National Academies Press: OpenBook
« Previous: 3 Non-Rodent Models for Microbiome Research
Suggested Citation:"4 Modeling Human Microbiota in Animal Systems." National Academies of Sciences, Engineering, and Medicine. 2018. Animal Models for Microbiome Research: Advancing Basic and Translational Science: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/24858.
×

4

Modeling Human Microbiota in Animal Systems

Animal models provide opportunities to define the contributions of members of the microbiota to community function and the mechanisms through which they affect various aspects of host biology. Six speakers addressed current approaches they are using in this regard and addressed how these approaches may promote further basic and translational research in this field. The six speakers were Federico Rey, assistant professor of bacteriology at the University of Wisconsin–Madison; Patrice Cani, a researcher from the Belgian Fund for Scientific Research and a group leader at the Université de Louvain Drug Research Institute; Wendy Garrett, professor of immunology and infectious diseases at the Harvard T.H. Chan School of Public Health; Richard Blumberg, professor of medicine at Harvard Medical School and co-director of the Harvard Digestive Diseases Center; Nancy Moran, the Leslie Surginer Endowed Professor in the Department of Integrative Biology at The University of Texas; and Tracy Bale, professor of neuroscience in the School of Veterinary Medicine and Department of Psychiatry at the Perelman School of Medicine, University of Pennsylvania.

CONNECTING MICROBES TO METABOLISM USING GNOTOBIOTIC MODELS

Microbes in the gut produce thousands of metabolites that affect mammalian physiology through interactions with host receptors and microbial community dynamics (Krishnan et al., 2015). As an example, Rey noted how the human digestive system cannot absorb the beneficial polyphenols and flavonoids in red wine until gut microbes first metabolize these compounds. At the same time, the choline and carnitine in a steak are not only essential nutrients for humans but also substrates for microbes that ferment them and produce chemicals, such as trimethylamine, that are associated with cardiovascular disease (Romano et al., 2015).

The large interpersonal differences in microbiota composition likely mean that nutrient metabolism and absorption from food will vary from one person to the next, which Rey believes may have differential effects on individual health. Understanding this phenomenon, he said, requires knowing what each of the

Suggested Citation:"4 Modeling Human Microbiota in Animal Systems." National Academies of Sciences, Engineering, and Medicine. 2018. Animal Models for Microbiome Research: Advancing Basic and Translational Science: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/24858.
×

myriad species in the gut are doing and the effects they are having on each other and on the individual. His approach to untangling this complexity is to colonize germ-free mice with species that are representative of the native community’s phylogeny and function.

Bacterial metabolism in the human gut converts choline into trimethylamine, which the liver then converts into trimethylamine-N-oxide (TMAO). In 2011, researchers showed that high plasma levels of TMAO were a good predictor of cardiovascular disease and that gut microbial metabolism was involved in producing this compound (Wang et al., 2011). Subsequently, a number of groups have found associations between plasma TMAO levels and other diseases, including adipose tissue inflammation, heart failure (Tang et al., 2014), renal failure (Tang et al., 2015), diabetes (Dambrova et al., 2016), colorectal cancer (Bae et al., 2014), and non-alcoholic fatty liver disease (Chen et al., 2016). Experiments in mice have shown that TMAO is causally associated with the development of atherosclerosis through its inhibition of reverse cholesterol transport and platelet activation (Warrier et al., 2015).

Starting from those observations, Rey and his colleagues screened some 100 sequenced human gut isolates representing 91 different species in 37 genera for their capacity to convert choline to trimethylamine. Of these 100 isolates, only 8 consumed any choline at all, and each produced trimethylamine (Romano et al., 2015). Fortunately, said Rey, the human microbiome project identified one strain of E. coli that uses the same pathway most of these eight organisms use to convert choline to trimethylamine and a mutant version of this strain that lacks the key enzyme involved in this conversion. This enabled his group to create a community of five gut organisms plus either the wild-type or mutant E. coli. Measuring TMAO levels in mice colonized with one of the two communities showed that TMAO was present only in the blood of the mice colonized with wild-type E. coli, whose serum levels of choline were lower as was the amount of DNA methylation observed in multiple tissues from these animals.

