Induction of Cross-Neutralizing Antibody Against H5N1 Virus After Vaccination with Seasonal Influenza Vaccine in COPD Patients

Abstract

Archival serum samples from elderly individuals with underlying chronic obstructive pulmonary disease (COPD) who were enrolled in a double-blind case-control study of seasonal influenza vaccine efficacy were assayed for cross-neutralizing antibody formation to avian influenza A (H5N1) virus. Of 118 serum samples, 58 were collected from influenza vaccinees (mean age 68.5 y), and 60 from placebo controls (mean age 68.4 y) who received vitamin B injections. Blood samples were collected before and at 1 mo after seasonal influenza vaccination from all subjects; in addition, for a longitudinal follow-up period of 1 y paired-blood samples were collected again from subjects who developed acute respiratory illness. Hemagglutination inhibition assay for antibodies to influenza A (H1N1), influenza A (H3N2), and influenza B viruses was carried out to determine the serological response to vaccination, and to diagnose influenza viral infection, while microneutralization assays were performed to detect cross-reactive antibody to H5N1 virus. Pre-existing cross-reactive H5N1 antibody at reciprocal titer 10 was found in 6 (10.3%) vaccinees and 4 (6.7%) placebo controls. There was no change in H5N1 antibody titer in these subjects after vaccination. On the other hand, 3 (5.2%) vaccinees developed seroconversion to H5N1 virus at 1 mo after vaccination, even though they had no pre-existing H5N1 antibody in their first blood samples. No cross-neutralizing antibody to H5N1 virus was detected in the placebo controls or in the 22 influenza patients, suggesting that influenza vaccination, but not influenza virus infection, induces cross-neutralizing antibody against avian influenza H5N1 virus.

Introduction

D irect transmission of highly pathogenic avian influenza (HPAI) H5N1 virus from poultry to humans was first reported in Hong Kong in 1997, and it caused 18 cases with 6 deaths, for a 33% fatality rate (3). Its re-emergence since 2003 has spread across continents, producing a higher fatality rate of about 60% in infected humans (24). Thailand reported the first human case on January 23, 2004, and the last case occurred in July 2006 (9). Nevertheless, human cases of avian influenza were seen in the People's Republic of China, Vietnam, Indonesia, and Egypt in 2009 (24). The elderly succumbed to avian influenza to a lesser extent than young children, which may be a result of cross-immunity elicited by repeated infection with other human influenza A virus subtypes (26).

A few pieces of evidence have suggested that heterosubtyping immunity may be able to confer protection across influenza subtypes, but the mechanism of this cross-protection, if it is mediated by cellular or humoral immunity, and from which antigenic domain, is not well defined (12,13). Theoretically, cross-immunity can be induced either by hemagglutinin (HA) or neuraminidase (NA) antigen. The hypothesis of cross-protection mediated by NA immunity was derived from the observation that the 1968 pandemic influenza caused by the H3N2 virus, which came after the 1957 pandemic caused by the H2N2 virus, produced fewer deaths (8,21). Increasing levels of anti-NA antibody have been associated with decreasing frequency of viral infection and suppression of clinical manifestations (11). In an animal model, mice immunized with DNA vaccine containing a human N1 NA gene survived lethal challenge with H5N1 virus (18). On the other hand, HA antigen might also play a role in heterosubtypic immunity. In a previous study researchers reported that children with primary infection by H1N1 or H3N2 virus developed HA antibody to the contemporary subtypes, as well as cross-reactive antibody to H8 as assayed by ELISA (1).

During 1997 and 1998, our group conducted a double-blind controlled trial to demonstrate the efficacy of human influenza vaccination in patients with chronic obstructive pulmonary disease (COPD) aged >60 y (10,22). Serum samples were collected from these subjects before and at 1 mo after vaccination to determine the serological response to influenza vaccine. These vaccinees were longitudinally followed up for a year, and paired-blood samples were collected from the individuals who developed acute respiratory illness (ARI) for serodiagnosis of influenza by hemagglutination inhibition (HI) assay. In the present study we aimed to detect neutralizing antibody against HPAI H5N1 virus in the aforementioned serum samples by microneutralization (microNT) assay. In this study we demonstrated that the elderly subjects developed cross-neutralizing H5N1 antibody as the result of vaccination, but not from influenza illness. Since our archival serum samples were collected before the occurrence of the avian influenza outbreak, H5N1 infection in the subjects was excluded.

