(C) PLOS One
This story was originally published by PLOS One and is unaltered.
. . . . . . . . . .
Maternal emulsifier consumption programs offspring metabolic and neuropsychological health in mice [1]
['Maria Milà-Guasch', 'Neuronal Control Of Metabolism', 'Neucome', 'Laboratory', 'Institut D Investigacions Biomèdiques August Pi I Sunyer', 'Idibaps', 'Barcelona', 'Sara Ramírez', 'Sergio R. Llana', 'Júlia Fos-Domènech']
Date: 2023-08
Modern lifestyle is associated with a major consumption of ultra–processed foods (UPF) due to their practicality and palatability. The ingestion of emulsifiers, a main additive in UPFs, has been related to gut inflammation, microbiota dysbiosis, adiposity, and obesity. Maternal unbalanced nutritional habits during embryonic and perinatal stages perturb offspring’s long–term metabolic health, thus increasing obesity and associated comorbidity risk. However, whether maternal emulsifier consumption influences developmental programming in the offspring remains unknown. Here, we show that, in mice, maternal consumption of dietary emulsifiers (1% carboxymethyl cellulose (CMC) and 1% P80 in drinking water), during gestation and lactation, perturbs the development of hypothalamic energy balance regulation centers of the progeny, leads to metabolic impairments, cognition deficits, and induces anxiety–like traits in a sex–specific manner. Our findings support the notion that maternal consumption of emulsifiers, common additives of UPFs, causes mild metabolic and neuropsychological malprogramming in the progeny. Our data call for nutritional advice during gestation.
Funding: This study was funded by the European Research Council under the European Union’s Horizon 2020 research and innovation programme (grant agreement no. 725004) and supported by: ‘la Caixa’ Foundation (ID100010434) under agreement LCF/PR/HR19/52160016 and the CERCA Programme/Generalitat de Catalunya (to M.C.); Marie Skłodowska-Curie Action fellowship (H2020-MSCA-IF) NEUROPREG (grant agreement no. 891247; to R.H-T.); the Spanish Ministry of Science and Innovation, Juan de la Cierva fellowship (IJC2018-037341-I to S.R.); Miguel Servet contract (CP19/00083) from Instituto de Salud Carlos III co-financed by ERDF (to A.O.). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Within the context outlined above, the current study aimed to investigate the impact of emulsifier consumption during pregnancy and lactation on offspring’s long-term health using the mouse as an experimental model. Our data showed that maternal intake of emulsifiers induced mild metabolic and neuropsychological alterations in the progeny, thus calling for nutritional advice towards UPF consumption during gestation.
Epidemiological and experimental evidence show that a perturbed environment during early life results in developmental adaptations that predispose the offspring to health disturbances in adulthood in both humans and rodents (“The Developmental Origins of Health and Disease” (DOHaD)) [ 13 – 16 ]. In this context, maternal dietary insults during gestation and lactation interfere with the programming of multiple neurocircuits [ 17 – 22 ], thus contributing to the development of diverse metabolic and neuropsychological disorders [ 23 – 25 ]. Indeed, it is worth noting that such pre- and perinatal nutritional challenges compromise the adequate development of hypothalamic feeding systems, including pro-opiomelanocortin (POMC) and agouti-related peptide (AgRP) neurons [ 19 , 20 , 26 , 27 ], which are crucial for systemic energy and metabolic homeostasis [ 28 ].
Emulsifiers, one of the most common UPF additives, are used as stabilizers to form or maintain a homogenous mixture of 2 or more immiscible phases. They can be found in numerous UPF items, including margarines, mayonnaise, salad dressings, bread, ice creams, cake mixes, fruit juices, snacks, instant soups, and noodles among many others. The Food and Agriculture Organization/World Health Organization (FAO/WHO) allows the addition of emulsifiers up to 1%. Among the most used emulsifiers, sodium carboxymethyl cellulose (CMC), and polysorbate 80 (P80) have been extensively added to UPFs for over 30 years. Alarmingly, recent studies indicated that emulsifier consumption causes gut microbiota dysbiosis, intestinal inflammation and cancer, metabolic syndrome, and obesity [ 7 – 12 ].
