Dynamics of immunoglobulin G subclasses during the first two years of life in Malawian infants born to HIV-positive mothers

The process of immunoglobulins development and maturation starts during intrauterine life [1] however, the fetus can not produce IgGs, that are received from the mother in a complex mechanism of selective placental passage (preferential transport occurs for the IgG1 isotype followed by IgG4, IgG3 and IgG2 [2]. Neonates are therefore born with a functional immaturity of the immune system and early protection initially relies on the presence of maternal antibodies [3]. Only after the first months of life will infants start to produce their own IgGs, achieving the full immune competence only in late adolescence [4].

In maternal pathological conditions, such as infections and/or inflammatory status the bidirectional fetal-maternal immune cross-talk, including the passage of IgG from mother to fetus, can be altered with important consequences for offspring’s health [5, 6]. Clinical and epidemiological studies reported evidence that maternal HIV infection can deeply affect the maternal/fetal unit, interfering with the immunomodulatory factors which shape immune maturation in fetuses [7, 8]. Immunological abnormalities have been observed in HIV-exposed uninfected (HEU) children, including defects in CD4+ helper T cells and in immune regulatory function [9], and low responsiveness to vaccination [10]. In particular, maternal transplacental transfer of IgGs is inadequate in HIV-exposed children. In healthy pregnancies, full-term neonates have a cord blood IgG concentration often exceeding the maternal plasma concentration [11], but in HIV infection significant reduction of the IgG child/maternal ratio (CMR) has been observed [12]. Several studies have shown that HEU newborns have lower levels of Hib-, pertussis-, pneumococcus-, and tetanus-specific antibodies when compared to non-HIV exposed peers [13]. HIV studies on antenatal vaccine programs have also reported impaired passage through the placenta [14‐16].

However, while the decreased transplacental passage has been extensively demonstrated, only a few studies have investigated the subsequent development and maturation of total IgG and IgG isotypes in HEU infants. Immunoglobulins have a key role in the response against pathogens and in the development of adequate responses to vaccinations [17] and the determination of their levels can provide useful information on the status of the humoral immune system. IgGs ranges are well established in adult populations from different geographical areas [18], but the reference intervals are still uncertain in infants since many external factors, such as in utero stimuli, genetic and environmental influences, and exposition to pathogens, could impact on the dynamic process of immunoglobulin development and maturation [2, 11, 19]. Because of the limited number of studies reporting the dynamics of IgG levels in African children, there is a need for a better characterization of the immunoglobulin profile in these populations. The present study is therefore aimed to assess the IgG and IgG subclasses levels during the first 2 years of the life of Malawian infants born to HIV+ mothers.

In our cohort, the HIV-exposed uninfected infants were born with high IgG levels with an over-representation of the IgG1 and a low IgG2 concentration. At 2 years of life, the HEU infants still had high levels of total IgG and IgG1 and a substantial disproportion in isotype distribution with respect to reference intervals of same-aged infants. It has to be noted that, to contextualize our findings, the evaluation of IgGs maturation in our cohort was made using reference values from other studies on infants from different ethnic groups.

The longitudinal changes observed in our cohort in IgA and IgM, synthesized by the fetus from early intrauterine life [28], seem to present development within the physiological ranges reported by the literature.

In our study the HEU infants 1-month-old had higher total IgG and IgG1 levels with respect to reference intervals reported in other studies [19, 24‐27]; this finding is not surprising since the neonatal IgG profile reflects the maternal IgG levels and distribution. In HIV infection the decline of T-cell functionality is paralleled by the hypergammaglobulinemia and the polarization of immunoglobulins towards IgG1 subclass, indices of B-cells dysregulation [29, 30]. At the same age, however, IgG2 isotype in the HEU infants of our cohort was underrepresented. This finding can be attributable to the lowest affinity of IgG2 isotype with FcRn placental receptors (preferential transport: IgG1 > IgG4 > IgG3 > IgG2) that causes a selective and temporary impairment of IgG2 levels also in healthy neonates [2, 31]. The one-month-old HEU infants of our cohort had a more pronounced deficit with respect to the reference values. Recently, our group reported that the IgG2 deficit in HEU infants could also be attributable to the low levels of circulating IgG2 in HIV-positive pregnant women, suggesting that the low affinity with FcRn receptors of placenta together with maternal low levels could synergically contribute to the IgG2 deficit in neonates [32].

