Serologic Evidence of Avian Metapneumovirus Infection Among Adults Occupationally Exposed to Turkeys

Abstract

Genetically similar, the avian metapneumovirus (aMPV) and the human MPV (hMPV) are the only viruses in the Metapneumovirus genus. Previous research demonstrated the ability of hMPV to cause clinical disease in turkeys. In this controlled, cross-sectional, seroepidemiological study, we examined the hypothesis that aMPV might infect humans. We enrolled 95 adults occupationally exposed to turkeys and 82 nonexposed controls. Sera from study participants were examined for antibodies against aMPV and hMPV. Both in bivariate (OR=3.2; 95% CI: 1.1–9.2) and in multivariate modelling adjusting for antibody to hMPV (OR=4.1; 95% CI: 1.3–13.1), meat-processing workers were found to have an increased odds of previous infection with aMPV compared to controls. While hMPV antibody cross-reactivity is evident, these data suggest that occupational exposure to turkeys is a risk factor for human infection with aMPV. More studies are needed to validate these findings, to identify modes of aMPV transmission, and to determine risk factors associated with infection.

Introduction

B uys and Du Preez first isolated the avian metapneumovirus (aMPV) from turkeys in South Africa in the 1970s (Njenga et al. 2003). Thereafter, the virus was isolated in Europe, Africa, the Middle East, South America, and, most recently, the United States. This virus causes an acute, highly contagious infection of the upper respiratory tract in turkeys (Njenga et al. 2003). Four subgroups of aMPV have been identified and an effective live-attenuated vaccine is available and is normally administered through the turkeys' drinking water (Gulati et al. 2001, Patnayak et al. 2002). aMPV type C (aMPV/C) is unique to North America and genetically different from the European strains (Seal 2000). It was first reported in the United States in 1996.

The human MPV (hMPV) was first isolated in the Netherlands in 2001 (van den Hoogen et al. 2001). Since then, it has been identified in numerous countries throughout the world, including North America (Stockton et al. 2002, Bastien et al. 2003, Boivin et al. 2003, Ebihara et al. 2003, Falsey et al. 2003, Hamelin et al. 2004, Ludewick et al. 2005, Williams et al. 2005a, 2005b). The virus is now recognized as an important etiologic agent of human acute respiratory tract illnesses that manifests with influenza-like symptoms, often accompanied by high fever, myalgia, bronchiolitis, and pneumonia. Children under 5 years of age are the most susceptible to infections, and it is among this group that the virus was initially isolated (van den Hoogen et al. 2001, Heikkinen et al. 2008, Mansbach et al. 2008, Regamey et al. 2008, Von Linstow et al. 2008). Further studies have demonstrated the infectivity of hMPV in other population groups, including immunocompromised individuals, the elderly, and young adults (O'gorman et al. 2006, Boivin et al. 2007, Johnstone et al. 2008, Ljubin Sternak et al. 2006, Louie et al. 2007). No vaccine is available against hMPV, though animal studies investigating a live-attenuated vaccine model are showing promise (Herfst and Fouchier 2008).

Molecular comparison of the genomic sequences of hMPV and aMPV/C reveals a close relationship (van den Hoogen et al. 2002, Njenga et al. 2003, Yunus et al. 2003). In fact, aMPV/C is more closely related to hMPV than to other aMPV subgroups. Research shows that there is an overall 80% shared amino acid identity between hMPV and aMPV/C (van den Hoogen et al. 2002). Recently, researchers at the University of Minnesota conducted a controlled experiment, in which 80 two-week-old turkey poults were inoculated with one of four different genotypes of hMPV (Velayudhan et al. 2006). All four genotypes caused transient, clinically evident disease among the birds. Results from this study demonstrated that hMPV can cause clinical disease in turkeys.

In this study, we investigated the hypothesis that aMPV could infect humans. Through a cross-sectional serosurvey, we examined U.S. turkey-exposed workers for evidence of previous aMPV infection.