In other experiments, Rey and his colleagues examined the effect these two communities had on metabolic disease in mice fed a high-fat diet, which is known to increase the body’s need for methyl donors. The mice with the wild-type, choline-consuming E. coli accumulated more fat and higher levels of triglycerides in their livers compared to mice colonized with the community that cannot metabolize choline. Additional experiments with pregnant mice showed that methylation levels in the brains of the pups were higher in those whose mothers were colonized by the mutant strain of E. coli. In addition, when the pups grew to young adulthood, those who were born of mothers colonized by the mutant strain of E. coli displayed lower levels of anxiety and obsessive-compulsive behavior than did mice born of mothers colonized with wild-type E. coli. One conclusion from these studies, said Rey, is that the microbial choline utilization pathway may limit choline availability during pregnancy and affect fetal brain development. “This is something to think about because current dietary guidelines do not consider interpersonal difference in choline-consuming bacteria in the gut,” said Rey.

Suggested Citation:"4 Modeling Human Microbiota in Animal Systems." National Academies of Sciences, Engineering, and Medicine. 2018. Animal Models for Microbiome Research: Advancing Basic and Translational Science: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/24858.
×

Like Rey, Cani is interested in finding the mechanistic links between gut microbial communities and human disease. Building off the observation Jeff Gordon and his colleagues made that axenic mice fed a high-fat diet are more resistant to bodyweight gain and fat mass development compared to mice that were not germ-free (Backhed et al., 2007), Cani and his collaborators are using gnotobiotic mice to study how gut microbial communities affect the development of obesity and type 2 diabetes. Several groups have shown that transferring gut microbes from genetically obese or diabetic mice into germ-free recipient mice transfers at least part of the phenotype to the recipient mice (Everard et al., 2014; Geurts et al., 2015; Vijay-Kumar et al., 2010). Similarly, gut microbiota transferred from mice with gastric bypass reduced the weight and fat mass in recipient animals fed a high-fat diet (Liou et al., 2013). Gordon and his colleagues have shown that germ-free mice gain weight when they receive gut microbes transplanted from an obese identical twin human, but not when they received gut microbes from the non-obese identical twin (Ridaura et al., 2013). Some of these studies, said Cani, suggest that cross talk between microbes and hosts may involve short-chain fatty acids binding to specific G-protein-coupled receptors.

Cani and his colleagues have been investigating this link through the lens of low-grade inflammation caused by the administration of bacterial lipopolysaccharides (LPSs) (Cani and Delzenne, 2009). Plasma LPS levels are increased across different strains of mice fed obesity-inducing diets or that were genetically obese (Cani et al., 2007). In these animals, blocking the LPS receptor prevented serum LPS levels from increasing and subsequent inflammation from occurring. Additional experiments showed that intestinal LPS triggered metabolic endotoxemia, insulin resistance, and macrophage infiltration (Cani et al., 2008). The deletion of the enzyme N-acyl phosphatidylethanolamine phospholipase D (NAPE-PLD) from adipocytes causes mice to develop spontaneous obesity (Geurts et al., 2015); thus germ-free mice that received microbes from the genetically modified obese mice gained both weight and fat mass, while their adipose tissue displayed metabolic changes similar to those observed in the donor mice. The mechanism linking adipose tissue to the microbiome remains unknown, said Cani.

His collaborators have also been studying the inverse correlation between the presence of the bacterium Akkermansia muciniphila in the gut microbiome and weight gain. Numerous experiments have shown that this bacterium, which accounts for 1 to 5 percent of the human gut microbiome, can reduce the metabolic endotoxemia and insulin resistance normally associated with a high-fat diet (Everard et al., 2013), perhaps by increasing the thickness of the intestinal mucous layer. He noted that, in humans with gastric bypass, as well as in type 2 diabetics on the drug metformin, the proportion of A. muciniphila in the gut microbiome increases significantly. Moreover, individuals on a low-calorie diet who lost weight and whose cholesterol levels, inflammatory tone, and insulin sensitivity improved also had a higher proportion of A. muciniphila, as well as 16 other metagenomic species, in their gut microbiota (Dao et al., 2016).

Suggested Citation:"4 Modeling Human Microbiota in Animal Systems." National Academies of Sciences, Engineering, and Medicine. 2018. Animal Models for Microbiome Research: Advancing Basic and Translational Science: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/24858.
×

These results, said Cani, suggest that A. muciniphila might be a therapeutic candidate for treating obesity, type 2 diabetes, and other associated disorders. While the bacterium itself is too sensitive to oxygen to be contained within a pill, a protein from the bacterium’s outer membrane produces the same effect in obese and diabetic mice as the live bacterium (Plovier et al., 2017). Preliminary experiments have shown that humans can safely take pasteurized A. muciniphila, and Cani and his colleagues are planning further clinical studies in humans.

