study duration, and patients' demographic information.
Regarding patient demographic data, there was no signi cant difference observed in HAdV detections between genders in this study. This is consistent with the ndings of previous studies conducted in China [3,20]. However, this was in contrast to some previous reports that male children were more likely to be infected with HAdV [24,25]. The highest HAdV-positive rate was observed among children between the ages of 5 and 10 years, which is similar to the ndings of a recent study in the same city [3]. However, most previous reports found that the HAdV infections occurred more often among children under ve years of age [19, 26-28].
Our study revealed that although HAdV infections were detected throughout the year, the prevalence peaked in the summer. This is consistent with ndings of Chen et al [28] during 2012-2013. But it was not consistent with that of studies conducted in Northern China [22] and Mexico [29]. In Tanzania and Switzerland, HAdV infections were observed during all seasons of the year with no clear seasonality demonstrated [21,27]. These difference in HAdV prevalence between seasons are interesting and bear future study as seasonal risk could in uence future employment of HAdV vaccines which are in development in China [30].
The majority (94.6%) of HAdV types detected in this study were of species B. HAdV-B (e.g., 3, 7, and 55) has been continuously reported to be associated with more severe acute respiratory disease than other HAdVs species [2,14,18]. Only seven patients were found to have HAdV-C and only eight patients were found to have HAdV-E infection during the study period.
Previously, three types of HAdV-C (e.g., 1, 2, and 5) were identi ed in China, although studies of HAdV-C species have been limited [33]. HAdV-C species viruses were identi ed as the primary pathogens responsible for respiratory tract infections among hospitalized children, particularly among infants under two years of age [33]. Our study is consistent with this in that more than half of HAdV-C-positive patients were observed in children less than two years old. Previous research suggests that recombination events are commonly observed among HAdV-C types [33]. As recombinant HAdV strains have caused epidemics, it seems prudent to monitor for changes in HAdV-C types in China.
Although our study provides crucial molecular evidence regarding the epidemiology and clinical features of HAdV infections in Guangzhou, China, it has several limitations. First, we did not study outpatients and they could have had a different distribution of HAdV types. Second, study samples were rst identi ed using a commercial assay which is not thought to be as sensitive as the qPCR we employed. Hence, we likely only captured data on the specimens with higher viral titers. Thus, the true prevalence of HAdV among hospitalized patients may have been higher and the distribution of HAdV types different. Third, we were unable to rule out a number of possible viral coinfections that were not surveilled for with the commercial product (e.g., parain uenza 4, rhinoviruses, enteroviruses, etc.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download. BMCSupplementaryMaterialXWGCG202104.docx