Modern lifestyle promotes the disproportionate consumption of sugar and saturated fats together with a sedentary life, leading to the development of obesity (and its associated comorbidities), which has reached pandemic proportions [ 1 ]. In recent years, the so-called ultra-processed foods (UPFs) have become remarkably popular in the market due to their convenience and palatability. As defined by the NOVA classification, UPFs are industrial formulations with little or no whole food, poor nutritional quality, high glycemic load, low dietary fibers, and substantial amounts of additives (colorants, flavorings, sweeteners, thickeners, emulsifiers, etc.) [ 2 ]. Importantly, scientific evidence has associated UPF consumption with the development of obesity, type 2 diabetes (T2D), cardiovascular disease, cancer, depression, and gastrointestinal disorders [ 3 – 6 ].
(A–D) Open field performance in 23–week–old male (A and B) (n = 6 CTRL–CTRL; n = 6 CTRL–Emul; n = 6 Emul–CTRL; n = 5 Emul–Emul) and female (C and D) (n = 6 mice/group) offspring born of control and emulsifier–exposed mothers, including time spent per zone (A and C) and total distance traveled (B and D) after 12 weeks of WD exposure. (E, F) Time spent in the light compartment during the dark–light box test in 23–week–old male (E) (n = 6 CTRL–CTRL; n = 6 CTRL–Emul; n = 6 Emul–CTRL; n = 5 Emul–Emul) and female (F) (n = 6 mice/group) offspring born of control and emulsifier–exposed mothers after 12 weeks of WD exposure. (G–J) Short–term memory parameters in 24–week–old male (G and H) (n = 6 CTRL–CTRL; n = 6 CTRL–Emul; n = 6 Emul–CTRL; n = 5 Emul–Emul) and female (I and J) (n = 6 CTRL–CTRL; n = 5 CTRL–Emul; n = 5 Emul–CTRL; n = 4 Emul–Emul) offspring born of control and emulsifier–exposed mothers, after 13 weeks of WD exposure, including discrimination index (G and I) and exploratory time (H and J). Data are derived from 1 single experiment. Data are expressed as mean ± SEM. Statistical analysis was performed by two–way ANOVA followed by Sidak’s post hoc analysis. *p < 0.05; **p < 0.01. The data underlying this figure can be found at DOI: 10.6084/m9.figshare.22742759 . WD, western–style diet.
CMC or P80 consumption have been shown to cause anxiety-like behaviors [ 11 ]. We next asked if maternal consumption of emulsifiers was sufficient to induce neuropsychological deficits in the offspring. To test this, we conducted a behavioral screening of anxiety-like phenotypes by exposing the offspring to the open field and dark-light box paradigms. Offspring of emulsifier-treated dams presented no changes in locomotor activity in an open field test ( S5A–S5D Fig ) but male offspring exhibited a significant decrease in the time spent in the light compartment of a dark-light box paradigm ( S5E and S5F Fig ). Anxiety-like behaviors were intensified upon WD challenge during adulthood, particularly in females ( Fig 7A–7F ). These results suggested increased anxiety-like states in the offspring of dams exposed to CMC+P80 during gestation and lactation. In addition, dietary insults during pregnancy have been linked to cognitive dysfunction in the offspring [ 25 , 31 ]. While cognition was not impaired in the offspring born to emulsifier-treated dams fed with normal chow diet ( S5G–S5J Fig ), life-long exposure to emulsifiers in combination with WD led to cognitive impairments in a novel object recognition test (NORT) in males ( Fig 7G and 7H ). Female offspring did not present memory recognition deficits (Figs 7I and 7J and S5I and S5J ). Together, these results demonstrate that emulsifier consumption during pregnancy could induce life-long consequences in offspring neuropsychological and metabolic health.
