Detection of Progeny Immune Responses after Intravenous Administration of DNA Vaccine to Pregnant Mice.

A number of factors influence the development of tolerance, including the nature, concentration and mode of antigen presentation to the immune system, as well as the age of the host. The studies were conducted to determine whether immunizing pregnant mice with liposome-encapsulated DNA vaccines had an effect on the immune status of their offspring. Two different plasmids (encoding antigens from HIV-1 and influenza virus) were administered intravenously to pregnant mice. At 9.5 days post conception with cationic liposomes, injected plasmid was present in the tissues of the fetus, consistent with trans-placental transfer. When the offspring of vaccinated dams were immunized with DNA vaccine, they mounted stronger antigen-specific immune responses than controls and were protected against challenge by homologous influenza virus after vaccination. Moreover, such immune responses were strong in the offspring of mothers injected with DNA plasmid 9.5 days after coitus. These results suggest that DNA vaccinated mothers confer the antigen-specific immunity to their progeny. Here we describe the methods in detail as they relate to our previously published work.

Introduction

Most vaccines intended for human use are administered to infants and children. Due to the immaturity of their immune system, newborns exposed to foreign antigens are at risk of developing tolerance rather than immunity (2-7). For example, if antigen is administered shortly after birth, forbidden clones can emerge and induce such tolerance (2-7). A number of factors influence the development of tolerance, including the nature, concentration and mode of antigen presentation to the immune system, as well as the age of the host (8,9). Over the past decade, there has been considerable interest in the use of DNA vaccines to prevent infection by pathogenic viruses, bacteria and parasites, with phase I clinical trials being initiated against malaria, HIV-1 and hepatitis B virus.

In the present study, we confirmed that plasmid DNA administered to pregnant mice could reach the fetus through the placenta. This was true both of DNA vaccines encoding the env gene of HIV-1 as well as those encoding the influenza virus matrix (M) and nucleoprotein (NP) genes. Analysis of the immune response of offspring whose mothers were immunized with the influenza DNA vaccine indicate that these progenies had enhanced level of protection against the same virus infection.

Materials and methods

Animals

We used 6-10-week old BALB/c female mice purchased from Japan SLC, Inc. (Shizuoka, Japan). All mice were allowed free access to sterile food and water.

Viral protein expression plasmids and antibodies

A pME18S-M expression plasmid was constructed with the pME18S expression vector into which M region cDNA from influenza virus strain A/PR/8/34 (H1N1) had been inserted (13). pME18S empty vector was used as a control plasmid for A/PR/8/34 challenge. The expression of the proteins was confirmed by Western blot analysis (10). pCMV160IIIB encoding the env gene of HIV-1 strain IIIB has been described in detail in our previous report (11). DNA vaccines of NP (A/pCMV-V1NP) and HA (V1J-HA (PR8)) genes of the A/PR/8/34 strain (14-16) were the kind gifts from Drs. J. J. Donelly and D. Montgomery, Merck Research Lab., West Point, PA. To confirm that plasmid DNA was transferred through the placenta, a lacZ expression plasmid containing a chicken β-actin promoter was also used.

Virus Preparation (Protocol I)

Mouse-adapted influenza A/PR/8/34 (H1N1) viruses were used in this study. Maudin-Darby canine kidney (MDCK) cells were cultured with Eagle's minimum essential medium (MEM, Nissui Corp. Tokyo, Japan) containing 10% FCS at 5% CO2, 37oC. Viruses were harvested from infected MDCK cells and titrated according to the plaque formation method.

DNA immunization (Protocol II)

Mothers were injected intravenously (i.v.) with DNA vaccine before or after coitus. Preparations containing various doses of the DNA vaccine were encapsulated into liposomes (12).