Materials and Methods

Subjects

During March 2007 to April 2008, we enrolled 57 turkey growers, 38 turkey-processing plant workers, and 82 controls with no turkey exposure. Study participants were asked to complete a questionnaire that included demographic, occupational, and general health questions. We asked the turkey growers about their use of the live-attenuated aMPV vaccine for their turkeys. All participants completed the study questionnaire and serum was successfully obtained from 170 of 177 participants (96%). This study was approved by an institutional review board, and all subjects signed informed consent documents.

aMPV enzyme immunosorbent assay

The assay was performed as previously described with minor modifications (Chiang et al. 2000, Goyal et al. 2003). The aMPV isolate APV/Minnesota/Turkey/2a/1997 was used as antigen. Sera were heat inactivated at 56°C for 30 min and diluted 1:40 in blocking buffer (0.01M PBS containing 0.1% Tween 20 and 1% BSA). Diluted sera were added to four virus-coated wells and four negative antigen-coated wells at 50 μL/well, and the plates were incubated at room temperature for 1 h. Each assay included turkey aMPV antiserum and uninfected turkey serum as controls. After a 3-times wash, 50 μL of goat anti-human horseradish peroxidase (HRP) conjugate, reactive to heavy and light chains of human IgG (Millipore, Billerica, MA), was diluted 1:5000 in blocking buffer and added to wells containing human sera, whereas 50 μL of goat anti-turkey HRP conjugate, reactive to heavy and light chains of turkey IgG (KPL, Gaithersburg, MA), diluted 1:1500 was added to wells containing turkey serum controls. After 1-hour incubation, the plates were washed five times and 100 μL of substrate solution was added. After the substrate was added, the plates were incubated in the dark at room temperature for 10 min and the reaction was stopped with 25 μL of 2.5M sulfuric acid. The optical density (OD) was read at 490/405 nm dual wavelengths using an automated plate reader (VERSAmax; Molecular Devices, Sunnyvale, CA). Sera were considered positive when the average OD reading in the viral antigen wells was two standard deviations higher than the OD in the negative control antigen wells (Effendy et al. 2005).

aMPV Western blot assay

The aMPV isolate APV/Minnesota/Turkey/2a/1997 was propagated in Vero cells for 4 days. The supernatant was then centrifuged at 6000 rpm for 30 min and then concentrated by overlaying it over 12.5 mL of 25% sucrose and ultra-centrifuged at 24,000 rpm for 1 h. The virus pellet was resuspended in 200 μL of sterile PBS. The protein concentration was 2300 μg/mL as determined by a Quant-iT protein assay kit (Invitrogen, Carlsbad, CA). The virus was then treated with 2×sample buffer (Sigma-Aldrich, St. Louis, MO). The treated virus and Precision Plus protein ladder (Bio-rad, Hercules, CA) were then heated at 95°C for 5 min. Virus and ladder were loaded to Bio-rad's Ready Gels in 1×Tris-Glycine-SDS buffer (Sigma) and run at 90V for 10 min then 200V for 1 h. Transfer to nitrocellulose membrane (GE-Amersham, Piscataway, NJ) was performed on Bio-Rad Trans-Blot semidry transfer cell using 3MM blotting paper (Whatman, Kent, United Kingdom) and was run at 30V for 1 h. After transfer was complete, each blot was stained with Ponceau stain (Boston BioProducts, Ashland, MA) for both the observation of protein presence and as a measure of transfer efficiency. To each blot, 10 mL of 5% nonfat milk in PBS w/0.05% Tween (PBST) was added and blots were allowed to block at room temperature with shaking for 1 h. The milk block was then removed and serum samples diluted 1:50 in 10 mL 5% nonfat milk in PBST were added. Samples were allowed to incubate overnight at 4°C with shaking. Each blot was then washed twice with PBST. The secondary antibody, anti-human IgG HRP-conjugated (GE-Amersham), reactive against heavy and light chains of IgG, was diluted 1:2000 in 10 mL 5% milk in PBST, and was then added to each individual blot and shaken at room temperature for 1 h. After another wash cycle, blots were stained using peroxidase substrate (Vector Laboratories, Burlingame, CA) and were considered positive when bands appeared on the blot.

hMPV microneutralization assay

Antibody titers against hMPV were determined by a modified microneutralization assay (Skiadopoulos et al. 2004). The assay was conducted as specified, but the end titers were determined by using an enzyme immunosorbent assay (EIA) that was conducted according to the following protocol. The microneutralization plates were fixed with 80% acetone and washed four times with wash buffer and blotted dry. A mouse anti-hMPV antibody (Millipore, Chicago, IL) was diluted 1:1000 in a solution of 5% milk and 100 μL was transferred onto all the wells of the plates. The plates were incubated at room temperature for 1 h after which they were washed four times and 100 μL of HRP conjugated goat anti-mouse (KPL, Gaithersburg, MD) diluted 1:2000 was added. The plates were incubated at room temperature for 1 h and were then washed four times and 100 μL of tetramethylbenzidine peroxidase substrate 2 (TMB) (KPL) was added. Reaction with TMB was stopped by adding 1N sulfuric acid after a 20-min incubation in the dark at room temperature. Absorbance was read within 30 min at 450 nm wavelength using an automated plate reader (VERSAmax). Average absorbance was determined for quadruplicate wells of virus-infected (VC) and uninfected (CC) control wells. The serum titer result was considered positive at a 1:10 dilution if its absorbance value was less than X, where X=[(average absorbance of VC wells) + (average absorbance of CC wells)]/2. The virus back titer was run in quadruplicate and was accepted when it produced positive results in 5–7 wells containing the lowest dilution of test virus (i.e., positive in rows A-E, positive or negative in F and G, and negative in row H).

