The Yin-Yang Concept of Pediatric Obesity and Gut Microbiota

The era of pediatric obesity is no longer a myth. Unfortunately, pediatric obesity has reached alarming incidence levels worldwide and the factors that contribute to its development have been intensely studied in multiple recent and emerging studies. Gut microbiota was recently included in the wide spectrum of factors implicated in the determination of obesity, but its role in pediatric obese patients is far from being fully understood. In terms of the infant gut microbiome, multiple factors have been demonstrated to shape its content, including maternal diet and health, type of delivery, feeding patterns, weaning and dietary habits. Nevertheless, the role of the intrauterine environment, such as the placental microbial community, cannot be completely excluded. Most studies have identified Firmicutes and Bacteroidetes as the most important players related to obesity risk in gut microbiota reflecting an increase of Firmicutes and a decrease in Bacteroidetes in the context of obesity; however, multiple inconsistencies between studies were recently reported, especially in pediatric populations, and there is a scarcity of studies performed in this age group.


Introduction
Pediatric obesity has reached epidemic proportions worldwide regardless of age. The prevalence of obesity varies by age, sex, ethnicity, race, and socio-economic status [1]. Obesity prevalence is higher in females of non-Hispanic black and of Hispanic origin when compared to non-Hispanic whites [1,2]. According to the World Health Organization, approximately 40 million children below the age of five years are diagnosed with obesity with particular prevalence in younger children [3]. The global prevalence of pediatric overweight and obesity was reported to vary from 5.7% to 40% depending on the studied population [4][5][6][7]. The multifactorial determination of this condition has been assessed in multiple studies which have shown that obesity occurs in genetically predisposed individuals who fulfill 'obesogenic' conditions, such as improper dietary habits, poverty or limited wealth, and even certain cultural traditions [1,8]. Nevertheless, no interdependence between genetic factors and 'obesogenic' factors has been established. Without doubt, genetic factors play a major role since multiple studies have shown that certain maternal gene polymorphisms are associated with increased risk for obesity in offspring [9][10][11][12], suggesting that predisposition to the development of obesity begins during intrauterine life. Moreover, the involvement of certain genes in weight gain by regulating feeding behavior, appetite, metabolism and energy expenditure has been clearly demonstrated in multiple studies [8]. Thirty-two loci of the human genome were found to be associated with increased body weight, but only 2% of the population was found to carry these genes [13]. In terms of the exogenous determination of obesity, energy intake that exceeds energy expenditure, along with a wide-spectrum of factors which interact with this process in multiple ways, represent the most important triggers [14]. The long term-persistence of obesity from childhood to adulthood is definitely a hallmark of decreased life expectancy

