Evidence for in utero sensitization to Ascaris lumbricoides in newborns of mothers with ascariasis

Abstract

Human infections with Ascaris lumbricoides may have important effects on allergy and susceptibility to infectious diseases that start in early life. To investigate if sensitization to Ascaris occurs in utero, we measured IFN-γ and IL-4 responses in Ascaris antigen-stimulated cord blood of newborns of infected and non-infected mothers using flow cytometry. There was evidence of elevated frequencies of IFN-γ and IL-4-expressing CD4+ T cells in newborns of infected mothers compared to those of non-infected mothers. Our data provide evidence of in utero sensitization to A. lumbricoides, and raise the possibility that the immunological effects of infection start in the fetus.

Introduction

Illness caused by the geohelminth or intestinal helminth parasites, Ascaris lumbricoides, Trichuris trichiura, and hookworm is recognized as a major cause of disability in resource poor regions of the tropics where they are estimated to infect approximately 2 billion humans.¹

Infections with geohelminths including A. lumbricoides, the most prevalent geohelminth, may influence both the development of allergy,¹ and susceptibility to helminth and non-parasitic infectious diseases of childhood.^2-4 Such effects are likely to be immunologically mediated, long-lasting, and programmed in early life – either in utero through trans-placental exposures from maternal infections or in early childhood by actual infection.¹

Evidence for in utero sensitization to A. lumbricoides infection would indicate that the induction of host immune responses starts before birth, while the absence of sensitization would be suggestive of the acquisition of immune responsiveness later in life.

There is strong evidence that maternal infections with filarial and schistosome helminth infections have profound effects on immunity to helminths and other pathogens.^2,4,5 For example, BCG-vaccinated children in Kenya born of mothers infected with filarial or schistosome parasites had significantly decreased IFN-γ-responses to purified protein derivative than children of mothers who were not infected.² Similarly, newborns exposed in utero to filarial infection but without evidence cord blood T cell reactivity to filarial antigen and presumed to be ‘tolerized’, had a 12-fold greater risk of later filariasis infection compared to newborns of uninfected mothers.⁴ To our knowledge, there is no published evidence that maternal infections with A. lumbricoides, alters the immune response in utero, even though this helminth is estimated to infect over 1 billion people.

To investigate possible in utero effects on the immune response by maternal A. lumbricoides infections, we compared cord blood CD4+ T cell responses in newborns of A. lumbricoides-infected and non-infected mothers.

Materials and Methods

Study design and subject recruitment

The study was designed to address the specific question of whether the presence of maternal A. lumbricoides infection may sensitize the fetus in utero to A. lumbricoides antigens. To investigate this, we evaluated the effects of stimulation of cord blood with A. lumbricoides antigens on the intracellular expression of a typical Th1 cytokine, IFN-γ, and a typical Th2 cytokine, IL-4. The design was case-control with cases being newborns of mothers infected with A. lumbricoides and controls being newborns of mothers with no evidence of geohelminth infection. Mothers were recruited between June and July 2006 in the maternity department of the Hospital ‘Padre Alberto Buffoni in the town of Quinindé in Esmeraldas Province. Inclusion in to the study was based on the mother providing 1-2 stool samples around the time of birth, informed written consent, vaginal delivery, gestation of 34 or more weeks, and negative alkaline denaturation tests performed at the time of sample collection to detect contamination of cord blood with maternal hemoglobin. The study protocol was approved by the Ethical Committee of the Hospital Pedro Vicente Maldonado, Ecuador.

Data and sample collection

A questionnaire was administered to the mother to obtain information on relevant variables including mother's educational level, monthly household income, and obstetric history. Maternal stool samples were examined using the modified Kato-Katz and formol-ether concentration methods.⁶ Maternal and cord blood were collected into plastic tubes (BD Vacutainer, Plymouth, UK) containing sodium heparin. Cord blood was collected by cannulation of the umbilical vein after the newborn was delivered but before delivery of the placenta.

