Apart from the influence of genetic differences, certain populations of individuals might be at higher risk of CNS disturbances owing to increased exposure associated with food intake patterns related to lifestyle. In this regard, Arctic populations have higher than average (compared populations living in Southern Canadian locations) concentrations of several toxins (including MeHg, pesticides and PCBs), largely owing traditional diets consisting of wildlife that bioaccumulate these environmental chemicals. The high levels of metals, PCBs, and pesticides (e.g. DDT) consumed from fish and other species of northern Canada [[16–18]] would be expected to readily penetrate multiple organs and eventually enter the brain.
We presently report that in utero + gestational exposure to a mixture of chemical contaminants (MeHg, pesticides, PCBs), based on blood contaminant profiles in Northern Canadian Inuit Arctic mothers, produced long term elevations of several cytokines within the hypothalamus of female rats. These effects on hypothalamic brain cytokines were evident in adulthood long after dosing ceased (about 120 days after dosing) and were observed when toxin exposure produced blood levels in rat dams near those of humans. Indeed, analysis of tissue residue data from previous studies using identical dosing procedures showed that blood levels in rat mothers were comparable to those of Arctic human mothers . Yet, one should still exercise caution when extrapolating between human and rat samples. While early life exposure to MeHg had little impact on cytokines in the present study, the PCBs and the full mixture (containing PCBs, OCs and MeHg) both elevated basal cytokine levels when assessed at five months of age. Although systemic exposure to LPS at five months of age increased most hypothalamic cytokines, prior developmental exposure to the contaminants did not alter the impact of LPS on adult cytokine levels. Thus, developmental exposure to realistic levels of environmental chemicals provoked long-term inflammatory cytokine elevations within the brain but did not sensitize animals to the impact of endotoxin exposure at adulthood.
These findings are consistent with the evidence indicating that exposure to environmental toxins during neurodevelopment can influence central nervous system (CNS) functioning long after exposure has occurred. For instance, gestational exposure to PCBs, mercury, lead, and organic pollutants has been associated with later cognitive disturbances in infants and children and may contribute to disorders of attention and activity [[9, 43–45]]. Yet, such cognitive effects are generally mediated by hippocampal and cortical brain regions, whereas hypothalamic brain changes (as observed in the present investigation) are typically associated with stress responses and hormonal output. Indeed, a plethora of data indicates that psychological and immunological (particularly LPS) stressors promote marked hypothalamic neurochemical alterations, often coupled with signs of sickness (e.g. fever, piloerection, ptosis, curled body posture) or depressive-like symptoms, such as anhedonia [[38, 40, 46–49]]. Similarly, we and others have reported that a range of stressful conditions (particularly psychosocial stressors stemming from changes in housing conditions), cytokines (including IL-1b, TNF-α and IFN-α) and immune agents that mimic bacterial or viral infections (LPS and poly I:C, respectively) increase hypothalamic cytokine expression and promote microglial-dependent neuroinflammatory activity [[39, 50–53]].
Although scant data exists regarding the impact of chemical toxins and hypothalamic functioning, one recent report did indicate that the pesticide, dieldrin, increased hypothalamic expression of an array of genes that are known to be responsible for oxidative functions and cell survival . Similarly, MeHg was found to reduce hypothalamic dopamine levels and induce anxiety-like effects in exposed fish . Hypothalamic and limbic brain circuits, along with a shift towards increased production of pro-inflammatory Th1 cytokines, were even posited to be responsible for the sickness symptoms provoked by smells associated with previous chemical toxin exposure .
Further rationale for focusing upon the hypothalamus (besides it being a key stress integrative brain region that is known to express a higher level of cytokines than most brain regions), stems from the substantial evidence showing that several pesticides and PCBs have well known endocrine disruption effects and have been reported to affect HPA and immune functioning. For instance, systemic administration of the PCB mixture, Aroclor 1248, altered glucocorticoid levels and the mitogenic response of peripheral immune cells . When an alternate PCB mixture (Aroclor 1254) was orally administered to female monkeys, dose-dependent alterations of T cell activity and antibody production were observed [58, 59]. Intriguingly, perinatal exposure to PCB congers 126 and 153 (as in the present study) appeared to sensitize the HPA axis, such that a much greater and prolonged cortisol response was evident with mild stress application at nine months of age . Similar to the PCBs, several different classes of pesticides were reported to affect HPA functioning in a number of different species, including male and female rats, as well as bears and fish [[61–64]]. Although scant data exist for MeHg, one recent study did report that Beluga sturgeon fed MeHg rich diets displayed elevated cortisol and glucose levels .
While the blood levels of contaminant in these animals were not available, we have conducted previous studies using identical dosing methodology and have shown that this dose of the mixture produces blood levels of PCB and OC pesticides in rat dams that are comparable to maternal blood levels in Canadian Arctic human population  Table 1. Other studies using the same mixture and dosing regimen has shown that PCBs and MeHg can both alter cerebellar gene expression patterns [66, 67]. Taken together with the present results, the available evidence suggests that, at exposure levels relevant to human populations, the chemical mixture likely affects neurodevelopmental processes and has long-term consequences upon cytokines that are known to fundamentally shape neuroinflammatory functioning.
