INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

Neonates are deficient in several components of inflammatory, innate, and specific immune responses. The presence of high-titer maternal antibody in newborns is the major form of protection from disease in early life. Maternal immunoglobulin G (IgG) crosses the placenta from mother to fetus during development (12) and typically exceeds titers of the same antibody in the mother. This passive antibody slowly declines over the first year of life, a period during which the infant's immune system matures, becomes more experienced, and develops its own repertoire of protective memory immune responses. However, maternal antibody can also interfere with active immunization of the offspring (1). Immunization protocols are often delayed several months and/or require multiple booster immunizations to achieve the desired protective immune response. Thus, a window of time exists when maternal antibody levels are too low to reliably protect an infant from infectious disease but are high enough to prevent responses to vaccines.

DNA vaccination is an attractive method for immunization in the presence of maternal antibody. Maternal antibody is thought to interfere with traditional vaccine efficacy by reducing the amount of antigen available for processing and presentation by antigen-presenting cells. The ability of DNA vaccines to directly transfect cells bypasses this problem. The maternal antibody will not inhibit the DNA vaccine itself because antigen is not available until de novo synthesis occurs. Both DNA and subsequent antigen expression persists for several weeks (4, 6). Thus, DNA-raised immune responses could occur as maternal antibody titers wane. Some groups have reported success following neonatal DNA immunization in the presence of maternal antibody (14), while others have failed (11, 15, 21, 25).

We have previously shown that intramuscular (i.m.) and gene gun (g.g.) immunization of mice as neonates or adults with an influenza hemagglutinin (HA)-expressing DNA generates long-lasting protective IgG responses (18). In this study, we address the ability of DNAs expressing HA and nucleoprotein (NP) to generate humoral and cellular responses in the presence of maternal antibody. Our results show an inhibition of DNA-raised antibody responses to HA that correlates with the amount of maternal antibody present at the time of immunization. However, the presence of maternal antibody did not affect the generation of antibody to NP or the generation of long-lived cellular immune responses to HA or NP.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Mice. BALB/c mice (Harlan Sprague-Dawley, Indianapolis, Ind.) were housed in microisolator cages at the Emory University Winship Animal Facility (Atlanta, Ga.). Six- to eight-week-old female mice were infected intranasally (i.n.) with a sublethal dose of influenza A/PR/8/34 and allowed to recover from infection. Approximately 3 months later, these influenza virus-immune mice, as well as naive females, were bred. Pregnant females were separated into individual cages and monitored daily for births. Birth dates were recorded as the dates the litters were discovered. Pups were weaned and sex separated at 3 to 4 weeks of age.

Plasmid DNA. pJW4303/H1 (HA DNA) and pCMV/NP (NP DNA) plasmid vector construction and purification procedures have been previously described (8, 17). Both vectors are under the transcriptional control of the cytomegalovirus (CMV) immediate-early promoter. The empty pJW4303 vector was used as a negative control. Plasmids were grown in either Escherichia coli DH5 or HB101 and purified using Qiagen (Chatsworth, Calif.) UltraPure-100 columns.

DNA immunizations. Twelve-week-old young adult mice were anesthetized with 0.03 to 0.04 ml of a mixture of 5 ml of ketamine HCl (100 mg/ml) and 1 ml of xylazine (20 mg/ml). i.m. DNA immunizations involved the injection of 0.04 ml of sterile 0.9% saline containing 50 µg of total DNA into a surgically exposed quadriceps muscle (17). One-day-old unanesthetized neonatal mice were injected with an equivalent DNA-saline injection combination into the gluteus maximus muscle. Surgical exposures were not performed in the neonatal animals. g.g. immunizations were performed on abdominal skin using the hand-held Accell gene delivery system as described previously (17). Adult mice were anesthetized, and abdominal skin was shaved with electric clippers. Neonatal mice were neither anesthetized nor shaved. Both groups of mice were immunized with a single g.g. dose containing a total of 2 µg of DNA per 0.5 mg of 1-µm gold beads (Bio-Rad, Hercules, Calif.) at a helium pressure setting of 400 lb/in². Neonatal g.g. immunization parameters were optimized prior to experiments to determine the proper target location and appropriate pressure for bead penetration into the epidermal skin layer (data not shown). The doses of DNAs given i.m. were as follows: 25 µg of pJW4303/HA plus 25 µg of pJW4303 (HA DNA), 25 µg of pCMV/NP plus 25 µg of pJW4303 (NP DNA), and 25 µg of pJW4303/HA plus 25 µg of pCMV/NP (HA + NP DNA). The doses of DNA given via g.g. were as follows: 1 µg of pJW4303/HA plus 1 µg of pJW4303 (HA DNA), 1 µg of pCMV/NP plus 1 µg of pJW4303 (NP DNA), and 1 µg of pJW4303/HA plus 1 µg of pCMV/NP (HA + NP DNA).

