THE JOURi'JAL OF Ii'JFECTIOUS DISEASE. VOL. 136, SUPPLEMENT. DECEMBER 1977 © 1977 by the University of Chicago. All rights reserved.

Effect of Priming Infection on Serologic Response to Whole and Subunit Influenza Virus Vaccines in Animals Colin McLaren, Martha W. Verbonitz, Sylvester Daniel, George E. Grubbs, and Francis A. Ennis

From the Division of Virology, Bureau of Biologics, Food and Drug Administration, Bethesda, Maryland

characteristics of a secondary response [3] and indicates that the animals have been "primed" by the first infection. Further studies in primed ferrets and hamsters also showed a reduced response of these animals to split-product vaccines compared with that to whole-virus vaccine [4, 5]. Experiments were therefore designed to study the effect of priming infections with influenza A viruses representative of the different major strains (HswlNI, HON1, H2N2, H3N2) on the subsequent serologic responses of mice and hamsters to immunization with whole or subunit A/ NJ /76 (HswINI) influenza virus vaccines. Advantage was taken of the opportunity to use the vaccine preparations that had been administered to volunteers and had been well characterized in laboratory tests.

The results presented elsewhere in this issue demonstrated that the serologic responses of' humans to A/New Jersey (NJ)/76 virus vaccine varied according to the age of the vaccinee; after immunization higher titers of antibody were evident in the older groups. The increased responsiveness of persons older than 51 years to the vaccinesis probably the result of childhood infections with influenza A/swine-like viruses in these individuals and thus represents a secondary antibody response. Younger adults have had initial exposure to viruses that are less closely related antigenically to A/NJ /76 virus but also have had a smaller cumulative exposure to influenza A viruses. Enhanced responses to a vaccine may therefore stem either from a specific secondary response to a homologous virus or from a less well-defined "priming" phenomenon induced by one or more heterologous viruses. Studies in animals have shown that animals infected with one strain of influenza A virus show an enhanced serologic response to subsequent immunization with a heterologous strain of virus having different hemagglutinin and neuraminidase antigens [I, 2]. The antibody response to the second, heterologous virus has some of the

Materials and Methods

A nimals. Swiss mice were obtained from the Small Animal Section of the National Institutes of Health (Bethesda, Md.), Mice weighing 9-12 g or 23-25 g were used in separate experiments as described below. Syrian hamsters two to three months old were obtained from Charles River, Lakeview, N.]. Viruses and vaccines. The following strains of virus were grown in hens' eggs by standard procedures: A/swinejl976/31 (HswINl); A/

Please address requests for reprints to Dr. C. McLaren, Bureau of Biologics, Food and Drug Administration, 8800 Rockville Pike, Bethesda, Maryland 20014.

5706

Downloaded from http://jid.oxfordjournals.org/ at Indiana University Library on July 29, 2015

Results from studies in humans demonstrated that the serologic responses to Aj New Jersey (NJ)j76 virus vaccines varied according to the age of the vaccinee and appeared to be related to previous exposure of vaccinees to the different strains of influenza A virus. Experiments in animals were therefore performed to investigate the role of previous infection on responses to AjNJ /76 virus vaccines. Mice were infected with influenza A viruses representative of the major strains (HswINI, HONI, H2N2, H3N2) and later inoculated with various doses of whole or subunit A/NJ /76 virus vaccines. Results from experiments in mice demonstrated a low antigenicity of the subunit vaccine in unprimed animals compared with that of whole-virus vaccine; however, the antigenicity of both vaccines was enhanced in mice primed by previous infection with HO and HI viruses. In hamsters previously infected with an H3N2 virus, however, the response to only the whole-virus vaccine was enhanced by the priming infection. The responses of the mice were qualitatively more similar to those of humans than were the responses of hamsters.

