Humoral immunity in the ewe 2. The effect of pregnancy on the primary and secondary antibody response to protein antigen

Veterinary Immunology and Immunopathology, 25 ( 1990 ) 155-166 Elsevier Science Publishers B.V., Amsterdam - - Printed in The Netherlands

155

Humoral I m m u n i t y in the E w e 2. The Effect of P r e g n a n c y on the P r i m a r y and S e c o n d a r y Antibody Response to Protein Antigen G.E. REYNOLDS and J.F.T. GRIFFIN

Microbiology Department, University of Otago, Dunedin (New Zealand) (Accepted 14 November 1989 }

ABSTRACT Reynolds, G.E. and Griffin, J.F.T., 1990. Humoral immunity in the ewe. 2. The effect of pregnancy on the primary and secondary antibody response to protein antigen. Vet. Immunol. Immunopathol., 25: 155-166. This study examines the effect of pregnancy on the quantity and isotype of antibody in ewes immunised with a novel primary antigen. The primary and secondary antibody levels to bovine serum albumin (BSA) in alum adjuvant, were compared between non-pregnant ewes and ewes immunised at different stages of pregnancy. Anti-BSA isotype specific responses were measured using an indirect ELISA. Results show the levels of immunoglobulin M (IgM) increased and persisted in response to BSA during pregnancy (P < 0.05 ). Secondary immunoglobulinG 1 {IgG 1 ) titres were significantly impaired in late pregnancy and during lactation {P<0.05). The lower levels of immunoglobulin G2 (IgG2) were unaffected by pregnancy under these experimental conditions. Ewes were also immunised with BSA in alum adjuvant for their primary inoculum and BSA in different adjuvants for the secondary inoculation. Primary IgM persisted at higher levels during pregnancy compared with the response in non-pregnant ewes {P < 0.05). Following the secondary injection, lower levels of anti-BSA specific IgM, IgG1 and IgG2 antibodies were produced in late pregnancy and during lactation compared with the levels in non-pregnantcontrol ewes (P < 0.05 ). Alterations in regulatory T-cell function or effector B-cell activity would most readily explain the qualitative changes in antibody titre observed following primary injection during pregnancy. The results also suggest an associated impairment of immunological memory following primary immunisation during pregnancy.

INTRODUCTION

There is considerable evidence that pregnancy produces an altered state of immunoreactivity. Increased susceptibility to infection with bovine viral diarrhoea and infectious bovine rhinotracheitis viruses are reported in pregnant cattle (Lloyd, 1983). Pregnant ewes have a marked increase in susceptibility to gastrointestinal helminths during pregnancy and lactation (Connan, 1976). 0165-2427/90/$03.50

© 1990 Elsevier Science Publishers B.V.

156

G.E. REYNOI,DS AND J.F.T. GRIFFIN

Soluble factors in pregnant serum such as sex hormones and pregnancy associated proteins are thought to influence this material immunocompetence through their impact on regulatory T-cell function or effector B-cell activity (Gill, 1985). This reduced effector activity may result in decreased risk of rejection of the fetus by the maternal immune system. Preparturient vaccination has become a common practice in farm animals to enhance passive immunity and increase newborn survival (Fleenor and Stott, 1983). Vaccination of pregnant animals with Escherichia coli or rotavirus induce high antibody titres in colostrum and milk and protect the offspring against enteric disease (Myers, 1978; Wells et al., 1978; Acres et al., 1979). Vaccination of the dam confers immunity against mastitic organisms responsible for reduced milk production (Watson, 1980). As pregnancy has been shown to influence immune function in the ewe, it seems appropriate to examine immunisation during pregnancy to ensure that the maternal response is being recruited optimally for protection of the newborn. As shown earlier (Reynolds et al., 1990) the "upstream" regulatory TH cell and effector B-cell action can be evaluated by examining the different isotypic classes of antibody produced in response to a given antigen. The first experiment of this study was performed to critically examine qualitative aspects of antibody production during pregnancy in response to the novel primary protein immunogen, BSA. Bovine serum albumin was used in the sheep so that true primary immunisation could be studied because there was no pre-existing immune memory to this antigen. Animals were inoculated with a primary and secondary injection of BSA in alum at different times preceding, during and following pregnancy and their antibody responses were compared with nonpregnant controls. The second phase of this study examines the induction of cellular memory following primary antigenic activation with BSA. When animals were given a primary injection with BSA in alum and secondary inoculation of BSA with different adjuvant combinations, variable levels of antibody were produced. The adjuvants with their differential ability to produce altered effector activity made it possible to detect the degree of activation that had occurred in the primary reaction in prgnant ewes compared with non-pregnant controls. Thus the secondary antibody response is used to probe the regulatory TH and effector B-cell events in the primary cellular reaction in pregnant and non-pregnant ewes. MATERIALSAND METHODS Animals A flock of 96 virgin Perindale ewes were used in these experiments. Progesterone implants were used to synchronise oestrus in breeding females.