Detection of serum IgG by ELISA. At various times postimmunization, mice from each group were bled and individual mouse serum was assayed by standard quantitative enzyme-linked immunosorbent assays (ELISAs) to assess anti-HA- and anti-NP-specific IgG levels in immune serum, as described elsewhere (8). Data are shown as geometric mean titers (GMT).

Preparation and stimulation of splenocytes for cytokine production. Spleens were harvested from groups of immunized mice (n = 2 to 3) and pooled in p60 petri dishes containing ~4 ml of RPMI 10 medium (RPMI, 10% fetal bovine serum, and 20 µg of gentamicin per ml). All steps in splenocyte preparations and stimulations were done aseptically. Spleens were minced with curved scissors into fine pieces and then drawn through a 5-ml syringe attached to an 18-gauge needle several times to thoroughly resuspend the cells. Then cells were expelled through a nylon mesh strainer into a 50-ml polypropylene tube. Cells were washed with RPMI 10, and red blood cells were lysed with ACK lysis buffer (Sigma, St. Louis, Mo.) and washed three more times with RPMI 10. Cells were then counted by trypan blue exclusion and resuspended in RPMI 10 plus 80 U rat interleukin-2 (Sigma) per ml to a final cell concentration of 2 × 10⁷ cells/ml. Cells to be used for intracellular cytokine staining were stimulated in 96-well flat-bottom plates (Becton Dickenson Labware, Lincoln Park, N.J.), and cells to be used for cytokine analysis of bulk culture supernatants were stimulated in 96-well U-bottom plates (Becton Dickinson Labware, Lincoln Park, N.J.). Then, 100-µl portions of cells were dispensed into wells of a 96-well tissue culture plate to a final concentration of 2 × 10⁶ cells/well. Stimulations were conducted by adding 100 µl of the appropriate peptide diluted in RPMI 10. CD8⁺ T cells were stimulated with either a K^d-restricted HA peptide (IYSTVASSL) (26) or a K^d-restricted NP peptide (TYQRTRALV) (20). Negative control stimulations were done with media alone. Cells were then incubated as described below to detect extracellular cytokines by ELISA or intracellular cytokines by fluorescence-activated cell sorter (FACS) staining.

Detection of IFN- in bulk culture supernatants by ELISA. Pooled splenocytes were incubated for 2 days at 37°C in an humidified atmosphere containing 6% CO₂. Supernatants were harvested, pooled, and stored at 80°C until assayed by ELISA. All ELISA antibodies and purified cytokines were purchased from Pharmingen (San Diego, Calif.). Then, 50 µl of purified rat anti-mouse gamma interferon (IFN-) monoclonal antibody diluted to 5 µg/ml in coating buffer (0.1 M NaHCO₃, pH 8.2) was distributed per well of a 96-well ELISA plate (Corning, Corning, N.Y.) and incubated overnight at 4°C. Plates were washed three times with phosphate-buffered saline (PBS)-0.025% Tween 20 (PBS-T) and blocked with 250 µl of 2% dry milk-PBS for 90 min at 37°C. Plates were washed three times with PBS-T. Standards (recombinant mouse cytokine) and samples were added to wells at various dilutions in RPMI 10 and incubated overnight at 4°C for maximum sensitivity. Plates were washed three times with PBS-T. Biotinylated rat anti-mouse cytokine detecting antibody was diluted in PBS-T to a final concentration of 2 µg/ml, and 100 µl was distributed per well. Plates were incubated for 1 h at 37°C and then washed three times with PBS-T. Streptavidin-AP (Gibco BRL, Grand Island, N.Y.) was diluted 1:2,000 according to manufacturer's instructions, and 100 µl was distributed per well. Plates were incubated for 30 min and washed an additional three times with PBS-T. Plates were developed by adding 100 µl of AP Developing Solution (Bio-Rad, Hercules, Calif.) per well and incubating the mixtures at room temperature for 50 min. Reactions were stopped by addition of 100 µl of 0.4 M NaOH and read at an optical density at 405 nm. Data was analyzed using Softmax Pro version 2.21 computer software (Molecular Devices, Sunnyvale, Calif.).