Gut Microbiota and Obesity-To Be or Not to Be?
The gut microbiota might be defined as a shield of protection for mucosal permeability which regulates the fermentation and absorption of dietary polysaccharides, accounting for its role in the regulation of fat accumulation and subsequent obesity development [31]. The gut microbiota and host are in a symbiotic relationship with the gut microbiota contributing to the host's physiological homeostasis and impairment of the gut microbiota resulting in the development of a wide spectrum of disorders [32]. In terms of obesity, alteration in the gut microbiota leads to an increase in energy harvesting from the diet. Aside from its role in increasing energy harvesting and the regulation of fat storage [32,33], the gut microbiota has also been shown to contribute both to the modulation of substrate formation for storable fat synthesis [34], and the production of compounds that, once absorbed into the systemic circulation, will promote the development of obesity-associated complications by increasing tissue inflammatory injuries and insulin resistance [35].
The gut microbiota might be considered as a true standalone organ with each individual harboring hundreds of different species with only a few dozen conserved between individuals, forming a stable core in a normal healthy state [36]. Based on the definition of a standalone organ, the gut microbiota has multiple functions, including digestion of certain otherwise indigestible dietary fibers, and interaction with several cells, such as immune, metabolic or nervous system cells involved in protection against pathogens, contributing to creation of a barrier against infections [37]. It has long been known that the colonization process begins at birth and is strongly influenced by multiple factors up to the age of three years, such as delivery type, antibiotic administration during early childhood, feeding habits, time of weaning, food composition, and environmental hygiene, which will be discussed further.
Gut microbiota development continues during childhood and adolescence becoming increasingly stable over time, but it is generally accepted that it closely resembles that of an adult by around the age of three years [38,39]. The gastrointestinal tract of a healthy adult has been found to harbor approximately 10 2 microbial cells within the stomach, duodenum and jejunum, while the distal ileum hosts between 10 7 and 10 8 microbial cells, the highest proportion being found in the colon which contains around 10 11 -10 12 microbial cells [32]. The human gut contains approximately 1-2 kg of microorganisms with >105-fold more genes compared to the human genome itself [32]. The gut microbiome is crucial in maintaining the host's homeostasis by contributing to the digestive process and energy production, preventing pathogen colonization and regulating the activity of the immune system [40]. Studies have demonstrated that the gut microbiome has the ability to modulate the individual's susceptibility to obesity by increasing caloric extraction from indigestible dietary fibers as well as by augmenting its storage in adipose tissue [40]. Studies have also revealed that the gut microbiome influences whole body metabolism since germ-free mice displayed alterations in liver, kidney and bowel homeostasis [41][42][43].
Diet is a crucial factor involved in both obesity development and gut microbiota regulation. Thus, it has been demonstrated that dietary patterns shape the appearance of distinct combinations of bacteria within the gut, defined as enterotypes [44], by providing different growth-promoting and growth-inhibiting factors for the development of specific phylotypes [45]. It is well-known that obesity is associated with subclinical inflammation, even in pediatric patients, which may lead to a wide spectrum of long-term complications [21,22]. The gut microbiota seems also to contribute to the development of this inflammatory profile. According to a study performed on overweight and obese subjects, individuals that consumed the most unhealthy food, consisting of confectionary and sugary drinks with low consumption of yoghurt, fruit and water, displayed a significantly more severe inflammatory profile when compared to those with the healthiest diet based on increased consumption of yoghurts, fruits and soups [46]. The same authors concluded that the healthiest dietary pattern was strongly related to the highest microbial diversity within the gut [46]. The complex interaction between obesity and gut microbiota is further supported by other findings that have shown a decrease in microbial community richness in the context of overweight and obesity [47]. As mentioned, not only does obesity seem to be shaped by the gut microbiota, but also obesity-associated inflammation. Thus, increased consumption of fat is associated with a more severe degree of white adipose tissue inflammation and its related complications, while polyunsaturated fatty acid-based diets were found to counteract this inflammatory state in overweight and obesity, promoting a metabolically healthy phenotype [48,49]. In terms of protein consumption, it was shown that, if coupled with exercise, this might exert a positive effect on gut microbiota development. Clarke et al. demonstrated a positive correlation between high protein consumption and gut microbial diversity, but they also showed that athletes with a low body mass index harbor significantly higher levels of Akkermansia, which is inversely correlated with obesity in both animals and humans [50,51]. Although the relative abundance of each member of the gut microbial community is highly dynamic during an individual's lifespan, the overall composition of the gut microbiome can be stable for years [52,53]. Thus, we might assume that host genetic factors might also be involved in shaping the gut microbiota apart from dietary patterns, but the extent of this involvement remains unclear. Studies performed on monozygotic and dizygotic twins failed to demonstrate significant associations in gut microbiota profiles at various ages from infancy to adulthood, suggesting that genetic factors have only a minor role in shaping the human gut microbiota [54,55]. In addition, the influence of dietary patterns on gut microbiota has been found to be sex-dependent with diet affecting gut microbiota development differently in males and females [56]. These findings are in contradiction to the previously mentioned study indicating that host genetic factors remain active participants in shaping gut microbiota, and the precise role of genetic factors requires further clarification.
A wide spectrum of inconsistencies between studies is evident in terms of gut microbiota in obese individuals. Thus, although most studies have reported that obese individuals possess reduced richness of microbes or their genes within the gut [47,54,57,58], decreased diversity was not constant in all obese individuals [47]. Another inconsistency is related to the ratio of Bacteroidetes to Firmicutes in obese versus lean individuals, which was observed to increase, decrease or not change at all [59]. Ethnicity is also related to these inconsistencies based on two studies that assessed Prevotella-dominated communities showing a positive correlation of these communities with body mass index in HIV positive subjects and controls originating from Mexico City [60], but not in older Amish individuals [2]. Another recent study also suggested a role of race/ethnicity in triggering distinct gut microbiota subtypes in the context of obesity [61].