THE EFFECTOF PREGNANCYONTHE PRIMARYANDSECONDARYRESPONSEIN THE EWE

157

Immunisation Animals were injected with 500/tg of BSA (Sigma, U.S.A.) for both the primary and secondary inoculation. Equal volumes of BSA were emulsified in various adjuvants and a 1 ml inoculum was given to each animal via the s.c. route. The three adjuvants used in this study were aluminium hydroxide (alum), diethylaminoethyl dextran (DEAE dextran) and Freund's complete adjuvant (FCA), prepared as previously described (Reynolds et al., 1990). Blood samples A blood sample was taken from each sheep prior to inoculation and further samples were taken at 1-2 weekly intervals. The serum was removed and stored at - 20 ° C until required for testing. B S A immunisation throughout pregnancy A group of 36 ewes were used in this longitudinal study. The sampling period began approximately 42 days preconception and ended in the postpartum period. Pregnant ewes were grouped according to the three stages at which the primary inoculation of BSA in alum (500/zg) was administered; preconception, early and late pregnancy (Table 1 ). An age matched group of non-pregnant control ewes, inoculated at different stages throughout the experiment was included for comparison. The secondary inoculation with BSA in alum was given 56 days post primary. The influence of adjuvants on the response to B S A during pregnancy A group of 30 pregnant and 30 ages-matched non-pregnant ewes were used in this study. At day 80 of gestation, the 30 pregnant animals were inoculated TABLE 1 The BSA in alum inoculation schedule for pregnant and non-pregnant ewes Group

Number of animals

Time of primary inoculation

Interval between primary and secondary inoculation (gestational status )

Preconception

10

42-56 days preconception (Non-pregnant) 28-35 days pregnant (Pregnant) 105-112 days pregnant (Pregnant)

56 (Pregnant) 56 (Pregnant } 56 (Lactating) 56 ( Non-pregnant )

Early pregnancy

9

Late pregnancy

7

Non-pregnant

10 ( Non-pregnant )

158

G.E. REYNOLDSAND J.F.T. GRIFFIN

TABLE 2 The immunisation regime for pregnant and non-pregnant ewes inoculated with BSA in alum for their primary injection and BSA in different adjuvants for their secondary inoculation Pregnant and nonpregnant group

Number of animals

Primary inoculation

Secondary inoculation

1

10

2

10

0.5ml BSA ~+ 0.5ml alum same as above

3

10

same as above

0.5ml BSA ~+ 0.5ml alum 0.Sml BSA ~+0.Sml I)EAE dextran solution 0.5ml BSA ~+ 0.5ml FCA -

-

~1000pg/ml.

to produce primary immune cellular activation with 500 pg of BSA in alum, subcutaneously. The non-pregnant animals were similarly inoculated for their primary inoculation. At day 56 post primary inoculation (when the ewes were 136 days pregnant ), animals were randomised into 3 groups (Table 2), for their secondary inoculation. Each group received 500 pg of BSA in different adjuvants to induce significant antibody production. The three adjuvants used were, alum, DEAE dextran or FCA. The ELISA assay A four step indirect ELISA system was used for the detection of class-specific antibodies to BSA using predetermined optimal concentrations of antigen, immune sheep serum, anti-sheep monoclonal antibody and anti-mouse peroxidase labelled conjugate (Reynolds et al., 1990). RESULTS