Intracellular cytokine staining and FACS analysis. Pooled splenocytes were incubated for 5 to 6 h at 37°C in a humidified atmosphere containing 6% CO₂. A Golgi transport inhibitor, Monensin (Pharmingen, San Diego, Calif.), was added at 0.14 µl/well according to the manufacturer's instructions, and the cells were incubated for an additional 5 to 6 h (24). Cells were thoroughly resuspended and transferred to a 96-well U-bottom plate. All reagents (GolgiStop kit and antibodies) were purchased from Pharmingen unless otherwise noted, and all FACS staining steps were done on ice with ice-cold reagents. Plates were washed two times with FACS buffer (1× PBS, 2% bovine serum albumin, 0.1% [wt/vol] sodium azide). Cells were surface stained with 50 µl of a solution containing 1:100 dilutions of rat anti-mouse CD8-allophycocyanin (APC), CD8-CD69-phycoerythrin (PE), and CD16/CD32 (FcIII/RII; "Fc Block") in FACS buffer. Cells were incubated in the dark for 30 min and washed three times with FACS buffer. Cells were permeabilized by thoroughly resuspending them in 100 µl of Cytofix/Cytoperm solution per well and incubating them in the dark for 20 min. Cells were washed three times with Permwash solution. Intracellular staining was completed by incubating 50 µl per well of a 1:100 dilution of rat anti-mouse IFN--fluorescein isothiocyanate (FITC) in Permwash solution in the dark for 30 min. Cells were washed two times with Permwash solution and one time with FACS buffer. Cells were fixed in 200 µl of 1% paraformaldehyde solution and transferred to microtubes arranged in a 96-well format. Tubes were wrapped in foil and stored at 4°C until use (<2 days). Samples were analyzed on a FACScan flow cytometer (Becton Dickinson). Compensations were done using single-stained control cells stained with rat anti-mouse CD8-FITC, CD8-PE, CD8-TriColor, or CD8-APC. Results were analyzed using FlowJo version 2.7 software (Tree Star, San Carlos, Calif.).

Influenza virus A/PR/8/34 challenge. Metofane-anesthetized mice were challenged by i.n. inoculation of 50 µl of influenza virus A/PR/8/34 (H1N1) containing allantoic fluid diluted 10⁴ in PBS (50 to 100 50% lethal dose; 0.25 hemagglutinating unit). Mice were weighed daily and sacrificed following >20% loss of prechallenge weight. At this dose of challenge virus, 100% of naive mice succumbed to influenza virus infection by 4 to 6 days. Sublethal infections were done similarly using a 10⁷ dilution of virus.

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Maternal antibody inhibits IgG responses to HA but not to NP. Mice born to influenza virus-immune mothers and vaccinated on the day of birth were deficient in the generation of DNA-raised IgG to HA but not to NP (Fig. 1, Table 1). One-day-old mice born to naive mothers or influenza virus-immune mothers were immunized on the day of birth by i.m. or g.g. inoculation with HA, NP, or HA + NP DNA. At the time of DNA immunization, the GMT of anti-influenza virus maternal IgG in the neonates was 128.5 µg/ml. By 12 weeks of age, the titers had fallen to background levels. At 30 weeks postpriming, the neonates immunized in the presence of maternal antibody showed different patterns of IgG responses to HA and NP than those immunized in the absence of maternal antibody. In the maternal-antibody-positive groups, the highest antibody responses were seen for NP, the second highest antibody responses were seen for HA + NP, and the lowest antibody responses were seen for HA (Fig. 1A and B). In contrast, in the maternal-antibody-negative groups, the highest antibody responses were seen for HA plus NP, the second highest antibody responses were seen for HA, and the lowest antibody responses were seen for NP (Fig. 1A and B).