Priming Infection and Serologic Response

Results

Antigenicity of vaccines in mice. Results from the first series of experiments are shown in figure 1. Since quantitative immunoelectrophoretic studies showed a twofold difference in hemagglutinin content between the MSD and W vaccines

[6], the serologic responses are shown as a function of the amount of viral hemagglutinin rather than CCA-unit content. Unprimed control mice produced serum HAl antibody after immunization with 0.24 p.,g of hemagglutinin of the wholevirus vaccine (MSD), and titers of HAl antibody increased with higher doses of vaccine. In contrast, the HAl responses of the un primed mice to the split-virus vaccine (W) were lower. Mice previously infected with A/swine/1976/ 31 virus showed a much higher final titer of antibody after immunization with the antigenically related AjNJ /76 virus vaccine than did normal mice. Antibody levels from the Ajswinej1976j31 infection were increased fourfold or more by immunization with either MSD or W vaccine. As can be seen in figure 1, the response to the W vaccine was not dose-dependent; there were equal HAl titers after administration of 0.24-5.8 p.g of AjNJ /76 virus hemagglutinin. Higher titers of antibody were produced with the whole-virus MSD vaccine, and a plateau was reached with 2.2 and 11.8 p.g of hemagglutinin of this vaccine. Responses to both vaccines were also enhanced in animals primed by infection with either AjPRj 8/34 or A! F"NI /1/47 virus. These viruses are antigenically closely related to each other [8], and A/ PR j 8 j 34 virus shows a low level of cross-reaction with A/NJ j76 virus, as shown by low titers of HAl antibody to A/NJ /76 virus in unvaccinated, convalescent-phase mice infected with A/PR/8j34 virus. Responses to MSD vaccine were similar in both A/PR/8/34 virus- and A/FMjl/47 virus-primed mice; these animals had HAl titers that were enhanced compared with those of unprimed mice, but titers did not reach the level found in A/swine/1976j31 virus-primed mice (figure 1). Priming with A/FMjlj47 virus greatly enhanced the response to W vaccine so that the response to this vaccine was similar to the response induced by comparable amounts of MSD vaccine. Slightly different results were obtained with mice primed with the more recent H2N2 and H3N2 viruses, which are less closely antigenically related to A/NJ j76 virus. Whereas a significant enhancement of the response to W vaccine was found in the Aj Aichij68 virus-primed group, responses to MSD vaccine were not enhanced. Priming with A/Singapore/lj57

Downloaded from http://jid.oxfordjournals.org/ at Indiana University Library on July 29, 2015

PR/8/34 (HONI); A/FM/lj47 (HINI); A/Singaporejl/57 (H2N2); A/Aichi/2/68 (H3N2); and A/Hong Kong/ I j68 (H3N2). The preparations of AjNJ j76 virus vaccine used in these studies were also used in the human dose-response studies. Formalin-inactivated Merck Sharp and Dohme (MSD, West Point, Pa.) whole-virus vaccine containing 740 chick cell-agglutinating (CCA) units/0.5 ml and the Wyeth Laboratories (W, Philadelphia, Pa.) triN-butyl phosphate-disrupted vaccine containing 660 CCA units j 0.5 ml were used. These vaccines contained 118 p.,g and 58 p.,g of hemagglutininj 0.5 ml, respectively, as described previously [6]. Dilutions of these vaccines were prepared in phosphate-buffered saline, and animals were inoculated ip with 0.5 ml as described below. Serologic tests. Samples of blood were collected from the retroorbital plexus or by axillary slash. Sera were treated with receptor-destroying enzyme and titrated by standard methods [7] for measurements of HAl antibody using the pools of virus listed above. HAl titers to A/NJ j76 virus were measured using the AjNJ j76 (X53) recombinant virus. Experimental design. Two similar experiments were performed. In the first, groups of 150-200 Swiss mice (9-12 g) were inoculated intranasally with a nonlethal dose (as determined by prior titration) of one of the influenza A viruses described previously. Equal numbers of mice served as uninfected control groups. Twenty-eight days later animals were bled, and groups of 15-20 were inoculated with 0.5 ml of a 1:10, 1:50, or 1:250 dilution of vaccine or with 0.5 ml of saline. Samples of blood were again collected 14 days after vaccination. A second experiment for comparison of the serologic responses of mice and hamsters used the same time table but a wider range of dilutions of vaccine and smaller groups of animals (10 mice per group, six hamsters per group).

5707

McLaren et ai.

5708

6

5 N

Cl

g

I~

A'/FM/1/47

A/PR/8/34 (Ho N,)

A/SWINE/31 (HswN,)

~C

4

NIL.