Gut Microbiota Composition and Obesity
The composition of the gut microbiota consists of the phyla Firmicutes, Bacteroidetes, Proteobacteria, Actinobacteria and Verrucomicrobia, with Firmicutes and Bacteroidetes accounting for more than 90% of total bacterial species [62]. The gut microbiota in healthy individuals comprises a high ratio of Bacteroidetes to Firmicutes, while this ratio is reversed in obese individuals with a higher abundance of Firmicutes [63]. Moreover, obese individuals were observed to possess a high level of Lactobacillus species and a relatively low abundance of Bacteroides vulgates [64]. Metagenomic studies have suggested that the ratio of Bacteroidetes to Firmicutes might be used as a biomarker of obesity susceptibility [54,65] (Figure 1). Metagenomic analysis of the gut microbiome revealed that 40% of the intestine microbial gene pool is common among all humans around the world, indicating the presence of a core microbiome [66]. Obesity metabolic changes were shown to be influenced by an increase in short-chain fatty acid producers (Actinobacteria) and pathogenic Proteobacteria, with increase in their abundance in the context of obesity [67]. Interestingly, these alterations in gut microbiota were found to be reversible with weight loss [68]. Several mechanisms have been proposed to account for the influence of the gut microbiota on obesity development. One suggested mechanism is related to the ability of microorganisms to ferment dietary polysaccharides otherwise not digested by humans [35]. Thus, the formation process of dietary fibers results in an increased production of shortchain fatty acids (e.g., butyrate, acetate, propionate) identified in fecal or cecal samples, which, after absorption, induce lipogenesis and promote triglyceride storage through various molecular pathways, such as the activation of carbohydrate responsive element-binding protein and sterol regulatory element-binding transcriptor 1 [35,69]. Another peculiarity of short-chain fatty acids is their ability to suppress the fasting-induced adipocyte factor leading to the inhibition of lipoprotein lipase and subsequent accumulation of triglycerides in host adipocytes [35]. Nevertheless, several positive effects of short-chain fatty acids on human metabolism were suggested by Morrison and Preston in a recent review, such as a potential role in maintaining the integrity of the gut mucosa (butyrate), beneficial effects on glucose homeostasis, a role in appetite regulation (propionate), and the regulation of immune system and inflammatory response [70]. Considering the inconsistencies reported in the literature related to this topic, further studies are needed to clarify the precise role of short-chain fatty acids in human homeostasis, especially in terms of lipid metabolism.

Maternal Factors
The gut microbiota shaping process begins in utero and the most important factor that influences fetal wellbeing is maternal diet [71,72]. According to Gibson et al., offspring born to mothers consuming fish oil were found to have a higher level of opportunistic bacteria, such as Enterococcus faecium, Bilophila wadsworthia and Bacteroides fragilis, with a negative impact on the infant's immune responses [73]. In contrast, the study of Chu et al. [71] found a reduced level of Bacteroidetes in infants whose mothers consumed a high-fat diet during pregnancy. Habitual maternal intake of probiotic-containing food leads to a reduced risk for both spontaneous preterm delivery [74], and allergic disorders during childhood [75]. In addition, maternal alcohol consumption during pregnancy was demonstrated to negatively influence both maternal and neonatal gut microbiome, and, in terms of neonatal outcomes, it resulted in an increased risk for infections and other disorders later in life [76,77].
Maternal health during pregnancy is another key factor in infant wellbeing. Several studies aiming to identify the relationship between maternal health and infant wellbeing concluded that the microbial composition of infants born to obese mothers differs markedly from those born to lean mothers due to increased exposure of the offspring to oxidative stress [78][79][80]. It is well-known that infants born to mothers diagnosed with diabetes, or those with an increased body mass index, are more prone to developing diabetes and obesity in later life [81][82][83] due to alterations in the maternal gut microbiome during both pregnancy and lactation [84]. A recent study concluded that the microbiome in overweight or obese mothers is associated with an increased risk for early obesity in infants since early life microbial composition predicts obesity development later in life [85]. Nevertheless, the precise composition of pregnant women's gut microbiota is not yet clearly established and reported findings remain contradictory. Collado et al. observed increased levels of Clostridium, Bacteroidetes and Staphylococcus, with a decreased level of Bifidobacterium, in mothers with overweight/obesity [78], while the study of Santacruz et al. found an abundance of Staphylococcus, Enterobacteriaceae and Staphylococcus, with a significantly lower level of Bifidobacterium and Bacteroides, when assessing a similar group [86].
Dysbiosis of the placental microbiota might represent another important factor for neonatal health. Despite the fast colonization during the first three years of life, the hypothesis that the fetus is seeded with bacteria during the intrauterine life from the maternal placental microbial community cannot be ruled out [87]. The few studies that have assessed placental microbial composition observed an increase in the order Pseudomonadales, and the genera Actinobacteria and Proteobacteria, along with lower levels of Bacteroidetes and Firmicutes in pregnant women with gestational diabetes mellitus [88,89]. Furthermore, studies that compared meconium samples of neonate subjects born to mothers with dia-betes to those whose mothers were non-diabetic found an abundance of the Bacteroidetes and Lachnospiraceae families and the Parabacteriodes genus along with decreased levels of Proteobacteria in meconium samples from the diabetes group [90].
Based on the above, we might hypothesize a complex interaction between maternal diet and obesity, gestational diabetes mellitus, placental microbial composition and the risk of childhood obesity through important alterations in the newborn's gut microbiome, beginning from intrauterine life.

