Influenza C infections in Western Australia and Victoria from 2008 to 2014

Background Influenza C is usually considered a minor cause of respiratory illness in humans with many infections being asymptomatic or clinically mild. Large outbreaks can occur periodically resulting in significant morbidity. Objectives This study aimed at analyzing the available influenza C clinical samples from two widely separated states of Australia, collected over a 7‐year period and to compare them with influenza C viruses detected in other parts of the world in recent years. Patients/Methods Between 2008 and 2014, 86 respiratory samples that were influenza C positive were collected from subjects with influenza‐like illness living in the states of Victoria and Western Australia. A battery of other respiratory viruses were also tested for in these influenza C‐positive samples. Virus isolation was attempted on all of these clinical samples, and gene sequencing was performed on all influenza C‐positive cultures. Results and conclusions Detections of influenza C in respiratory samples were sporadic in most years studied, but higher rates of infection occurred in 2012 and 2014. Many of the patients with influenza C had coinfections with other respiratory pathogens. Phylogenetic analysis of the full‐length hemagglutinin–esterase–fusion (HE) gene found that most of the viruses grouped in the C/Sao Paulo/378/82 clade with the remainder grouping in the C/Kanagawa/1/76 clade.


| INTRODUCTION
Influenza C is often ignored compared with the other human pathogens in the Orthomyxoviridae family (influenza A and influenza B).
With respect to other influenza types, it causes a milder disease with fewer complications and, as it has not been included in routine testing, there is little new information being generated about its impact. In etiologic studies of respiratory illnesses that have included influenza C testing, it usually accounts for a low proportion of acute respiratory pathogens identified. For example, a Canadian study 1 found 2.32% of respiratory samples tested from children identified influenza C, while a study in Japan spanning 18 years 2 (1996-2013) found influenza C in 0.43%-1.73% of samples from children; this result was similar to a Spanish study that reported influenza C in 0.7% of children's samples. 3 Low rates were reported in a retrospective study Scottish study that screened 3300 respiratory samples collected from children and adults during the period from August 2006 to June 2008 with only six positive influenza C-positive samples (0.2%) identified (with 4/6 from children ≤2 year) compared to 3.2% influenza A detections and 0.9% influenza B detections. 4 Higher rates of influenza C infection have been reported in a Nigerian study, where 4.8% of the respiratory samples from children were positive for influenza C 5 and Finnish study among young adult male military recruits, 6 in which influenza C was identified in 4.2% of all samples. Influenza C can also be identified in a significant proportion of children hospitalized with lower respiratory tract infections, as reported by Shimizu et al. 7 2015, who found approximately 10% of children had influenza C present in samples collected from four Japanese hospitals during 2009-2010. Often these cases of influenza C also have other viruses or bacteria detected in the specimen making it difficult to attribute the contribution of influenza C infections to the clinical illness manifestations. 5,8 In contrast to most viral etiology studies, serological studies often find high seroprevalence rates in the community which rise rapidly up to 10 years of age to around 50%-60% of the population, indicating that widespread transmission has occurred among children in many different countries. 9 Unlike influenza A and B infections where early treatment with neuraminidase inhibitor drugs can be effective, there are no effective antiviral treatments for influenza C.
This study analyzed influenza C viruses detected in respiratory samples collected from two influenza illness surveillance programs operating in the state of Western Australia (WA) from 2001 to 2014: one covering patients of all ages presenting to general practitioners with an influenza-like illness (ILI), and the other covering young children presenting with a respiratory illness to a metropolitan pediatric hospital emergency department or a general practitioner. These were compared with influenza C viruses detected from routine screening of respiratory samples from Melbourne, Victoria from 2011 to 2014. This report is the first to isolate and characterize (including sequencing) influenza C viruses collected in Australia.

| Real-time (RT-PCR) assay
Prior to sequencing or culturing, due to the age of some of the clinical samples, a real-time PCR assay was designed and run so that samples could be retested at the WHO Centre to confirm that viral RNA was still detectable. Primers and a minor groove binding probe were designed using an alignment of influenza C hemagglutinin-esterase-fusion (HE) gene sequences from GenBank. Sequences were downloaded and aligned using the Megalign program of DNAStar (Madison, WI, Australia), and HE primers and probes were obtained (GeneWorks, Adelaide, SA, Australia) and used for the real-time PCR of clinical samples and also for the confirmation of influenza C virus isolates (Table 1)

| Gene sequencing and phylogenetic analysis
Purified RNA extracted from influenza C-positive samples was amplified using the BIOLINE MYTAQ one-step RT-PCR kit (Bioline, Australia) using in-house-designed HE gene-specific primers supplied by Bioneer (Melbourne, Australia; Table 1). RT-PCR was per-

| Patient details and initial testing
For the 16 influenza C-positive samples from the SPN(WA) GP surveillance, the age range was 1-40 years with the median age of 22 years (25% male, 75% female), while for the 63 influenza C-positive WAIVE samples the age range was 0.6-5.1 years with a median age of 1.7 year (46% male, 54% female). Note that the WAIVE program only recruits children aged between 6 months and 59 months who attended the Emergency Department or were admitted to the Princess Margaret Hospital, Perth (WA) with an ILI (defined by the presence or history of fever and a respiratory symptom). This study group was therefore made up exclusively of young children and they had higher rates of influenza C than the SPN(WA) group (Table 1).
During the WAIVE study, the overall prevalence of influenza C cases was 1.6% and varied from 0% (2009) Table 3). The most common multiple infections along with influenza C were with human enteroviruses/ rhinoviruses (13 samples), influenza A (seven samples; four with influenza A H1N1pdm09 and three with influenza A(H3N2)) followed by RSV (six samples) and adenovirus (six samples) (

| Sequencing and phylogenetic analysis
Hemagglutinin-esterase-fusion gene sequence was obtained for all samples that yielded influenza C virus isolates. Thirty-five influenza C full HE sequences (28 from Perth/WA and seven from Victoria) and one partial HE sequence from Perth were obtained from the virus isolates.
These sequences were compared with publically available HE sequences from reference viruses including those that represent the six antigenically distinct influenza C groups from recently circulating influenza C viruses (Fig. 1) Victorian samples were evenly distributed in both of these clades (Fig. 1).
The bootstrap values for all of these clade assignments were high. The

| DISCUSSION
This is the first report characterizing influenza C viruses obtained from Australia. Influenza C viruses were detected sporadically in two widely in the WAIVE study, 11.1% vs 16.8%, respectively. The frequency of large influenza C epidemics in Australia is unknown, but looking at the WAIVE data (Table 2) there is some indication that they may occur approximately every 2 years, as the influenza C cases detected were the highest in 2010, 2012, and 2014. In Japan, it also appears that influenza C epidemics could be detected every couple of years. 14 Overall, we were moderately successful in isolating influenza C, with 51.5% of original clinical samples yielding isolates. These isolates and 2012 from children of less than 10 years of age. In a Paediatric Clinic of the University of Milan, a study over 4 years (2008-9 to 2011-2) in children under 15 years of age with influenza C and radiographically confirmed community-acquired pneumonia, five cases were identified with two viruses of the C/Kanagawa/1/76-lineage and three from the C/Sao Paulo/378/82-lineage. 18 Clearly, viruses from these two influenza C lineages have been the dominant clades seen in many countries in recent years even though they appear to have changed very little based on their HE phylogeny, since they were first reported.
Unfortunately, our sample size was not large enough to be able to gain an in depth understanding of influenza C within Australia. Nor can it determine the level of reassortment that has occurred as the six internal genes have yet to be sequenced. However, Odagiri et al.