REVISITING KOCH’S POSTULATES1FROM A MICROBIAL COMMUNITY PERSPECTIVE

When Garrett first began exploring how microbes contribute to colon cancer, she and her colleagues found that the bacterium Fusobacterium nucleatum, an anaerobic gram-negative macrobacterium, was enriched in human colorectal tumors and stools (Kostic et al., 2012). Feeding a human isolate of F. nucleatum to a strain of mice predisposed to develop intestinal adenomas increased the rate at which these mice developed tumors (Kostic et al., 2013). Additional experiments showed that specific strains of F. nucleatum greatly expanded the number of multiple types of myeloid immune cells at the earliest stages of colorectal tumor development. To determine if these results had any bearing on what was happening in human colorectal cancer, Garrett and her collaborators examined RNAseq data generated as part of The Cancer Genome Atlas program and found the same signatures in humans that they saw in mice. The researchers also found that the presence of certain strains of F. nucleatum correlated with different T cell subsets in human colorectal cancer patients and that the presence of these strains correlated with a poorer prognosis (Mima et al., 2015, 2016).

Screening of a library of F. nucleatum strain 23726 revealed two clones that did not impair natural killer (NK) cell–induced cytotoxicity, and detailed molecular studies conducted with collaborators at the Hebrew University of Jerusalem identified the protein Fap2, an adhesin, as the protein that impairs NK cytotoxicity. Experiments using human cell lines then identified TIGIT, an immune checkpoint inhibitor, as the binding partner on NK cells for Fap2 and showed that Fap2TIGIT binding protected tumors from immune cell attack (Gur et al., 2015).

Meanwhile, Garrett and her collaborators found that a different binding region of Fap2 interacted with the tumor-expressed sugar galactose-N-acetyl galactose, and this interaction mediated the enrichment of F. nucleatum in colo-

___________________

1 Koch’s postulates are a set of four criteria for judging whether a given microbe is the cause of a given disease: (1) the bacteria must be present in every case of the disease, (2) the bacteria must be isolated from the host with the disease and grown in pure culture, (3) the specific disease must be reproduced when a pure culture of the bacteria is inoculated into a healthy susceptible host, and (4) the bacteria must be recoverable from the experimentally infected host (see https://www.medicinenet.com/script/main/art.asp?articlekey=7105 [accessed March 2, 2018]).

Suggested Citation:"4 Modeling Human Microbiota in Animal Systems." National Academies of Sciences, Engineering, and Medicine. 2018. Animal Models for Microbiome Research: Advancing Basic and Translational Science: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/24858.
×

rectal tumors (Abed et al., 2016). They also found that other bacteria were capable of producing immune system changes to ensure their own survival while promoting cancer growth and spread. Experiments in mice identified a chemokine called CCL2 that enriches myeloid-derived suppressor cells in colon tumors. Taken together, said Garrett, these results suggest that, in the human mouth, F. nucleatum is an innocuous symbiont, but occasionally it will colonize the gut and create an immune microenvironment that is permissive for microbes and tumors.

With regard to animal welfare, Garrett said, animal protocols and monitoring are often constructed to catch signs of distress secondary to procedures; therefore simple complementary workflows that minimize labor but help maintain the animals’ health are crucial. She also believes simple methods for recognizing signs of suffering and for taking action are needed, such as using paper-based bedding for housing an animal developing prolapse from a tumor. Also, her group is developing noninvasive monitoring methods, such as luminescence-based in vivo imaging, to replace invasive monitoring procedures, such as colonoscopy, and refine experimental procedures.

In addition to obesity and cancer, avenues of research have linked a lack of early exposure of the human microbiome to microbes in the environment to a set of conditions known as atopic diseases, a set of allergic hypersensitivities that include food allergy, atopic eczema, allergic rhinitis, and asthma (Bach, 2002; Carpenter et al., 1989; Ege et al., 2011). Research has also suggested that inflammatory bowel disease (IBD), which is not thought of as an atopic disease but shares immunological characteristics and pathways, may also result from a lack of early life exposures to environmental microbes (Benchimol et al., 2009; Shaw et al., 2010).