Briefly, a 1:1 volume of 6.01 mg/ml 3β[N-(N'N'-dimethylaminoethane) carbamoyl] cholesterol (DC-Chol) and 5.99 mg/ml dioleoylphosphatidylethanolamine (DOPE) in chloroform was mixed with an evaporator set at 40oC for 1 hr. Five volumes of 20 mM HEPES buffer (pH 7.8) was added into the mixture, mixed and kept at 4oC overnight to completely dissolve the pellet. Then, the resultant solution was sonicated with a Sonifier 250 (Branson, Danbury, CT) set at output control 2-3; duty cycle 30, for 5 min. The cationic liposomes were stocked at 4oC and used within 2-3 months. Prior to administration, an appropriate amount of DNA in phosphate-buffered saline (PBS), pH 7.2, was mixed with the liposome solution at a volume ratio of 1:1. The pregnant mice were i.v. administered the DNA vaccine or empty vector on various days after post conception (p.c.) to assess immunogenicity. Six weeks after birth, their offspring were injected intramuscularly (i.m.) in gastrocnemius muscles with 20-50 μg of the same expression plasmid or the empty vector. For the influenza virus challenge experiment 50 μg each of plasmids expressing influenza M and NP protein was administered into pregnant mice by i.v. route. The same amount of DNA plasmids was administered to offspring by i.m. route.

X-gal staining

Mouse fetus was cut in two and washed with PBS once followed by incubation at 37oC for 30 min with X-gal staining buffer (5 mM K4Fe(CN)6⋅3H2O, 5 mM K3Fe(CN)6, 20 mM MgCl2, 1 mg/ml X-gal).

FISH analysis (Protocol III)

FISH analysis was performed essentially by the method of El-Naggar et al (17). A 564-bp HIV env region fragment (nt 1569-2133) was amplified from pCMV160IIIB plasmid (10) using primers (5-ATGTGTAACACCTCAGTCATTAC and TTATCTTTTTTCTCTCTGCACCAC-3). The PCR product was purified from 1.5% agarose gel using QIAquick gel extraction kit (Qiagen) and was labeled with digoxigenin-11-dUTP by nick translation. The labeled product (300-560 bp DNA) was confirmed on 2% agarose gel and used for a hybridization probe. The tissue samples were taken from mice to which 50 μg of HIV plasmid pCMV160IIIB (11) or influenza plasmid V1J-HA(PR8) (14) with liposomes were administered and sliced to prepare histological examination. The samples were denatured and hybridized with digoxigenin-labeled probes using a previously described method (17). After hybridization, the slides were washed and stained with an anti-digoxigenin rhodamine (a red fluorochrome: Boehringer Mannheim, Germany). The slides were counterstained with 4,6-diamino-2-phenylindole dihydrochloride (DAPI). Images were made with a Nikon SA fluorescence microscope (Nikon Corp., Tokyo, Japan) and a Charge Coupled Device (CCD) camera interfaced with a Cyto Vision (Applied Imaging, Sunderland, UK).

Cytokine ELIspot assay (Protocol IV)

The cytokine ELIspot assay was performed with minor modifications, as previously described (18,19). Briefly, 96-well microplates (MAIPS4510, Millipore, Bedford, MA) were coated with anti-mouse IFN-γ rat mAb (PharMingen), and after adding cells isolated from the spleen 7 days after immunization, plates were incubated in a 5% CO2 atmosphere at 37°C with or without 10 μg/ml of V3 peptide. After a 24-hour culture, plates were washed and incubated again for 2 hours with biotinylated anti-mouse IFN-γ mAb (PharMingen). Then, after staining with alkaline phosphatase, the spots in each well were counted using a computer assisted video image analysis system (Zeiss Co., Germany). By applying the proper dilution factor the total number of cytokine-secreting cells was calculated.

Virus challenge

Under light diethyl ether anesthesia, the offspring were simultaneously infected with virus at day 10 after immunization with the same plasmid DNA as that administered to their mothers. Five lethal doses (LD50) of influenza A/PR/8/34 (H1N1) in 30 μl of PBS were administered by the intratracheal route using a 24-gauge stainless steel animal feeding tube (Popper & sons, New York, NY). The mortality rate was determined after 20 days.

Statistical analysis

Statistical analysis for comparison of two groups was conducted using an unpaired t-test or one-way factorial analysis of variance (ANOVA) for distribution parameters. Significance was defined as p<0.05 in both analyses.