Delivery Mode
It has been well-documented that the delivery mode shapes the neonatal gut microbiome. Thus, when born vaginally, the neonate comes in contact with the maternal vaginal and gut microbiota and the infant's gut microbiome will be similar to the vaginal microbial composition, being dominated by the genera Lactobacillus, Sneathia and Prevotella [91]. Otherwise, if the baby is born by cesarean section, the major component of the neonate's gut microbiome resembles the mother's skin microbiome and nosocomial surroundings [92]. Recent studies have revealed an important relationship between cesarean section delivery and the risk of obesity during childhood [93,94]. It was found that infants born by cesarean section possess a lower diversity of the gut microbial community compared to vaginally delivered infants, with lower levels of both Bifidobacteria and Escherichia-Shigella, along with a complete absence of Bacteroides [95][96][97][98]. These infants display a gut microbial community mainly composed of the genera Staphylococcus, Propionibacterium and Corynebacterium, resembling the mother's skin microbial community [91].

Feeding Patterns
Breastfeeding remains the most appropriate alimentation pattern for infants since human milk contains important bioactive compounds, such as oligosaccharides, which are essential for infant development and offer protection against several childhood disorders, such as obesity [99], type 2 diabetes mellitus [100], celiac disease [101], allergies [102], diarrhea [103], and other metabolic diseases [104]. Moreover, human oligosaccharides have the ability to promote Bifidobacteria and Bacteroides growth [105], and to regulate infant wellbeing due to their action as prebiotics, modulators of innate immune responses and their anti-inflammatory properties [106]. Additional evidence suggests that Bifidobacterium species, such as B. longum, B. breve, B. infantis and B. pseudocatenulatum, represent the most common Actinobacteria in breastfed infants [107][108][109], along with lactic acid bacteria, such as Enterococcus, Lactobacillus and Clostridium species [110]. Maternal health is another crucial modulator of human milk content resulting in alterations of the human milk microbiome and subsequent negative changes in the infant gut microbiome [92]. Maternal obesity was found to alter the human milk microbiome by decreasing bacterial community diversity, leading to lower levels of Bifidobacterium and higher levels of Staphylococcus [111,112]. Recent evidence has also underlined the role of human milk in maintaining infant protein balance [113]. In terms of formula-fed infants, a recent study, which assessed the composition of formula-fed and breast-fed infants gut microbiota at the age of 40 days, found that α diversity was lower in a breast-fed group compared to formula-fed infants [114]. These findings were supported by other studies which concluded that the gut microbiota of formula-fed infants is more diverse compared to that of breastfed infants, being similar to that of older children [115,116]. The most predominant genus in both breastfed and formula-fed infants are Bifidobacterium and Enterobacteriaceae which seem to decrease with age to almost zero by 18 months [114,117]. In addition, formula-fed infants were found to have increased levels of both Clostridia and Veillonella spp. compared to breastfed infants [108,114,116]. Considering that human milk is incontestably the most suitable feeding pattern for infants, further studies are required to identify the precise role of each bacterium within the gut microbial community and the differences between feeding patterns.