Blumberg’s approach to testing the hypothesis that early exposure to specific microbes affects immune function and susceptibility to atopic diseases, IBD, and other ailments has been to study the effect that early exposure has on natural killer T (NKT) cells. Doing so, he explained, requires exposing germ-free animals to microbes early in life and looking for the development of a phenotype that does not appear if germ-free animals are exposed to the same microbes later in life. So-called invariant NKT cells recognize host and microbial lipid antigens presented by the molecule CD1d and play a critical role in the early immune response as orchestrators of downstream events. Invariant NKT cells, said Blumberg, are important regulators of bacterial commensalism, whether it involves a pathogen, such as Pseudomonas aeruginosa, or a commensal organism, such as E. coli or Lactobacillus gasseri, a normal inhabitant of the lower reproductive tract in healthy women (Nieuwenhuis et al., 2002, 2009).

Suggested Citation:"4 Modeling Human Microbiota in Animal Systems." National Academies of Sciences, Engineering, and Medicine. 2018. Animal Models for Microbiome Research: Advancing Basic and Translational Science: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/24858.
×

Research in multiple laboratories, on both mice and humans, supports the involvement of CD1d and invariant NKT cells in the pathogenesis of IBD (Boirivant et al., 1998; Fuss et al., 2004; Heller et al., 2002; Iijima et al., 2004; Jostins et al., 2012; Liao et al., 2012). Blumberg and his colleagues have shown that exposure to microbial colonization in the early stages of life protects germ-free mice from high invariant NKT cell infiltration in the colon and lung (Olszak et al., 2012). Furthermore, they showed that germ-free mice are highly susceptible to oxazolone-induced colitis associated with triggers that activate invariant NKT cells, and that this susceptibility is eliminated if the mice are exposed to what he called microbial programming during the first couple of weeks of their lives.

Invariant NKT cells are also involved in the development of asthma (Akbari et al., 2003; Albacker et al., 2013; Iwamura and Nakayama, 2010), and Blumberg and his colleagues have shown that early life exposure to microbes protects germ-free mice from allergic asthma. “So two different diseases, one an atopic disease, the other a complex disease, could only be normalized in terms of their sensitization to later triggers of those diseases, if they received the microbial education during the early part of life,” said Blumberg. The mechanistic connection, he explained, may be the chemokine ligand CXCL16. This ligand is important for NKT cell recruitment, and germ-free mice exposed to microbes early in life have low levels of CXCL16, whereas germ-free mice not exposed to microbes have high levels of this ligand in serum, the colon, and the lung (Lexmond et al., 2014). Similarly, when the offspring of antibiotic-treated pregnant mice are given antibiotics during the first two weeks of life, they become quite susceptible to oxazolone-induced colitis in a CD1d-dependent and invariant NKT-dependent manner, said Blumberg.

Other experiments may have identified at least one symbiotic organism—Bacteroides fragilis—and one specific molecule—a glycosphingolipid called GSL-Ff717—that can normalize invariant NKT levels in the colon of germ-free mice (An et al., 2014) through a pathway that is CKCL16 independent. This molecule, said Blumberg, represents a new class of microbial immunomodulatory molecules. Though early bacterial colonization normalizes invariant NKT cell levels in the lung, other organisms must be involved, he noted, because B. fragilis was not able to normalize NKT cell levels in that tissue.

Since publishing the results of these studies, other investigators have found a similar effect from early microbial colonization in germ-free mice on other immune system components, including IgE (Cahenzli et al., 2013) and regulatory T cells in the skin and lungs (Gollwitzer et al., 2014; Scharschmidt et al., 2015). Taken together, said Blumberg, these studies support the hypothesis that atopic disorders and numerous complex diseases, including IBD, originate from inappropriate microbial exposure during early life through pathways that he believes are developmentally regulated. He explained that his current hypothesis suggests that NKT cell infiltration into the colon is a developmentally regulated process influenced by microbes. As a result, it is likely there exist age-dependent pathways linked to later life sensitivity to environmental events. Iden-

Suggested Citation:"4 Modeling Human Microbiota in Animal Systems." National Academies of Sciences, Engineering, and Medicine. 2018. Animal Models for Microbiome Research: Advancing Basic and Translational Science: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/24858.
×

tifying and understanding these pathways in humans are challenging, but perhaps Drosophila or C. elegans could serve as appropriate models for their study.