The BSA in alum study When BSA in alum was injected into ewes at different stages of pregnancy, alterations in IgM and IgG1 levels were observed. The IgM responses to BSA are represented in Fig. 1. At day 14, both groups given their primary antigen injection during pregnancy (early and late gestation groups ), had significantly higher IgM titres than non-pregnant control animals (P < 0.05). The secondary response produced "spikes" of IgM activity in groups inoculated first during pregnancy. Three of nine animals immunised first in early pregnancy produced significant IgM (Fig. 1) and one of seven animals of the group first injected in late pregnancy produced an IgM antibody response after day 35

THE EFFECT OF PREGNANCY ON THE PRIMARY AND SECONDARY RESPONSE IN THE EWE 5O

4°1 30

159

A .~- Non-pregnant •-e- Pre-pregnancy ~ Earlypregnancy

\

/ I

I ~r

.r

-,,;

~

.

~0-

" i

0

1

~

I

'

I

•

T

•

~

•

I

•

i

•

I

•

i

•

I

•

1000 ,~ 2 /

800I 600 • 400 - ~

•~- Non-pregnant ~ Pre-pregnancy -~- Earlypregnancy

, ~ [~ ~

L~

200

: •

oi loo •

60-

•~" -e,-~-o-

•

60-

Non-pregnant Pre-pregnancy Earlypregnancy Latepregnancy

4020-

o ~ - ' ~ ~ 1,*

1°

BSA-alum

t6

"

"~

~r2

"

~ 2°

BSA-alum

?_r---"" 30

'-

Days

Figs. 1,2 a n d 3. T h e a n t i - B S A IgM, IgG1 a n d IgG2 antibody response in ewes given primary ( 1 ~ ) a n d s e c o n d a r y (2 ° ) inoculations with B S A in a l u m while n o n - p r e g n a n t or a t different stages of p r e g n a n c y for their p r i m a r y inoculation•

post secondary inoculation. There was little detectable secondary IgM response in animals first immunised prior to conception or non-pregnant controls. During the primary response, IgG1 levels (Fig. 2) were not significantly different. All groups except non-pregnant controls produced peak titres at day 21, whereas non-pregnant animals produced a peak response at day 28. Following secondary inoculation, all groups produced a peak in IgG1 titre at day 14.

160

G.E. REYNOLI),q AND J.F.T. GRIFFIN

Animals immunised first in late pregnancy and boosted postpartum had significantly lower titres followed boosting, compared with all other groups {P < 0.05). At the end of lactation, this group has an increased IgG1 antibody titre from day 35 post secondary inoculation and was significantly higher than all other groups by day 63 {P<0.05). The group given a primary inoculum of BSA in early pregnancy and boosted later in pregnancy had a significantly lower IgG1 titre at day 35 post secondary inoculation compared with non-pregnant controls {P<0.05). Except for the late pregnancy group, the peak primary antibody responses were significantly lower than peak secondary antibody responses (P<0.025). In all groups IgG2 levels were insignificant throughout the primary and secondary phase (Fig. 3 ).

The BSA in different adjuvants studies Immunising ewes with BSA in alum and other adjuvants, produced variable IgM, IgG1 and IgG2 levels relative to the adjuvant used and gestational status of the animal. The antibody response in animals injected with BSA in alum for the primary and secondary immunisation. Consistent with the previous experiment, the primary day 14 IgM response (Fig. 4) was found to be significantly higher in pregnant ewes when compared to non-pregnant ewes (P < 0.01 ). The peak primary IgG1 and IgG2 titres to the BSA antigen (Figs. 5 and 6), showed no significant difference between pregnant and non-pregnant animals. Following the boost with BSA in alum, both pregnant and non-pregnant animals had little or no detectable anti-BSA IgM but the levels of IgG1 and IgG2 at 14-21 days post secondary inoculation were significantly higher in non-pregnant ewes (P < 0.05). No boost in IgG1 titres was found in pregnant animals when compared to primary titres, but non-pregnant ewes showed a significant boost in IgG1, 14 days post secondary inoculation compared to 14 days post primary inoculation (P < 0.015 ). The antibody response in animals injected with BSA in D E A E dextran for the secondary immunisation. The IgM response (Fig. 7) peaked at day 14 post secondary inoculation and non-pregnant animals had significantly higher titres than pregnant ewes on this day ( P < 0 . 0 5 ) . The day 14 IgG1 response (Fig. 8) followed similar trends with significantly higher antibody titres in the nonpregnant controls (P < 0.05 ). Significant IgG1 boosting was seen in both groups at day 14 compared to primary titres (P<0.005). There was minimal antiBSA IgG2 boosting (Fig. 9) at day 14 in both pregnant and non-pregnant groups.