Materials and Methods

Subjects

A total of 118 COPD patients who attended the COPD clinic at Siriraj Hospital, Bangkok, Thailand, were enrolled in a double-blind case-control study to evaluate the efficacy of seasonal influenza vaccine between 1997 and 1998. Fifty-eight subjects (mean age 68.5 y) received vaccine, whereas 60 subjects (mean age 68.4 y) received vitamin B injection as the placebo control. Venous blood samples (10 mL) were collected from each subject prior to vaccination or vitamin B injection; the second blood samples were collected 1 mo thereafter to determine the serological response to vaccination. No subject developed influenza-like illness during this 1-mo period. Furthermore, the subjects were longitudinally followed-up for 1 y for ARI. In those cases, paired-blood samples at 4- to 6-wk intervals were collected for serodiagnosis of influenza by HI assay using the vaccine strains as the test antigens. Serum samples were kept frozen at −20°C until testing. The study was approved by the Institutional Review Board of the Committee on Ethics, Faculty of Medicine, Siriraj Hospital, Mahidol University.

Vaccines

The seasonal influenza vaccine used in this study was the split-typed trivalent vaccine from Pasteur Merieux, Lyon, France. A 0.5-mL dose of vaccine contained influenza A/Texas/36/91 (H1N1), A/Nanchang/933/95 (H3N2), and B/Harbin/07/94, at concentration of 15 μg of HA for each virus. A 0.5-mL dose of vitamin B₁ was used as placebo.

Hemagglutination inhibition test

The procedure for the HI assay was as described elsewhere (10,23). Briefly, non-specific inhibitor in the test sera was eliminated by treatment with receptor-destroying enzyme from Vibrio cholerae (Denka Seiken, Niigata, Japan) overnight at 4°C, followed by inactivation at 56°C for 30 min. Then non-specific agglutinator was removed by absorption with 50% chick red blood cells. The test antigen panel, influenza A/H1N1, influenza A/H3N2, and influenza B viruses, was kindly provided by the World Health Organization (WHO). Serum with HI antibody titer ≥1:10 was considered positive for influenza antibody. A positive seroresponse to influenza vaccine was obtained when paired-blood samples collected before and after vaccination showed a fourfold or greater rise in HI antibody titer against any of the test antigens. Similarly, influenza diagnosis by HI assay was given based on the same criteria.

Microneutralization assay

The ELISA-based microNT assay protocol was based on that described in Kitphati et al. (9) and the WHO manual (23). The experiments were conducted in a laboratory with biosafety level 3. Influenza A/Thailand/1 (KAN-1)/2004 (H5N1) was used as the test virus. Briefly, a serum sample was twofold serially diluted from the dilution of 10 to 1280; then 60 μL of each serum dilution was mixed with an equal volume of the test virus at a concentration of 200 TCID₅₀/100 μL and incubated at 37°C for 2 h. A 100-μL volume of the virus-serum mixture was inoculated onto a Madin-Darby canine kidney (MDCK) cell monolayer in a well of a microtiter plate and further incubated for 18–20 h in a CO₂ incubator. The assays were run in duplicate. The reaction plate was tested by ELISA to determine the amount of influenza nucleoprotein produced in the infected MDCK cells. Mouse monoclonal antibody (Chemicon International, Temecula, CA) was used as the primary antibody, anti-mouse Ig conjugated with horseradish peroxidase (Southern Biotechnology Associates, Birmingham, AL) was used as the second antibody, and TMB (KPL Inc., Gaithersburg, MD) was used as the chromogenic substrate. The reaction plate was read under a spectrophotometer at the wavelengths of 450 and 630 nm. Wells with uninfected cells and virus back-titration were included as the control sets in every test plate. The serum dilution that reduced more than 50% of the amount of viral nucleoprotein compared to the virus control was considered to be positive for anti-H5N1 antibody. The cut-off titer for positive H5N1 antibody was set at 10.

