Detection of antibodies against influenza D virus in swine veterinarians in Italy in 2004

Abstract Influenza D virus (IDV) was first isolated from a swine with respiratory disease symptoms in 2011 in the United States. Epidemiological and serological studies support the hypothesis that cattle represent the natural reservoir of IDV with periodical spillover events to other animal hosts. Little is known about the seroprevalence in humans and in specific target groups such as veterinarians in Italy. This study was designed to assess the prevalence of antibodies against two influenza D lineages (D/660 and D/OK) in Italy in archived serum samples from veterinarians working with swine collected in 2004. Serum samples were tested by haemagglutination inhibition (HI) and virus neutralization (VN) assays. Results showed that 4.88% (4/82) of tested samples were positive for D/660 and 2.44% (2/82) for D/OK by HI assay. Three out of four samples showed positivity when tested by VN assay. Our data suggest undetected IDVs might have circulated and/or been introduced in Italy as early as 2004 at least in some animal species such as swine. In addition, it seems that the virus was circulating among veterinarians before the first isolation in 2011. This finding highlights the importance to continue monitoring the IDV spread in animals and humans for more detailed surveillance.

IDV, as influenza C virus (ICV), has seven RNA segments and only one major surface glycoprotein, the hemagglutinin-esterase-fusion, which is responsible for binding, receptor destroying, and fusion. The homology of amino acid sequences between IDV and ICV is roughly 50%, however, the distance between the two influenza viruses is similar to the one found between influenza A and B viruses (IAVs, IBVs). 2,3 No cross-reactivity has been detected between IDV and ICV. 4 Although the virus has been first isolated in swine, several epidemiological and serological studies support the hypothesis that cattle represent the natural reservoir of IDV with periodical spillover events to other animal hosts (i.e., camel, sheep, swine, horse, goat). [4][5][6] The viral genome has been detected in some animal species while only specific antibodies have been detected in horses without evidence of viral genome or virus isolation. 7 So far, there is no evidence of IDV infection in chickens and turkeys. 8 The virus has been identified in different countries across the world (i.e., France, Italy, Luxemburg, Canada, Mexico, China, Mississippi, Japan, Nebraska). 5,6,[9][10][11][12] To date, different lineages have been identified, D/OK-, D/ 660-, D/Japan-lineages (D/Yama2016 and D/Yama2019) and the recently identified D/CA2019. 13  show a prevalence of 6.3% and 3.1%, respectively, in 2016-2017. 15,16 The live trade seems to play a key role in viral spread considering that Italy, together with Spain, is one of the most important cattle importers in Europe from France. Data report higher IDV seroprevalence in importing countries (i.e. Italy) than in exporting countries suggesting that cattle may come in contact with the virus during transportation or just after. 15 IDV seroprevalence in Italy has been studied on general population only, ranging from 5.1% in 2005 to 46.0% in 2014. 17 International studies on cattle and farming workers performed in Florida and Malaysia showed a seroprevalence of 94% and 1.3%, respectively, suggesting that cattle-exposed people could be infected with IDV through occupational zoonotic transmission. 18,19 The aim of this study was to assess the prevalence of antibodies against two IDV lineages (D/660 and D/OK) in Italy in archived serum samples from veterinarians working with swine collected in 2004. Human serum without immunoglobulin A, immunoglobulin M, and immunoglobulin G was used as a negative control (Sigma-Aldrich).

| Influenza viruses
All serum samples were tested by the haemagglutination inhibition (HI) assay.
Positive samples with a sub-set of negative samples were tested by the virus neutralization (VN) assay.

| HI assay
The HI assay was performed as previously described. 17 All serum samples, including positive and negative controls, were pretreated with receptor-destroying enzyme (ratio 1:5) from Vibrio cholerae (Sigma-Aldrich) for 18 h at 37°C in a water bath followed by heat inactivation for 1 h at 56°C in a water bath with 8% sodium citrate (ratio 1:4). All serum samples were tested in duplicate using turkey red blood cells (0.35%). The antibody titer was expressed as the reciprocal of the highest serum dilution showing complete inhibition of agglutination. Since the starting dilution was 1:10, a titer below the detectable threshold was conventionally expressed as 5 for calculation purposes.

| Virus neutralization
The MDCK cell cultures were grown at 37°C in 5% CO 2 and preincubated for 4 h.
Serum samples, including positive and negative controls, were previously heat-inactivated at 56°C for 30 min. Samples twofold diluted with EMEM culture medium supplemented with 0.5% fetal bovine serum were mixed with an equal volume of virus (100 TCID50/well). After 1 h of incubation at 37°C in 5% CO 2 , 100 µl of the mixture was transferred to a plate containing 1.5 × 10 4 MDCK cells/well. Plates were read for haemagglutination activity in the supernatant after 5 days of incubation at 37°C in 5% CO 2 . The VN titer was expressed as the reciprocal of the highest serum dilution showing the absence of haemagglutination.

| RESULTS
Three out of 4 samples were further tested by the VN assay and showed positivity for both lineages with titers ranging from 10 to 40 (Table 1). Unfortunately, we did not have enough serum for one sample to be tested by the VN assay (Table 1). Table 2

| DISCUSSION
In this study, we tested 82 archived serum samples from swine