Results

Gene transfer into fetuses

Initial studies examined whether DNA plasmids could be transmitted through the placenta of pregnant mice. To evaluate plasmid uptake and expression in fetal tissue, a plasmid expressing the lacZ gene was utilized. To increase the uptake

of this plasmid, it was liposome-encapsulated prior to i.v. delivery (8). Tissues from newborn mice from mothers injected with the lacZ plasmid and liposomes 9.5 days p.c. showed strong expression of that gene (Fig. 1). Of particular interest was the intense staining in the umbilical region of the fetuses. Examination by the FISH method confirmed that abundant plasmid DNA had been transmitted to the fetuses (Fig. 2). We found abundant plasmid DNA in spleen, liver, lung, and other tissues (data not shown).

Fig. 1

Gene expression of lacZ in mouse fetus.

A and B, Stained tissue of a fetus cut in two, whose mother received 30 μg of lacZ plasmid with liposomes at day 9.5 postcoitus (p.c.); C and D, stained tissue of a fetus whose mother received 30 μg of empty plasmid with liposomes at day 9.5 p.c.

Fig. 2

FISH analysis.

Vertebra of a fetus whose mother received HIV plasmid pCMV160IIIB (A) or influenza plasmid V1J-HA (PR8) (B) with liposomes at day 9.5 p.c. These samples were reacted with HIV env region fragment labeled with digoxigenin-11-dUTP, followed by staining with an antidigoxigenin rhodamine. Red fluorochrome indicates the presence of HIV-IIIB DNA.

ELIspot analysis using spleen cells from immunized offspring of vaccinated mothers was performed (Table 1). When stimulated in vitro with vaccine-encoded antigen, a significant increase in the number of spleen cells secreting IFN-γ was observed.

Table 1

ELIspot analysis of IFN-γ producing spleen cells from DNA-vaccinated mice whose mothers had been injected with the same vaccine during pregnancy.

Immunogen administered to		IFN-γ producing cells
Pregnant mothers	Progenies	(Spot/106spleen cells)
pCMV160IIIB**	pCMV160IIIB	42.6±9.6*
pCMV160IIIB**	Non-immune	20.6±6.7
Empty vector**	pCMV160IIIB	28.2±3.5*
Empty vector	Empty vector	16.7±6.3
Non-immune	Non-immune	13.9±5.9

At day 9.5 p.c., pregnant BALB/c mice were i.v. injected with 50 μg of pCMV160IIIB or empty vector with liposomes. Six weeks after birth the offspring were immunized i.m. with 50 μg of the same plasmid or empty vector as received by their mothers. After 7 days, spleen cells were collected and cocultured with V3 peptide for 24 hours. Data represent means±SE of 6-8 mice. *Indicates significant difference (p<0.05) compared to the empty vector (control). **Means a significant difference between the two indicated values. Data from two other experiments showed similar results.

Challenge test with influenza virus

To examine the immunoprotective effect of maternal vaccination with a DNA vaccine against influenza virus, offspring were immunized with 50 μg of the same vaccine at 6 weeks of age. Seven days later they were challenged with influenza virus A/PR/8/34. More than 70% of the offspring of vaccinated mothers survived (Fig. 3), whereas only 20% of the offspring of non-vaccinated mothers survived. All of the non-immunized offspring whose mothers received liposomes alone or were not immunized had died.

Fig. 3

Protection of offspring against a lethal A/PR/8/34 influenza virus challenge.

Day 9.5 p.c. pregnant BALB/c mice were i.v. injected with 50 μg each of pME18S-M and pCMV-V1NP with or without liposomes. Six weeks after birth, their offspring were immunized i.m. with a total of 50 μg of the same plasmid DNA with liposomes. In one group, offspring of mothers that had received DNA vaccine with liposomes were not administered vaccine. In another group, the mother and progenies received only empty vector. Non-immunized normal mice were used as the other control. After 7 days, all mice were challenged with 5xLD50 of A/PR/8/34 virus. The percentage of survival of these mice was studied for another 15 days. n, number of mice.

The timing of maternal DNA vaccination on the capacity of offspring to develop protective immunity was then examined. Whereas <20% of normal vaccinated mice (and offspring of mothers vaccinated 20 days prior to mating) survived the challenge, >50% of the offspring of immunized mothers survived (Table 2).