THE INTERFACE BETWEEN MICROBES AND NEUROSCIENCE: TWO CASE STUDIES

Bee Microbiome

When Moran began working with Apis mellifera, the Western or European honeybee, about 6 years ago, her goal was to use this species to study different aspects of how its distinct microbiome comes together, how the microbes within that microbiome interact, and how those interactions affect the behavior of bees. The honeybee gut microbiota comprises nine bacterial species that form dense, spatially organized communities in the hindguts of adult workers (Kwong and Moran, 2016) (see Figure 4-1). Any honeybee in the world will have these nine species, said Moran, and only one of these species, an Acetobacteriaceae, is found outside of the bee in nectar. In many ways, this is similar to what occurs in the mammalian gut. “The things that live in our gut for the most part only live in the gut. We do not find them in our food or in the environment,” said Moran. A major difference between the mammalian and honeybee gut microbiomes is that the mammalian microbiome comprises hundreds of species in contrast to the nine species that make up more than 95 percent of the honeybee microbiome.

Image
FIGURE 4-1 The honeybee gut microbiome. SOURCES: Moran slide 4 (Kwong and Moran, 2016).
Suggested Citation:"4 Modeling Human Microbiota in Animal Systems." National Academies of Sciences, Engineering, and Medicine. 2018. Animal Models for Microbiome Research: Advancing Basic and Translational Science: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/24858.
×

Honeybee and mammalian gut microbiota are similar, though, in that they are both transmitted socially early in life, mostly within family groups, and both host immune systems that can modulate community composition. High levels of strain diversity within symbiont species exist in honeybee and mammalian microbiota. For example, there can be up to 100 strains of Gilliamella apicola within an individual bee. Both honeybee and mammalian gut microbiomes include mixtures of gram-negative and gram-positive bacteria living in the oxygen- and nutrient-poor distal gut, and both can utilize complex plant polymers. Stress, age, antibiotics, and pathogens, such as Enterobacteriaceae, viruses, and trypanosomatids, can disrupt both honeybee and mammalian microbiomes.

When a new adult bee emerges from a capped hexagonal cell 21 days from when the queen lays an egg, its gut has few bacteria, but over the next 5 days a stable community of approximately one billion organisms establishes itself. Moran explained that, when the bee larva turns into a pupa, it sheds its entire gut lining so that when it emerges as an adult it is essentially germ-free. This enables Moran to take the pupae out of the hive when they are still in the capped pupal cells and allow them to emerge in the lab as germ-free individuals that she and her colleagues can manipulate as an experimental system. She also noted that her group developed methods to grow the nine species of bacteria in the honeybee gut microbiome as pure cultures in the laboratory and to introduce fluorescence or luminescence genes that enable them to monitor microbiome composition.

Inoculating the bees is simply a matter of feeding a sucrose solution or pollen laced with one or more of the bacterial species. Imaging of naturally and experimentally colonized honeybee gut tissue showed that experimentally introduced bacteria colonize the same niches in the gut as do the naturally colonized species. In addition, experimentally introduced microbiota can be passaged and replicated by co-housed bees. These experiments also revealed that colonization appears to occur in a particular sequence. For example, Snodgrassella needs to colonize the gut before Gilliamella can establish itself.

Genomic and metabolomic studies (Engel et al., 2012, 2014; Kwong et al., 2014; Powell et al., 2016) have shown that the members of the honeybee microbiome cross-feed and are interdependent on each other. They have also revealed that many of the interactions of the different species are antagonistic, involving toxins and bacteria type VI secretion systems. These studies have shown that some G. apicola strains can degrade pollen cell wall components, such as pectin, and use the resulting sugars as an energy source. This bacterium and others in the bee microbiome can also metabolize sugars that would otherwise be toxic to the bees (Zheng et al., 2016). Moran noted that, just as in the human gut, different bacterial strains break down different plant polysaccharides, a phenomenon that may be important nutritionally for the host given that the bee itself is incapable of metabolizing many of these polysaccharides.