T H E E F F E C T O F P R E G N A N C Y O N T H E P R I M A R Y A N D S E C O N D A R Y R E S P O N S E IN T H E E W E

161

50 40 "E

-~- Pregnant ~- Non-pregnant

30 20

0

1000

I

I

I

T

I

I

I

I

I

I

I

¥

,~

;~

42

56

70 84 Days

5

800

8

-~- Pregnant -e.- Non-pregnant

600

m

LD 2OO 0 100 80 •~

60

,~ ~D "~ t~

40

-~- Pregnant -~- Non-pregnant

20

o

T

0 14 28 1° BSA-alum

42

I 80

.--'F

,

,

0 14 28 2 ° BSA-alum l

136t

Parturition

Days of gestation Figs. 4 , 5 and 6. T h e a n t i - B S A [~M, I~O] and I~G2 a n t ~ b ~ y response ~n pregnant and nonpregnant ewes immunised w i t h B S A ~n alum for both the p r i m a r y (1 ~ ) and seconda~ (2 : ) Jnocu]ath)n.

The antibody response in animals injected with BSA in FCA for the secondary immunisation. Although the antibody response to BSA in FCA showed different kinetics to that seen with BSA in other adjuvants, there were significantly

16~

G.E. REYNOLDSANDJ.F.T. GRIFFIN 50

7

40 30 w

-o- Pregnant

//~

~

~

20 10 0 1000 8

800 600

Pregnant

400 200 0

I

I

I

I

I

I

I

~

I

i

I

i

I

100 9

80 ~ ~ < co _

~

60

~ Pregnant -~- Non-pregnant

40

20

~

0

~

I

14 28 1° BSA-alum I 80

I

42

0 14 28 42 2° BSA-DEAEdextran l

136t

I

I

56

70 84 Days

Parturition

Days of gestation Figs. 7, 8 and 9. The anti-BSA IgM, IgG1 and IgG2 antibody in pregnant and non-pregnant ewes immunised with BSA in alum for both the primary ( 1 ¢ ) injection and BSA in DEAE dextran for the secondary (2 :: ) inoculation.

lower IgM, IgG1 and IgG2 titres post secondary inoculation in pregnant animals compared with non-pregnant controls ( P < 0.005). The secondary IgM titre (Fig. 10) rose gradually to a peak at day 56 in pregnant animals and day

THE EFFECT OF PREGNANCY ON THE PRIMARY AND SECONDARY RESPONSE IN THE EWE so

10

163

=

40 ~ Pregnant -e- Non-pregnant

30

~ J

/

u.l

0

~-

,

•

,

•

,

1000

800

11 ~ Pregnant

~

600 400 200 0

i

100]

•~

617

,,~ ~,~ -n

40

~.1

!

!

i

!

i

!

i

!

12

.~- Pregnant

20 0

•

!

.

!

0 14 28 1° BSA-alum

.

i

42

t 80

•

i

•

i

•

i

0 14 28 2° BSA-alum l

•

i

42

.

i

56

•

!

,

70 84 Days

1361 Parturition

Daysof gestation Figs. 10, 11 a n d 12. T h e a n t i - B S A IgM, IgG1 a n d IgG2 a n t i b o d y response in p r e g n a n t a n d nonp r e g n a n t ewes i m m u n i s e d w i t h B S A in alum for b o t h t h e p r i m a r y ( 1 ° ) injection and B S A in FCA for t h e s e c o n d a r y (2 ° ) inoculation.