(A) Schematic illustration of offspring treatment until adulthood. (B) Six–hour fasting blood glucose levels at 22 weeks of age, after 11 weeks of WD exposure (n = 8 CTRL–CTRL; n = 10 CTRL–Emul; n = 8 Emul–CTRL; n = 8 Emul–Emul). (C) Plasma insulin levels after 6 h of fasting at 22 weeks of age, after 11 weeks of WD exposure (n = 8 CTRL–CTRL; n = 10 CTRL–Emul; n = 8 Emul–CTRL; n = 8 Emul–Emul). (D) Plasma leptin levels after 6 h of fasting at 22 weeks of age, after 11 weeks of WD exposure (n = 8 CTRL–CTRL; n = 10 CTRL–Emul; n = 8 Emul–CTRL; n = 8 Emul–Emul). (E) Body weight at 22 weeks of age, after 11 weeks of WD exposure (n = 8 CTRL–CTRL; n = 10 CTRL–Emul; n = 8 Emul–CTRL; n = 8 Emul–Emul). (F) gWAT weight normalized by total body weight and represented as % of control animals at 22 weeks of age, after 11 weeks of WD exposure (n = 8 CTRL–CTRL; n = 10 CTRL–Emul; n = 8 Emul–CTRL; n = 8 Emul–Emul). (G) GTT and (H) AUC (n = 8 CTRL–CTRL; n = 10 CTRL–Emul; n = 8 Emul–CTRL; n = 8 Emul–Emul) at 19 weeks of age, after 8 weeks of WD exposure. (I) ITT and (J) AUC (n = 8 CTRL–CTRL; n = 10 CTRL–Emul; n = 8 Emul–CTRL; n = 8 Emul–Emul) at 19 weeks of age, after 8 weeks of WD exposure. Data are derived from 1 single experiment. Data are expressed as mean ± SEM. Statistical analysis was performed by two–way ANOVA followed by Sidak’s post hoc analysis. *p < 0.05; **p < 0.01; ***p < 0.001. The data underlying this figure can be found at DOI: 10.6084/m9.figshare.22742759 . AUC, area under the curve; GTT, glucose tolerance test; gWAT, gonadal white adipose tissue; ITT, insulin tolerance test; WD, western–style diet.
(A) Schematic illustration of offspring treatment until adulthood. (B) Body weight at 22 weeks of age, after 11 weeks of WD exposure (n = 8 CTRL–CTRL; n = 7 CTRL–Emul; n = 7 Emul–CTRL; n = 5 Emul–Emul). (C) eWAT weight normalized by total body weight and represented as % of control animals at 22 weeks of age, after 11 weeks of WD exposure (n = 8 CTRL–CTRL; n = 7 CTRL–Emul; n = 7 Emul–CTRL; n = 5 Emul–Emul). (D) GTT and (E) AUC (n = 8 CTRL–CTRL; n = 7 CTRL–Emul; n = 7 Emul–CTRL; n = 5 Emul–Emul) at 19 weeks of age, after 8 weeks of WD exposure. (F) ITT and (G) AUC (n = 8 CTRL–CTRL; n = 7 CTRL–Emul; n = 7 Emul–CTRL; n = 5 Emul–Emul) at 19 weeks of age, after 8 weeks of WD exposure. (H) Six–hour fasting blood glucose levels at 22 weeks of age, after 11 weeks of WD exposure (n = 8 CTRL–CTRL; n = 7 CTRL–Emul; n = 7 Emul–CTRL; n = 5 Emul–Emul). (I) Plasma insulin levels after 6 h of fasting at 22 weeks of age, after 11 weeks of WD exposure (n = 8 CTRL–CTRL; n = 7 CTRL–Emul; n = 7 Emul–CTRL; n = 5 Emul–Emul). (J) Plasma leptin levels after 6 h of fasting at 22 weeks of age, after 11 weeks of WD exposure (n = 8 CTRL–CTRL; n = 7 CTRL–Emul; n = 7 Emul–CTRL; n = 5 Emul–Emul). Data are derived from 1 single experiment. Data are expressed as mean ± SEM. Statistical analysis was performed by two–way ANOVA followed by Sidak’s post hoc analysis. The data underlying this figure can be found at DOI: 10.6084/m9.figshare.22742759 . AUC, area under the curve; eWAT, epididymal white adipose tissue; GTT, glucose tolerance test; ITT, insulin tolerance test; WD, western–style diet.
