Influenza viral infection can cause serious illness among young infants 0-6 months of age. However, inactivated influenza vaccine is not recommended for this age group but pregnant women can be vaccinated during 2nd - 3rd trimester to induce passive immunization of their infants. Nevertheless vitamin A deficiency is highly prevalent among pregnant women in Bangladesh, >50% pregnant women consume less vitamin A than the recommended level. Given the fact that both clinical and sub-clinical vitamin A deficiency impair vaccine specific immunity, in this proposed study, we aim to investigate whether maternal vitamin A supplementation improve influenza vaccine specific immune responses among pregnant women and the passive protection of their infants.

In a placebo controlled clinical trial, sixty six mothers will be randomly assigned to receive either 10,000 IU vitamin A or placebo capsules weekly from second trimester to 6 month postnatal period. At 26-28 weeks of gestation, all mothers will be vaccinated with inactivated, trivalent influenza virus vaccine. Maternal and cord blood will be collected for vitamin A and influenza virus specific IgG assessment. Colostrum and breast milk at 6-month will be collected for vitamin A and influenza virus specific secretory IgA assessment. Venous blood (2-3 ml) will be obtained from all infants at the age of 6 months for vitamin A and influenza virus specific IgG assessment as well as infants' nasal swab for influenza virus specific secretory IgA.

Influenza morbidity affects the entire population and has an enormous impact upon the economic burden and the health care system, particularly causing serious illness among young infants 0-6 months of age. Recent studies show high rates of influenza illness and hospitalization among infants at this age group in industrialized countries (Chiu et al., 2002; Poehling et al., 2006) and influenza related childhood mortality is highest in young infants before 6 months of age (Bhat et al., 2005). Unfortunately, inactivated influenza vaccine is not recommended for this age group, but pregnant women can be vaccinated during 2nd - 3rd trimester. Thus the strategy of passive immunization (from mother) of infants is the effective way to combat this infection among young infants. This approach adopted in US since 1997 and recommended by WHO since 2005.

Upper respiratory tract is the portal of virus entry and primary site of replication (Gluck et al., 1994; Muszkat et al., 2000). Thus the importance of the presence of effective mucosal antibodies at the entry site of virus is essential to neutralize the virus and prevent infection (Meitin et al., 1994). Secretory IgA (sIgA) is the primary mucosal antibody, accounting for ~70% of the body's total antibody production (Brandtzaeg, 1994). sIgA has been shown to mediate nasal anti-influenza mucosal immunity (Renegar & Small, 1991) and therefore indicates the importance of vaccination strategies that trigger specific antibody production at mucosal surfaces as well as the importance of dietary nutrients to trigger the release of mucosal neuropeptides necessary for maintenance of the mucosal immune system (Renegar et al., 2001). A number of clinical trials examined the effect of vitamin A supplementation in humans on indicators of mucosal immunity. Supplementation to human immunodeficiency virus infected pregnant women is associated with improved gut permeability in their infants at 14 weeks as measured by lactulose/mannitol (L/M) urinary excretion test (Filteau et al., 1999). While vitamin A supplementation during pregnancy (Semba et al., 1999) or the early postpartum period (Filteau et al., 2001) show no effect on total sIgA in breast milk, rather vitamin A in milk has been shown to inhibit the growth of both enveloped and non-enveloped viruses (Clarke & May, 2000). However, total sIgA level of vitamin A-sufficient children has been detected significantly higher than that of vitamin A-deficient children (Lin et al., 2007). In animal model, vitamin A supplementation increases both total and antigen specific sIgA-containing cells in the mucosa, which in turn improves the survival rate (Nikawa et al., 2001) During early infancy breast milk is the predominant source of vitamin A and to date no study investigates the effect of prenatal vitamin A supplementation and vaccination on the vaccine specific mucosal sIgA responses in offspring. Although maternal IgA is not transplacentally transferred to the infants, but increased vitamin A and vaccine specific sIgA in breast milk as a result of maternal vitamin A supplementation may enhance infant's nasal sIgA responses that in turn may provide improved passive protection among young infants.

In our recent study (Mother'SGift Project, funded by Thrasher Research Fund), we have detected that maternal influenza vaccination can reduce influenza illness substantially in young infants up to 6 months of age in Bangladesh (Zaman et al., 2008). On the other hand the prevalence of vitamin A deficiency among pregnant women in Bangladesh as well as in other developing countries is very high. More than half of the pregnant women in Bangladesh consume less vitamin A than the recommended dietary allowance and have low vitamin A status (serum retinol <1•05 μmol/L) with approximately one-fifth of them classified as deficient (serum retinol <0•70 μmol/L) (Ahmed et al., 2003; Lee et al., 2008). Nonetheless, there is conclusive scientific evidence suggesting several aspects of both innate and adaptive immunity are compromised by clinical and subclinical vitamin A deficiency (Stephensen, 2001). In another study, we observed a significant positive association between primary yellow fever viral vaccine specific lymphocyte blastogenesis / antibody responses and body vitamin A store (Ahmad et al., 2008). Thus in view of the potential immune regulatory roles of vitamin A, it is of interest to investigate whether maternal vitamin A supplementation along with influenza vaccination could be used as public health intervention strategy to combat both influenza infection and vitamin A deficiency simultaneously among mothers and young infants in developing countries.

Research Design and Methods:

Study Site: This study will be carried out at ICDDR,B; Dhaka, Bangladesh. Three urban maternity clinics located at Mirpur area of Dhaka city will be selected for approaching mothers who visit the clinic in the first trimester and who are the residence of the nearby community.