Blood culture and flow cytometry

Blood samples were processed within 4 hours of collection as described previously.⁶ Briefly, blood was diluted in an equal volume of culture medium, and incubated in the absence of antigen or in the presence of PBS-soluble antigen extract of A. lumbricoides adult female worms⁶ at 10 μg/mL, and Staphylococcus enterotoxin B (SEB; Sigma Aldrich, St Louis, MO) at 10 μg /mL in a humidified atmosphere of 5% CO₂ at 37°C. After 2 hours, 20 μg/mL of Brefeldin A (Sigma) was added to each culture well and cultures were incubated for a further 12 hours. After incubation, stimulated whole blood was resuspended in BD Lysing solution (BD Pharmingen), fixed, cryopreserved and stored in liquid nitrogen. The cells were transported to the USUHS on dry ice, where they were prepared for flow cytometry. Cells were stained with fluorescein isothiocyanate (FITC)–conjugated mouse anti-human CD3 (Immunotech), Peridin Clorophylla Protein-conjugated (PerCP) mouse anti-human CD4 (Pharmingen), R-phycoerythrin (PE)-conjugated mouse anti-human IL-4 (Pharmingen) and allophycocyanin (APC)-conjugated rat anti-human IFN-γ (Pharmingen). Data were collected on a LSRII flow cytometer (BD Biosciences) and analyzed using CellQuest software (BD Biosciences).

Total IgE and Ascaris-specific antibodies

Levels of Ascaris-specific IgE were measured using the CAP system (Pharmacia Diagnostics), and levels of polyclonal IgE, and A. lumbricoides-specific IgG, IgG1, and IgG4 were measured as described.⁶ The lower detection limit of the polyclonal IgE assay was 0.2 IU/mL. Positive values for the A. lumbricoides IgG1 and IgG4 assays were defined as greater than 3 standard deviations above the mean value obtained from sera collected from 10 non-infected individuals living in the town of Quinindé, Esmeraldas Province.

Statistical analysis

Continuous and categorical variables were compared between infection group using the ranksum and chi-squared tests, respectively. Paired comparisons (i.e. mother vs. newborn) used Wilcoxon signed rank test. All analyses were performed using Stata 10 (Statacorp, College Station, TX).

Results

Population characteristics

Data from fourteen cases and fourteen controls are presented. Relevant baseline and confounding variables for newborns and mothers are shown in Table 1; cases and controls were balanced for these factors. All mothers in the case group were infected with A. lumbricoides and none of the control mothers had evidence of geohelminth infection. Other maternal geohelminths detected among cases were T. trichiura (86%) and hookworm (29%).

Table 1.

Baseline and immunological characteristics of newborns of A. lumbricoides-infected and non-infected mothers. Values are frequencies (separated by /), percentages (%), and median (range in brackets). Geohelminth infections intensities represent median eggs per gram of stool (range).

Characteristic	Infected (N=14)	Non-infected (N=14)
Race/Ethnicity
Mestizo/Afro-Ecuadorian/Amerindian	10/3/1	10/3/1
Gender
Male/Female	8/6	6/8
Maternal age, years	28 (16-30)	24 (16-36)
Birth weight, kg	3,350 (2.400-4,800)	3,300 (2,800-4,000)
Gestational age, weeks	40 (37-42)	40 (34-42)
APGAR at 5 minutes	9 (8-9)	9 (8-10)
Monthly family income (US$)	150 (80-450)	120 (30-400)
Maternal educational level
% completed primary	79%	86%
Crowding (persons/household)	6.5 (4-12)	5 (2-12)
Maternal infection status
Any	100%	0%
A. lumbricoides	100%	0%
Infection intensity	6,650 (70-28,280)	0 (0-0)
T. trichiura	79%	0%
Infection intensity	630 (70-1,890)	0 (0-0)
Hookworm	29%	0%
Maternal antibodies
Total IgE (IU/mL)	742 (33-3,029)	448 (28-3,494)
Ascaris-specific IgE (kU/L) [%]	2.4 (0.4-12.8) [78.6%]	(0.4-4.1) [57.1%]
A. lumbricoides-specific IgG1	2 (1-445) [100%]	1 (0-73) [86%]
A. lumbricoides-specific IgG4	5 (0-533) [93%]	1 (1-107) [100%]
Cord blood antibodies
Total IgE (IU/mL) [%]	0.2 (0.2-32.7) [43%]	0.2 (0.2-0.5) [36%]
Ascaris-specific IgE	<0.35 [0%]	<0.35 [0%]
A. lumbricoides-specific IgG1	2 (1-541) [100%]	4 (0-155) [92%]
A. lumbricoides-specific IgG4	2 (0-455) [57%]	2 (0-167) [71%]
Frequencies of cytokine-expressing
CD4+ T cells*
Newborn
IL-4	20 (2-320)	13 (5-156)
IFN-γ	70 (19-4,120)	31 (4-651)
Ratio IFN-γ/IL-4	7.2 (0.4-118.9)	2.9 (0.3-8.1)
Maternal
IL-4	38 (8-1,366)	12 (2-524)
IFN-γ	63 (13-313)	29 (2-218)
Ratio IFN-γ/IL-4	1.3 (0.1-11.3)	1.8 (0.1-18.2)