Statistical methods

Pearson's Chi square and Fisher's exact tests were used to compare categorical variables. Student's t-test was used to compare continuous variables. Odds ratios and associated confidence intervals were calculated by traditional method or by Fisher's exact method as appropriate. A manual, backward elimination logistic regression was then used to determine risk factors associated with positive titer. Analysis was performed using the SPSS 15.0 and WinPepi 1.8 software.

Results

Demographic characteristics of the study subjects are shown in Table 1. Turkey growers were older and more likely to be of the male sex than processing plant workers or controls. Most plant workers were Hispanic, whereas most growers were White non-Hispanic. Growers were also more likely to be current users of tobacco products. Processing plant workers reported more influenza-like illness in the previous year. There was no significant difference in the reporting of chronic conditions among all groups. The median hours of work per week for the turkey-processing plant workers was 40 h, whereas the farmers had a median work hours of 14 h.

Table 1.

Demographic Characteristics of the Study Subjects

	Controls	Meat-processing plant workers	Turkey growers
Variable	(n=82)	(n=38)	(n=57)	p-Value
Age
18–32	32 (39.0%)	14 (36.8%)	15 (26.3%)	0.001
33–44	28 (34.1%)	14 (36.8%)	10 (17.5%)
45–89	22 (26.8%)	10 (26.3%)	32 (56.1%)
Mean age (SD)	37.0 (13.0)	36.0 (10.5)	45.6 (15.3%)
Gender
Male	43 (52.4%)	23 (60.5%)	46 (80.7%)	<0.001
Female	39 (47.6%)	15 (39.5%)	11(19.3%)
Race/ethnicity
White, non-Hispanic	34 (41.5%)	1 (2.6%)	54 (94.7%)	<0.0005
Hispanic	43 (52.4%)	33 (86.8%)	3 (5.3%)
African American	2 (2.4%)	0 (0%)	0 (0%)
Asian	3 (3.7%)	4 (10.5%)	0 (0%)
Current user of tobacco
Yes	6 (7.3%)	11 (28.9%)	16 (28.1%)	<0.001
No	76 (92.7%)	27 (71.1%)	41 (71.9%)
Chronic disease
Yes	15 (18.3%)	8 (21.1%)	11 (19.3%)	NS
No	67 (81.7%)	30 (78.9%)	46 (80.7%)
Influenza-like illness
Yes	13 (15.9%)	10 (26.3%)	2 (3.5%)	0.015
No	69 (84.1%)	28 (73.7%)	55 (96.5%)

NS, not significant, p-value <0.05 was considered significant.

The seroprevalence of aMPV antibodies among turkey-processing plant workers was 86.5% as compared to 66.7% and 67.3% among controls and growers, respectively (Table 2). When compared to the controls, the plant workers had an OR of 3.2 (95% CI: 1.1–9.2) of being seropositive. To validate the results of the aMPV EIA, we selected 31 sera with positive or negative EIA results and conducted a western blot. We calculated the kappa statistic to be 0.844, indicating high agreement between the western blot and EIA results (Table 3).

Table 2.

Seropositivity for Antibodies Against the Avian Metapneumovirus and the Human Metapneumovirus Among Study Groups

Variable	Controls (n=82)	Meat-processing plant workers (n=38)	Turkey growers (n=57)	p-Value
aMPV antibody^a
Positive	52 (66.7%)	32 (86.5%)	37 (67.3%)	0.067
Negative	26 (33.3%)	5 (13.5%)	18 (32.7%)
hMPV Titer^b
Positive	76 (97.4%)	34 (97.1%)	53 (96.4%)	NS
Negative	2 (2.6%)	1 (2.9%)	2 (3.6%)

Antibody titers against the aMPV ≥1:40 were considered positive.

Antibody titers against the hMPV ≥1:10 were considered positive.

aMPV, avian metapneumovirus; hMPV, human MPV.

