Introduction
The worldwide prevalence of obesity is now recognized to have reached epidemic proportions. The World Health Organization reports that at least 1.9 billion adults are overweight while 600 million are clinically obese. 1 Obesity is especially pronounced in young individuals, including children, and women of reproductive age.Reference Hedley, Ogden and Johnson 2 , Reference Ogden, Carroll, Kit and Flegal 3 Maternal obesity is associated with an increase in both maternal and fetal complications during pregnancy;Reference Catalano and deMouzon 4 where maternal obesity not only impacts the health of the mother,Reference Athukorala, Rumbold, Willson and Crowther 5 , Reference Crane, Murphy, Burrage and Hutchens 6 it also has short- and long-term ramifications for her offspring.Reference Portela, Vieira, Matos, de Oliveira and Vieira 7
It is now well established that disease risk strongly correlates with early life environment, and offspring responses to the postnatal environment are modulated by intrauterine adaptations.Reference Gluckman and Hanson 8 , Reference Gluckman and Hanson 9 The mechanistic signaling pathways linking maternal to offspring disease risk remain unclear. With the emergence of the gut microbiota as a factor influencing disease risk,Reference Turnbaugh, Ley and Mahowald 10 , Reference Henao-Mejia, Elinav and Jin 11 new concepts governing how the colonization of the infant gut occurs,Reference Aagaard, Ma and Antony 12 – Reference Azad, Konya and Maughan 15 and the role of infant gut colonization in mediating the link between maternal and child obesity are evolving. In this paper, we review the current animal literature investigating gut colonization early in life, and the role of the gut microbiota in mediating obesity risk in the context of maternal overweight and obesity.
Early life nutritional programming of offspring obesity and metabolic dysfunction
Since the advent of the Developmental Origins of Health and Disease (DOHaD) hypothesis established through the early investigations of epidemiological data sets,Reference Barker 16 – Reference Barker 18 many animal models have consistently produced data to support the hypothesis that the intrauterine environment plays a central role in offspring metabolic compromise and obesity risk. Studies in both large animals, such as the sheep and pig, and small animals including the mouse, rat and guinea pig, have been instrumental in advancing our understanding of the relationship between early life adversity and the risk of developing disease later in life. In rodents, diet-induced maternal obesity results in offspring that are often growth restricted,Reference Howie, Sloboda, Kamal and Vickers 19 , Reference Zheng, Xiao and Zhang 20 but not always,Reference Shankar, Harrell and Liu 21 , Reference Page, Malik, Ripple and Anday 22 that display postnatal catch-up growth,Reference Howie, Sloboda, Kamal and Vickers 19 , Reference Page, Malik, Ripple and Anday 22 exhibit hyperinsulinemia and hyperleptinemia, insulin resistance as well as defects in pancreatic signaling and insulin secretion,Reference Howie, Sloboda, Kamal and Vickers 19 , Reference Samuelsson, Matthews and Argenton 23 – Reference Taylor, McConnell and Khan 26 and hepatic glucose and lipid handling dysfunctions.Reference Zheng, Xiao and Zhang 20 , Reference Borengasser, Lau and Kang 27 – Reference Bruce, Cagampang and Argenton 29 These effects are often magnified when offspring are themselves fed a postweaning high fat diet, highlighting interactive effects between the pre- and postnatal environments.Reference Howie, Sloboda, Kamal and Vickers 19 , Reference Shankar, Harrell and Liu 21 , Reference Page, Malik, Ripple and Anday 22 , Reference Connor, Vickers, Beltrand, Meaney and Sloboda 30 These observations are similar to those observed in large animals, where maternal obesity induces impaired peripheral insulin signaling in offspring,Reference Yan, Huang and Zhao 31 defects which appear present already during fetal life.Reference Zhu, Han and Tong 32 , Reference Yan, Zhu and Xu 33 Although less numerous, studies in non-human primates are consistent with those done in small and large animal models, showing significant metabolic impairments in offspring born to obese/high fat fed mothers.Reference McCurdy, Bishop and Williams 34 – Reference Suter, Chen and Burdine 39 Despite great advances in our understanding of the molecular and physiological pathways that govern the relationship between our early life environment and long-term disease risk, the fundamental mechanisms underpinning the link between maternal obesity and childhood obesity are still unclear. As such new lines of investigation are constantly explored in an effort to identify targets for intervention. One such system is the role of the gut microbiota and the host–microbe relationship in mediating gut development, nutrient absorption and immune function.
