Previous Article | Next Article ![]()
Clinical and Diagnostic Laboratory Immunology, March 2005, p. 375-379, Vol. 12, No. 3
1071-412X/05/$08.00+0 doi:10.1128/CDLI.12.3.375-379.2005
Copyright © 2005, American Society for Microbiology. All Rights Reserved.
Division of Parasitic Diseases, National Center for Infectious Diseases, Centers for Disease Control and Prevention, U.S. Department of Health and Human Services,1 Atlanta Research and Education Foundation, Atlanta, Georgia,2 Center for Vector Biology and Control Research, Kenya Medical Research Institute,3 Ministry of Health, New Nyanza Provincial Hospital, Kisumu, Kenya,4 Roll Back Malaria, World Health Organization, Geneva, Switzerland5
Received 22 September 2004/ Returned for modification 22 October 2004/ Accepted 4 January 2005
|
|
|---|
4 to 12 months, and the difference was statistically significant for four of the seven epitopes (P < 0.0035). These findings suggest that PM can negatively influence the development of immune responses to malaria in infants. |
|
|---|
According to immunologic theories, early exposure to an antigen in utero could induce immunologic tolerance (19). It is evident from numerous studies that in utero exposure to malarial antigens occurs in fetuses born to mothers with PM (4, 10, 11, 30). Experimental studies conducted in neonatal mice have shown that exposure to peptides representing T-cell epitopes of circumsporozoite protein (CSP) leads to tolerance (21). Further, mice born to immune mothers fail to produce antibodies in response to vaccination with formalin-fixed malaria parasites, a result which has been attributed to immunosuppression mediated by maternal antibodies (7). In another study, cord blood lymphocytes from parasitized placentas, compared to nonparasitized placentas, have been found to produce low levels of gamma interferon (8). Altogether, these studies raise the possibility that in utero exposure to malaria can have important consequences for the development of immune responses, especially at early stages in an infant's life. In this study, we determined if PM could alter the development of antibody responses to seven P. falciparum epitopes in a cohort of infants born to PM-positive and PM-negative mothers.
|
|
|---|
|
View this table: [in a new window] |
TABLE 1. Morbidity outcome by placental malaria status and predicted mean of antibody responses to seven peptides in infants from PM-positive and PM-negative mothersa
|
These peptides were synthesized at the Biotechnology Core Facility, National Center for Infectious Diseases, Centers for Disease Control and Prevention. 9-Fluorenylmethoxycarbonyl chemistry was used to produce peptides. The peptides were 80 to 90% pure and used without further purification.
Antibody assays. Total immunoglobulin G (IgG) antibody levels were measured in a standard enzyme-linked immunosorbent assay technique (28). The plates were coated with 100 µl of individual peptides (10 µg/ml) in 0.01 M phosphate-buffered saline (PBS; pH 7.2) overnight at 4°C. Plates were blocked with 200 µl of 3% bovine serum albumin/well in PBS for 1 h at room temperature and washed with PBS containing 0.05% Tween 20. A test sample of 100 µl (diluted 1:100) was added per well in duplicate and incubated for 1 h at room temperature. Plates were washed three times with PBS-Tween 20 and incubated with 100 µl of 1:3,000-diluted peroxidase-conjugated goat anti-human IgG antibodies (Southern Biotech, Atlanta, Ga.). After a 1-h incubation, plates were washed five times and developed with 100 µl of 3,3',5,5'-tetramethylbenzidine for 10 min. Reactions were stopped, and plates were read at 450 nm with an enzyme-linked immunosorbent assay reader (Molecular Devices SpectraMax 340; Sunnyvale, Calif.). Pooled human negative control plasma (from North American adults who had no history of travel to regions where malaria is endemic) was included in each plate, and the optical density values from these control samples were used to subtract for the background optical density in the test plasma. The positive control plasma (pooled sera of clinically immune Kenyan adults) was titrated subsequently with twofold serial dilutions starting with 1:100 to generate a standard curve, and each dilution corresponded to a preassigned antibody unit value. By using this standard curve, the relative antibody unit in each test sample was determined.