Gestational and lactational transfer of environmental toxins would be expected to place the developing fetus or young offspring at risk. These would be especially evident during in utero and perinatal stages, when neuronal migration and synaptic pruning are occurring, neurons are especially sensitive to perturbations caused by environmental agents. At the same time, biological detoxification systems involved in metabolism and clearance of toxic substances are not fully developed in fetuses, infants and young children [68, 69]. Indeed, it is likely of particular importance that toxin exposure in the present investigation occurred during times of rapid neural development, when the blood-brain-barrier (BBB) is not fully functional and the brain is exquisitely sensitive to toxic chemicals that can affect neuronal migration and differentiation, as well as synapse formation [70, 71]. Some of these same chemicals, including MeHg and the various pesticides, can cause deficits in BBB functioning, evident as a long term increased permeability [[72–75]]. Hence, the protracted hypothalamic cytokine changes presently observed could conceivably have stemmed from deficiencies in BBB functioning induced by the Arctic chemicals, resulting in enhanced infiltration of peripheral immune cells. Yet, it is important to consider that some aspects of the hypothalamus (median eminence) actually lack a fully functional BBB and may facilitate penetration of the toxins. Besides any effects of peripheral immune cells, it seems likely that the chemical insults could have directly affected central glial activity, as has been observed following bacterial endotoxin challenge , thereby promoting local cytokine production .
One of the primary mechanisms through which toxins may promote CNS pathology is by inducing inflammatory immune factors. Indeed, neurodegeneration and CNS pathology in general, often have a prominent neuroinflammatory component, which is typically characterized by excessive microglial activation and accumulation of pro-inflammatory cytokines and oxidative factors [[35, 78–80]]. Similarly, pesticides have been reported to increase superoxide production from circulating neutrophils, as well as promote cortical astrocyte expression and induce the expression of the pro-inflammatory cytokines IL-6, IL-8 and IFN-g [[66, 80–82]]. Our own work has also shown that the acute adult exposure to the pesticide, paraquat, provoked neuroinflammatory changes, including an elevation of microglial cell reactivity that was closely tied to the neuronal loss provoked by the pesticide . The current results further show that exposure to a combination of environmental pollutants containing OCs, when given at realistic concentration/ratios, increase hypothalamic IL-6 and IL-10, and to a certain degree, IL-1b, IL-12 and TNF-α concentrations.
The lack of statistically significant differences between the Arctic chemical treated groups that received LPS in adulthood was somewhat surprising. Indeed, it was reported that early life exposure to LPS promoted an enhanced neurodegenerative effect, coupled with increased central TNF-α levels, upon exposure to the pesticide, rotenone, later in life . However, the failure to presently detect cytokine differences in response to the acute adult LPS challenge likely stems from a ceiling effect. In fact, the endotoxin did generally augment most cytokines in all groups (relative to the endotoxin naive rats of the initial study) and such an effect might have made it especially difficult to detect any subtle effects of the early life chemical treatments. Along these lines, there was a definite trend of increased hypothalamic IL-1b levels in the PCB and full Arctic mixture perinatally treated mice that received LPS in adulthood. The variability in the response to LPS apparent in these mice suggests that some animals were "responders" and some "non-responders" to the early life chemical priming. Future studies aimed at better characterizing this effect would benefit from assessing the impact of a variety of LPS doses. Along with not having a dose-response for LPS, another caveat of this second study is the lack of a "pure" control group (owing to the availability of animals) that did not receive LPS.
The cytokine changes observed within the hypothalamus could have substantial behavioral implications. For instance, IL-1b, IL-6 and TNF-α have well documented sickness effects (e.g. ptosis, piloerection, curled body posture) that are related to hypothalamic neurochemical activity [[37, 84, 85]]. These inflammatory cytokines have also been implicated in a number of clinical conditions involving a primary component of fatigue or malaise, including chronic fatigue syndrome and multiple chemical sensitivity [86, 87]. In fact, disturbances of hypothalamic neuroendocrine activity and elevations of brain cytokines which are evident following challenge with the viral mimic, poly I:C (double stranded RNA), have been proposed to be common mechanisms leading to chronic fatigue and sickness [88, 89]. Interestingly, pesticide exposure has likewise been implicated in multiple chemical sensitivity syndromes and general sickness symptoms [90, 91]. We have also reported that the pesticide, paraquat, induced behavioral changes reminiscent of Parkinson's disease and depression . In effect, it is possible that the present hypothalamic cytokine changes induced by early life chemical treatments could have important consequences for behavioral and neuroendocrine functioning.
It is unclear whether the enduring CNS cytokine alterations provoked by the perinatal chemical treatments (observed months after exposure) stemmed from cumulative/progressive time-dependent effects or a long lasting more acute impact of these treatments. In any case, it is interesting to note that several studies have indicated that stressor exposure had neurochemical effects that increased with the passage of time [92, 93]. Our own work likewise demonstrated that the cytokine, TNF-α, time-dependently, sensitized CNS processes, such that re-exposure to the cytokine one month following a previous single injection induced greatly augmented behavioral (sickness symptoms), corticoid and central monoamine (NE within the hypothalamus) changes [38, 94]. Regardless of the mechanisms responsible for the central cytokine variations, these immunotransmitters are ultimately able to act upon their receptors (found predominately on glial cells and to a lesser degree neurons) to induce the activation of JAK-STAT (IL-6, IL-10 and IL-12) and NFkB (IL-1b, TNF-a) signaling pathways. While these signaling pathways mediate the anti-tumor and immunological functions of cytokines in the periphery, increasing evidence has also indicated an important role in a range of neurological conditions ranging from depression to Parkinson's and Alzheimer's disease [[95–97]].