The gut microbiota
Trillions of microbes reside within the human gut; collectively these microbes constitute our microbiota, and the genes they encode and the environment that they inhabit are known as our microbiome.Reference Whitman, Coleman and Wiebe 40 Coined as our ‘forgotten organ’, these commensal bacteria live symbiotically within the host, having coevolved with vertebrates over many millennia.Reference Chow, Lee, Shen, Khosravi and Mazmanian 41 The intestinal commensals play an essential role in the maintenance of host health as they are responsible for the metabolism of indigestible polysaccharides, the production of essential nutrients, the regulation of fat storageReference Backhed, Ding and Wang 42 and protection from pathogenic bacterial colonization (for review, see Nieuwdorp et al.Reference Nieuwdorp, Gilijamse, Pai and Kaplan 43 ). Commensal bacteria and their metabolites are essential for the normal development and function of the mammalian immune system, and the maintenance of gut epithelial barrier integrity.Reference Hooper and Macpherson 44 , Reference Turner 45 Studies in germ-free mice have shown that the gut microbiota is essential for the growth and development of the gastrointestinal system.Reference Reyniers 46 These gnotobiotic mice have reduced intestinal weight and surface area and poor nutrient absorption,Reference Thompson and Trexler 47 as commensal bacteria are required for the breakdown of nutrients.Reference Duncan, Lobley and Holtrop 48 Perturbations in this ecosystem, including those associated with diet, have been linked with an increased risk of developing non-communicable disease including obesity (for review, seeReference Ley 49 , Reference Ley, Backhed and Turnbaugh 50 ). The most cited microbial dysbiosis that characterizes an obesogenic phenotype is a shift in the two most prominent phyla in the mammalian gut; Firmicutes and Bacteroidetes. There is currently no sound evidence characterizing the mechanisms that establish a microbial profile commiserate those observed in a lean ‘healthy’ phenotype v. those of an obese ‘diseased’ phenotype. Furthermore, there exists a large discrepancy in the literature as to which bacterial genera and species are associated with deleterious outcomes. What constitutes a ‘healthy’ microbiota from the time of gut establishment, to maintenance throughout life, fails to exist in the literature. Despite these gaps, however, emerging studies have begun to shed some light on host–bacterial relationships and their role in metabolic phenotype.
Early life gut colonization: how early is early?
The hypothesis that the fetus develops within a sterile intrauterine environment was originally proposed over 100 years ago by French pediatrician Tissier,Reference Tissier 51 and since then has gone largely unchallenged, primarily due to the fact that many obstetric complications including preterm birth have been associated with intrauterine infection.Reference Lamont 52 As such, it has long been proposed that under normal circumstances, the infant acquires a bacterial inoculation at the time of birth, where exposure to maternal fecal and vaginal microbes establishes the budding neonatal gut microbiota. Early animal studies supported this contention.Reference Schaedler, Dubos and Costello 53 However, improvements in large-scale sequencing and computational analyses of metagenomicsReference Methe 54 have revealed alternative theories. Increasing evidence suggests that the developing fetal gut may be seeded before birth. Bacterial populations associated with the maternal gastrointestinal tractReference Jiménez, Marín and Martín 55 have been isolated using both culture-dependant and independent techniques, from meconium,Reference Jiménez, Marín and Martín 55 , Reference Moles, Gómez and Heilig 56 fetal membranesReference Jiménez, Marín and Martín 55 and from umbilical cord blood of healthy neonates.Reference Jiménez, Fernández and Marín 57 Pioneering animal studies demonstrated that oral administration of pregnant mice with genetically labeled Enterococcus fecium resulted in the presence of this bacterium in fetal meconium after sterile C-section,Reference Jiménez, Marín and Martín 55 supporting the hypothesis that the fetus may be exposed to maternally gut-derived bacteria in utero and that this exposure may prime gut development. Thus, it is possible that the maternal microbiota provides the first fetal bacterial inoculum and then from birth, microbial diversity increases to converge to an adult-like profile. We understand little about the developing ‘fetal microbiota’. However, factors including maternal gut bacterial load, bacterial composition, diet, genetics,Reference Collado, Isolauri, Laitinen and Salminen 58 mode of delivery, antibiotic useReference Fak, Ahrne, Molin, Jeppsson and Westrom 59 , Reference Mueller, Whyatt and Hoepner 60 intestinal tract development,Reference Pacha 61 bile acids, pHReference Fak, Ahrne, Molin, Jeppsson and Westrom 59 and host–microbial interactions,Reference Mackie, Sghir and Gaskins 62 which influence microbial establishment have allowed researchers to begin to understand more about how our gut microbiota is established, and how it is maintained over the life course.
