Tip:
Highlight text to annotate it
X
Johanna Lampe:
Great, good afternoon, and thanks much, Cindy, and also many thanks to
Lita and the organizing committee for the opportunity to participate in this meeting.
It really has been fantastic.
By way of introduction, my background is in nutrition, so I'm going to take more of a
nutrition-centric view of our discussion about relationships between the gut microbial community
and diet. And also, before I start, I'd also like to acknowledge my colleague Meredith
Hullar, who's a microbial ecologist, who's been working with us for a number of years
now, who really has helped in our group beginning to think about some of the interactions and
the issues with regards to these relationships.
So from a nutritional standpoint, when we think about health and risk of disease, often
this is the very broad and general paradigm, is that aspects of diet are influencing risk
of disease. And in many cases, we focus primarily on a number of chronic diseases that we're
faced with from a public health standpoint. And we can think about this, not only from
the standpoint of fuel availability, which certainly, in the last number of years, has
come to the fore front with our concerns over adiposity and the implications of our increase
in obesity in the Western world, but also the aspect of other dietary constituents,
whether they be recognized as central nutrients, or a whole host of other components of diet
that can influence various aspects of disease risk.
In the context of our discussion today, then, we're really thinking about the gut microbial
community as a potential modifier of effective diet. Certainly we've heard a lot of discussion
about how various microbes can have direct effects on tissues and on the human system,
but also the potential for the gut microbial community to influence fuel availability and
modulation of signaling that affects energy balance, but also the impact of the gut microbes
on various other dietary constituents. So I'll really focus on this piece today. We've
also heard, in a number of presentations, the capacity for diet to directly affect the
gut microbial community. So there's always that back-and-forth connection.
So as far as discussion today, I'd like to address a couple of questions in a very broad
sense. And I think this is really more to bring us back to thinking about the complexity
of studying the gut microbial community and the various other aspects of exposures in
humans or other animals, and how this may impact on the community structure.
So, first of all, what are the gut microbes doing with our food? What is the effect of
the gut microbiome on host dietary exposures, particularly in the context of a lot of population-based
studies, where we often rely heavily on a food frequency questionnaire to try to identify
dietary exposure. But to what extent is that really capturing what's going on in vivo?
And how might this influence disease risk in the cases where we have some examples of
this? So I really -- certainly, this is very broad. It's also capturing a couple of examples,
by no means comprehensive. And many of you can bring up addition examples that we can
discuss, too. And then I'll finish with gaps, needs, and challenges.
So there's no question about it, the human diet is extremely complex, which means, when
we come to think about it, we're not only dealing with exposure to thousands of compounds
on a daily basis, but also how we handle food, whether it be processing in a factory, whether
it be how we are preparing a meal in the kitchen. But various aspects of this can affect structure
and particle size, bioavailability of compounds to the host. And with reduced bioavailability
of compounds to the host, what you're actually, likely seeing is bioavailability of substrates
to the microbiome.
Tend to think about this with regards to gut microbial metabolism designed to make the
most of the situation. And in the context of flow of food through the gastrointestinal
tract, really first line of handling is through human digestion. And often, when we go back
to the old nutrition and medical textbooks, we're really focusing only on what we can
handle in the way of enzymatic reactions that allow for availability of nutrients and energy
sources for the human.
But there are a whole host of components. One could consider the indigestible -- that
is the dietary fiber we've heard about; resistance starch that's not easily metabolized by humans;
as well as a number of other components, what we might call the leftovers, which are other
bits that we cannot handle, either because from structural standpoint they are encased
in a non-digestible coding, as, for example, in seeds, some nuts, depending how well you
chew your large chunks of meat, availability of protein; all these things have then become
part of the bacterial substrate mix. And really a lot of the reactions that we come to consider
in relation to gut microbial metabolism are designed to handle these fermentation. We've
heard about reduction of nitrates and sulfates, esterification, aromatic fission, et cetera,
that come into play.
When we look at this in relation to the contribution of genes to -- that are related to metabolic
pathways, a good majority of these are focused on carbohydrate and amino acid metabolism,
as well as xenobiotic degradation, which Pete talked a little bit about in relation -- well,
quite a bit, in relation to drugs. But when we think about xenobiotics from the dietary
standpoint, we're also talking about a whole host of other compounds, such as many of the
plant compounds that, from a human standpoint, we also view as xenobiotic, but are also major
substrates for gut microbes, as well as many pyrolysis products that come with how we prepare
food by cooking.
So we've heard a lot already about production of short-chain fatty acids in relation to
carbohydrate fermentation. So I'm really not going to spend much time on this, beyond just
to remind ourselves that acetate, propionate, and butyrate, besides serving as energy sources
both for the colonic epithelium in the case of butyrate, as well as capacity for lipogenesis
and gluconeogenesis for acetate and propionate, but they also -- the importance of these compounds
as signaling molecules: butyrate acting as a histone deacetylase inhibitor in the colonic
epithelium, and therefore having effects on epigenetic modulation of apoptosis and proliferation,
as well as signaling further systematically, as we heard from Gary, the potential to have
even an impact in the lung. So I think this is an area of -- that we'll continue to hear
a lot about.
Looking at microbial metabolism of proteins and amino acids, again, this is a range of
different types of compounds from the standpoint of substrates. And it -- really depending
on the type of amino acid, when it's generating a variety of different products, the bacteria
are certainly in the context of the aromatic amino acids, a range of phenols and indoles.