Weaning and Diet
Diet definitely plays a crucial role in modulating the microbial community with it being observed that, once weaned, infants begin to develop a similar microbial composition to that of adults with an increase in diversity and significant changes in taxonomic groups [108,109]. Most changes in gut microbiota composition are observed between the ages of 9 to 18 months, consisting of a decrease of Lactobacilli, Bifidobacteria and Enterobacteriaceae and an increase in Bacteroides and Clostridium species, reinforcing the major role of weaning in triggering these changes [110]. Moreover, the type of diet contributes to the composition of the gut bacterial community. Assessing the role of dietary habits in children with obesity, Nakayama et al. observed that a high-fat westernized diet results in a decrease in Prevotellaceae in the gut microbiota which might contribute to obesity development during childhood [118]. However, the authors stated that children with a carbohydrate-based diet showed an increase in Prevotellaceae and a decrease in Bacteroidaceae [118]. Therefore, weaning and diet probably represent decisive factors in gut microbiota development since they are the last factors shaping the microbial community before the age of three years, when this process achieves relative stability.
The influence of the above-mentioned factors involved in shaping gut microbiome during early childhood is summarized in Table 1. The precise role of each factor in the development of pediatric obesity is a topic that definitely requires further research ( Figure 2).

Summary
The gut microbiota comprises a wide range of bacteria with complex physiological functions which extend to extraintestinal tissues and seem to be related to excessive weight gain [33]. It has been demonstrated that the gut microbial community might have a dichotomous role influenced by environmental factors, with the ability to promote energy storage or, in contrast, to favor leanness [110,119,120]. The shaping of the gut microbiota begins in utero and continues up to the age of 3-4 years being influenced by the maternal diet, amniotic fluid and placental microbial community, type of delivery, feeding practices and hygiene habits, such as brushing the teeth after meals [121]. The most important players related to obesity risk in the gut microbiota are the Firmicutes and Bacteroides. Studies on both animal and human subjects have found that obesity is associated with an increase in Firmicutes and a decrease in Bacteroidetes, independently of dietary caloric intake [34,122]. A recent systematic review [121] supported this statement, highlighting that several studies have found a positive relationship between higher levels of certain microbes belonging to the Firmicutes (e.g., Eubacterium halli, Clostridium leptum and Lactobacillus spp.) and overweight or obesity [110,119]. Nevertheless, the findings assessing the relationship between gut microbial composition and obesity remain contradictory, since other studies have found the contrary, with observed decrease in bacteria belonging to the same phyla (e.g., the genera Staphylococcus and Clostridium difficile [33,120]). Moreover, several studies have found higher levels of Bacteroides fragilis in children with overweight/obesity compared to normal-weight children [119,120,123], while others have found that overweight/obesity is associated with a decrease in the entire phylum Bacteroidetes, along with a decrease in the Bacteroides/Prevotella groups [124,125]. In addition, a meta-analysis that included mostly adults failed to find any association between the Bacteroidetes/Firmicutes ratio and obesity [65]. A recent study performed on 3-year-old children found inverse relationships between both Parabacteroidetes and Peptostrecoccae U and overweight/obesity [126]. The same study found a tendency of overweight/obese children to exhibit relative abundance of Ruminococcus and Akermansia [126]. Thus, the authors concluded that several differences in gut microbial composition between obese and lean adults are similar to the 3-year-old age group suggesting obesity predisposition has its origins in early childhood [126].
Regardless of the inconsistencies identified between studies, we must acknowledge the major role of gut microbiota in the development of childhood overweight/obesity. Additional studies to clarify the precise role of each bacterium from the gut microbiota would represent a major development in the era of pediatric obesity. Based on the wide spectrum of long-term complications associated with increased body mass index, filling the existing gaps regarding the relationship between gut microbial composition and pediatric obesity might be the cornerstone for the development of innovative prophylactic and/or therapeutic strategies targeted at decreasing the incidence of overweight/obesity in both children and adults.