Factors influencing early life microbial colonization
Maternal bacterial translocation during pregnancy
The concept that maternal gut microbes may prime fetal immune development has recently gained interest in the current literature. Pioneering studies in mice show that during pregnancy and lactation there exists bacterial translocation to extra-intestinal tissues, where pregnant mice have gut-derived bacterial presence in mesenteric lymph nodes, and within 24 h after delivery, had viable bacteria in mammary tissue.Reference Perez, Doré and Leclerc 63 In contrast to pathologic conditions in which the translocation of microbes are mainly Gram-negative, penetrating species in pregnant and lactating mice included Streptococcus, Lactobacillus and Bifidobacterium, whose numbers subsided gradually over time.Reference Perez, Doré and Leclerc 63 Recent human data suggest that the placenta may harbor a microbiota,Reference Aagaard, Ma and Antony 12 and although not yet shown in animal studies, the maternal translocation of gut microbes to the placenta has biological plausibility. Despite the fact that translocation through the gut epithelium is prohibited by a well-established barrier,Reference Turner 45 dendritic cells can paracellularly sample the gut lumen and its bacteria/bacterial components to facilitate bacterial transport to the periphery through lymph.Reference Rescigno, Rotta, Valzasina and Ricciardi-Castagnoli 64 , Reference Rescigno, Urbano and Valzasina 65 Whether pregnancy changes gut barrier function and/or dendritic sampling is unclear, however, data in mice show that a significant proportion of pregnant animals harbor bacteria in mesenteric lymph nodes during pregnancy compared with their non-pregnant counterparts.Reference Perez, Doré and Leclerc 63 Dendritic cell sampling may possibly explain bacterial translocation from the gut to peripheral sites such as the placenta, resulting in the presence of bacteria in fetal meconium.Reference Jiménez, Marín and Martín 55 These data are consistent with those derived from a non-human primate model where the bacterial species isolated from the monkey milk were found to be gut commensals.Reference Jin, Hinde and Tao 66 These early studies support the notion that during pregnancy, bacterial translocation to peripheral sites can occur – indeed a recent review has highlighted the fact that maternal bacterial translocation is a well-known phenomenon in animals outside of the mammalian class.Reference Funkhouser and Bordenstein 67
Coincident with the notion of in utero gut colonization, has been the concept that with pregnancy, the composition of the maternal gut microbiome changes. It is well established that the mother makes physiological adaptations during pregnancy that are essential to meet the metabolic demands of pregnancy, and maternal intestinal adaptation is an area of budding research. Over the course of healthy human pregnancy two main adaptations have been demonstrated to occur in the maternal human gut; intestinal microbial load increasesReference Koren, Goodrich and Cullender 68 and bacterial composition shifts.Reference Collado, Isolauri, Laitinen and Salminen 58 In a landmark paper, Koren et al. Reference Koren, Goodrich and Cullender 68 demonstrated that the pregnant microbiome shifts from first to third trimester of pregnancy, characterized by an increase in the abundance of Proteobacteria and Actinobacteria. Furthermore, colonization with first and third trimester human microbiota showed that third trimester microbiota induced insulin insensitivity and elevated adiposity in gnotobiotic mice compared with first trimester microbiota.Reference Koren, Goodrich and Cullender 68 These unique observations suggest that the maternal gut microbiota may facilitate the necessary metabolic adaptations that occur late in pregnancy. Whether maternal gut barrier function and dendritic bacterial translocation similarly shift with normal advancing gestation to facilitate fetal gut colonization is currently unknown. Further, Koren et al. noted no significant changes in microbiota composition between the third trimester of pregnancy and 1 month postpartum, an observation that was repeated in one other study.Reference Jost, Lacroix, Braegger and Chassard 69 However, differences in microbial populations appear to differ between studiesReference Koren, Goodrich and Cullender 68 , Reference Jost, Lacroix, Braegger and Chassard 69 and highlight the need for stringent and consistent molecular techniques and computational analyses when comparing data from different cohorts and studies. To date there are no animal data describing pregnancy-induced changes in the maternal gut microbiota over the course of pregnancy.