(A) Schematic illustration of offspring treatment until adulthood (10 weeks of age). (B) Body weight at 10 weeks of age (n = 5 CTRL–CTRL; n = 4 CTRL–Emul; n = 6 Emul–CTRL; n = 7 Emul–Emul). (C) eWAT weight normalized by total body weight and represented as % of control animals at 10 weeks of age (n = 5 CTRL–CTRL; n = 4 CTRL–Emul; n = 6 Emul–CTRL; n = 7 Emul–Emul). (D) Body length at 10 weeks of age (n = 5 CTRL–CTRL; n = 7 CTRL–Emul; n = 6 Emul–CTRL; n = 5 Emul–Emul). (E) ITT and (F) AUC (n = 11 CTRL–CTRL; n = 12 CTRL–Emul; n = 13 Emul–CTRL; n = 12 Emul–Emul) at 10 weeks of age. (G) GTT and (H) AUC (n = 14 CTRL–CTRL; n = 13 CTRL–Emul; n = 13 Emul–CTRL; n = 13 Emul–Emul) at 10 weeks of age. (I) Six–hour fasting blood glucose levels (n = 5 CTRL–CTRL; n = 4 CTRL–Emul; n = 6 Emul–CTRL; n = 7 Emul–Emul) at 10 weeks of age. (J) Plasma insulin levels after 6 h of fasting at 10 weeks of age (n = 4 CTRL–CTRL; n = 4 CTRL–Emul; n = 5 Emul–CTRL; n = 5 Emul–Emul). (K) Plasma leptin levels after 6 h of fasting at 10 weeks of age (n = 4 CTRL–CTRL; n = 4 CTRL–Emul; n = 5 Emul–CTRL; n = 5 Emul–Emul). Data in B, C, D, I, J, and K are derived from 1 single experiment. Data in E, F, G, and H are pools from 2 different experiments. Data are expressed as mean ± SEM. Statistical analysis was performed by two–way ANOVA followed by Sidak’s post hoc analysis. *p < 0.05; **p < 0.01. The data underlying this figure can be found at DOI: 10.6084/m9.figshare.22742759 . AUC, area under the curve; eWAT, epididymal white adipose tissue; GTT, glucose tolerance test; ITT, insulin tolerance test.
Next, we interrogated whether early life exposure to CMC+P80 induces long-term metabolic health disruptions. To differentiate between maternal programming consequences and emulsifier consumption after birth, we divided control and emulsifier offspring into 4 groups at weaning: maternal control + water (CTRL—CTRL), maternal control + emulsifiers after weaning (CTRL—Emul), maternal emulsifier + water (Emul—CTRL), and maternal emulsifier + emulsifier (Emul—Emul) treatment after weaning until 10 weeks of age ( Fig 4A ). In males, maternal consumption of emulsifiers, independent of its exposure after weaning, led to body weight reduction ( Fig 4B ), without significant changes in adiposity ( Fig 4C ) or body length ( Fig 4D ). Male offspring born from dams treated with emulsifiers showed higher insulin sensitivity ( Fig 4E and 4F ), without changes in food intake ( S4A Fig ), glucose tolerance, blood glucose, insulin, and leptin levels ( Fig 4G–4K ). Of note, life-long exposure to emulsifiers (intrauterine and postnatal life) induced stronger glucose homeostasis impairments leading to glucose intolerance ( Fig 4H ) (maternal control + emulsifier: 14,326 ± 5,834 area under the curve (AUC) versus maternal emulsifier + emulsifier: 19,084 ± 3,039 AUC, p = 0.0099). These effects were sex-specific since female offspring from emulsifier-treated dams did not show alterations in any of the analyzed parameters ( S3A–S3J and S4B Figs). The delayed leptin surge seen in male offspring from emulsifiers-treated dams ( Fig 2L and 2M ) prompted us to investigate their ability to respond to the anorexigenic effects of leptin. Maternal emulsifier consumption, or emulsifier consumption during life, did not perturb the weight- and food intake-reducing effects of exogenously administered leptin ( S4C and S4D Fig ). These results suggest that maternal consumption of emulsifiers affects the metabolic programming of male offspring, triggering mild alterations in glucose metabolism in adulthood.