Turning to the subject of bee behavior, Moran noted that there is an extensive literature on bee behaviors, and researchers have developed a number of assays for learning, motility, aggression, sociality, gustatory response, buzzing re-

Suggested Citation:"4 Modeling Human Microbiota in Animal Systems." National Academies of Sciences, Engineering, and Medicine. 2018. Animal Models for Microbiome Research: Advancing Basic and Translational Science: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/24858.
×

sponse, and other behaviors (Fahrbach and Robinson, 1995; Zayed and Robinson, 2012). In one learning experiment, Moran and her colleagues, using the plasmid system they created to introduce genes into honeybee microbiome species, studied the effect on bee behavior of adding L-dOPA, a precursor to dopamine, into bee gut bacteria. Other investigators had shown previously that bees fed dopamine learn to associate a color with punishment, as measured by a sting extension response, faster than control bees, and that feeding them a dopamine antagonist diminishes that learned response (Agarwal et al., 2011). This experiment found that bees inoculated with the L-dOPA-producing bacteria learned faster and demonstrated better memory than bees inoculated with bacteria engineered to produce green fluorescent protein.

Though this result was not surprising, Moran said it serves as proof of the principle that gut bacteria can produce chemicals that alter their host’s behavior and that this process can be studied in germ-free honeybees. Ongoing studies in her laboratory include examining how insulin produced by the bee microbiome affects hunger, as measured by proboscis extension, and the effect of isopentyl acetate, an alarm pheromone, on aggression, stinging, and cohort alert.

With regard to animal welfare issues, Moran said that Institutional Animal Care and Use Committee policies do not cover bees. Nonetheless, she and her collaborators strive to avoid procedures that might result in prolonged suffering of the bees. At the end of an experiment, the bees are killed by freezing, a common way for them to die in nature. In addition, because bees are an agriculturally important species and the subject of many types of studies, there are established protocols for using bees in research.

In closing, Moran discussed some of the challenges her group has encountered working with bees as a model organism. Given that honeybees have complex social lives in large colonies, studying them in the laboratory environment outside of the context of a colony and without a queen being present is highly artificial, she said, and nobody has been able to establish a germ-free colony. Bee behavior also varies genetically and with age, which requires controlling for numerous sources of variability. In addition, despite homologies in endocrine systems, immune systems, and nervous systems, many aspects of human biology do not apply to bees. Still, she said, one motivation for studying bees is the bees themselves. “They are important and a lot of them are dying so we are hopeful that some of what we find out will actually be helpful in improving the health of bees as pollinators,” said Moran. She added that the bee microbiome does protect against pathogens to some extent, though the mechanism is not known.

Maternal Microbiome

Bale’s interest in the microbiome stems from her work on how events that occur during pregnancy affect brain development. In particular, Bale was curious as to how maternal stress might affect the mother’s vaginal microbiome, and

Suggested Citation:"4 Modeling Human Microbiota in Animal Systems." National Academies of Sciences, Engineering, and Medicine. 2018. Animal Models for Microbiome Research: Advancing Basic and Translational Science: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/24858.
×

hence the initial inoculum that seeds the infant’s microbiome, and whether changes in the maternal vaginal microbiome had different effects on neurodevelopment in female and male newborns. She noted that neurodevelopmental disorders occur more frequently in males than females, and that research has identified many factors that influence male vulnerability during the neonatal window compared to females (Bale, 2015).

Over the past 14 years, Bale and her collaborators have cataloged a host of effects from maternal stress during early pregnancy that pass through at least two generations of offspring. These effects include changes in the behavioral stress response, the hypothalamic–pituitary–adrenal axis response to stress, activity of stress regulatory genes, cognitive deficits, and reduced post-pubertal weight gain (Howerton and Bale, 2014; Howerton et al., 2013; Morgan and Bale, 2011; Mueller and Bale, 2008). The effects occur in male but not female offspring developing in the same uterus. “Could there be an aspect by which the stress that is influencing Mom changes the contents or the composition of the vaginal microbiome such that when the babies are born they are getting a different inoculant than they would otherwise, which is influencing brain development?” asked Bale.

To answer that question, Bale and her colleagues looked at whether maternal stress during early pregnancy in mice changed the vaginal microbiome in a manner that persisted until the time of birth. In fact, early stress changes both the bacterial and the viral composition of the vaginal microbiome that persists through at least two days after birth (Jasarevic et al., 2015b) (see Figure 4-2). Proteomic analysis revealed significant changes in the vaginal tissues after exposure to stress, particularly among proteins involved in the immune response. Next, they showed that the mother passes these changes in the maternal microbiome to her offspring’s microbiome and that these changes result in metabolic reprogramming in the offspring’s gut and brain (Jasarevic et al., 2015a). In particular, said Bale, there is a dramatic drop in Lactobacillus levels in the maternal vagina, though not in maternal feces, and a corresponding drop in the gut microbiome of both male and female offspring.

