Date Published: March 7, 2019
Publisher: Public Library of Science
Author(s): Ghislaine Poizat, Coralie Alexandre, Sarah Al Rifai, Laure Riffault, Delphine Crepin, Yacir Benomar, Mohammed Taouis, Jonathan M. Peterson.
Resistin promotes hypothalamic neuroinflammation and insulin resistance through Toll like receptor 4 (TLR4), this hormone is thought to be a link between obesity and insulin-resistance. Indeed, resistin plasma levels are higher in obese and insulin resistant subjects. However, the impact of maternal resistin on the predisposition of offspring to hypothalamic neuroinflammation is unknown. Here, female mice were treated with resistin during gestation/lactation periods, then hypothalamic neuroinflammation was investigated in male offspring at p28 and p90. At p28, resistin increased the expression of inflammation markers (IL6, TNFα and NFκB) and TLR4 in the hypothalamus and decreased both hypothalamic insulin and leptin receptors’ expression. The hypothalamic up-regulation IL6, TNFα and TLR4 was sustained until p90 promoting most likely hypothalamic inflammation. Maternal resistin also increased IL6 and TNFα in the adipose tissue of offspring at p90 associated with a higher body weight gain. In contrast, liver and muscle were not affected. These findings reveal that the augmentation of maternal resistin during gestation and lactation promotes hypothalamic and adipose tissue inflammation of offspring as evidenced by sustained increase of inflammation markers from weaning to adulthood. Thus, maternal resistin programs offspring hypothalamic and adipose tissue inflammation predisposing then offspring to body weight gain.
The low-grade inflammation in obese, insulin-resistant or type-2 diabetes subjects is an important issue in developmental programming of inflammation and its long-term consequences on offspring [1, 2]. Indeed, hypercaloric maternal diet programs inflammation in offspring tissues [3–6]. Moreover, accumulated evidences suggest that changes of inflammatory markers in early life could have a strong impact on the prevalence of metabolic and cardiovascular diseases [7–9]. It has been suggested that offspring inflammation is promoted by LPS-induced maternal inflammation. This treatment led to strong changes in offspring immune system evidenced by more pro-inflammatory macrophages M1 and a higher Interleukin-1β (IL1β) production [10, 11]. It is also well documented that maternal obesity increased pro-inflammatory cytokine in the placenta enhancing then inflammatory responses of offspring. Furthermore, fatty acids, cholesterol, and triglyceride plasma levels were increased in fetuses born to obese ewes that also exhibited up-regulated Toll Like Receptor 4 (TLR4), Nuclear Factor-kappa B (NFκB) and c-Jun N-Terminal kinase (JNK) . TLR4 is considered as the binding site for LPS and, more recently TLR4 was described as a receptor for resistin, an adipokine [13, 14]. Resistin is implicated in the onset of low-grade inflammation found in obese or insulin resistant rodent models [15, 16]. Indeed, resistin decreases insulin responsiveness and contributes to the onset of type 2 diabetes . Interestingly, intracerebroventricular (ICV) resistin treatment induces overall inflammation and insulin resistance through the activation of hypothalamic TLR4 signaling pathway. This treatment also altered both adiponectin and FGF21 signaling pathways known as insulin sensitizers . These findings are in line with previous studies suggesting that resistin links obesity to insulin resistance . The impact of resistin during pregnancy on the placenta and offspring has been suspected. Indeed, resistin is expressed in human placenta, and it has been suggested that placental resistin modulates insulin sensitivity during pregnancy . Furthermore, resistin plasma levels are increased in women with gestational diabetes [20–22]. However, this augmentation remains controversial, since other studies reported that gestational diabetes does not affect plasma resistin levels . Moreover, several studies reported high resistin plasma levels in pregnant women as compared to non-pregnant women [22, 24]. It has been also reported that TLR4 was highly expressed in the placenta of obese women as compared to lean women, suggesting the implication of TLR4 in placental inflammation that could affect the fetus . Based on our previous findings demonstrating the implication of hypothalamic resistin/TLR4 signaling pathway in the onset of inflammation and insulin resistance , we hypothesized that maternal resistin could be implicated in the predisposition of offspring to inflammation and insulin resistance especially when subjected to inappropriate diet such as HFD. Indeed, the role and the implication of maternal resistin on the development of inflammation in offspring are still unknown especially at the hypothalamic level. Thus, in the present study, we aim to investigate whether maternal resistin has long-term effects on the predisposition of offspring to develop inflammation and metabolic disorders when subjected to HFD, with a special focus on hypothalamic neuroinflammation, termed as perinatal programmed hypothalamic inflammation. Indeed, HFD is known to promote inflammation and in this study feeding CC and CR offspring with HFD from p60 to p90 will respond to the question whether offspring born to dams treated with resistin (CR group) are prone to hypothalamic inflammation and alterations of metabolic parameters as compared CC group. For this purpose, pregnant mice were treated with resistin during gestation and lactation and then we analyzed the expression of inflammation markers in the hypothalamus and adipose tissue together with body weight gain and metabolic parameters of male offspring at p28 and adult mice following a 30 days HFD challenge.