Maternal dietary insults affect the development of axonal projections to target areas [ 19 , 26 , 30 ]. The changes observed in the expression of anorexigenic genes (pomc and cart) in males ( Fig 3C ) prompted us to evaluate if maternal emulsifier consumption would influence melanocortin projections to the paraventricular nucleus of the hypothalamus (PVH). While PVH AgRP staining density showed a non-significant trend to increase ( Fig 3E ), there was a notable increase in the density of α-melanocyte stimulating hormone (α-MSH, a bioactive anorexigenic product of POMC processing) staining in male offspring from emulsifier-treated dams at weaning ( Fig 3F ). These data suggested that maternal emulsifier consumption per se, without other components usually present in UPFs, is sufficient to induce a rewiring of the melanocortin hypothalamic feeding circuit that might underlie the metabolic changes observed in male offspring.
(A) Volcano plot of transcript expression in the MBH between control and emulsifier offspring at weaning. Threshold for FC (±1.5) and FDR (p < 0.05) was considered. DEGs upon maternal emulsifier consumption are depicted in blue (down–regulated) and orange (up–regulated). Unchanged genes are represented in black (n = 4 CTRL and n = 5 Emul). (B) Cytoscape plot of the down–regulated enriched pathways (p < 0.05) in the offspring of emulsifier–exposed dams. (C) Transcript expression of orexigenic and anorexigenic peptides in the MBH in male offspring from control and emulsifier–exposed dams at weaning (n = 4 CTRL and n = 5 Emul). (D) Transcript expression of orexigenic and anorexigenic peptides in the MBH in female offspring from control and emulsifier–exposed dams at weaning (n = 7 CTRL and n = 7 Emul). (E) Representative immunofluorescence images showing AgRP staining density in the PVH of control and emulsifier male offspring at weaning and integrated density quantification (n = 6 mice/group). (F) Representative immunofluorescence images showing α–MSH staining density in the PVH of control and emulsifier male offspring at weaning and integrated density quantification (n = 5 mice/group). Data in C and D are derived from 1 single experiment. Data in E and F are pools from 2 different experiments. Data are expressed as mean ± SEM. Statistical analysis was performed by unpaired t test in C, D, E, and F. *p < 0.05; ***p < 0.001. The data underlying this figure can be found at DOI: 10.6084/m9.figshare.22742759 . Pomc, pro–opiomelanocortin; Cart, cocaine–and amphetamine–regulated transcript; Agrp, agouti–related peptide; Npy, neuropeptide Y; Pcsk1, proprotein convertase 1; Mcr3, melanocortin 3 receptor; Mcr4, melanocortin 4 receptor; α–MSH, alpha–melanocyte–stimulating hormone; PVH, paraventricular hypothalamic nucleus; 3V, third ventricle; DEG, differentially expressed gene; FC, fold change; FDR, false discovery rate; MBH, mediobasal hypothalamus.
Emulsifier consumption has been associated with changes in feeding-related neuropeptides in adult mice [ 11 ]. To understand the impact of maternal emulsifier consumption on hypothalamic development, we performed RNA sequencing (RNAseq) in the mediobasal hypothalamus (MBH) of male offspring at weaning. Principal component analysis (PCA) of individual samples identified 2 distinct clusters ( S2A Fig ). RNAseq analysis uncovered 83 differentially expressed genes (DEGs) upon maternal emulsifier consumption ( Fig 3A ). Out of the total amount of DEGs, 54% (45) of the genes were up-regulated and 46% (38) were down-regulated ( Fig 3A and S1 Table ). Enrichment pathway analysis of the down-regulated DEGs revealed significant changes in 4 main pathways: feeding behavior/neuropeptides, neuronal activity, metabolism, and transcriptional regulation ( Fig 3B and S1 Table ). To confirm that maternal consumption of emulsifiers affects the development of neuronal circuits controlling feeding behaviors, we analyzed the expression of key genes related to energy balance and food intake control in the MBH of the progeny of female mice exposed to emulsifiers during gestation and lactation. This analysis included the assessment of pro-opiomelanocortin (pomc), cocaine- and amphetamine-regulated transcript (cart), agouti-related peptide (agrp), neuropeptide Y (npy), proprotein convertase 1 (pcsk1), melanocortin 3 receptor (mc3r), and melanocortin 4 receptor (mc4r). Gene expression analysis revealed that maternal consumption of emulsifiers reduced pomc and cart expression at weaning exclusively in male offspring ( Fig 3C and 3D ), with no changes in the overall number and size of POMC neurons ( S2B–S2D Fig ).