Statistical analysis

Data were analyzed by SPSS version 11.5 software (SPSS, Inc., Chicago, IL). Comparison of geometric mean titer (GMT) values and fold increase in titers between the two groups was performed by the Mann-Whitney U test. Comparison of prevalence between two groups was performed by chi-square testing. The level of statistical significance was set at p < 0.05.

Results

H5N1-neutralizing antibody after seasonal influenza vaccination

Elderly subjects with COPD were immunologically intact, as shown by the fact that 85% of the vaccinees, but not the placebo controls, developed a fourfold or greater rise in HI antibody titer to the influenza A/H1N1 and H3N2 vaccine strains (Table 1). GMTs of H5N1-neutralizing antibodies prior to vaccination in both groups were not statistically significantly different (5.37 versus 5.24; p = 0.558). Meanwhile, pre-existing H5N1-neutralizing antibody at titer 10 was found in 6 (10.3%) of 58 vaccinees, and in 4 (6.7%) of 60 placebo controls (Table 2). The numbers of subjects who had pre-existing H5N1 antibody in the two groups were also not statistically significantly different (p = 0.637). After influenza vaccination, 3 (5.2%) vaccinees who had no pre-existing H5N1 antibody in their initial blood samples did develop a fourfold or greater rise in H5N1-neutralizing antibody titers. The increase in GMT of H5N1 antibody in the vaccine group was not statistically significantly different (5.37 versus 6.05; p = 0.10). However, a slight increase in GMT was observed in the vaccine group, but not in the placebo group (Table 2). Two of these three subjects seroconverted to both H1N1 and H3N2, while the third one seroconverted to H1N1 virus only (Table 3).

Table 1.

Distribution of Antibody Response to H1N1 and H3N2 Vaccine Strains in Paired-Blood Samples from Vaccine and Placebo Groups

			Antibody titer
Virus	Group	Vaccination	<10	10	20	40	80	160	320	640	1280	% Vaccine responders ^a
H1N1	Vaccine	Pre	33	5	9	8	2	0	1	0	0
	(n = 58)	Post	3	3	8	5	10	9	7	4	9	85
	Placebo	Pre	24	12	8	10	4	1	0	1	0
	(n = 60)	Post	23	14	7	11	2	2	0	1	0	0
H3N2	Vaccine	Pre	27	10	5	7	5	2	1	1	0
	(n = 58)	Post	3	1	7	7	10	13	4	2	11	85
	Placebo	Pre	28	8	7	4	5	4	1	0	3
	(n = 60)	Post	28	8	7	4	6	4	0	0	3	0

A responder is defined by at least a fourfold rise in antibody titers in the paired serum samples.

Table 2.

Distribution of H5N1 NT Antibodies in Paired-Blood Samples from Vaccine and Placebo Groups

		No. of positive samples at NT antibody titers
Group		<10	10	20	40	80	GMT	Number of subjects with pre-existing antibody
Vaccine	B1	52	6	0	0	0	5.37^a,c	6^b
(n = 58)	B2	49	6	0	2^d	1^d	6.05^c
Placebo	B1	56	4	0	0	0	5.24^a	4^b
(n = 60)	B2	56	4	0	0	0	5.24

p = 0.558 for the comparison of pre-vaccination GMT.

p = 0.637 for the comparison of frequencies of subjects with pre-exiting H5N1 antibodies.

p = 0.100 for comparison of GMT between pre- and post-vaccination in the vaccine group.

Without acute respiratory infection.

Abbreviations: B1, pre-vaccination; B2, post-vaccination; GMT, geometric mean titer; NT, neutralizing antibody.

Table 3.

Antibody Responses to H1N1, H3N2, and B Viruses in Vaccinees Who Developed A Fourfold Rise in Antibody to H5N1 Virus

			Antibody titers to influenza viruses
Code No.	Group	Antibody to subtype	Pre-vaccination (B1)	Post-vaccination (B2)
15	V	H5N1 NT	<10	80
		H1N1 HI	<10	40
		H3N2 HI	<10	160
		B HI	<10	20
29	V	H5N1 NT	<10	40
		H1N1 HI	<10	160
		H3N2 HI	<10	320
		B HI	<10	<10
37	V	H5N1 NT	<10	40
		H1N1 HI	10	640
		H3N2 HI	160	160
		B HI	20	80

Abbreviations: NT, neutralizing antibody; HI, hemagglutination inhibition antibody; V, vaccination; B1, pre-vaccination; B2, post-vaccination; H5N1, influenza A/Thailand/1 (KAN-1)/2004; H1N1, influenza A/Texas/36/91; H3N2, influenza A/Nanchang/933/95; B, influenza B/Harbin/07/94.