Then came some surprising results: by day 4 after birth, these differences went away, but then at day 28, when puberty begins in mice, dramatic changes appeared in the male gut, while only slight changes occurred in the female gut. These changes in the male gut microbiota were associated with many-fold increases in mitochondrial, carbohydrate, and energy metabolism, which together could be affecting the availability of nutrients in the brain. Going back to the postnatal day 2 offspring, Bale and her colleagues found that amino acid transport into the paraventricular nucleus at day 2 was markedly different in males than in females. The paraventricular nucleus is the part of the brain that regulates stress reactivity, and it plays a role in the brain’s homeostatic response to feedback from the periphery, Bale explained.

Suggested Citation:"4 Modeling Human Microbiota in Animal Systems." National Academies of Sciences, Engineering, and Medicine. 2018. Animal Models for Microbiome Research: Advancing Basic and Translational Science: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/24858.
×
Image
FIGURE 4-2 Early pregnancy stress changes the mouse vaginal microbiome when comparing embryonic day 7.5 with postnatal day 2. SOURCES: Bale slide 7 (Jasarevic et al., 2015b).

These associations are interesting, she noted, but the real goal was to demonstrate causality. To get at causality, she and her colleagues performed a difficult set of experiments in which they delivered the pups via cesarean section and inoculated them with vaginal lavages from stressed and control mothers, a procedure that appears to reproduce the bacterial load and diversity of the vaginal microbiome in the gut of the offspring. It also reproduces the elevated stress response in post-pubertal males delivered vaginally from mothers exposed to stress. However, males from stressed mothers delivered by cesarean section who were then inoculated with a vaginal lavage from a non-stressed mother still show an elevated stress response. Males from control mothers who received a vaginal lavage from stressed mothers showed a slightly elevated stress response.

One factor Bale had not considered was that prenatal maternal stress could be affecting the development of the gut during fetal development in a sex-specific manner and that any such differences could be interacting with the maternal inoculant. In fact, when she and her colleagues conducted a proteomic analysis of the male and female gut at embryonic day 18.5, they found huge sex-related differences. “So right before birth, there are huge differences in the development of the gut that likely are interacting with the mother’s microbiome as they pass through the same vagina,” said Bale. Further analysis showed that immune system genes in the male gut were activated in a manner similar to that seen in a response to Leishmania infection, though there was no infection by this parasite. “This was surprising to us because they have not been inoculated yet

Suggested Citation:"4 Modeling Human Microbiota in Animal Systems." National Academies of Sciences, Engineering, and Medicine. 2018. Animal Models for Microbiome Research: Advancing Basic and Translational Science: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/24858.
×

and you are already seeing differences in priming of the gut for these animals,” said Bale. These differences, present at birth, then interact with the initial vaginal inoculant.

These differences, though not by themselves, manifesting as disease or a behavioral disorder, are establishing how the male and female gut and brain respond to the environment later in life. For example, exposing these animals to a week of chronic stress lowers gut permeability in the males but not females born of mothers who experienced stress during early pregnancy. Bale noted, too, that changes in the microbiota of these animals are associated with changes in plasma levels of various metabolites and may affect their transport into the brain.

In closing, she pointed to the common practice among neuroscientists of shipping pregnant mice for use in research studies. Doing so, she said, exposes the pregnant mice to a variety of stressors and different environments. Given the results she and her colleagues have obtained, she wondered how the effects of early pregnancy stress might be affecting the results of neuroscience and behavioral studies involving those pregnant mothers and their offspring. She also commented on the possibility that these types of differences could be related to the increased vulnerability of human males during the prenatal and neonatal period that manifests in lower survival rates and increased rates of developmental disorders, such as autism, which occurs four to five times more frequently in males than females.