(A) Experimental design of maternal emulsifier consumption and offspring collection at weaning. (B) Body length at weaning of male (n = 9 CTRL and n = 6 Emul) and female (n = 13 CTRL and n = 6 Emul) offspring from control and emulsifier–exposed dams. (C) Body weight at weaning of male (n = 9 CTRL and n = 12 Emul) and female (n = 11 CTRL and n = 12 Emul) offspring from control and emulsifier–exposed dams. (D) Epididymal and gWAT weight normalized by total body weight and represented as % of control animals in male (n = 9 CTRL and n = 12 Emul) and female (n = 11 CTRL and n = 12 Emul) offspring from control and emulsifier–exposed dams at weaning. (E) GTT and (F) AUC in male (n = 20 CTRL and n = 19 Emul) offspring from control and emulsifier–exposed dams at weaning. (G) ITT and (H) AUC in male (n = 8 CTRL and n = 4 Emul) offspring from control and emulsifier–exposed dams at weaning. (I) Six–hour fasting blood glucose levels in male (n = 9 CTRL and n = 10 Emul) and female (n = 11 CTRL and n = 11 Emul) offspring from control and emulsifier–exposed dams at weaning. (J) Plasma insulin levels in male (n = 8 CTRL and n = 10 Emul) and female (n = 10 CTRL and n = 11 Emul) offspring from control and emulsifier–exposed dams at weaning after 6 h of fasting. (K) Plasma leptin levels in male (n = 8 CTRL and n = 8 Emul) and female (n = 10 CTRL and n = 11 Emul) offspring from control and emulsifier–exposed dams at weaning after 6 h of fasting. (L) Plasma leptin levels across postnatal development (P7–P10–P13–P21) (P7 n = 6 CTRL and n = 5 Emul; P10 n = 6 CTRL and n = 6 Emul; P13 n = 6 CTRL and n = 6 Emul; P21 n = 8 CTRL and n = 8 Emul) in male offspring from control and emulsifier–exposed dams. (M) Peak plasma leptin levels at P10 (n = 6 CTRL and n = 6 Emul) in male offspring from control and emulsifier–exposed dams. (N) GTT and (O) AUC in female (n = 19 CTRL and n = 21 Emul) offspring from control and emulsifier–exposed dams at weaning. (P) ITT and (Q) AUC in female (n = 5 CTRL and n = 5 Emul) offspring from control and emulsifier–exposed dams at weaning. (R) Plasma leptin levels across postnatal development (P7–P10–P13–P21) (P7 n = 6 CTRL and n = 4 Emul; P10 n = 6 CTRL and n = 4 Emul; P13 n = 6 CTRL and n = 4 Emul; P21 n = 10 CTRL and n = 11 Emul) in female offspring from control and emulsifier–exposed dams. Data in B, G, H, L, M, P, Q, and R are derived from 1 single experiment. Data in C, D, E, F, I, J, K, N, and O are pools from 2 different experiments. Data are expressed as mean ± SEM. Statistical analysis was performed by unpaired t test in B, C, D, F, H, I, J, K, M, O, and Q and two–way ANOVA followed by Sidak’s post hoc analysis in E, G, N, and P. Panels L and R were analyzed using a two–way ANOVA mixed effects. *p < 0.05; **p < 0.01. The data underlying this figure can be found at DOI: 10.6084/m9.figshare.22742759 . AUC, area under the curve; GTT, glucose tolerance test; gWAT, gonadal white adipose tissue; ITT, insulin tolerance test.
