| BackgroundIn the past,the world public health officials focused more attention on the threat of avian influenza viruses,particularly highly pathogenic H5N1 strains.However,in 2009,the novel swine H1N1 influenza virus emerged in Mexico and the United States,and spread rapidly around the world,causing a pandemic.Phylogenetic analysis indicated that the novel swine H1N1 influenza virus that caused the 2009 pandemic contained gene segments of swine,avian and human influenza viruses.Evolutionary analysis revealed that it was closely related to the swine influenza viruses(SIVs)and initially named the influenza as "Swine flu".The emergence of the novel swine H1N1 influenza virus has shocked people and has to re-think past knowledge and practices.SIVs belong to influenza A virus in the Orthomyxoviridae family.The genome of SIVs consists of eight single negative-strand RNA segments,this multi-segment genomic feature makes the virus easily to reassortant or recombine,and often results in emergence of novel viruses.Pigs have both ?-2,3 and ?-2,6 sialic acid receptors in their respiratory tract,which can not only infect SIVs,but also infect IAVs from both poultry and humans.The pigs,which are considered to be a "mixing vessel" of IAVs,plays an important role in the interspecies transmission chain of "avian-swine-human".Currently,the most prevalent subtypes of SIVs are H1,H3,N1,and N2,and mainly distributed in Europe,North America and East Asia.China's pig production industry has developed rapidly and now China is the largest pork producer in the world.In recent years,most of small pig farms have been replaced by confined animal feeding operations(CAFOs)in China.The SIVs are prevalent in CAFOs.In addition,China has a high population density,a large number of livestock,poultry breeding,and abundant migratory birds with a wide migration route.These provides good conditions for the recombination or reassortment of SIVs,human and avian influenza viruses,which improving the chances of generating a novel influenza virus.Due to the lack of systematic and prospective surveillance of swine influenza viruses,it has largely hampered the understanding of the evolution of influenza viruses.Most of the existing data are derived from samples of sick pigs from different regions,rather than from prospective studies in specific regions lacking of enough knowledge of the evolution of SIVs,cross-species transmission,and the risk of infection in humans.Therefore,our study has carried out CAFOs surveillance,prospective cohort study of swine-exposed population and influenza-like illness(ILI)surveillance in China to identify the prevalence and distribution of SIVs in CAFOs,the SIVs infection among swine-exposed population,the risk factors of SIVs infection and the evidence for SIVs cross-species transimission.Our study results are important for improving the ability to prevent and control influenza pandemics.ObjectivesTo identify the prevalence and distribution of SIVs through the monthly surveillance in CAFOs.To identify the infection rate of SIVs,biomarkers of infection and risk factors for SIVs infection through cohort study of swine-exposed populations and controls.To identiify the evidence for SIVs cross-species transmission through ILI surveillance.To clarify the transimission of influenza virus in humans,pigs and farm environment through gene sequencing and phylogenetic analysis.MethodsFrom March 2015 to September 2018,11 CAFOs were enrolled in Shandong and Jiangsu Province for active surveillance of SIVs.Pig oral secretions and pig farm environmental specimens(fecal/slurry,environmental water,environmental swabs,bioaerosol)were collected monthly,sampling registration form(time,location,pig age,etc.)and the farm monthly follow-up form(number,health status,weather conditions,etc.)were collected at the same time.Specimens were screened by quantitative real-time polymerase chain reaction(q RT-PCR),and the positive specimens(Ct value?38)were cultured by Madin-Darby canine kidney(MDCK)cells to isolate virues,and the whole genome sequence of influenza virus was amplified by one-step reverse transcription PCR(RT-PCR).Phylogenetic analysis was used to identify the evolutionary characteristics of SIVs.Between March 2015 and December 2017,a 2-year prospective cohort study was conducted.The swine-exposed subjects were recruited from the CAFOs,slaughterhouses,and pork markets,and the general populations were included as controls.During the enrollment,a questionnaire was used to collect demographic information,occupational exposure history,and disease history.Serum and nasal wash specimens were collected.At the two-year follow-up,the annual follow-up information,serum and nasal wash specimens of the study subjects were collected.Nasal wash specimens were screened by q RT-PCR,and positive samples were isolated by MDCK cells.The