^*Frequencies of cytokine-expressing CD4+ T cells (per 10⁵ lymphocytes) after stimulation with A. lumbricoides antigen

Total IgE and A. lumbricoides-specific antibodies in maternal and cord blood

All mothers had evidence of specific of IgG1 or IgG4 antibodies to A. lumbricoides and most had detectable levels of A. lumbricoides-specific IgE (>0.35 kU/L) indicating that exposure to ascariasis in this study population is universal (Table 1). Levels of polyclonal IgE were elevated also in both groups of mothers. Similar proportions of case and control newborns (43% vs. 36%) had detectable levels (>0.2 IU/mL) of polyclonal IgE, but none had detectable levels of Ascaris-specific IgE (>0.35 kU/L). Of the 28 newborns included in the study, only one newborn did not have detectable levels of A. lumbricoides IgG1 in cord blood and the majority had A. lumbricoides IgG4 antibodies. Cord blood IgG antibodies are likely to be of maternal origin. There were no significant differences by maternal infection status of A. lumbricoides IgG1 and IgG4 antibodies in cord blood (measured either as antibody levels or proportions with detectable antibodies).

Frequencies of CD4+ T cells expressing intracellular IFN–γ and IL-4

The frequencies of CD4+ T cells expressing IFN-γ was greater in blood stimulated with A. lumbricoides antigen compared to medium in samples from newborns of A. lumbricoides-infected mothers (P=0.005) and no significant difference was observed for newborns of non-infected mothers (P=0.31). Significantly greater frequencies of CD4+ lymphocytes expressing IFN-γ (P=0.02) and IL-4 (P=0.01) were observed in cord blood from newborns of infected compared to non-infected mothers (Table 1 and Figure 1).

Figure 1.

Frequencies of CD4+ T cells expressing IL-4 and IFN-γ in cord blood samples from newborns of A. lumbricoides infected (shaded boxes) and non-infected (clear boxes) mothers. Shown are box plots with median values (central line), inter-quartile range (box margins), 95% confidence intervals (bars), and outlying values (circles). The y-axis represents the number of cytokine-expressing CD4+ T cells per 100,000 lymphocytes.

Discussion

The data from the present study has provided evidence for antigen-specific T cell responses to A. lumbricoides antigens in utero suggested by the observation of increased frequencies of antigen-specific IFN-γ and IL-4-expressing CD4+ T cells in cord blood from newborns of A. lumbricoides-infected compared to those of non-infected mothers.