View larger version (29K): [in this window] [in a new window]   FIG. 1.   Effects of maternal antibody on DNA-raised IgG responses. Mice born to influenza virus-immune or naive mothers were immunized on the day of birth either i.m. (A) or g.g. (B) or as young adults either i.m. (C) or g.g. (D) with HA, NP, or HA-NP DNAs. Serum samples were tested over a course of 30 weeks for anti-influenza virus IgG levels by ELISA. Results are shown as the GMT for IgG for groups of 7 to 14 mice. The standard errors of the mean for the GMT at 30 weeks postpriming are given in Table 1.

Maternal antibody does not block the generation of long-lived CD8⁺ T-cell responses. We next determined if anti-influenza virus antibody was capable of preventing DNA-raised T-cell responses. Mice born to naive or influenza virus-immune mothers were DNA immunized as described for Fig. 1 i.m. or g.g. with HA- and/or NP-expressing DNA. One year later, mice were inoculated with a sublethal dose of influenza virus A/PR/8/34 to activate memory T cells. Mice were sacrificed 7 days later, and spleens from each group (n = 2 to 3) were pooled. Splenocytes were cultured for 10 to 12 h with K^d-restricted HA or NP peptides. The final 5 or 6 h of culture included monensin (Pharmingen), which inhibits protein transport and allows for the accumulation of intracellular cytokines. The cells were surface stained for CD8 and CD69 (an early activation antigen), permeabilized, stained for intracellular IFN-, and analyzed by flow cytometry.

View larger version (45K): [in this window] [in a new window]   FIG. 2.   Effects of maternal antibody on DNA-raised CD8+ T-cell responses, as determined by intracellular cytokine staining. Mice born to naive or influenza virus-immune mothers were immunized either i.m. (A) or g.g. (B) on the day of birth or in adulthood with HA and/or NP DNA or pJW4303 control DNA (C). One year later, all mice were inoculated with a sublethal dose of influenza virus. Mice were sacrificed 7 days later, and spleens from each group (n = 2 to 3) were pooled. Splenocytes were cultured for 10 to 12 h with a Kd-restricted HA or NP peptide as appropriate. The final 5 or 6 h of culture included monensin, which inhibits protein transport by the Golgi and allows for the accumulation of intracellular cytokines. The cells were surface stained for CD8 and CD69 (early activation antigen), permeabilized and stained for intracellular IFN-, and analyzed by flow cytometry. Data are shown as dot plots, with numbers in quadrants representing the respective percentage of the CD8+ cell population.

View larger version (31K): [in this window] [in a new window]   FIG. 3.   Effects of maternal antibody on the generation of DNA-raised CD8+ T-cell responses, as determined by IFN- production. Splenocytes from the same mice used in Fig. 2 were cultured for 2 days with the appropriate Kd-restricted HA or NP peptide. Bulk culture supernatants were harvested, pooled, and assayed for IFN- production by ELISA. Results are for mice immunized with HA plus NP DNA (A), HA DNA (B), or NP DNA (C).

Maternal antibody inhibits protection from influenza virus challenge. Maternal antibody inhibited the protective efficacy of DNA immunizations, with protection correlating with HA-specific IgG (Fig. 4, Table 3). All mice DNA immunized as neonates in the presence of maternal antibody succumbed to i.n. influenza virus challenge, as did mice immunized with the control DNA. In contrast, mice born to naive mothers and immunized with HA or HA + NP DNA were protected. Codelivery of HA and NP DNAs provided slightly better protection than HA DNA alone. NP DNA failed to protect any mice, regardless of the presence of NP-specific IgG and CD8⁺ T-cell responses.

