Periodic Reporting for period 1 - MEMORIS (Maternal Enteric Microbiota for Offspring's Repertoire development and Illness Susceptibility)
Reporting period: 2018-03-01 to 2020-02-29
Our intestine is colonized by a myriad of symbiotic microbes that constitute the intestinal microbiota. They feed on the nutrients that we ingest but also help us extract energy from food and stimulate our immune system. While babies are delivered from a sterile womb, they are exposed to microbes after birth. The newborn microbiota is still simple and consists of species adapted to metabolize sugars from breast milk. With the introduction of solid food, the gut microbiota matures and becomes more diverse. Disturbance of the microbiota is associated to many immune-mediated diseases such as inflammatory bowel disease.
Research from the last decade had elucidated that there is a narrow time window after birth, during which the immune system is sensitive to cues from the environment, especially the microbiota. It was shown that germ-free (GF) mice that lack a microbiota entirely are more prone to allergy. Colonization of GF mice during early life can prevent later life allergy while colonization in adult age fails to do so.
Our lab had previously shown that this ‘window-of-opportunity’ does not only open at birth but already during pregnancy. Metabolites from the maternal microbiota are transported to the offspring through the placenta and breast milk and stimulate the baby’s innate immune system and gut function (Gomez de Agüero & Ganal-Vonarburg et al., Science, 2016). This study was enabled by a model of gestational colonization: Pregnant dams are intestinally exposed to the E. coli strain HA107 which cannot replicate in vivo and is thus cleared from otherwise GF mice. Consequently, the mother is exposed to microbes in her intestine during pregnancy but returns GF before delivery thereby also preventing any direct microbial exposure of the pups.
In the context of the MEMORIS project (Maternal Enteric Microbiota for Offspring Repertoire development and Illness Susceptibility), we wished to address how these gestational cues influence the offspring as they become colonized with their own microbiota as is the case under physiological conditions. More specifically, our aims were to understand the consequences of maternal microbiota cues for (1) the pups’ own microbiota development, (2) their adaptive immune system and (3) disease susceptibility.
Research from the last decade had elucidated that there is a narrow time window after birth, during which the immune system is sensitive to cues from the environment, especially the microbiota. It was shown that germ-free (GF) mice that lack a microbiota entirely are more prone to allergy. Colonization of GF mice during early life can prevent later life allergy while colonization in adult age fails to do so.
Our lab had previously shown that this ‘window-of-opportunity’ does not only open at birth but already during pregnancy. Metabolites from the maternal microbiota are transported to the offspring through the placenta and breast milk and stimulate the baby’s innate immune system and gut function (Gomez de Agüero & Ganal-Vonarburg et al., Science, 2016). This study was enabled by a model of gestational colonization: Pregnant dams are intestinally exposed to the E. coli strain HA107 which cannot replicate in vivo and is thus cleared from otherwise GF mice. Consequently, the mother is exposed to microbes in her intestine during pregnancy but returns GF before delivery thereby also preventing any direct microbial exposure of the pups.
In the context of the MEMORIS project (Maternal Enteric Microbiota for Offspring Repertoire development and Illness Susceptibility), we wished to address how these gestational cues influence the offspring as they become colonized with their own microbiota as is the case under physiological conditions. More specifically, our aims were to understand the consequences of maternal microbiota cues for (1) the pups’ own microbiota development, (2) their adaptive immune system and (3) disease susceptibility.
We first colonized pups born to gestationally colonized versus GF mothers with a diverse microbiota and followed its composition over time by 16s sequencing. We found that our model system recapitulated previous human studies in that the microbiota of newborns was of low diversity and dominated by pioneering species such as Lactobacilli and became more diverse once mice start eating solid food. The differences in composition seemed to be rather subtly influenced by the maternal microbiota.
We next asked whether instead of microbiota composition, its function, i.e. gene expression, might be influenced by gestational cues. To be able to detect modest differences in microbiota function, we were in need of the most sensitive analysis tools available. We therefore established a collaboration with the group of Randall Platt from the ETH Zürich who had developed a strain of E. coli that records its own RNA expression into a DNA-based permanent storage. This strain keeps a cellular memory of past events and we argued that it might be suitable to detect subtle differences in microbiota function of newborns imposed by the maternal microbiota. In a collaborative effort, we were able to establish the usage of this recording E. coli strain in vivo to report on gut function in the context of different diets and inflammation. A main advantage of this innovation is that information about the gut condition can be obtained non-invasively from the feces and we envision that this tool will be of significant interest not only for research but also in the clinic as a living diagnostic able to report on intestinal pathology. We now aim to use it to address more subtle influences of the maternal microbiota on the offspring’s own microbiota development.
Our second aim was to elucidate the influence of the maternal microbiota on the offspring’s adaptive immune system. We performed comprehensive phenotyping of B and T cells from the offspring of gestationally colonized versus GF mothers. We also assessed the immunoglobulin repertoire used by various B cell subsets. We found evidence that both B- and T- cell responses in the offspring are indeed influenced by the maternal microbiota. We are now following up on how the maternal impact on adaptive immune responses relates to the colonization of the pups with their own microbiota as well as how these B- and T- cell responses are dependent on each other.