Lack of H5N1-neutralizing antibody after natural influenza illness

During 1 y of longitudinal follow-up, 102 episodes of ARI occurred among vaccinees and placebo controls. Influenza was diagnosed in 5 (8.6%) vaccinees and 17 (28.3%) placebo controls (Table 4). Nevertheless, no increase in cross-reactive neutralizing H5N1 antibody was detected in these influenza patients. There was one placebo control that had an H5N1 antibody titer of 20 in paired sera. The results demonstrated that acute influenza virus infection did not elicit detectable cross-H5N1 antibody in the study subjects.

Table 4.

Distribution of H5N1 NT Antibodies in Paired-Blood Samples from Patients with Acute Respiratory Illness

			No. of positive samples at NT antibody titers
Illness	Group (no. episodes/no. case)	Sample	<10	10	20	40	80	GMT
Flu	Vaccine	BA	5	0	0	0	0	5.00
	(5/5)	BC	5	0	0	0	0	5.00
	Placebo	BA	16	1	0	0	0	5.21
	(17/17)	BC	16	1	0	0	0	5.21
	Total	BA	21	1	0	0	0	5.16
	(22/22)	BC	21	1	0	0	0	5.16^a
Non-Flu	Vaccine	BA	34	4	1	0	0	5.56
	(39/35)	BC	34	4	1	0	0	5.56
	Placebo	BA	40	1	0	0	0	5.09
	(41/33)	BC	40	1	0	0	0	5.09
	Total	BA	74	5	1	0	0	5.31
	(80/68)	BC	74	5	1	0	0	5.31^a

p = 0.952 for comparison of GMT between influenza and non-influenza patient groups.

Abbreviations: BA, acute blood sample; BC, convalescent blood sample; GMT, geometric mean titer; NT, neutralizing antibody.

Discussion

In the present study we demonstrated pre-existing cross-neutralizing antibody against H5N1 virus in 10 (8.5%) of 118 subjects, as well as induction of a fourfold or greater rise of this cross-reactive antibody in 5.2% of elderly individuals with underlying COPD who were vaccinated with seasonal influenza vaccine. The data on the induction of cross-neutralizing antibody against H5N1 virus by seasonal influenza vaccine are controversial. In our previous unpublished work, cross-neutralizing antibody against H5N1 virus was not detected in 42 young people aged 25–40 y with a history of seasonal influenza vaccination, and also not in 140 healthy subjects aged 20–45 y, of which approximately 70% had neutralizing antibody to A/New Caledonia/20/99 (H1N1) and A/Fujian/411/02 (H3N2). Gioia et al. (4) demonstrated a rise in H5N1-neutralizing antibody titer to >20-fold over baseline in 13 (34.2%) of 38 vaccinees (aged 27–59 y, with an average age of 43 y), while their H5N1 HI antibody titers remained at undetectable levels. In contrast, Tang et al. (19) did not detect cross-reactive antibody to influenza H5N1 virus, either by microNT or HI assay, in serum samples collected at intervals from 10 vaccinees (aged 20–40 y) who received seasonal influenza vaccine, even though they could elicit a serological response with high HI antibody titers to the vaccine strains. During the 1997 avian influenza outbreak, Rowe et al. (17) found that an H5N1-neutralizing antibody titer of 80 was suggestive of avian influenza infection. However, this appeared to apply only to people aged <50 y. Moreover, the H5N1 microNT assay may be less specific in adults aged 60 y and older (17). Results of that study and ours suggest that elderly persons may have cross-reactive antibody to avian H5N1 virus as a result of seasonal influenza vaccination or repeated infection by human influenza viruses. The inducible H5N1 antibody may be the result of immunologic priming that occurred in children infected with seasonal influenza viruses bearing common epitopes to the H5N1 strain, as previously reported (16). Preschool-age children were more susceptible due to their lack of immunity against influenza virus infections (2). During the 2003–2004 influenza season approximately 43% of cases occurred in 10-year-old children, while only 12% occurred in those aged 65 and older (14). The attack rate of seasonal influenza is approximately 5–10% per year (25). Therefore elderly persons have most likely experienced natural influenza virus infections several times in their lives. Data from the current outbreak also indirectly support the theory that the elderly, who may have experienced influenza disease several times in their lives, died from H5N1 infection at lower rate than younger people (26).