Study Subject: The inclusion criterion is to enroll mothers at the beginning of second trimester (i.e. approximately 12 weeks of gestation), willing to stay in Dhaka during pregnancy and willing to admit in the clinic at delivery. The gestational age will be determined by self-reported LMP, which is likely be underestimated in most cases. A two-week variation will be acceptable in this proposed study. The exclusion criteria are: history of systemic disease, previous complicated pregnancies or of pre-term delivery, abortion, congenital anomaly, and hypersensitivity to influenza vaccine or receipt of the vaccine.

Overview of Experimental Design: This is a placebo-controlled, double-masked and randomized trial. The overall study design is outlined in Figure 1, timeline and hypothetical duration of the proposed protocol related activities are illustrated in Table 1. To allow scheduling of the sample collection, storage and assay with the number of hired field attendants and lab facilities that will be available, we plan to enter approximately five pregnant mothers per week into this study in the first three-month period of the study (Table 1). Informed written consent will be obtained from the mothers. Sixty-six pregnant women will be randomized to receive either weekly 10,000 IU vitamin A or placebo (corn oil) capsules and blood will be collected for serum vitamin A assessment. Randomization will be stratified by three clinics using balanced block design. This weekly supplementation will continue until 6 months postpartum. The supplementing vitamin A dose to be used in this study is similar to the doses used weekly among pregnant women in other studies (Semba et al., 1999; Cox et al., 2006) and shown to be safe. Field workers will deliver the supplements to mothers, in person and will directly observe their consumption. At 26-28 weeks of gestation, all mothers will receive inactivated, trivalent influenza virus vaccine (GlaxoSmithKline; Dresden, Germany). Standard inactivated trivalent influenza A/H3N2, A/H1N1, and B strains as recommended by WHO for the Northern Hemisphere will be obtained from commercial sources for this project. Study physician will administer the vaccine intramuscularly. At delivery, cord blood and colostrum will be collected for Influenza specific IgG and sIgA assessment respectively as well as vitamin A concentration. Birth weight and length of the newborn babies will also be recorded. At 6-month, breast milk will be collected for vitamin A and influenza virus specific sIgA assessment and venous blood (2-3 ml) from infants (6 mo of age) will be obtained for vitamin A and influenza virus specific IgG assessment as well as infants' nasal washings for influenza virus specific sIgA assessment. Our proposed study will not hamper the existing prenatal care for the mothers such as folate/iron supplementation or existing infant vaccination program in Bangladesh.

Randomization: Each mother will be given a unique randomly generated identification number, which will indicate the study group she is assigned to. The same ID will be labeled on the bottle containing 60 capsules. Field attendants will use these capsules for weekly supplementation ensuring the ID no of the mother and bottle ID are same. A third party (ACME laboratories, Dhaka Bangladesh) will prepare and supply vitamin A or placebo capsule in bottles, labeled with ID numbers from a list of numbers. The responsible personnel will keep it confidential until the end of study

Sample collection: Peripheral blood from mother, cord blood, colostrum, breast milk, peripheral blood and nasal washings from infants will be collected as described in Figure 1. Nasal secretions will be obtained by introducing wicks (cotton applicators) into the nostril and gently rotating for 90s as described previously (Wright et al., 1996). The wicks will be extracted into 1mL PBS, centrifuged, and the supernatants will be collected. All samples will be frozen until tested simultaneously.

Serological evaluation: We have chosen to investigate the humoral immune responses only to one influenza antigen because the purpose of this proposed study is to evaluate the effect of maternal vitamin A supplementation on the infants' passive antibody levels, thus it is prudent to investigate immune responses to one antigen. The following parameters of vitamin A and Influenza A specific humoral immunity will be tested.

Vitamin A assessment: Vitamin A content in plasma, serum, colostrum and breast milk will be measured by standard HPLC method in place in the Nutritional Biochemistry Lab of LSD.

Hemagglutination inhibition test [HAI]: HAI antibody titers will be measured by standard microtiter assay (Admon et al., 1997). 12 HA units of the selected Influenza A vaccine strain will be added to diluted pre-treated sera with receptor destroying enzyme (RDE, Sigma MO) and heat inactivation. Following incubation, washed chicken erythrocytes will be added until sedimentation ensues in the control wells. HAI titer is the inverse of highest dilution of serum exhibiting complete HAI.

Influenza-specific IgG and sIgA ELISA: Influenza virus A specific IgG in serum/plasma and sIgA in colostrum, breast milk and infant nasal swab will be analyzed by ELISA as described previously with minor modification (Bruhl et al., 2000; Greenbaum et al., 2004). Briefly immunoplates will be coated with purified influenza antigen obtained from NIBSC at 2 ug/mL for overnight at 40C. and then diluted samples will be added for another overnight incubation. HRP-conjugated anti-human IgG (BD-Pharmingen, CA) or mouse anti-sIgA, peroxidase-labeled (Alpco Diagnostics, NH) will be used as conjugates. To construct standard curve, a serial dilution of pooled samples assigned with an arbitrary value will be introduced in each ELISA plate. Total sIgA in colostrum, breast milk and infant nasal swab will also be determined by EIA kit (Alpco Diagnostics) to adjust influenza A specific sIgA response with respect to overall sIgA secretion.

Inclusion Criteria:

• mothers at the beginning of second trimester (i.e. approximately 12 weeks of gestation)
• willing to stay in Dhaka during pregnancy and willing to admit in the clinic at delivery
• the gestational age will be determined by self-reported LMP, which is likely be underestimated in most cases. A two-week variation will be acceptable in this proposed study

Exclusion Criteria:

• history of systemic disease
• previous complicated pregnancies or of pre-term delivery
• abortion
• congenital anomaly
• hypersensitivity to influenza vaccine or receipt of the vaccine