Suggested Citation:"4 Modeling Human Microbiota in Animal Systems." National Academies of Sciences, Engineering, and Medicine. 2018. Animal Models for Microbiome Research: Advancing Basic and Translational Science: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/24858.
×
Page 17
Suggested Citation:"4 Modeling Human Microbiota in Animal Systems." National Academies of Sciences, Engineering, and Medicine. 2018. Animal Models for Microbiome Research: Advancing Basic and Translational Science: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/24858.
×
Page 18
Suggested Citation:"4 Modeling Human Microbiota in Animal Systems." National Academies of Sciences, Engineering, and Medicine. 2018. Animal Models for Microbiome Research: Advancing Basic and Translational Science: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/24858.
×
Page 19
Suggested Citation:"4 Modeling Human Microbiota in Animal Systems." National Academies of Sciences, Engineering, and Medicine. 2018. Animal Models for Microbiome Research: Advancing Basic and Translational Science: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/24858.
×
Page 20
Suggested Citation:"4 Modeling Human Microbiota in Animal Systems." National Academies of Sciences, Engineering, and Medicine. 2018. Animal Models for Microbiome Research: Advancing Basic and Translational Science: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/24858.
×
Page 21
Suggested Citation:"4 Modeling Human Microbiota in Animal Systems." National Academies of Sciences, Engineering, and Medicine. 2018. Animal Models for Microbiome Research: Advancing Basic and Translational Science: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/24858.
×
Page 22
Suggested Citation:"4 Modeling Human Microbiota in Animal Systems." National Academies of Sciences, Engineering, and Medicine. 2018. Animal Models for Microbiome Research: Advancing Basic and Translational Science: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/24858.
×
Page 23
Suggested Citation:"4 Modeling Human Microbiota in Animal Systems." National Academies of Sciences, Engineering, and Medicine. 2018. Animal Models for Microbiome Research: Advancing Basic and Translational Science: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/24858.
×
Page 24
Suggested Citation:"4 Modeling Human Microbiota in Animal Systems." National Academies of Sciences, Engineering, and Medicine. 2018. Animal Models for Microbiome Research: Advancing Basic and Translational Science: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/24858.
×
Page 25
Suggested Citation:"4 Modeling Human Microbiota in Animal Systems." National Academies of Sciences, Engineering, and Medicine. 2018. Animal Models for Microbiome Research: Advancing Basic and Translational Science: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/24858.
×
Page 26
Suggested Citation:"4 Modeling Human Microbiota in Animal Systems." National Academies of Sciences, Engineering, and Medicine. 2018. Animal Models for Microbiome Research: Advancing Basic and Translational Science: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/24858.
×
Page 27
Suggested Citation:"4 Modeling Human Microbiota in Animal Systems." National Academies of Sciences, Engineering, and Medicine. 2018. Animal Models for Microbiome Research: Advancing Basic and Translational Science: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/24858.
×
Page 28
Next: 5 Experimental Reproducibility Using Gnotobiotic Animal Models »
Animal Models for Microbiome Research: Advancing Basic and Translational Science: Proceedings of a Workshop Get This Book
×
Buy Paperback | $45.00 Buy Ebook | $36.99
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

The surface of the human body and its mucous membranes are heavily colonized by microorganisms. Our understanding of the contributions that complex microbial communities make to health and disease is advancing rapidly. Most microbiome research to date has focused on the mouse as a model organism for delineating the mechanisms that shape the assembly and dynamic operations of microbial communities. However, the mouse is not a perfect surrogate for studying different aspects of the microbiome and how it responds to various environmental and host stimuli, and as a result, researchers have been conducting microbiome studies in other animals.

To examine the different animal models researchers employ in microbiome studies and to better understand the strengths and weaknesses of each of these model organisms as they relate to human and nonhuman health and disease, the Roundtable on Science and Welfare in Laboratory Animal Use of the National Academies of Sciences, Engineering, and Medicine convened a workshop in December 2016. The workshop participants explored how to improve the depth and breadth of analysis of microbial communities using various model organisms, the challenges of standardization and biological variability that are inherent in gnotobiotic animal-based research, the predictability and translatability of preclinical studies to humans, and strategies for expanding the infrastructure and tools for conducting studies in these types of models. This publication summarizes the presentations and discussions from the workshop.

  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    Switch between the Original Pages, where you can read the report as it appeared in print, and Text Pages for the web version, where you can highlight and search the text.

    « Back Next »
  6. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  7. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  8. ×

    View our suggested citation for this chapter.

    « Back Next »
  9. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!