(A) Experimental design of maternal emulsifier consumption highlighting the period of maternal characterization. (B) Daily water consumption of control and emulsifier–treated females before mating (n = 5/group). (C) Daily food intake of control and emulsifier–treated females before mating (n = 5/group). (D) Body weight of control and emulsifier–treated females before mating (n = 10 CTRL and n = 10 Emul). (E) GTT and (F) AUC of control and emulsifier females before mating (n = 10 CTRL and n = 10 Emul). (G) Fasting blood glucose levels of control and emulsifier–supplemented females before mating (n = 10 CTRL and n = 10 Emul). (H) Plasma leptin levels after 6 h in fasting of control and emulsifier–treated females before mating (n = 8 CTRL and n = 8 Emul). Data are derived from 1 single experiment. Data are expressed as mean ± SEM. Statistical analysis was performed with an unpaired t test in B, C, D, F, G, H, and by two–way ANOVA followed by Sidak’s post hoc analysis in E. *p < 0.05. The data underlying this figure can be found at DOI: 10.6084/m9.figshare.22742759 . AUC, area under the curve; GTT, glucose tolerance test.
Discussion
The consumption of UPFs has increased sharply over the last 2 decades. In fact, recent surveys have shown that UPF consumption contributes to 25% to 50% of the total daily caloric intake in adults [32] and more than 60% among school-age children in the United Kingdom and the United States [33,34]. UPF intake during pregnancy and its potential adverse effects on maternal–child health have started to be investigated in humans, identifying a positive association between UPF consumption and gestational weight gain, neonatal adiposity, and the development of attention deficit hyperactivity disorder [35–38]. The combination of distinct additives (emulsifiers, sweeteners, colorants, flavorings, etc.) concur with saturated fats and sugars in UPFs and, therefore, the precise contribution of each of these additives to maternal health and offspring developmental programming needs to be carefully assessed. In the present study, we investigated the transgenerational health impact of emulsifiers in mice. In our experimental model, dams and/or offspring mice were exposed to prolonged and continuous amount of CMC+P80, 2 extensively used emulsifiers in the UPF industry. This pattern of administration echoes to some extent the human environment, where exposure to emulsifiers is high and usually combined with other additives, although its consumption occurs in an intermittent manner. Given that access to UPF occurs throughout life (and not only during pregnancy and breastfeeding), we opted for an extended experimental design by providing CMC+P80 several weeks before fecundation. This approach considered the potential long-term effects of emulsifiers on the female intrauterine environment. We found that the maternal consumption of these emulsifiers was sufficient to induce mild metabolic, cognitive, and psychological impairments in male offspring (and to a lesser extent in females).
As regards metabolism, the mild decrease in food intake observed in the female mice before the onset of pregnancy (after 6 weeks of CMC+P80 treatment) disappeared upon prolonged exposure to emulsifiers. This could account to an adaptative response after short-term supplementation with emulsifiers. Our results also indicated that maternal ingestion of emulsifiers resulted in glucose intolerance, even in the absence of body weight gain. In addition, maternal emulsifier consumption delayed the postnatal leptin surge in male offspring. Alterations in leptin levels and glycemic fluctuations during pregnancy can disrupt neuronal specification, proliferation and wiring of hypothalamic circuits in the offspring. Indeed, leptin levels directly influence axonal outgrowth of POMC and AgRP neurons during lactation [23,39]. Therefore, the delayed leptin surge in male offspring of dams exposed to CMC+P80 could underlie the alterations in α-MSH staining innervating the PVH at weaning. Nevertheless, the leptin system seems to function correctly as plasma leptin levels and leptin response were not compromised in adult offspring.
Moreover, our gene expression and α-MSH staining density analysis suggested that anorexigenic brain circuits were more sensitive to the detrimental effects of emulsifier consumption during pregnancy. In addition, the reduction in pomc and cart expression in male offspring could reflect a compensatory mechanism for the higher α-MSH staining density reaching the PVH. This could underlie the decrease in body weight and adiposity observed in these animals at weaning and during adulthood. Together, our results suggest that emulsifiers can influence the development of hypothalamic neurocircuits during early life in a similar fashion as variations in nutrients do, including glucose, lipids, and food additives [40–42].