(H, N,) 6

5

~ 0----

Cl

/

0

o

3 II

I

I

o .48 .24

I

I

I

12 2.4 1.2 5.9

JLg

II

I

I

o .48 .24

I

I

:lJ

I

I

o .48

2.4 12 1.2 5.9

.24

I

1

I

I

2.4 12 1.2 5.9

:1(("1

I

o .48

I

I

I-

4

~

o

3

•

, 0"

..0"""

,

0'''''''

2

~I

I

o .41

I

I

I

I

2.4 12 .24 1.2 5.9 flog HA

~

I

II

I

I

0.41 t4 12 .24 1.2 5.9 flog HA

s::

/

/I

.24

4

1

-l

5"

3

"p_---o

o .48

o

I

I

I

2.4 12 1.2 5.9

if

2

flog HA

Figure 1. Serum antibody responses of mice to A/New Jersey (NJ)/76 virus vaccines. Animals were infected with influenza A viruses, and 28 days later they were inoculated ip with either Merck Sharp and Dohme, West Point, Pa. (whole, e e) or Wyeth Laboratories, Philadelphia, Pa. (disrupted, 0- - - -0) A/NJ /76 virus vaccines or with saline (x). Sera were collected 14 days later and tested individually for HAl antibody (HA). The results for unprimed mice (NIL) are shown twice. GMT = geometric mean titer.

virus also failed to enhance the response to MSD vaccine (with the exception of an apparently anomolous result with 11.8 IJ-g of hemagglutinin), but the responses of these animals to W vaccine was also not significantly different from that of unprimed mice. The results of the A/Singapore/ 57 virus-primed animals may in part, however, be the result of a lower severity of the priming infection with A/Singapore/57 virus compared with the other priming infections (results not shown). Experiments with normal mice and animals previously infected with A/Aichi/68 or A/ Singapore/57 viruses were therefore repeated. Results of these experiments confirmed the earlier results with A/Aichi/68 virus-primed mice but demonstrated enhanced responses to both Wand MSD vaccines in A/Singapore/57 virus-primed animals. Titers of antibody after immunization with W vaccine were similar in the A/Aichi/68 virus- and A/Singapore/57 virus-primed groups, whereas titers in A/Singapore/57 virus-primed

mice given 11.8 or 2.4 IJ-g of hemagglutinin of MSD vaccine were significantly higher than in the corresponding A/Aichi/68 virus-primed and unprimed groups (results not shown). Comparison of responses of mice and hamsters.

The experimental design was similar to that of the previous experiment except that A/Hong Kong/l/68 virus was used as the priming virus; this virus is antigenically identical to A/ Aichi/68 virus and produced a similar response in mice. The range of vaccine dilutions was extended to include the 200- and 800-CCA unit doses given to volunteers (30 and 120 IJ-g of hemagglutinin of MSD vaccine; 14.6 and 58 IJ-g of hemagglutinin of W vaccine). The results are shown in figure 2. Except for those animals given 58 IJ-g of hemagglutinin of W vaccine, the normal, unprimed mice again showed a significantly greater antibody response to MSD vaccine compared with that to W vaccine. The A/Hong Kong/I/68 virus-priming infection enhanced the response to all but the high-

Downloaded from http://jid.oxfordjournals.org/ at Indiana University Library on July 29, 2015

6

5

Cl

g

2

NIL.

5 N

l

flog HA

A/SING/57 IH 2 N2)

6

I

12 2.4 1.2 5.9

.24

flog HA

flog HA

HA

3

~ -l

""P---

x

2

4

Priming Infection and Serologic Response

S709

MICE

8

8 NORMAL

7

7

6

6 I

I N

gOJ

I

5

(:J

3 2

I

0----- ---I

I

.08

.15

~d

I

I

3:

4

IP--_~'

l

3

I

2 I

I

I

I

I

I

I

II

I

.76 3.8 15 59 1.5 7.5 30 119

I

.08

.IS

I

I

I

I

I

I

I

I

.76 3.8 15 S9 I.S 1.S 30 119

JL9 HA

JLQ HA HAMSTERS 7

7 NORMAL

A/HK/68

6

6 5

5 N

gOJ l~

G>

p-__4---