Table 2

The importance of timing the DNA immunization of pregnant mothers against A/PR/8/34 virus challenge.

DNA vaccination of		Survival after immunization
Pregnant mothers	Progeny	Survivors/total (%)
1. DNA vaccination
20 days before coitus	DNA vaccination	3/19 (15.8)
day 5.5 p.c.	DNA vaccination	5/20 (25.0)
day 9.5 p.c.	DNA vaccination	11/21 (52.3)*
day 14.5 p.c.	DNA vaccination	12/19 (63.2)*
day 9.5 p.c.	no vaccination	3/18 (16.7)*
2. Empty vector
day 9.5 p.c.	DNA vaccination	2/22 (9.1)
3. Non-immune control	Non-immune	1/19 (5.3)

Pregnant BALB/c mice were i.v. injected with 25 μg each of A/pCMV-V1NP+pME18S-M and liposomes on day 5.5, 9.5 or 14.5 p.c. Six weeks after birth, all offspring were immunized i.m. with 50 μg of the same plasmid DNA as received by their mothers. After 7 days, these mice were challenged with 5xLD50 of A/PR/8/34 and the per cent survival was determined after another 20 days. Data represent means±SE of 19-22 mice. *Indicates statistically significant difference (p<0.05) between non-immune control group. Data from another separate experiment showed similar results.

These findings indicate that immunization of mothers with a DNA vaccine against the influenza virus improves the ability of their offspring to develop protective immunity against viral challenge post vaccination.

Discussion

Two independent techniques were used to establish that plasmid DNA administered to pregnant mice could reach the fetus. One is the administration of a β-gal encoding plasmid which allowed for the direct identification of protein expression in neonates. Intense staining of the placenta was consistent with transplacental migration of the plasmid. The other technique is the FISH method which was used to directly detect plasmid DNA in fetal tissue (Fig. 2). Our results confirm and extend the previous finding (10) that in mice a β-gal plasmid can be transmitted through the placenta to the fetus. Of particular importance, we established that such trans-placental transfer influences the recipients' subsequent capacity to mount an immune response against the plasmid-encoded antigen. This was manifested by improved cellular immunity (Table 1) and higher levels of pathogen-specific protection (Fig. 3 and Table 2).

These studies were also performed to test the hypothesis that the administration of a DNA vaccine during pregnancy may induce antigen-specific tolerance in the offspring, as suggested by the clonal selection theory of Burnet (3,4,20). Although pregnant mice were immunized with various doses of several different DNA vaccines, we found that immunity but not tolerance was elicited in the fetus (Fig. 3 and Tables 1 & 2). Using this technique, we did not observe antigen-specific immune tolerance in progeny as reported by Mor et al. (21). This could reflect our use of a different plasmid (Mor et al. detected tolerance following neonatal immunization with a plasmid encoding the circumsporozoite protein of malaria), or the very limited amount of plasmid actually transferred trans-placentally. Indeed, Ichino et al. demonstrated that neonatal tolerance was dose-dependent, and could be reliably induced only when >10 μg of plasmid was injected into newborn mice (8). This is consistent with other reports showing that low dose antigen can induce immune responsiveness, while high dose immunization can induce tolerance in young recipients (9,22,23). This might be in support of Burnet's theory that high levels of neonatal antigen can trigger clonal deletion (4,8).

When offspring of vaccinated mothers were immunized at 6 weeks of age with the same DNA vaccine, they displayed significant anamnestic responses. Re-immunization was required, however, since trans-placental transport of plasmid alone did not trigger strong immune responses in the newborn, or provide adequate protection from infection (Fig. 3 and Table 2). On the other hand, re-exposure of these mice to vaccine at 6 weeks of age elicited a strong, protective immune response characterized by antigen-specific antibody, CTL and cytokine responses. The administration of DNA vaccine into amniotic fluid induced a high level of protective immunity (24). In the present study, when DNA vaccine was given to mothers, Ag-specific acquired immunity was induced in their offspring. Therefore, this method may be effective in the prevention of pertussis, hepatitis type B and C, mumps, rubella and various other infections occurring in infants as well as animals.