Emerging evidence suggests, in both mice and humans, that low concentrations of dietary emulsifiers are sufficient to impact intestinal barrier function and to induce gut inflammation, thus increasing the incidence of intestinal diseases and metabolic syndrome [8,43–46]. The fact that emulsifiers can modulate the gut microbiome [8,47], perturbing its function and generating inflammation, could also contribute to the metabolic impairments and cognition deficits observed in male offspring. In this context, maternal consumption of P80 during gestation and lactation leads to gut dysbiosis and predisposes to colitis in the offspring [48]. Moreover, dysbiosis of the maternal gut microbiome during pregnancy has been shown to modulate fetal thalamocortical axonogenesis [49], disrupt brain function and behavior in the offspring [50–52]. Alterations in gut microbiome could also play a part in the anxiety-like traits we observed in both sexes since emulsifier consumption has been linked to anxiety-like behavior in mice [11]. These observations suggest that emulsifiers may also interfere with the development of other brain regions related with diverse behavioral processes.
Our results also indicated that maternal emulsifier consumption had a sex-specific effect in the offspring. Indeed, male offspring were more susceptible to metabolic and neuropsychological disruptions caused by maternal emulsifier consumption, providing additional evidence that males and females respond differently to a suboptimal maternal environment [21,53]. In addition, at a first glance, the combined exposure of emulsifiers and western diet during adulthood seems to alleviate female mice from the effects of energy-dense diets on glucose metabolism. This alteration, however, could be the consequence of impaired glucose absorption by the gut, increased glycosuria or defective nutrient transport throughout the intestinal cavity. The sex-related developmental mechanisms underlying these metabolic and cognitive sexual dimorphisms require further investigation.
The effects of emulsifier ingestion on maternal programming were stronger than those from exposure after weaning, agreeing with the idea that intrauterine and early postnatal life are critical developmental periods that, if disrupted, can have profound metabolic consequences in adulthood. Some of these effects (e.g., glucose tolerance, cognition) seemed to be slightly worsen upon prolonged exposure to emulsifiers (maternal plus after weaning treatments). Whether the alterations in offspring metabolic and neuropsychological health arise from disturbances induced by emulsifiers in the mothers before pregnancy onset or directly derive from effects during gestation and lactation could not be addressed in the present study. Follow-up studies analyzing the consequences of emulsifiers at each stage separately would be necessary. In addition, how long-term emulsifiers (mimicking persistent exposure to UPFs throughout life) affect female reproductive status, embryonic survival as well as maternal and offspring health outcomes demand further evaluation.
Previous studies have showed that consumption of CMC or P80 individually during adulthood was sufficient to induce metabolic alterations and microbiota dysbiosis in mice [8,11]. However, mice from control dams exposed to CMC+P80 did not show metabolic alterations under our experimental settings. This discrepancy with previous studies may be due to differences in the experimental protocol implemented, as our approach differed in the onset and length of treatment as well as in the combination of compounds. We combined these 2 broadly used emulsifiers in order to maximize their effects and mimic, to some extent, the presence of diverse emulsifiers in most contemporary processed food items. It is plausible that the mixture of CMC+P80 induces a milder (rather than an additive or synergistic) effect, but this requires further investigation.
Food packaging labels provide null or scarce information regarding the actual content of additives in UPFs. This greatly limits consumer knowledge regarding the levels eaten and our ability to avoid the ingestion of a large and diverse array of food additives [32,54]. Even food items that are perceived as “healthy,” such as vegan/vegetarian products, contain large amounts of additives (including emulsifiers) that per se could induce long-term metabolic impairments. In this regard, it is important to bring about greater societal consciousness that some apparently “healthy” industrial formulations might induce metabolic alterations to a similar extent as products usually considered “unhealthy.”
Collectively, our study showed that maternal consumption of emulsifiers commonly present in UPF items induced mild metabolic, cognitive, and psychological programming effects in the offspring in a sex-dependent manner. Our findings emphasize the importance of a healthy developmental environment during gestation and endorse the idea that the amount of UPFs consumed during gestation should be taken into serious consideration. We call for awareness of UPF intake during pregnancy and lactation to avoid potential detrimental effects on the metabolic and neuropsychiatric health of the progeny, thus building adequate nutritional habits for mothers and infants.
[END]
---
[1] Url:
https://journals.plos.org/plosbiology/article?id=10.1371/journal.pbio.3002171
Published and (C) by PLOS One
Content appears here under this condition or license: Creative Commons - Attribution BY 4.0.
via Magical.Fish Gopher News Feeds:
gopher://magical.fish/1/feeds/news/plosone/