serum neutralizing antibody against swine H1N1,H3N2 and avian H5N1,H5N2,H7N9 and H9N2 subtypes were detected by microneutralization(MN)assay.The total Ig A,virus-specific(swine H1N1 and swine H3N2)Ig A of nasal wash specimens were detected by enzyme-linked immunosorbent assay(ELISA).Weekly follow-up was conducted by telephone or home visit.ILI questionnaire information,nasal washes and nasopharyngeal swabs were collected.Nasal washes and nasopharyngeal swabs were screened by q RT-PCR and the positive specimens were isolated by MDCK cells.The isolated strains were sequenced and phylogenetic analysis was performed.Serum and nasal wash specimens were detected by MN and ELISA assay to compare the difference between serum MN titers and nasal wash Ig A concentrations in samples among different phases.The categorical data were analyzed by ?2 test or Fisher's exact test.The quantitative data were analyzed by t test or analysis of variance.For surveillance data in CAFOs,univariate analysis was used to calculate the odds ratio(OR)and 95% confidence interval(CI)to identify risk factors that affect the positive detection of IAV in each type of specimen.For population data,univariate and multivariate logistic regression analysis were used to calculate OR and 95% CI to identify risk factors for the infection of swine and avian influenza viruses in the population.The Poisson regression model was used to calculate the incidence rate ratio(IRR)and the 95% CI to compare the difference between the incidence of different subtype of SIVs in the swine-exposed population and the general population.All statistical tests were two-sided,and P<0.05 was considered statistically significant.Entry and analysis of the data was performed using Epi Data 3.1 software(Odense,Denmark)and SPSS Statisitic 21.0 software(SPSS,Chicago,Illinois,USA).Results Through the surveillance in CAFOs from March 2015 to September 2018,a total of 17,525 specimens were collected,including 11864 specimens of pig oral secretions(POS),1427 specimens of fecal/slurry(F/S),1427 specimens of environmental water(EW),1427 environmental swab(ES)specimens and 1380 bioaerosol(BA)specimens.The positive rate of IAV was 7.7%(1344/17525),and the positive rates of IAV of POS,F/S,EW,ES and BA were 8.4%(1002/11864),4.9%(70/1427),5.9%(84/1427),8.1%(115/1427)and 5.3%(73/1380),respectively.Pig oral secretion specimens had the highest positive rate of IAV,and the fecal/slurry specimens had the lowest positive rate.Influenza A virus was characterized by seasonal bimodal epidemics in summer 6/7 and winter 1/2,and pigs with smaller age were at high risk of SIVs infection.Sequencing and alignment analysis of 34 isolates and 113 specimens with Ct value?30 showed that SIVs were prevalent in CAFOs,including Eurasian avian-like H1N1,pdm09 H1N1,swine H1N2 and swine H3N2 subtypes of IAV,in which Eurasian avian-like H1N1 was the most prevalent strain.Reassortment of influenza virus gene segments existed among Eurasian avian-like H1N1,pdm09 H1N1,and swine H3N2 subtypes.The H1N1 and H1N2 influenza viruses had E190 D,G225E/D and R226 Q mutations while H3N2 influenza viruses had Q226 I and G228 S mutations on the HA protein,suggesting that the receptor binding capacity had improved.Fourteen H1N1 subtype influenza viruses had N248 D resistance mutations on the NA protein.H1N2 and all H3N2 subtype influenza viruses had I122 V resistance mutations on the NA protein.One H1N1 influenza virus had a D701 N mutation on the PB2 protein,which was associated with increased viral pathogenicity and transmission,while all other viruses had mutations at T271 A and Q591 on PB2 protein.No PB1-T296 R and PB2-E627 K mutation sites were found in all viruses.A prospective cohort study of human SIVs infection was conducted between March 2015 and December 2017.For swine H1N1 virus,at baseline,the positive rate of serum MN antibody against swine H1N1 virus was higher in the exposed group than in the control group(13.1% vs 6.0%,P=0.046);the seroconversion rate was 19.9% in swineexposed population and 12.1% in the control population upon the paired serum samples collected in different years,and there was a significant difference(P=0.019);the seroconversion density in swine-exposed group(13.93/100 person-years)was significantly higher than that in the control group(8.64/100 person-years)(IRR=1.713,95% CI 1.002-2.926,P=0.047).Further factor analysis found that exposure to pigs and the outbreak of pigs during work(adjusted OR=3.349,95% CI 1.170-9.592)were risk factors for infection of swine H1N1 virus.For swine H3N2 virus,the swine-exposed population had a lower positive rate against swine H3N2 virus than the control population(36.8% and 52.0%,P=0.007)at baseline;the seroconversion rate was 12.2% in the swineexposed population and 20.7% in the control population upon paired serum samples,and there was a