The third aim of the MEMORIS project was to understand how cues from the maternal microbiota shape the offspring’s disease susceptibility. We compared the course of disease models between offspring of gestationally colonized versus control mothers. To model inflammatory bowel disease, we exposed mice to dextran sulfate sodium which induces colitis. To study type 1 diabetes we used the non-obese diabetes (NOD) mouse model wherein mice progressively develop hyperglycemia. Under the tested conditions, there was no major role for the maternal microbiota as we observed similar disease parameters in offspring of gestationally colonized versus GF mothers.
We hypothesized that our inability to detect an impact of the maternal microbiota on disease susceptibility might be because transient exposure to E. coli does not fully recapitulate the effects of a diverse microbiota. As a mitigation strategy, we aimed to develop further auxotrophic bacterial strains that would transiently colonize GF mice and could be used for gestational colonization experiments. The effects of gestational colonization are largely mediated via bacterial metabolites that activate the aryl hydrocarbon receptor (AhR). As E. coli produces such metabolites at rather moderate levels, we chose a bacterium that synthesizes such ligands at greater rates. We engineered a novel auxotrophic strain of a Gram+ bacterium that is a natural colonizer of the mammalian intestine known to abundantly synthesize AhR ligands. It reversibly colonizes GF mice and induces specific immune responses. We aim to use this strain for gestational colonization experiments hypothesizing that the effects on the offspring will be different from and stronger than those exerted by E. coli. This novel tool will be exploited by immunologists beyond the scope of the MEMORIS project for the characterization of microbiota-directed immune responses.
We next asked whether instead of microbiota composition, its function, i.e. gene expression, might be influenced by gestational cues. To be able to detect modest differences in microbiota function, we were in need of the most sensitive analysis tools available. We therefore established a collaboration with the group of Randall Platt from the ETH Zürich who had developed a strain of E. coli that records its own RNA expression into a DNA-based permanent storage. This strain keeps a cellular memory of past events and we argued that it might be suitable to detect subtle differences in microbiota function of newborns imposed by the maternal microbiota. In a collaborative effort, we were able to establish the usage of this recording E. coli strain in vivo to report on gut function in the context of different diets and inflammation. A main advantage of this innovation is that information about the gut condition can be obtained non-invasively from the feces and we envision that this tool will be of significant interest not only for research but also in the clinic as a living diagnostic able to report on intestinal pathology. We now aim to use it to address more subtle influences of the maternal microbiota on the offspring’s own microbiota development.
Our second aim was to elucidate the influence of the maternal microbiota on the offspring’s adaptive immune system. We performed comprehensive phenotyping of B and T cells from the offspring of gestationally colonized versus GF mothers. We also assessed the immunoglobulin repertoire used by various B cell subsets. We found evidence that both B- and T- cell responses in the offspring are indeed influenced by the maternal microbiota. We are now following up on how the maternal impact on adaptive immune responses relates to the colonization of the pups with their own microbiota as well as how these B- and T- cell responses are dependent on each other.
The third aim of the MEMORIS project was to understand how cues from the maternal microbiota shape the offspring’s disease susceptibility. We compared the course of disease models between offspring of gestationally colonized versus control mothers. To model inflammatory bowel disease, we exposed mice to dextran sulfate sodium which induces colitis. To study type 1 diabetes we used the non-obese diabetes (NOD) mouse model wherein mice progressively develop hyperglycemia. Under the tested conditions, there was no major role for the maternal microbiota as we observed similar disease parameters in offspring of gestationally colonized versus GF mothers.
We hypothesized that our inability to detect an impact of the maternal microbiota on disease susceptibility might be because transient exposure to E. coli does not fully recapitulate the effects of a diverse microbiota. As a mitigation strategy, we aimed to develop further auxotrophic bacterial strains that would transiently colonize GF mice and could be used for gestational colonization experiments. The effects of gestational colonization are largely mediated via bacterial metabolites that activate the aryl hydrocarbon receptor (AhR). As E. coli produces such metabolites at rather moderate levels, we chose a bacterium that synthesizes such ligands at greater rates. We engineered a novel auxotrophic strain of a Gram+ bacterium that is a natural colonizer of the mammalian intestine known to abundantly synthesize AhR ligands. It reversibly colonizes GF mice and induces specific immune responses. We aim to use this strain for gestational colonization experiments hypothesizing that the effects on the offspring will be different from and stronger than those exerted by E. coli. This novel tool will be exploited by immunologists beyond the scope of the MEMORIS project for the characterization of microbiota-directed immune responses.
Overall, the MEMORIS project yielded fresh insights into the role of the maternal microbiota especially for the adaptive immune system of the offspring which are now subject to further investigation. It also yielded a series of novel research tools exploitable by scientists from various fields such as the new auxotrophic bacterial strain that reversibly colonizes GF mice. Until the end of the project, we expect that mimicking the maternal microbiota with our novel auxotrophic strain will uncover the implications of gestational cues from the mother's intestinal microbiota for the offspring's disease susceptibility. The in vivo application of Record-seq E. coli to assess gut function in a non-invasive manner will not only facilitate host-microbe mutualism research but also have direct implications for human health when applied as a living diagnostic in the future.