Table 3.

Agreement Between Enzyme Immunosorbent Assay and Western Blot Avian Metapneumovirus Assays

		Western Blot
		Positive	Negative	Total
Enzyme immunosorbent assay	Positive	20	3	23
	Negative	1	7	8
	Total	21	10	31

Due to potential cross-reactivity between avian and human, MPVs, we measured antibody titers against hMPV using a microneutralization assay. There was no significant difference in seropositivity against hMPV among all three exposure groups (Table 2). Only 8% of the growers reported using the live attenuated aMPV vaccine for their turkeys. Since the turkey farmers did not have a significantly higher seroprevalence than the controls, the effect of the vaccine on aMPV antibody seropositivity was not further analyzed.

Further analysis was conducted focusing on meat-processing plant workers as this group had the highest antibody titers compared to the control group. In bivariate analysis among meat-processing plant workers and nonexposed controls, being of the Hispanic ethnicity was significantly associated with elevated antibody titers against aMPV. Age was not associated with antibody titers against aMPV. A set of occupational risk factors such as the proper use of protective equipment and tasks performed at work were analyzed. None of these variables had a significant association with a positive antibody titer.

Multivariable logistic regression was used to control for potential confounders. Adjusting for antibody titers against hMPV, we detected an OR of 4.1 (95% CI: 1.3–13.1) for being seropositive for aMPV antibodies among turkey-processing plant workers compared to the unexposed controls (Table 4).

Table 4.

Odds Ratios for an Elevated Antibody Titer ^a Against Avian Metapneumovirus Among Meat-Processing Plant Workers Using Logistic Regression

Risk factor	Unadjusted OR (95% CI)	Saturated model OR (95% CI)	Final model OR (95% CI)
Exposure to turkeys
Processing plant worker	3.2 (1.1–9.2)^b	1.6 (0.4–6.7)	4.1 (1.3–13.1)^b
Control	ref	ref	ref
hMPV titer
≥1:10	6.3 (0.3–140.2)	5.3 (0.3–95.5)	5.9 (0.5–67.1)
<1:10	ref	ref	ref
Race/ethnicity
Other	9.9 (3.9–25.5)	10.7 (3.4–33.5)	NI
White	ref	ref
Gender
Male	1.4 (0.6–3.2)	1.6 (0.6–4.4)	NI
Female	ref	ref
Current smoker
Yes	1.9 (0.5–7.02)	0.5 (0.1–2.7)	NI
No	ref	ref
Influenza-like illness
Yes	1.3 (0.4–3.9)	1.6 (0.4–5.9)	NI
No	ref	ref
Age (continuous)	1.0 (0.97–1.04)	1.0 (0.95–1.0)	NI

Antibody titers against the aMPV ≥1:40 were considered positive.

Statistically significant.

NI, not included.

Discussion

To our knowledge, this is the only study that has investigated infection with aMPV among individuals exposed to turkeys. Evidence provided by this study suggests that occupational exposure to turkeys in meat-processing plants increases the risk of infection with aMPV. There, workers spend more time in contact with turkeys than turkey farmers, and this may explain the lack of evidence of infection among turkey growers. Workers have a wide spectrum of activities within the meat-processing plant. Exposure to live birds occurs when unloading turkeys from transport vehicles, handling birds, and cleaning and decontaminating trucks and cages where live birds have been contained. Workers are also exposed to bird carcasses and body fluids when the turkeys are butchered or when they clean and decontaminate working areas.

Elevated antibody titers against avian viruses among poultry meat packers and slaughterhouse workers have been previously reported. Netto and Johnson documented evidence of infection of poultry-processing plant workers with avian leukosis/sarcoma virus and reticuloendotheliosis virus (Johnson 1994, Netto and Johnson 2003). Choudat et al. (1996) reported an OR of 4.54 (95% CI: 2.27–9.10) for serologic evidence of previous infection with the avian Marek disease virus among poultry-processing plant workers as compared to nonexposed controls. Similar to aMPV, avian leukosis/sarcoma virus, reticuloendotheliosis virus, and Marek disease virus are transmitted among birds by airborne particulates or direct contact. Our findings contribute to the increasing evidence of zoonotic transmission of avian viruses to processing plant workers.