Based on the in-vitro microNT assay results, it is possible that the H5N1-neutralizing antibodies detected may be induced by the common epitopes present in HA proteins, but less likely to those in NA proteins. It is generally accepted that antibody to HA blocks viral entry and confers neutralizing activity and protective immunity, whereas antibody to NA blocks virus release and confers only partial protective immunity. Cross-neutralizing antibodies induced by common epitopes in HA from different influenza subtypes has been reported. Monoclonal antibodies against conserved antigenic sites on H1 or H2 could neutralize both the H1 and H2 subtypes (15). Using the phage display library technique together with panning by recombinant H5 HA, two sets of researchers showed that gene pools from peripheral blood mononuclear cells of healthy subjects could produce heterosubtypic monoclonal antibodies directed against epitopes in the HA2 domain of the HA molecule that were able to neutralize HPAI H5N1 virus, as well as other influenza virus subtypes (5,6). Although seasonal influenza vaccine induced a fourfold or greater rise in cross-neutralizing H5N1 antibody in the three vaccinees who did not have pre-existing H5N1 antibody in their initial blood samples, the vaccine failed to boost the anamnestic response in vaccinees who had pre-existing H5N1 antibody. It is unclear whether the population of memory lymphocytes and the common epitopes present in vaccines are well matched; perhaps pre-existing H5N1 antibody had bound the common epitopes in the vaccine and diminished its boosting effect.

Regarding serologic surveys for H5N1 virus infection in Hong Kong, Treanor et al. (7,20) reported that the an H5N1-neutralizing antibody titer of 40 was the primary immunogenic threshold to distinguish between infected and uninfected persons. Neutralizing H5N1 antibody titers of 40–80 were also found in our three vaccinees. Since this study was carried out on archival blood samples collected before the avian influenza outbreak, it appears that the cross-H5N1-neutralizing antibody that developed in our subjects was induced by vaccination, not by natural H5N1 infection.

Footnotes

Acknowledgments

This work was supported by grants from the National Research Council of Thailand, the Thailand Research Fund for Senior Research Scholars, and the Chalermphrakiat Grant, Faculty of Medicine, Siriraj Hospital, Mahidol University. The authors thank the WHO for kindly providing the influenza antigens, and special thanks are due to all of the subjects who participated in this study.

References

Burlington

, Wright

, van Wyke

, Phelan

, Mayner

. Development of subtype-specific and heterosubtypic antibodies to the influenza A virus hemagglutinin after primary infection in children. J Clin Microbiol, 1985; 21:847–849.

Brownstein

, Kleinman

, Mandl

. Identifying pediatric age groups for influenza vaccination using a real-time regional surveillance system. Am J Epidemiol, 2005; 162:686–693.

Claas

, Osterhaus

, van Beek

et al. Human influenza A H5N1 virus related to a highly pathogenic avian influenza virus. Lancet, 1998; 351:472–477.

Gioia

, Castilletti

, Tempestilli

et al. Cross-subtype immunity against avian influenza in persons recently vaccinated for influenza. Emerg Infect Dis, 2008; 14:121–128.

Gocnyk

, Fislova

, Mucha

, Sladkova

, Russ

, Kostolansky

, Vareckova

. Antibodies induced by the HA2 glycopolypeptide of influenza virus haemagglutinin improve recovery from influenza A virus infection. J Gen Virol, 2008; 89:958–967.

Gocnyk

, Fislova

, Sladkova

, Mucha

, Kostolansky

, Vareckova

. Antibodies specific to the HA2 glycopolypeptide of influenza A virus haemagglutinin with fusion-inhibition activity contribute to the protection of mice against lethal infection. J Gen Virol, 2007; 88:951–955.