Postnatal gut colonization: mode of delivery and nutrition
Consistent with the Developmental Origins hypothesis that the prenatal and postnatal environments interact to establish adult phenotype, it may be that prenatal priming of the fetal gut modulates postnatal acquisition of an established microbiome. Unlike several mechanistic processes that have largely been studied in animal models, many studies describing the vertical transmission of maternal microbes at birth (vaginal microbiome v. C-section) and during lactation through breast milk, have been done in human cohorts.Reference Azad, Konya and Maughan 15 , Reference Makino, Kushiro and Ishikawa 70 – Reference Bezirtzoglou, Tsiotsias and Welling 72 Infants delivered vaginally harbor microbial communities similar to maternal vaginal populations, while microbes similar to the maternal epithelium dominates the intestinal tract of animals born by C-section.Reference Dominguez-Bello, Costello and Contreras 73 Lactobacillus, Prevotella and Sneathia species dominate vaginally born infants while C-section infants harbor Staphylococcus, Corynebacterium and Propionibacterium.Reference Dominguez-Bello, Costello and Contreras 73 This difference based on mode of delivery is speculated to be a result of the failure of contact between maternal vaginal and fecal flora.Reference Murono, Fujita and Yoshikawa 74 Furthermore, differences between C-section and vaginally born infant microbial populations has been documented to persist 4Reference Azad, Konya and Maughan 15 and 6 months after birthReference Gronlund, Lehtonen, Eerola and Kero 75 and up to 7 years later.Reference Salminen, Gibson, McCartney and Isolauri 76 Animal studies also show distinct differences in neonatal gut colonization as a result of mode of delivery. Within the first 2 weeks of life, vaginally delivered piglets shown higher proportions of Bacteroides Prevotella and Clostridium XIVa associated with higher levels of propionate and butyrate in the ascending colon.Reference Wang, Radlowski and Monaco 77 Interestingly, gut colonization was independently modulated by postnatal nutrition, where sow-reared (breast fed) piglets showed higher levels of butyrate, isobutyrate, valerate and isovalerate concentrations v. formula-fed piglets,Reference Wang, Radlowski and Monaco 77 suggesting that gut microbial metabolic activity differs between sow-reared and formula-fed piglets. Although not measured in this study, the presence of elevated proportions of Bacteroides-Prevotella, Clostridium, Bifidobacterium spp. and Lactobacillus spp. may explain the differing proportion of short-chain fatty acids (SCFAs). This result is consistent with human data that report distinct microbial profiles between breast and formula-fed infants.Reference Bezirtzoglou, Tsiotsias and Welling 72 , Reference Fallani, Young and Scott 78 , Reference Gomez-Llorente, Plaza-Diaz and Aguilera 79 Mice born by C-section, compared with spontaneous vaginal delivery, displayed a distinct bacterial profile at weaning characterized by a higher abundance of Bacteroides and Lachnospiraceae and reduced abundance of Rikenellazeae and Ruminococcus; however, no taxonomic clustering according to delivery method persisted to adulthood.Reference Hansen, Andersen and Krych 80 This appears to be in contrast to recent human data demonstrating that adults born by C-section have a distinctly different gut microbial composition.Reference Goedert, Hua, Yu and Shi 81
Postnatal gut colonization: exposure to prenatal antibiotics
Antibiotic treatment during pregnancy is commonly used to combat urinary and gastrointestinal infections,Reference Le, Briggs, McKeown and Bustillo 82 , Reference Chan, Hohmann, Barshak and Pukkila-Worley 83 bacterial vaginosis,Reference Sheehy, Santos, Ferreira and Berard 84 and to prevent maternal and infant infection during gestation and after delivery.Reference Turrentine, Greisinger, Brown, Wehmanen and Mouzoon 85 Antibiotic administration during pregnancy alters vaginal microbial populations,Reference Stokholm, Schjorring and Eskildsen 86 contributes to aberrant microbial colonization of the infant gutReference Fallani, Young and Scott 78 , Reference Jauréguy, Carton and Panel 87 , Reference Mangin, Suau, Gotteland, Brunser and Pochart 88 and has recently been associated with an increased risk of developing childhood obesity.Reference Mueller, Whyatt and Hoepner 60 Given the close relationship between the gut microbiota and risk of obesity,Reference Ley, Backhed and Turnbaugh 50 animal studies investigating broad-spectrum antibiotics during gestation have been used to understand the link between gut dysbiosis and long-term obesity risk (see next section). In rats, a single course of an antibiotic-cocktail of metronidazole, neomycin and polymyxin administered in drinking water 3 to 4 days before delivery resulted in higher stomach pH, lower stomach weight and increased intestinal permeability in neonates.Reference Fak, Ahrne, Molin, Jeppsson and Westrom 59 Treatment also promoted antibiotic-resistant Enterobacteriaceae in the offspring at birth, which persist beyond lactation.Reference Fak, Ahrne, Molin, Jeppsson and Westrom 59 Similar changes in long-term gut permeability have been demonstrated in pigs.Reference Arnal, Zhang, Erridge, Smidt and Lallès 89 Increased gut permeability has been associated with increased bacterial translocation to metabolic tissues and in the promotion of non-alcoholic fatty liver disease (for review, seeReference Scarpellini, Lupo and Iegri 90 ), potentially linking aberrant neonatal gut colonization with long-term metabolic complications. Administration of similar antibiotic cocktails to non-obese diabetic (NOD) mice in early and late pregnancy induced a persistent reduction in offspring microbial diversity, distinct clustering of bacterial taxa and an increased risk of diabetes incidence observed at 20 weeks of age, but this disappeared at 30 weeks of age.Reference Tormo-Badia, Hakansson and Vasudevan 91 Collectively, these studies suggest that prenatal antibiotic administration induces permanent changes in both gut microbial populations and gut function, potentially contributing to an increased risk of developing metabolic complications in adulthood.