We detected a high seroprevalence of antibodies against hMPV in our study population. Ljubin-Sternak et al. reported an overall prevalence of 77.4% in a Croatian patient population (Ljubin Sternak et al. 2006). This prevalence was 100% among those older than 20 years. Similarly, Ebihara et al. (2003) detected a 72.5% seroprevalence among a general Japanese population aged between 1 month and 36 years. In this study, the researchers used a virus isolate from the Netherlands as antigen in their serologic assays; however, the prevalence may have been higher had they used an isolate endemic to Japan. In our study, we used an hMPV type A isolate that is endemic to the geographic area where the study was conducted and prevalent in other parts of North America, including Mexico (Noyola et al. 2005).

The aMPV/C and hMPV are closely related as they share at least 80% of amino acids and have similar genomes (van den Hoogen et al. 2002, Yunus et al. 2003). The possible cross-reactivity between antibodies against hMPV and aMPV was given considerable attention. This cross-reactivity may explain the high percent of positive individuals in the control group but does not nullify the fact that exposed individuals had significantly higher seropositivity to aMPV. We validated our aMPV EIA by comparing it to a western blot assay. The results of this comparison showed that the EIA had high agreement with western blot. Further, we controlled for the potential confounding effects of exposure to hMPV in our statistical analysis. We were not able to provide proof of aMPV infection among the turkey farms in our study areas at the time of enrollment. However, there is evidence that aMPV circulates in the American Midwest (Bennett et al. 2004).

None of the measured risk or protective factors was associated with an elevated antibody titer against aMPV. This could be due to the small sample size of processing plant workers or to self-selection of a specific homogeneous category of these workers into the study. In conclusion, our study suggests that occupational exposure to turkeys in processing plants is a risk factor for infection with aMPV. These findings support similar evidence of zoonotic transmission of avian viruses in similar occupational groups. More studies are needed to better understand the modes of transmission of these viruses and their associated risk factors.

Footnotes

Acknowledgments

We especially thank Binu T. Velayudhan, Ph.D., Virginia Tech, Blacksburg, VA, and Sharon F. Setterquist formerly of the University of Iowa's Center for Emerging Infectious Diseases for their advice and assistance in the adaptation of serologic techniques. We also thank Ana W. Capuano, M.P.S. M.S., for her initial statistical demonstration of association. This work was made possible in part by a grant from the University of Iowa Student Government and by the American Lebanese Syrian Associated Charities.

Disclosure Statement

No competing financial interests exist.

References

Bastien

, Ward

, Van Caeseele

et al. Human metapneumovirus infection in the Canadian population. J Clin Microbiol, 2003; 41:4642–4646.

Bennett

, Nezworski

, Velayudhan

et al. Evidence of avian pneumovirus spread beyond Minnesota among wild and domestic birds in central North America. Avian Dis, 2004; 48:902–908.

Boivin

, De Serres

, Cote

et al. Human metapneumovirus infections in hospitalized children. Emerg Infect Dis, 2003; 9:634–640.

Boivin

, De Serres

, Hamelin

et al. An outbreak of severe respiratory tract infection due to human metapneumovirus in a long-term care facility. Clin Infect Dis, 2007; 44:1152–1158.

Chiang

, Dar

, Goyal

et al. A modified enzyme-linked immunosorbent assay for the detection of avian pneumovirus antibodies. J Vet Diagn Invest, 2000; 12:381–384.

Choudat

, Dambrine

, Delemotte

et al. Occupational exposure to poultry and prevalence of antibodies against Marek's disease virus and avian leukosis retroviruses. Occup Environ Med, 1996; 53:403–410.

Ebihara

, Endo

, Kikuta

et al. Seroprevalence of human metapneumovirus in Japan. J Med Virol, 2003; 70:281–283.

Effendy

, Tambunan

, Reksodiputro

. Cut-off levels based on deviation from standard negative control is better than moderate level based on fixed cut-off for ACA assessed using ELISA for the diagnosis of antiphospholipid syndrome. Acta Med Indones, 2005; 37:12–19.

Falsey

, Erdman

, Anderson

et al. Human metapneumovirus infections in young and elderly adults. J Infect Dis, 2003; 187:785–790.

10.

Goyal

, Lauer

, Friendshuh

et al. Seroprevalence of avian pneumovirus in Minnesota turkeys. Avian Dis, 2003; 47:700–706.

11.

Gulati

, Patnayak

, Sheikh

et al. Protective efficacy of high-passage avian pneumovirus (APV/MN/turkey/1-a/97) in turkeys. Avian Dis, 2001; 45:593–597.

12.

Hamelin

, Abed

, Boivin

. Human metapneumovirus: a new player among respiratory viruses. Clin Infect Dis, 2004; 38:983–990.

13.