During pregnancy, maternal obesity is associated with metabolic inflammation attributed to increased adipose tissue and circulating proinflammatory cytokine levels [1, 27]. In addition, maternal obesity modifies the secretion of adipokines including resistin that is known to be involved in the onset of inflammation . Indeed, resistin is considered as a potential link between obesity and insulin resistance. However, the pro-inflammatory role of resistin during perinatal period is unknown. Furthermore, the correlation between circulating levels of resistin and the BMI of pregnant women are conflicting. Some of these studies revealed positive correlation between resistin plasma levels and the BMI in obese pregnant women and others showed similar plasma resistin levels between obese and non-obese women [22, 24, 28, 29]. However, there is a consensus concerning the deleterious effects of maternal inflammation, even though the specific effect of resistin has not been investigated independently of the controversy concerning its secretion by adipocyte in humans. Thus, the mouse model is of interest to investigate the effect of resistin in perinatal period. In this extensive preliminary study in the field, we investigated the long-term effects of maternal resistin on the predisposition of offspring to develop hypothalamic neuroinflammation and subsequent metabolic disorders. Here, we focused on the effects of resistin independently of systemic inflammation on normal dams and studied the long-term effects of this treatment on offspring and attempted to answer the question whether maternal resistin predisposes offspring to hypothalamic inflammation. For this purpose, the chosen resistin dose induced a slight but not significant increase of resistin plasma levels to avoid a high inflammation that could lead to abortion. However, resistin plasma levels of dams are higher when compared to offspring and this could be attributed to gender as previously reported . We demonstrate that resitin treatment during pregnancy and lactation periods induced long-term effects on hypothalamic inflammation of male offspring associated with the alteration of metabolic parameters. Resistin treatment of dams did not modify their metabolic status or inflammatory markers as evidenced by unchanged glycemia, insulinemia and body weight. Furthermore, IL6 and TNFα plasma levels were not affected. Thus, in these conditions, we selectively investigated the impact of resistin perinatal treatment independently from systemic pro-inflammatory cytokine changes. At weaning (p28), offspring born to dams treated with resistin, CR group, exhibited lower resistin plasma levels without any changes in the other circulating parameters. This could be attributed to the alteration of adipose tissue resistin secretion consequently to perinatal resistin treatment that acts as a negative feedback regulatory loop, however the mechanism is unknown. This decrease is most likely compensated by the significant increase of TLR4, which is the binding site of resistin, and this could explain the increased expression hypothalamic inflammatory markers. Indeed, the expression of pro-inflammatory markers (IL6, NFκb, TNFα) was significantly increased in the hypothalamus of CR group as well as microgliosis markers (ICAM, CD68, IBA1) and astrogliosis (GFAP) markers. This suggests a potential hypothalamic neuroinflammation resulting from the maternal resistin treatment. Importantly, the long-term consequences of maternal resistin seem to differentially affect other brain regions such as the cortex. Indeed, inflammatory markers were not affected in the cortex of dams and p28 offspring, the effect is nevertheless observed in p90 CR group with increased IL6, TNFα and TLR4 expression. This indicates that the hypothalamus is affected earlier than the cortex. Additionally, p28 CR group exhibits a significant diminution of hypothalamic insulin and leptin receptor associated to the up-regulation of SOCS3 a negative regulator of insulin/leptin signaling. However, the expression level of PTP-1B was reduced in CR group as compared to CC group. These data show most likely the impairment of both IR and ObRb early in life. This finding suggests a strong impact later in life as previously described where leptin plays a crucial role in the neuronal organization of hypothalamic nuclei and especially ARC to PVN neuronal projections . Indeed, the alteration of leptin action early in life has long-term impact on energy homeostasis [32, 33] and contributes to metabolic diseases. We have also investigated the impact of resistin perinatal treatment on the predisposition of offspring to inflammation and glucose intolerance when challenged with HFD. In fact, we have anticipated that using HFD will exacerbate the difference between CC and CR offspring and determine whether offspring born to dams treated with resistin are prone to inflammation and metabolic disorders. Indeed, HFD is known to promote these disorders. However, in future experiments It will be interesting to perform identical studies using Chow diet. At p90, CR group showed a significant increase in body weight gain and plasma insulin levels associated with slight glucose intolerance. Indeed, we have shown that AUC of GTT curves are significantly different between CC and CR, but non parametric ANOVA test showed no significant difference between the two groups when considering the time course curves. Thus, we conclude that maternal resistin led to a slight glucose intolerance, but when combined to the hyperinsulinemia found in CR group this fairly reveals most likely a potential insulin resistance. Plasma levels of IL6 and TNFα were not modified. However, as at p28, TLR4 was up-regulated in the hypothalamus despite a stable resistin plasma level this may contribute to hypothalamic inflammation. Indeed, hypothalamic expression of IL6, resistin, TLR4 and TNFα is increased in CR group at p90 as compared to CC group, revealing a sustained expression of inflammation markers in the hypothalamus. The expression levels of reactive astrogliosis and microgliosis markers were similar between CR and CC except for ICAM as well as IR and ObRb. This could be attributed to the fact that both groups were fed HFD and this most likely attenuated the differences except for the hypothalamic neuroinflammation markers. Our findings also show the up-regulation of IL6 and TNFα in the adipose tissue of CR group as compared to CC but not in liver or muscle. This indicates that the hypothalamus and adipose tissue of offspring are the most affected tissues in response to the augmentation of maternal resistin during gestation and lactation.