Statistical analyses. Group comparisons assessing the differences in frequencies of morbidity outcomes between children from PM-positive and PM-negative mothers were done with SUDAAN (version 8.02; RTI, Research Triangle, N.C.) to adjust the standard errors for correlations of repeated measures within the data set for each child. In this analysis, chi-square tests of no association were conducted on children 0 to 12 months of age, and a two-sided P value of <0.05 was considered statistically significant.
All antibody data were standardized and normalized by logarithmic transformation before the analysis. Analyses to assess whether children from PM-positive or -negative mothers had different antibody responses were evaluated by using a linear mixed model with a random intercept and an adjusted standard error to account for the correlation of measurements for each child. Analyses were conducted at two time intervals (0 to 4 months of age and
4 to 12 months of age) for the seven antibody measurements. Predicted means were calculated for children born to PM-positive and PM-negative mothers.
Age was assessed as a predictor of antibody response, but the slope of age was not significantly different from zero. Similarly, low birth weight (LBW) was also assessed univariately as a predictor of antibody response and in a multiple regression with PM at the two time intervals. There was no correlation between antibody response and LBW (data not shown). Therefore, age and LBW were not included in the final model.
Lastly, we assessed the association of antibody levels and infant morbidity by using logistic regression with generalized estimating equations (GEE) to adjust the standard errors for correlation within the set of repeated measurements for each individual. In order to assess the association of a prior antibody measurement and a current morbidity outcome, we selected antibody measurements taken 14 and 34 days prior to the morbidity measurement and assessed the association with the odds of a child having that morbidity outcome. Similar analyses using GEE logistic regression were conducted for the current antibody measurement and the odds of a child having a morbidity outcome at the time of the antibody measurement. All observations were used in this analysis. All missing visit data was assumed to be missing at random.
For the antibody-related analyses which contained multiple comparisons, a two-sided adjusted Bonferroni P value of <0.0035 was deemed statistically significant. All data analyses were conducted by using SAS software (version 8.02; SAS Institute, Cary, N.C.).
|
|
|---|
Differences in the antibody responses throughout the first year of life.
The differences between infants born to PM-positive and PM-negative mothers in the monthly measures of mean IgG antibody levels to seven epitopes were compared (Fig. 1). It is apparent from Fig. 1 that, overall, infants born to PM-positive mothers had lower levels of antibodies to all of the epitopes, especially after the first 4 months of life. To exclude the potential confounding effect of maternally acquired antibodies, we compared the differences in PM status as stratified by infant age (summarized in Table 1). First, we analyzed the differences between the groups in the first four months of life, since maternal antibodies are still present and could largely account for total IgG at this age range. For infants 0 to 4 months of age, there was no significant difference in the overall mean antibody levels between the two groups, except that infants born to PM-positive mothers had significantly lower level of antibody responses to epitope EBA-175 (PL890) than infants born to PM-negative mothers. With infants aged
4 to 12 months, antibody levels to CSP (PL876), RAP-1 (PL885), and EBA-175 (PL890, PL893) epitopes were significantly lower in those from PM-positive mothers in comparison to those from PM-negative mothers (P < 0.0035). Although antibody responses to the remaining three epitopes showed the same trend, the differences were not statistically significant (P > 0.0035). In summary, it is evident that infants born to PM-positive mothers generated lower levels of antibodies to malarial epitopes than infants born to PM-negative mothers.