Early life modulation of gut colonization and risk of obesity and metabolic compromise
Maternal obesity and nutritional excess in large and small animal models
With our knowledge that different bacterial species utilize and produce distinct nutritional substrates and metabolites, and vary in their capacity to harvest energy,Reference Turnbaugh, Ley and Mahowald 10 the concept that obesity risk can be modulated by gut microbial communities has become plausible. An ‘aberrant’ gut microbiota is hypothesized to be equally as detrimental to organism health as having no microbiota at all.Reference Fak, Ahrne, Molin, Jeppsson and Westrom 59 Early genomic studies in male mice examining 16S ribosomal RNA gene sequences of cecal microbiota harvested from genetically obese ob/ob,Reference Ley, Backhed and Turnbaugh 50 leptin deficient db/db mice, and models of diet-induced obesityReference Turnbaugh, Ridaura and Faith 92 , Reference Turnbaugh, Backhed, Fulton and Gordon 93 have collectively demonstrated shifts in the proportions of Bacteroidetes and Firmicutes relative to lean counterparts. These models showed a marked reduction and proportional increase in the levels of the Bacteroidetes and Firmicutes phyla, respectively.Reference Ley, Backhed and Turnbaugh 50 Shotgun sequencing of the intestinal flora harvested from ob/ob mice display an enrichment of genes encoding enzymes involved in energy extraction, supporting the notion that the prominent bacterial populations of the obese gut specialize in energy harvest. Furthermore, upon colonization of germ-free mice with the distal gut microbiota harvested from conventionally raised mice, a dramatic increase in adiposity is observed within 10–14 days despite a decrease in food consumption {Turnbaugh, 2006 #11}. This change in adiposity is hypothesized to involve several related mechanisms including the microbial fermentation of indigestible dietary polysaccharides, subsequent intestinal absorption of monosaccharides and SCFAs, their conversion to more complex lipids in the liver and microbial regulation of host genes to promote the deposition of lipids into adipocytes {Turnbaugh, 2006 #11}. Collectively, these findings led researchers in the field to propose that the microbiota of obese individuals may be more efficient at extracting energy from a given diet than the microbiota of lean counterparts. Consistent results have been drawn from human work.Reference Turnbaugh, Ley and Mahowald 10
Similar studies have yet to be conducted examining the evolution of the maternal obese microbiota throughout pregnancy in experimental animal models. In humans, two main factors have been identified that are associated with shifts in commensal populations in the maternal gut; pre-pregnancy weight and gestational weight gain. Examination of the microbiota of 36 normal weight [body mass index (BMI)<25] and 18 overweight (BMI>30) women demonstrated division-specific blooms of Bacteroides and Staphylococcus in overweight women, where proportions of Bacteroides was correlated to gestational weight gain.Reference Collado, Isolauri, Laitinen and Salminen 58 Other studies have identified increases in Staphylococcus, Enterobacteriaceae and Escherichia coli, which have been reported to be strongly associated with metabolic endotoxaemia and inflammation. Bifidobacerium and Akkermansia muciniphila, responsible for the production of metabolites such as butyrate and mucin degradation, respectively, decrease in relative abundance in overweight pregnant women.Reference Santacruz, Collado and Garcia-Valdes 94 These results are in accordance with a recent study that displayed elevated levels of Staphylococcus and reduced levels of Bifidobacterium in the feces of infants, whom at the age of 7, were classified as clinically overweight.Reference Santacruz, Collado and Garcia-Valdes 94 Although elevated pre-pregnancy weight and gestational weight gain was associated with altered proportions of intestinal commensals, currently there are no data describing how the maternal microbiota evolves over the course of pregnancy in the context of obesity. Although existing data suggest a strong relationship between maternal obesity, gestational weight gain and gut microbial shifts, research has yet to reveal mechanisms explaining why these shifts occur and how microbial shifts are altered during obese pregnancy.