significant difference(P=0.003);the seroconversion density(8.81/100 person-years)in the swine-exposed group was significantly lower than that in the control group(14.09/100 person-years)(IRR=0.589,95% CI 0.360-0.964,P=0.034).Further factor analysis found that the high frequency use of protective shoes(adjusted OR=0.391,95% CI 0.156-0.980)was a protective factor.For avian influenza viruses,serum MN antibodies were performed against H5N1,H5N6,H7N9 and H9N2 subtype avian influenza viruses.For H5N6 influenza virus,only one subject in exposed group had positive serum MN antibodys at 12 th and 24 th month follow-up,and seroconverted.For H9N2 virus,3 swine-exposed subjects and 1 control subject were positive for serum MN antibody at different time points,and 2 of exposed subjects and 1 control seroconverted at annual follow-up.All serum were tested negative for both H5N1 and H7N9 avian influenza viruses.At the same time,the total Ig A and the swine specific Ig A antibodies in nasal wash specimens were analyzed in this study.At baseline,total Ig A antibody concentrations(1754 ng/m L and 585 ng/m L),swine H1N1 specific Ig A antibody concentrations(653 ng/m L and 97 ng/m L),and swine H3N2 specific Ig A antibody concentrations(1074 ng/m L and 434 ng/m L)were higher in the exposed group than in the control group(P<0.001).At the next two years of follow-up,the total Ig A antibody and swine specific Ig A antibody concentrations were also higher in the exposed group than in the control group.For swine H1N1 specific Ig A antibodies,exposure to pigs(adjusted OR=2.607,95% CI 1.674-4.059)and smoking history(adjusted OR=1.385,95% CI 1.066-1.800)were risk factors for elevated concentrations.For swine H3N2 specific Ig A antibodies,exposure to pigs(adjusted OR = 2.167,95% CI 1.434-3.274),smoking history(adjusted OR=1.428,95% CI 1.092-1.868)and medication history within 30 days(adjustment OR=1.685,95% CI 1.250-2.270)were the risk factors for elevated concentrations.The study also carried out ILI surveillance in the cohort,with a total of 56 reports during the study period,including 41 swine-exposed population and 15 control population.32(57.1%)ILI groups of serum(baseline investigation or follow-up/acute/convalescent)were associated with at least one 4-fold rise in MN titers against a swine virus.22(39.3%)and 17(30.4%)groups of serum had at least a 4-fold rise in MN titers against swine H1N1 and H3N2 influenza viruses,respectively.The total Ig A,swine H1N1 specific Ig A,and swine H3N2 specific Ig A antibody concentrations of ILI nasal washes were lower than those of the nasal washes at the time of inclusion or follow-up.Phylogenetic analysis showed that the two H1N1 viruses isolated from the ILI cases and the SIVs in CAFOs had relatively high identity with the nucleotide and amino acid sequences of HA(99.17-100%/98.40-100%),NA(98.93-100%/98.28-100%),and M(99.59-100%/99.38-100%)gene segments and the key amino acid mutation sites were consistent,suggesting that SIVs cross-species transmission between swine and human had occurred.Conclusions(1)SIVs were prevalent in CAFOs,in which various subtypes of IAVs such as Eurasian avian-like H1N1,pdm09 H1N1,swine H1N2 and swine H3N2 were cocirculating.The Eurasian avian-like H1N1 subtype influenza virus gradually became the predominant lineage.Reassortment of gene segments within the same subtype or among different subtypes was common.(2)Swine-exposed population had higher serum MN antibody positive rate,seroconversion rate and seroconverted density against swine H1N1 virus than the controls,suggesting that swine-exposed subjects were more likely to infect swine H1N1 influenza virus.(3)Swine-exposed population had higher total Ig A,swine H1N1 specific Ig A and swine H3N2 specific Ig A antibody concentrations in nasal washes than controls,suggesting that Ig A may be a biomarker of SIVs infection.(4)ILI serum had at least a 4-fold rise in MN titers against a swine virus and ILI virus isolates had highly identity with influenza viruses in swine,suggesting that SIVs cross-species transmission between the humans and pigs had occured.Significants and innovationThis study was unique for its prospective One Health approach of concomitantly studying swine workers,pigs,and swine farm environment.The systematic and prospective surveillance found that SIVs were prevalent in CAFOs with diverse genotypes and widespread reassortment.IAV detected in BA specimens from CAFOs would be helpfulled to understand the transmission of SIVs.Serological study suggested that swine-exposed population were more likely to get infection of swine H1N1 influenza virus.We firstly detected Ig A antibody concentrations in nasal washes,and the results showed that Ig A may be a biomarker of SIVs infection.The evidence of cross-species transmission for influenza virus was found in the study. |
PDF Full Text Request