Heikkinen

, Osterback

, Peltola

et al. Human metapneumovirus infections in children. Emerg Infect Dis, 2008; 14:101–106.

14.

Herfst

, Fouchier

. Vaccination approaches to combat human metapneumovirus lower respiratory tract infections. J Clin Virol, 2008; 41:49–52.

15.

Johnson

. Poultry oncogenic retroviruses and humans. Cancer Detect Prev, 1994; 18:9–30.

16.

Johnstone

, Majumdar

, Fox

et al. Human metapneumovirus pneumonia in adults: results of a prospective study. Clin Infect Dis, 2008; 46:571–574.

17.

Ljubin Sternak

, Vilibic Cavlek

, Falsey

et al. Serosurvey of human metapneumovirus infection in Croatia. Croat Med J, 2006; 47:878–881.

18.

Louie

, Schnurr

, Pan

et al. A summer outbreak of human metapneumovirus infection in a long-term-care facility. J Infect Dis, 2007; 196:705–708.

19.

Ludewick

, Abed

, van Niekerk

et al. Human metapneumovirus genetic variability, South Africa. Emerg Infect Dis, 2005; 11:1074–1078.

20.

Mansbach

, McAdam

, Clark

et al. Prospective multicenter study of the viral etiology of bronchiolitis in the emergency department. Acad Emerg Med, 2008; 15:111–118.

21.

Netto

, Johnson

. Mortality in workers in poultry slaughtering/processing plants: the Missouri poultry cohort study. Occup Environ Med, 2003; 60:784–788.

22.

Njenga

, Lwamba

, Seal

. Metapneumoviruses in birds and humans. Virus Res, 2003; 91:163–169.

23.

Noyola

, Alpuche-Solis

, Herrera-Diaz

et al. Human metapneumovirus infections in Mexico: epidemiological and clinical characteristics. J Med Microbiol, 2005; 54:969–974.

24.

O'Gorman

, McHenry

, Coyle

. Human metapneumovirus in adults: a short case series. Eur J Clin Microbiol Infect Dis, 2006; 25:190–192.

25.

Patnayak

, Sheikh

, Gulati

et al. Experimental and field evaluation of a live vaccine against avian pneumovirus. Avian Pathol, 2002; 31:377–382.

26.

Regamey

, Kaiser

, Roiha

et al. Viral etiology of acute respiratory infections with cough in infancy: a community-based birth cohort study. Pediatr Infect Dis J, 2008; 27:100–105.

27.

Seal

. Avian pneumoviruses and emergence of a new type in the United States of America. Anim Health Res Rev, 2000; 1:67–72.

28.

Skiadopoulos

, Biacchesi

, Buchholz

et al. The two major human metapneumovirus genetic lineages are highly related antigenically, and the fusion (F) protein is a major contributor to this antigenic relatedness. J Virol, 2004; 78:6927–6937.

29.

Stockton

, Stephenson

, Fleming

et al. Human metapneumovirus as a cause of community-acquired respiratory illness. Emerg Infect Dis, 2002; 8:897–901.

30.

van den Hoogen

, Bestebroer

, Osterhaus

et al. Analysis of the genomic sequence of a human metapneumovirus. Virology, 2002; 295:119–132.

31.

van den Hoogen

, de Jong

, Groen

et al. A newly discovered human pneumovirus isolated from young children with respiratory tract disease. Nat Med, 2001; 7:719–724.

32.

Velayudhan

, Nagaraja

, Thachil

et al. Human metapneumovirus in turkey poults. Emerg Infect Dis, 2006; 12:1853–1859.

33.

von Linstow

, Hogh

, Nordbo

et al. A community study of clinical traits and risk factors for human metapneumovirus and respiratory syncytial virus infection during the first year of life. Eur J Pediatr, 2008; 165:1125–1133.

34.

Williams, JV, Crowe, JE, Jr., Enriquez, R, et al. Human metapneumovirus infection plays an etiologic role in acute asthma exacerbations requiring hospitalization in adults. J Infect Dis, 2005a. 192:1149–1153.

35.

Williams

, Martino

, Rabella

et al. A prospective study comparing human metapneumovirus with other respiratory viruses in adults with hematologic malignancies and respiratory tract infections. J Infect Dis, 2005b. 192:1061–1065.

36.

Yunus

, Govindarajan

, Huang

et al. Deduced amino acid sequence of the small hydrophobic protein of US avian pneumovirus has greater identity with that of human metapneumovirus than those of non-US avian pneumoviruses. Virus Res, 2003; 93:91–97.