Despite our lack of understanding of the obese pregnant microbiome, studies describing gut microbial shifts and changes in gut function in offspring born to obese or diet-induced obese mothers have begun to emerge. Early studies have shown that maternal obesity induces ectopic adipose deposition in skeletal muscleReference Yan, Huang and Zhao 31 as well as enhanced expression of proinflammatory cytokines in fetal gut and young adult ovine offspring.Reference Yan, Huang and Wang 95 Indeed, free fatty acids can increase gut permeability to microbial products such as lipopolysaccharide (LPS), triggering gut inflammationReference Stenman, Holma and Korpela 96 or can directly induce pattern recognition receptors (TLRs) and downstream proinflammatory pathways.Reference Huang, Rutkowsky and Snodgrass 97 Despite the lack of microbial data, these early data collectively support the hypothesis that maternal obesity induces adaptive changes in fetal gut development that could predispose to gastrointestinal inflammatory disorders and to obesity.
In a murine model of maternal high fat diet, offspring born to mothers fed a high fat/high sugar diet throughout pregnancy had increased incidence of infection, autoimmune disorders and allergic sensitization associated with enhanced colonic inflammatory responses and increased circulating bacterial LPS.Reference Myles, Fontecilla and Janelsins 98 These offspring also demonstrated decreased fecal bacterial diversity with an increased ratio of Firmicutes to Bacteriodetes and an over representation of the genera Lachnospiraceae and Clostridiales.Reference Myles, Fontecilla and Janelsins 98 Although similar results were found in offspring of mice fed high omega-3 fatty acids during pregnancy, it appears that different fatty acids promote distinct shifts in gut microbiota as omega-3 pups demonstrated increases in Blautia, Oscillibacter, Clostridales, Robinsoniella, Lactococcus and Eubacterium, but a significant reduction in the abundance of Lachnospiraceae, Anaerotruncus and Roseburia,Reference Myles, Pincus, Fontecilla and Datta 99 highlighting the need to compare similar diets when comparing shifts in offspring microbial populations as a result of early life manipulations.
Despite increasing numbers of studies demonstrating that intrauterine exposure to an obesogenic environment not only predisposes offspring to systemic and organ-specific inflammation and gut dysbiosis, data on gut development and colonization derived from animal models must be interpreted with caution. Many model systems now exist contributing to our current understanding of host–microbiota interactions.Reference Kostic, Howitt and Garrett 100 Rodents display unique gut microbial populations where the relative abundances of most of the dominant genera in mouse and human are quite different; genera with high abundance in human gut microbiota include Prevotella, Faecalibacterium and Ruminococcus, whereas Lactobacillus, Alistipes and Turicibacter are more abundant in mouse gut microbiota.Reference Nguyen, Vieira-Silva, Liston and Raes 101 Rodents are also postnatal developers where rats and mice are born with immature gastrointestinal tracts that mature during a critical window within the first 10–14 days of postnatal life. However, despite these limitations, rodent models have contributed significantly to our understanding of the gut microbiome and its relationship to metabolic function and risk of obesity. Microbial manipulations essential to research are easily accomplished in small animals. Transgenic mice, gnotobiotic mice, as well as antibiotic administration and fecal colonization are easily performed in small animals that breed reasonably well in a short time period. Studies investigating ruminant species such as the sheep, which exhibit unique gut functionReference Hofmann 102 further contribute to our knowledge of species-specific shifts in microbial populations and gut susceptibilities. Thus, the advantages of using these types of animal models, as well as our detailed understanding of their biology, continue to make them indispensible in our investigations of early life programming of gut colonization and its relationship to long-term disease risk.