In the first 6 months of life, we observed a temporary decline in total IgG levels, associated with the waning of maternal antibodies, slowly replaced by the HEU infants IgG own production [33]. However, a different pattern was observed for the 4 IgG subclasses: while IgG1 levels showed a decline similar to that of total IgG, IgG2 and IgG3 isotypes increased, and the levels of IgG4 did not change significantly. This finding could be correlated to the role that maternal IgG could have on the neonate’s immune system maturation [34]. Many studies have indeed reported that for several vaccines, such as live attenuated, toxoid and conjugated vaccines, high titers of maternal acquired IgG inhibit the infant’s humoral immune response after infant vaccination [35]. We hypothesize that a similar mechanism could also affect the onset of the infant’s own IgG isotype production.

In our study, neither maternal viroimmunological conditions during pregnancy (CD4 count and ART duration), nor infant’s birth weight was predictive of HEU infants’ IgG maturation over time, but we cannot exclude that the small sample size could have affected the statistical analysis.

The trend we observed in IgGs development during 24 months in the HEU infants of this longitudinal study was not different from those observed in reference intervals reported in the literature; however, at 24 months the level of total IgG was almost double to the reference values, and IgG1 was overexpressed among the subclasses. IgG2 subclass levels at 24 months were slightly lower than the normality range but their proportion over the total IgG levels accounted for 7.4% of total IgG, corresponding to half of the proportion reported for healthy populations (12–19%) [19, 24, 25, 27]. The higher IgG3 levels found in our study with respect to the reference intervals were probably linked to their role in the immunological response to malaria, which is endemic in the area where the study took place [36]. In this view, it has to be considered that the exposition to endemic pathogens is considered one of the major force driving immunological maturation [37]; unfortunately, in our study, the lack of accurate diagnosis during the scheduled medical visits prevented us to determine the real impact of viral, parasitic and bacterial diseases on developing and maturation of immunoglobulin in these children.

Although this study was designed to determine the dynamics of immunoglobulins in HEU children, we had the opportunity to test the IgG levels in infants infected by HIV during breastfeeding. Two children were infected before the physiological initiation of IgG synthesis, and in both cases, at 6 months we found a significant hypergammaglobulinemia, with values similar to those found in HIV+ symptomatic adults [38], and a strong discrepancy between the total IgG levels and the sum of subclasses levels. The nephelometric analysis revealed for both infants that the sum of the subclass measurements accounted for only 41% of the total IgG, with a partial reduction of IgG1 subclass. The result was confirmed in both cases in a repeat test. Although we cannot rule out a technical problem (i.e. reduced interaction of IgG1 protein with the IgG1 antiserum), the finding could indicate a virus interference with the production of IgG subclasses by B cells in acute infection, determining an early impairment of the B cell compartment. The finding in our study could be considered an anecdotal observation however, this topic could be investigated in larger cohorts.

The major limitation of this study is the lack of an appropriate control group (i.e. age-matched children from the same geographical area) which does not allow us to draw definite conclusions on the levels that we found. The availability of reference values in Sub-Saharan African countries remains a challenge due to many factors, including organization aspects and lack of laboratory facilities. We are aware that the use of reference values from populations of different ethnic origin and exposed to different environmental factors, can be problematic [39], but we tried to minimize the problem using reference data selected from very recent studies, in which immunoglobulin determinations were performed using the nephelometric method, which is considered the gold standard for serum protein detection with a high standard of inter- and intra- laboratory precision and reproducibility.

The low number of subjects included is another limitation of this study, although we determined the intra-individual variation over time and observed very low inter-individual variability, that did allow us to be confident with the validity of our results.

Here we reported the dynamics of IgG development and maturation in HEU infants, suggesting that their IgG profile at 24 months may still present anomalies (mainly represented by hypergammaglobulinemia and low IgG2 levels) probably as a consequence of the early-life exposure to maternal HIV-related immune alterations.

In our study, performed before the adoption of the Option B+ strategy, we found a significant association between a low maternal CD4+ cell count and the levels of IgG in neonates, confirming the impact that compromised viro-immunological maternal conditions could have on the transplacental passage of IgG in neonates [2, 14, 40]. The women of this study received only a median of 10 weeks of ART during pregnancy, a duration that was inadequate to restore the HIV-related maternal immunological dysfunctions which can deeply impact neonates’ health. Future studies will determine if these anomalies in infants may be corrected by the adoption of the current strategy of universal and life-long antiretroviral therapy administration.