Katz

, Lim

, Bridges

et al. Antibody response in individuals infected with avian influenza A (H5N1) viruses and detection of anti-H5 antibody among household and social contacts. J Infect Dis, 1999; 180:1763–1770.

Kilbourne

. Influenza pandemics of the 20th century. Emerg Infect Dis, 2006; 12:9–14.

Kitphati

, Pooruk

, Lerdsamran

et al. Kinetics and longevity of antibody response to influenza A H5N1 virus infection in humans. Clin Vaccine Immunol, 2009; 16:978–981.

10.

Kositanont

, Kanyok

, Wasi

, Wongsurakiat

, Suthamsmai

, Maranetra

. Occurrence and protective level of influenza infections using serology in patients with COPD in vaccination study. J Med Assoc Thai, 2004; 87:964–969.

11.

Monto

, Kendal

. Effect of neuraminidase antibody on Hong Kong influenza. Lancet, 1973; 1:623–625.

12.

Nguyen

, Moldoveanu

, Novak

et al. Heterosubtypic immunity to lethal influenza A virus infection is associated with virus specific CD8(+) cytotoxic T lymphocyte responses induced in mucosa-associated tissues. Virology, 1999; 254:50–60.

13.

Nguyen

, Ginkel van

, Vu

, McGhee

, Mestecky

. Heterosubtypic immunity to influenza A virus infection requires B cells but not CD8+ cytotoxic T lymphocytes. J Infect Dis, 2001; 183:368–376.

14.

North Dakota Department of Health, Division of Disease Control: 2003–2004 Influenza Season Summary. http://www.ndhealth.gov/disease/Documents/pump%20handle/01–09.pdf

15.

Okuno

, Isegawa

, Sasao

, Ueda

. A common neutralizing epitope conserved between the hemagglutinin of influenza A virus H1 and H2 strains. J Virol, 1993; 67:2552–2558.

16.

Remarque

, Bruijn de

, Boersma

, Masurel

, Ligthart

. Altered antibody response to influenza H1N1 vaccine in healthy elderly people as determined by HI, ELISA, and neutralization assay. J Med Virol, 1998; 55:82–87.

17.

Rowe

, Abernathy

, Hu-Primmer

et al. Detection of antibody to avian influenza A (H5N1) virus in human serum by using a combination of serologic assays. J Clin Microbiol, 1999; 37:937–943.

18.

Sandbulte

, Jimenez

, Boon

, Smith

, Treanor

, Webby

. Cross-reactive neuraminidase antibodies afford partial protection against H5N1 in mice and are present in unexposed humans. PLoS Med, 2007; 4:e59:265–272.

19.

Tang

, Karry

, Chan

PKS

. Lack of cross-immune reactivity against influenza H5N1 from seasonal influenza vaccine in humans. J Med Virol, 2008; 80:1992–1996.

20.

Treanor

, Campbell

, Zangwill

, Rowe

, Wolff

. Safety and immunogenicity of an inactivated subvirion influenza A (H5N1) vaccine. N Engl J Med, 2006; 354:1343–1351.

21.

Viboud

, Grais

, Lafont

, Miller

, Simonsen

. Multinational impact of the 1968 Hong Kong influenza pandemic: Evidence for a smoldering pandemic. J Infect Dis, 2005; 192:233–248.

22.

Wongsurakiat

, Maranetra

, Wasi

, Kositanont

, Dejsomritrutai

, Charoenratanakul

. Acute respiratory illness in patients with COPD and the effectiveness of influenza vaccination: a randomized controlled study. Chest, 2004; 125:2011–2020.

23.

World Health Organization. WHO manual on animal influenza diagnosis and surveillance. Geneva: World Health Organization, 2002(document WHO/CDS/CSR/NCS/2002.5)28–39.

24.

World Health Organization. Cumulative number of confirmed human cases of avian influenza A/(H5N1), 2009. http://www.who.int/csr/disease/avian_influenza/country/cases_table_2009_04_08/en/index.html

25.

World Health Organization. Influenza (seasonal) fact sheet No 211, 2009. http://www.who.int/mediacentre/factsheets/fs211/en/index.html

26.

, Gao

, Feng

et al. Clinical characteristics of 26 human cases of highly pathogenic avian influenza A (H5N1) virus infection in China. PLoS ONE, 2008; 3:e2985.