Although difficult to perform and costly, studies in non-human primates offer a close evolutionary relative and will significantly contribute to our understanding of early life impacts on human gut colonization and obesity risk. One such study has investigated the developmental programming of offspring gut colonization through maternal diet. As demonstrated by distinct clustering with principle coordinate analysis, maternal high-fat diet intake resulted in offspring gut dysbiosis, where the data that cluster tightly together and display the most significant dysbiosis was observed in offspring who experienced a high fat diet during fetal and postnatal life. Offspring demonstrated a depletion of Proteobacteria, specifically Campylobacter spp. and Helicobacter spp, and enrichment of the Firmicutes, Ruminococcus and Dialister.Reference Ma, Prince and Bader 103 It was proposed that the maternal diet, rather than maternal obesity per se, played a major role in shaping the offspring gut microbiome at 1 year of age.Reference Ma, Prince and Bader 103
Although human studies are still lacking, a recent report has shown that maternal obesity is related to differences in the composition of the gut microbiota in children between 18 and 27 months of age even after accounting for paternal BMI and indicators of child body composition, suggesting that exposure rather than genetics may account for this shift in composition.Reference Galley, Bailey, Kamp Dush, Schoppe-Sullivan and Christian 104 Interestingly, effects of maternal obesity on the composition of the gut microbiota in offspring were stronger and more consistent among those born to mothers of higher socioeconomic status (SES) as defined by income and/or education. Also, when PiCRUST was used to examine metagemone function based on obesity status in the higher income group only, no significant differences were found. Although the mechanisms underlying the interaction between maternal obesity and SES in predicting the composition of the child gut microbiota are not known, research investigating the relative contribution of genetic v. behavioral factors in obese adults may differ between individuals of high and low SES. Despite the fact that no differences in the children from obese and non-obese mothers were found in terms of breastfeeding behavior, age at which solid foods were introduced, or the current frequency of consumption of meat, vegetables and cereals, this study did not include detailed food diaries and it is possible that subtle changes in behavior and lifestyle could account for the differences in high v. low SES status in this population. The authors suggest that the extent to which maternal obesity confers measurable changes to the offspring gut microbiome may differ based on the etiology of maternal obesity. No doubt this area of research will continue to expand and provide us with a clearer understanding of the inter-relationship between lifestyle, behavior, maternal obesity, childhood obesity and their impacts on early life gut colonization.
Early life nutritional interventions: modifying gut colonization and obesity risk
Studies of host–microbe interactions early in life offer a new approach to investigate the mechanistic basis of obesity and related metabolic disorders. Although a number of nutritional interventions during the perinatal time window have emerged as beneficialReference Lillycrop, Phillips, Jackson, Hanson and Burdge 105 – Reference Benaissa, Merzouk, Merzouk and Narce 110 , few have investigated impacts on early life gut colonization and function. Maternal dietary supplementation of methyl donors has been found to be effective in preventing early life-induced metabolic compromise and obesity, likely through changes in DNA methylation.Reference Lillycrop, Phillips, Jackson, Hanson and Burdge 105 , Reference Waterland, Travisano, Tahiliani, Rached and Mirza 106 , Reference Carlin, George and Reyes 111 , Reference Cordero, Milagro, Campion and Martinez 112 However, in rodents, contrary to the improvements seen in metabolic profiles, maternal methyl donor supplementation just before and throughout pregnancy and lactation (5 mg/kg folic acid; 0.5 g/kg vitamin B12; 5 g/kg betaine and 5.76 g/kg choline) significantly increased colitis susceptibility in offspring. This was associated with colonic mucosal DNA methylation.Reference Schaible, Harris, Dowd, Smith and Kellermayer 113 In young adulthood, the offspring born to supplemented mothers show altered profiles of several microbial genera found in the colon.Reference Schaible, Harris, Dowd, Smith and Kellermayer 113 Roseburia,Reference Willing, Dicksved and Halfvarson 114 a genera which has previously been associated with Crohn’s disease in humans, was also depleted. Furthermore, a maternal gluten-free diet during pregnancy and lactation in a NOD mouse model resulted in a shift in microbial populations of mothers and offspring. Offspring gut microbiota in Verrcomicrobia and Proteobacteria including A. muciniphila and Proteus.Reference Hansen, Krych and Buschard 115 Interestingly, these effects were seen only in offspring at weaning and not later in life, highlighting the role of nutrition in mediating gut colonization during the neonatal period. Long term, offspring showed a reduction in the incidence of diabetes, reduced pancreatic inflammation and increased gut tight junction and mucin expression levels, suggestive of enhanced gut barrier integrity.Reference Hansen, Krych and Buschard 115 Thus, in utero gluten-induced changes in early neonatal gut colonization appeared to confer improved gut function as well as long-term protective effect from diabetes in this model.