![]() View larger version (23K): [in a new window] |
FIG. 1. Comparison of IgG antibody responses to seven different P. falciparum epitopes in infants born to PM-negative and to PM-positive mothers. The mean log values of antibody units are reported.
|
|
|
|---|
Our finding is consistent both with earlier observations that maternal infection with other parasites during pregnancy can negatively influence the development of immune responses in young children (13, 21) and with rodent studies that demonstrated immunologic tolerance to CSP in neonatal mice (21). We have also found that infants born to PM-infected mothers had a higher risk of symptomatic malaria and anemia (mild and moderate) than infants born to uninfected mothers, which is consistent with a previous study (14) and with earlier findings from the same cohort as that used in the present study (29). However, it is important to point out here that although our findings may casually suggest that a lowering of antibody response to malarial epitopes may have contributed to the increased risk of malaria in infants of maternally infected mothers, we do not have definite evidence to confirm such a hypothesis. In this context, it is also noteworthy that we have not measured antibodies to variant surface antigens that have been implicated in protective immunity (12, 18). In the placentae of infected mothers, the predominant parasites are mainly chondroitin sulfate A binding (5), and it would be interesting to test whether infants born to these mothers have different levels of variant surface antigen-specific antibodies compared to infants of PM-negative mothers.
Different immunologic explanations for the present findings can be offered. According to current immunologic theory (19), early fetal priming of an antigen to the immune system can induce tolerance. Prenatal exposure to microfilariae has been shown to induce immune tolerance as manifested by impaired cellular and humoral responses to filarial antigens in experimental rat studies (6). In the same study, it was also shown that the induction of tolerance due to prenatal exposure allowed the development of the infective larvae to maturity in Fisher rats, animals which are otherwise innately resistant to this parasite (6). Subsequent human studies have also confirmed that prenatal exposure to filarial parasites resulted in filarial antigen-specific impairment of immune responses, especially cellular responses (27). Previous studies have shown that PM results in the exposure of the fetus to malarial antigens and leads to prenatal sensitization to malarial antigens, as demonstrated by the presence of malarial antigen-specific IgM antibodies in cord blood (4, 10, 11, 21, 30). Given the particular susceptibility of the fetus to tolerogenic signals during prenatal life, it is conceivable that an immune tolerance theory could at least partially explain the results of the present study.
It had been shown previously that persisting maternal malaria antibodies can affect the development of antibodies in neonatal mice by generating suppressor T cells (7). We do not know if such a mechanism could explain our findings. However, it is important to point out that several studies have shown that persisting maternal antibodies to various pathogens can interfere with the development of pathogen-specific antibody responses in the offspring after vaccination (24). Recently, Stanisic et al. (26) have shown that the presence of Plasmodium yoelii MSP-119kDa-specific maternal antibodies inhibited the pups from developing antibody responses to this antigen after immunization. This inhibition was specific to antibody development, and it did not affect T-cell responses. It has been proposed that the presence of maternal antibodies can mask the immunodominant epitopes from B-cell recognition by binding to the antigen. At the same time, the resulting antigen and antibody complexes may be taken up by the antigen-presenting cells for processing and presentation to prime T cells. Such events may allow the activation of T cells but prevent generation of antibody by B cells (24, 26). Given that maternal antibody levels to malarial epitopes were at similar levels in the first few months of life in infants born to PM-positive and PM-negative mothers, it is unlikely that the suppressor cell (7) or epitope hypotheses (24) could adequately explain our findings. However, further studies are necessary to confirm our results in other settings of malaria endemicity and to explain why maternal exposure to malaria negatively influences the generation of antibody responses to malarial epitopes.
In the present study, we did not find any significant association between antibodies to selected epitopes from CSP, MSP-2, RAP-1, and EBA-175 antigens and protection against clinical malaria. Since this study involved active monthly follow-up and treatment for malaria, there were only a few cases of symptomatic malaria and malaria-related morbidity in the subcohort, and this result might have limited our ability to assess protection. We would like to point out that this study was not designed to assess the protective effect of antibodies to these epitopes; such studies would require large sample sizes. Further studies to fully assess the impact of PM on the development of cellular immune responses and the mechanistic basis of diminished antibody responses are required.
|
|
|---|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»