Although the data were collected during a small case-control study, we were able to show statistically significant effects between newborns of infected and non-infected mothers for the two cytokines measured. The fact that mothers were recruited from the same maternity department over a short period of 2 months, the analyses were performed blind to maternal infection status, and the newborns were balanced with respect to potential confounders, make the data more compelling. The absence of detectable anti-Ascaris IgE in cord blood in the newborns of uninfected mothers and negative alkaline denaturation tests makes significant admixture of maternal and fetal blood unlikely. We measured CD4+ T cell responses because a previous study has shown that cytokine production by helminth antigen-stimulated cord blood lymphocytes is primarily dependent on CD4+ T cells.²

Maternal A. lumbricoides infections may sensitize fetal immunity either by passage of parasite antigens or maternal anti-idiotype IgG antibodies across the placenta. A. lumbricoides antigens can be detected in the peripheral blood of infected individuals⁷ even though adult parasites are confined to the intestinal lumen. The absence of anti-Ascaris IgE in cord blood samples may indicate a lack of appropriate signals for IgE production by fetal B cells. The fetus is capable of producing IgE from the eleventh week of gestation.⁵ Previous studies have indicated that maternal schistosomiasis,^2,5 lymphatic filariasis,^2,5,8 and hookworm infection⁹ but not ascariasis¹⁰ are associated with the presence of parasite-specific IgE in cord blood. Analysis of an additional 86 cord blood plasma samples from newborns of 43 Ascaris-infected and 43 non-infected mothers at the same Hospital failed to detect anti-Ascaris IgE in any of the samples (authors' unpublished data). Adult filarial and schistosome parasites live in the lymphatics and venous system, respectively, and greater quantities of antigen from these parasites may reach the placental circulation compared to adult geohelminths. Similarly, greater quantities of antigen are likely to be released into the circulation by the large quantities of filarial larvae and schistosome eggs that are typically present in infected individuals compared to the larvae of A. lumbricoides or hookworm undergoing systemic migration. For this reason, it is more likely that maternal filarial and schistosome parasites will sensitize the fetus for specific IgE. The single study that reported hookworm-specific IgE in newborns of infected mothers appeared to measure only low-titer antibodies.⁹ The discrepant findings between newborns of Ascaris and hookworm-infected may be explained by parasite-specific effects, false positive serologic reactions, or the inability to detect very low titer IgE by the assay we used (detection limit of 0.35 kU/L).

Previous studies have shown that cord blood lymphocytes from newborns of helminth- infected mothers produce substantial amounts of IFN-γ⁸ but also IL-4² and IL-5.^2,9 In the present study, we observed a possible Th1-skewing of the A. lumbricoides-specific response - suggested by higher frequencies of CD4+ T cells expressing IFN-γ compared to IL-4 and the absence of anti-Ascaris IgE - among newborns compared to their mothers. Previous studies of newborn infants of mothers with schistosomiasis showed that skin testing with parasite extracts did not induce immediate hypersensitivity responses but did induce delayed type hypersensitivity and migration inhibition responses that may suggest strong anti-helminth Th1 responses at birth.^11,12

The risk of allergic disease is low in areas of the rural tropics where geohelminth infections are highly prevalent and it has been suggested that these infections may suppress allergic inflammation.¹ Early exposures and infections with helminths may be important in mediating these effects and such a hypothesis is supported by the observations of early effects on the risk of allergy of maternal helminth¹³ and infant infections.¹⁴ It has been suggested that these suppressive effects may be mediated by IL-10,¹⁵ a cytokine not measured in the present study.

In addition to effects on allergy, sensitization of the fetus to maternal ascariasis may have other important consequences including: 1) enhanced or reduced susceptibility to childhood infections with A. lumbricoides depending on the type of immune response generated in the newborn;⁴ 2) increased susceptibility to other infectious diseases such as HIV³ and tuberculosis;² and 3) possible effects on morbidity caused by ascariasis or other helminth infections.

The data from the present study provide evidence for specific sensitization to A. lumbricoides antigens in newborns of infected mothers. We demonstrated increases in the frequencies of CD4+ T cells expressing IFN-γ and IL-4 after in vitro stimulation with Ascaris antigens. The findings suggest that the immune modulatory effects on host immunity attributed to A. lumbricoides infection may start in utero. An understanding of how these parasites mediate such regulatory effects may have important implications for our understanding of the regulation of inflammation in childhood in the rural tropics.