Despite our general lack of understanding of the specific signaling pathways involved, studies investigating perinatal probiotic supplementation have reinforced the notion that a balanced or ‘healthy’ microbiota early in life is critical in the maintenance of health (we use the term ‘healthy’ microbiota here to refer to those populations most similar to those that have been observed in lean, healthy individuals and animal models). Indeed, studies in human have shown that probiotic intake during pregnancy and lactation have positive outcomes in children including reduced incidence of atopy including eczema.Reference Rautava, Kainonen, Salminen and Isolauri 116 , Reference Rautava, Kalliomäki and Isolauri 117 It appears, however, that strict administration during the prenatal period may not be enough and continued administration during lactation is required.Reference Boyle, Ismail and Kivivuori 118 Apart from its role in nutrition, breast milk provides immune-modulatory componentsReference Field 119 and promotes neonatal gut colonization,Reference Gueimonde, Laitinen, Salminen and Isolauri 120 and thus could be a direct route for the delivery of prebiotics and probiotics to the infant in cases of ongoing maternal supplementation.Reference Abrahamsson, Sinkiewicz, Jakobsson, Fredrikson and Björkstén 121 These data are supported by animals studies, where probiotic supplementation during pregnancy and lactation in pigs showed the presence of probiotic strains in the suckling piglets without direct administration,Reference Starke, Pieper, Neumann, Zentek and Vahjen 122 , Reference Taras, Vahjen, Macha and Simon 123 although in animals probiotic exposure may also be through ingestion of the maternal feces.Reference Jadamus, Vahjen, SchÄFer and Simon 124 In rats, probiotic supplementation to mothers late in pregnancy, throughout lactation and to their offspring demonstrated long-term effects on maternal and offspring microbial composition, weight gain and adiposity.Reference Karlsson, Molin and Fak 125 In contrast, a high protein or high-prebiotic-fibre diet throughout pregnancy and lactation in rats resulted in lower circulating LPS and higher Bifidobacteria in both mothers and offspring.Reference Hallam, Barile, Meyrand, German and Reimer 126 This result is not only attributed to the reported negative correlations between circulating LPS and Bifidobacteria levels but is explained by the introduction of critical growth substrates including human milk oligosaccharide (HMO), neutral oligosaccharide and reconstructed HMO metabolic pathways encoded by beneficial bacterial species.Reference Sela, Li and Lerno 127 It is known that specific endogenous bacteria such as Bifidobacterium species commonly found in the distal gastrointestinal tract of infantsReference Sela, Li and Lerno 127 such as Bifidobacterium longum subsp. Infantis utilize HMOsReference Oozeer, van Limpt and Ludwig 128 , thus it is possible that dietary combinations utilizling pre/probiotics that target these species, increase maternal transfer to breast milk and are beneficial in establishing early gut colonization. However, this has yet to be proven.
Interestingly, it has recently been shown that probiotic-induced changes in pup survival and sex ratio in mice was associated with endocrine-mediated changes in the mother. In mice lacking oxytocin, probiotic-induced microbial changes in the mother were absent.Reference Ibrahim, Kearney and Levkovich 129 Although the mechanisms are unknown, this observation is suggestive of an interaction between endocrine signaling and gut colonization during pregnancy.
Concluding remarks
The startling rise in the prevalence of obesity, particularly in women of reproductive age and children has made it a pressing global health concern. In order to combat this epidemic, current research targeting the early life environment has exploded in hopes of improving offspring metabolic outcome. Research on the maternal gut microbiota and fetal gut colonization is only in its infancy. The mechanisms driving maternal gut microbial shifts during healthy pregnancy and pregnancies complicated by obesity are currently unknown. Further investigation of the maternal gut microbiota throughout pregnancy is required to understand its influence on fetal gut establishment and development and whether disruption in the maternal microbiota due to obesity can confer increased risk of disease later in life in the offspring.
Given the recent association between bacterial colonization and metabolic/obesogenic outcomes, the current paradigm describing these diseases as ‘non-communicable’ may soon face formidable challenges. Although it has yet to be proven empirically whether the relationship between commensal bacterial transmission and obesity risk is direct or indirect, emerging evidence regarding modulation of maternal systems including the innate/adaptive immune response, nutrient uptake and utilization and/or inflammatory-mediated changes in gut function have demonstrated plausible mechanisms by which obesity may be considered ‘communicable’ between a mother and her offspring. The years of research to follow will doubtlessly shed much light on this new black box of microbial-induced phenotype modulation, and reveal new horizons for understanding the nature of metabolic disease transmission.
Acknowledgements
DMS is supported by the Canada Research Chairs Program.
Conflicts of Interest
None.