Developmental immunotoxicity of early-life arsenic exposure

Arsenic exposure, mainly via drinking water and certain food, is a major public health concern because of its association with various types of cancers and other non-cancer adverse health effects. Worldwide, more than one hundred million people are exposed to elevated levels of arsenic on a regular basis. Bangladesh is one of the most severely affected countries, due to the natural occurring arsenic contamination in ground water that is used as drinking water. Although numerous studies have focused on investigating arsenic carcinogenicity, growing evidence indicates that arsenic may also affect the immune system. However, the specific effects of arsenic on immune function, particularly on the developing immune system in humans, are poorly understood. The overall aim of the present thesis was to elucidate potential effects of arsenic exposure on the developing immune system, including underlying mechanisms and critical windows of exposure. The four papers included in the thesis were nested into a randomized, community-based food and micronutrient supplementation trial, the Maternal and Infant Nutrition Interventions, Matlab (‘MINIMat’) trial, conducted in Matlab, a rural area of Bangladesh. Mother-child pairs were followed from early gestation to 9 years postpartum. More than 95% of the population in the study area use groundwater, retrieved from hand pumped tube-wells, as their main drinking water. Screening of all functioning tube-wells showed a wide range of arsenic contamination; about 70% exceeded the WHO guideline value of 10 μg/L, and 50% exceeded the national standard of 50 μg/L in drinking water. Arsenic exposure in the present studies was assessed based on the concentrations of the sum of inorganic arsenic and its methylated metabolites (monomethylarsonic acid and dimethylarsinic acid) in maternal urine during pregnancy and child urine at 4.5 and 9 years of age (Urinary arsenic: U-As), using high-performance liquid chromatography online with hydride generation and inductively coupled plasma mass spectrometry (HPLC-HG-ICPMS). In a sub-sample, we measured arsenic concentrations in maternal blood, placenta, and cord blood using ICPMS after acid digestion. Immune and inflammatory markers were measured in placenta (number of T cells, 8-oxoguanine (8-oxoG), cytokines, and leptin) and cord blood (cytokines, signal joint T cell receptor excision circles (sjTRECs), 8-hydroxy-2'-deoxyguanosine (8-OHdG), oxidative-stress defense and apoptosis related genes). Cell-mediated immunity was measured by delayed type hypersensitivity in response to purified protein derivative (PPD) injection and plasma cytokines were measured in children at 4.5 years of age. Plasma total IgG and measles-mumps-rubella vaccination-specific IgG concentrations were measured in children at 9 years of age. The median U-As concentration was 77 μg/L (range 2-2,064 μg/L) during early pregnancy, 57 μg/L (range 12-1,228 μg/L) in children at 4.5 years, and 53 μg/L (range 9-1,268 μg/L) at 9 years of age. The concentrations at different time points were significantly correlated, indicating somewhat reduced exposure over time. Arsenic exposure in early pregnancy was associated with reduced number of T cells in the placenta. Maternal arsenic exposure was also associated with decreased thymic function (number of naïve T cells) in newborns (measured by sjTRECs in umbilical cord blood lymphocytes), particularly in response to maternal infections. Maternal U-As was positively associated with oxidative stress markers (8-oxoG in placenta and 8-OHdG in cord blood). Oxidative stress responsive genes were down-regulated and apoptosis related genes were up-regulated in cord blood in relation to elevated arsenic concentrations. Placental inflammatory cytokines (IL-1β, TNF-α, and IFN-γ) and leptin increased with increasing maternal arsenic exposure. Cord blood cytokines (IL-1β, IL-8, IFNγ, and TNFα) showed a U-shaped association in relation to maternal arsenic exposure. Thus, prenatal arsenic exposure seems to impair vital immune responses in newborns. In children at 4.5 years of age, arsenic exposure was related to a weak response to the injected purified protein derivative. Particularly, the association was stronger in children with recent infections, indicating that arsenic impaired T cell-associated immune response to infections. In addition, the children’s arsenic exposure was associated with reduced plasma concentrations of Th1 cytokines (IL-2 and TNF-α) without alteration of the Th2 cytokines. In children at 9 years of age, prenatal arsenic exposure was positively associated with plasma total IgG, and the association was apparent mainly in boys and in malnourished children. Furthermore, we also found preliminary evidence of decreasing mumps-specific vaccination response with increasing arsenic exposure during childhood. In summary, elevated arsenic exposure appeared to adversely affect the developing immune system, including both T cell and B cell-associated immune functions. Maternal arsenic exposure seemed to reduce T cells in placenta and cord blood and increase nonspecific antibody production by B cells in children at 9 years of age. Early childhood arsenic exposure appeared to reduce cell-mediated immune function and vaccination response. These effects may have consequences for susceptibility to various immune-related diseases later in life, and thus these finding are of high public health relevance since the main source of arsenic exposure is via drinking water and food. Therefore, more effective mitigation strategies for arsenic exposure are needed