42 in non-pregnant controls and both remained at plateau levels for the remainder of the experiment. The IgG1 titre (Fig. 11 ) reached a plateau at day 14 for non-pregnant animals and day 28 for pregnant ewes post secondary

164

G.E. REYNOLDS AND J.F.T. GRIFFIN

inoculation. At day 28 post secondary inoculation, the IgG2 response (Fig. 12 ) in non-pregnant ewes was significantly higher than in pregnant animals (P<0.005). The response also persisted throughout the remainder of the experiment. DISCUSSION Pregnancy was found to influence the isotypic class and quantity of antibody produced following immunisation with a novel primary immunogen. When BSA in alum was injected into ewes at different phases of pregnancy, there was a persistence of IgM titres and diminished IgG1 levels. The anti-BSA IgM production was higher following primary injection and persisted into the secondary period when ewes were given a primary inoculation during pregnancy. This profile did not follow classical immune responsiveness and suggests alterations in the maturation of the isotypic class of antibody produced during pregnancy. The persistence of IgM antibody has been reported in pregnant ewes inoculated with Aspergillus [umigatus (Corbel and Day, 1978), mice inoculated with E. coli (Kenny and Diamond, 1977) and pregnant sows injected with foot and mouth disease vaccine (Francis and Black, 1986). Such persistence suggests an impairment of immunological switch from IgM to IgG associated with the classical anamnestic responses produced in animals re-exposed to microbial antigen. Although anti-BSA IgG2 levels were not major contributors to the response during pregnancy, IgG1 titres were significantly reduced in late gestation and during lactation (Fig. 2). This is consistent with literature on immunity in the peri-parturient period (Lloyd, 1983). The transfer of IgG1 from serum to mammary tissue is the most obvious explanation for the antepartum lowered response (Brandon et al., 1971). A decrease of 50% has been shown to occur in IgG1 levels 2-3 weeks before parturition in the ruminant (Williams and Millar, 1979). The lower response postpartum may be due to immunosuppression associated with late pregnancy, parturition and lactation (Lloyd, 1983). Transfer of IgG1 antibody into the colostrum pool and subsequent transfer to the suckling offspring makes it difficult to assess the true status of serum IgG1 levels in ewes during the last 4 weeks of pregnancy. The experiment involving primary immunisation with BSA in alum and secondary inoculation with BSA in different adjuvant combinations attempted to evaluate the degree of activation of memory cells during pregnancy. The results are consistent with the earlier experiment with persistent IgM levels during pregnancy suggesting impaired IgM to IgG switch following primary inoculation. This increase in IgM levels post primary immunisation further suggests an associated impairment of memory. Hibma and Griffin (1987) described similar findings using a guinea pig model which showed impaired IgM to IgG switch and low affinity antibody produced during pregnancy. After a booster