View larger version (18K): [in this window] [in a new window]   FIG. 4.   Protection correlates with antibody to HA. Mice born to naive or influenza virus-immune mothers were immunized on the day of birth or in adulthood by i.m. or g.g. with HA plus NP (A), HA (B), or NP DNA (C). Thirty weeks later, groups of four to eight mice were challenged i.n. with a lethal dose of influenza A/PR/8/34 and weighed daily. Mice were sacrificed following >20% loss of prechallenge body weight. Results are shown as the percent survival at 7 days postchallenge versus the prechallenge GMT of IgG.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

Recent studies from our lab have shown that immunization of neonatal mice with influenza HA DNA generates strong, long-lasting, protective IgG responses (18). Here we extend these studies to evaluate the effects of maternal antibody on the efficacy of neonatal DNA immunizations. Our results show that maternal antibody inhibits the magnitude of IgG responses to HA but not to NP (Fig. 1, Table 1). They also show that maternal antibody does not interfere with the raising of CD8⁺ T-cell responses (Fig. 2 and 3, Table 2). Finally, we show that protection against a lethal challenge correlates with the presence of antibody to HA (Table 3, Fig. 4). Cellular immune responses in the absence of antibody to HA failed to protect against the highly virulent A/PR/8/34 influenza virus challenge (Fig. 4, Table 3).

Interestingly, the presence of maternal antibody at the time of DNA immunization inhibited the generation of antibody responses to HA more strongly than to NP (Fig. 1, Table 1). DNA-raised IgG responses to HA were reduced by at least 16-fold, whereas antibody responses to NP were not significantly affected by the presence of maternal antibody (Table 1). This was true both for g.g. (P < 0.001) and i.m. (P < 0.02) deliveries of DNA. This difference is unlikely to be due to the absence of maternal antibody to NP because approximately 40% of the total antibody response to influenza virus is directed against NP (3). When expressed in a DNA vaccine, the influenza virus HA localizes to the plasma membrane, whereas the influenza virus NP is intracellular. This suggests that maternal antibody inhibits the generation of IgG responses to plasma membrane proteins more strongly than to intracellular proteins.

The ability of maternal antibody to inhibit DNA-raised IgG responses to plasma membrane, but not intracellular proteins, is consistent with some, but not all, prior literature (Table 4). The one other study of a DNA-expressed intracellular protein, LCMV NP, is consistent with our findings with influenza virus NP (10; Table 1). Thus, in general, intracellular proteins may be protected against the blocking activity of maternal antibody. Five other studies have been done on the ability of maternal antibody to block humoral responses to DNA-expressed plasma membrane proteins (Table 4). These studies varied with regard to the ability to raise IgG in the presence of maternal antibody. Similar to our findings with influenza HA, maternal antibody blocked DNA-raised IgG responses to measles virus HA and pseudorabies virus gD. In contrast, maternal antibody failed to block DNA-raised IgG responses to herpes simplex virus gB, bovine herpesvirus gD or rabies virus glycoprotein (13, 14, 23, 25; Table 4). The differences in the ability of plasma membrane proteins to raise antibody may be due to differences in the levels of maternal antibody in the different model systems. This is suggested in the rabies virus studies, where the level of passive antibody affected the extent of blocking (25). A second, more interesting possibility would be that plasma membrane proteins differ in how antibody affects their interactions with the immune system. If this is so, plasma membrane proteins that are sensitive to blocking by maternal antibody may be able to be engineered for resistance.

In contrast to antibody responses, all DNA-expressed proteins have raised cell-mediated immunity in the presence of maternal antibody (Fig. 2 to 4, Tables 2 and 4). This ability to raise cellular immunity is independent of the ability to raise antibody (Table 4). These findings imply that maternal antibody does not block processing and presentation of DNA-expressed antigens.

In agreement with prior studies using the highly virulent A/PR/8/34 challenge, HA DNA immunization afforded protection whereas NP DNA immunization failed to protect against influenza challenge (Fig. 4 [5, 19]). The ability of HA to protect correlated with the level of anti-HA antibody (Fig. 4). Anti-HA antibody differs from anti-NP antibody or cell-mediated responses to HA or NP in that it can block virus entry. Thus, antibody to HA can limit the incoming infection. By contrast, cytotoxic T lymphocytes to HA and NP are unable to block infection but can play a role in the recovery from influenza virus infection. Other groups have shown a 50% greater survival rate than for controls using NP DNA immunizations (9, 22). These studies differed from ours in using multiple booster immunizations and challenging with a low dose (20% survival rate in nonimmunized controls) of the more attenuated A/HK/68 (H3N2) influenza virus (16, 22). In agreement with prior studies, coimmunization with HA + NP DNAs enhanced protective immunity (Table 3, Fig. 4 [2, 7]).