THE EFFECT OF PREGNANCY ON THE PRIMARY AND SECONDARY RESPONSE IN THE EWE

165

immunisation with BSA in alum, DEAE dextran or FCA, non-pregnant responses were consistent with earlier studies (Reynolds et al., 1990). In pregnant animals, there were reduced IgM, IgG1 and IgG2 levels during late pregnancy and lactation. These alterations would reflect the endocrine influence during this period and the transfer of IgG1 to mammary tissue {Williams and Millar, 1979; Lloyd, 1983). Evidence for memory impairment is difficult to determine in the present studies as high levels of secondary antibody tend to obscure subtleties of cellular regulation which are more accessible to probing during primary responses. Increased primary IgM and decreased secondary IgM during pregnancy does however suggest peturbations in immunoregulation associated with gestation. Evidence for altered T-cell function during pregnancy is reflected in an impaired ability to resist chronic infections (Baines and Pross, 1982; Brabin, 1985; Lloyd, 1983). An associated impairment of antibody production during pregnancy, involving defective IgM to IgG switch, would influence the levels of maternal IgG1 available to the offspring through passive transfer via colostrum. In this study the low levels of primary immunity produced within the last 6 weeks of gestation, associated with the change in quality of the antibody response during pregnancy suggests that primary vaccination during pregnancy would be less effective in providing passive antibody for protection of the newborn. The indications from this data are that primary immunisation of animals prior to conception to activate immune memory is a desirable prerequisite to ensure that optimal passive immunisation is achieved through the use of boosting vaccination regimes involving pregnant animals. ACKNOWLEDGMENTS We wish to thank Dr. K.J. Beh for access to the monoclonal antibodies used in this experiment, and the Animal Production Unit, Ministry of Agriculture and Fisheries, Invermay for organising experimental flocks of sheep and their continued access. G.R. gratefully acknowledges scholarship support from Southland Frozen Meats Limited. REFERENCES Acres, S.D., Isaacson, R.E., Babuik, L.A. and Kapilany, R.A., 1979. Immunisation of calves against enterotoxigenic colibaciilosis by vaccinating dams with purified K99 antigen and whole cell bacterins. Infect. Immun., 25: 121-126. Baines, M.G. and Pross, H.F., 1982. Impairment of the thymus-dependent humoral immune response by syngeneic or allogeneic pregnancy. J. Reprod. Immunol., 4: 337-348. Brabin, B.J., 1985. Epidemiology of infection in pregnancy. Rev. Infect. Dis., 7: 579-603. Brandon, M.R., Watson, D.L. and Lascelles, A.K., 1971. The mechanism of transfer of immunoglobulin into mammary secretions of cows. Aust. J. Exp. Biol. Med. Sci., 49: 613-623.

166

(;.E. REYNOI,DSANDJ.F.T. GRIPFIN

Connan, R.M., 1976. Effect of lactation on the immune response to gastrointestinal nematodes. Vet. Rec., 99: 476-477. Corbel, M.J. and Day, C.A., 1978. Examination of the immunoglobulin classes involved in the serological response of pregnant sheep to Aspergillus/umigatus. Sabouraudia, 16: 23-33. Fleenor, W.A. and Stott, G.H., 1983. Quantification of bovine IgG, IgM and IgA antibodies to Clostridium perfringens fl toxin by enzyme immunoassay II. Systemic effec~ of maternally derived antibodies on immunisation of newborn calves. Vet. Immunol. Immunopathol., 4:6:~,2 654. Francis, M.J. and Black, I,., 1986. Humora[ response of pregnant sows to foot and mouth disease vaccination. J. Hyg. Cambridge, 96:501 511. Gill, T.J., III, 1985. Immunity and pregnancy. CRC Crit. Rev. Immunol., 5:201-227. Hibma, M. and Griffin, J.F.T., 1987. Altered humoral immunity during pregnancy in the guinea pig. J. Reprod. Immunol., 10: 299-307. Kenny, J.F. and Diamond, M., 1977. Immunological responsiveness to Escherichia coli during pregnancy. Infect. Immun., 16: 174. 180. I~loyd, S., 1983. Effect of pregnancy and lactation upon infection. Vet. Immunol. Immunopathol., 4: 153-176. Myers, L.L., 1978. Enteric colibacilh)sis in calves: Immunogenicity and antigenicity of E. coli antigens. Am. J. Vet. Res., 39: 761-765. Reynolds, G.E., Suttie. J. and Griffin, J.F.T.. 1990. Humoral immunity in the ewe. I. The influence of adjuvancy and immunogenic substitution on immune reactivity following immunisation with protein antigens. Vet. Immunol. Immunopathol., 25: 1- 12. Watson, D.L., 1980. Immunological functions of the mammary gland and its secretion: Comparative review. Aust. J. Biol. Sci. 33: 403-422. Wells, P.W., Snodgrass, D.R., Herring, J.A. and McDawson, A., 1978. Antib()dy titres (o lamb rotavirus in colostrum and milk of vaccinated ewes. Vet. Rec., 103: 46-48. Williams, M.R. and Millar, P., 1979. Changes in serum immunoglobulin levels in Jerseys and Friesians near calving. Res. Vet. Sci., 26:81-84.