Sulfur amino acids can produce a whole host of different sulfur compounds. I'll talk a
little bit more about that in a minute. And the rest of the other amino acids can be deaminated
or decarboxylated to contribute to ammonia production amines, as well as, can also be
deaminated and fermented additionally to produce short-chain fatty acids. And in the case of
the branch-chain amino acids, to produce branch-chain -- short-chain -- branch-chain fatty acids,
excuse me.
So by way of example, in the context of the aromatic amino acid metabolism, L-Tryptophan
can be converted to indole and pyruvate. And Poster Number 18 has also another nice example
of an additional conversion of tryptophan to indole pyruvate, which is of concern in
relation to a metabolite that's been identified with autism, if you're interested in some
more metabolic pieces of that. But in the case of production of indole, it has received
some attention, both from the standpoint that it's present both in the human and rodent
gut lumen at low millimolar concentrations, and has been shown to modulate expression
of both pro- and anti-inflammatory genes, typically showing increases in the anti-inflammatory
genes and down-regulation of the pro-inflammatory genes, as well as strengthening some of the
epithelial cell barrier, and decreasing pathogen colonization. So I think this is just one
example, but it really points out the importance of so many of these small molecules, many
of which we probably don't even begin to understand their potential to impact on the host.
Another amino acid -- product of amino acid metabolism, in the case of sulfur amino acids,
is generation of hydrogen sulfide. This has been shown to be toxic to colonic sites both
in in vitro and in vivo models, and can contribute to inflammation, and therefore, is of concern
in relation to ulcerative colitis and colon cancer. And from the standpoint of handling
biogut bacteria, hydrogen sulfide can be generated both by fermentation of sulfur containing
amino acids, but also the action of the sulfate-reducing bacteria on inorganic sulfates; sp sulfate
and sulfites, many of which are used as preservatives in food, and also just present in foods in
general. So, again, another source there.
Just by way of an example, people have been saying today, it's often really hard to do
studies in humans, but one can do dietary interventions. And although we aren't necessarily
looking at who gets ulcerative colitis at the end, this is an example of a controlled
feeding study, in which case volunteers were sequestered in a metabolic ward, participated
in five different treatments of 10 days each of meat supplementation in a controlled diet.
And they were looking at fecal sulfate concentrations; they also looked at urinary sulfate excretion
in relation to these treatments. And as you can see, as you increase the amount of meat
in the diet, you see increases in fecal sulfate concentrations. So the -- human studies such
as these are small but they can often add important connecting bits of information in
relation to the impact of these compounds in human health and exposure to them.
Another story that's received quite a bit of attention in the last couple of years is
the production of trimethylamine oxide, which comes from the bacterial conversion of choline
to trimethylamine. Choline is derived from phosphatidylcholine, or lecithin, that is
a good part of the diet, particularly if you include meat products, and you've been eating
the hard-boiled eggs the last couple of mornings at breakfast. You've been receiving your phosphatidylcholine
challenge. And choline is released upon lipolysis in the gut; it undergoes metabolism by gut
microbes to trimethylamine, is absorbed, further metabolized by hepatic flavin mono-oxygenases,
and it's converted to trimethylamine oxide, which has been shown in animal models to be
atherosclerotic, and I'll show you a little data in a minute about potential risks in
humans.
A lot of work has been done in animal models already, but in the context of, again, a human
study, with dosing of 40 individuals with those phosphatidylcholine challenge, both
as an egg source as well as a deuterated phosphatidylcholine, they can show, you see increases in TMAO as
a result of the intervention. Provide a broad spec antibiotic: could show decreases as a
result, and then a reappearance of the TMAO upon withdrawal of the antibiotic, again,
showing the relationship between bacterial production of the trimethylamine and subsequent
TMAO.
As part of this study, they also evaluated prospectively the relationship between plasma
TMAO concentrations in 4,000 individuals who were then followed for three years for cardiovascular
events. And showed significant differences among quartiles of plasma TMAO concentrations
in relation to risk of myocardial infarction, stroke, and death, with certainly the highest
risk among those individuals in the highest quartile, again, showing this relationship
of production of a plasma component in relation to disease outcome.
So the other piece of metabolism that is a little near and dear to my heart is in the
area of phytochemicals. And there's an estimate that we are exposed to some 20,000 different
phytochemicals over the span of about 70 different plant foods. And many of these compounds are
also handled by gut microbes, certainly a lot of the phenolics, the terpenoids, which
include the carotenoids, various sulfur containing compounds, and the glucosinolates and indoles
also.
So what I'll do in the next few minutes is focus, as by way of example, on the glucosinolates
and the isoindoles. But certainly there are many other examples of the impact of gut microbes
on exposure to these compounds and to their metabolites.
First example is in relation to production of isothiocyanates from glucosinolates in
the area of cancer prevention. Cruciferous vegetable, that is the broccoli containing
-- or broccoli family, cabbage family, veggies, are major source of these in the human diet.
And as they exist in the plants, they are -- have a glucose moiety bound to them, aptly
named glucosinolates. And these are not bioactive in this form, but their release of the glucose
moiety to production of the isothiocyanate is production of the bioactive component that's
been shown to be important with regards to reduced carcinogenesis in animal models, as
well as associated with lower risk of several common cancers in human epidemiologic studies.
Now, as the compound exists, there is an enzyme in cruciferous vegetables that can cleave
this glucose moiety to release the isothiocyanate. And the thought is there that the primary
reason the plant does this is in the face of a threat -- or not a threat, but having
been chomped on or chewed that it actually releases the isothiocyanate, which has a bitter
and pungent feel in the mouth, and will cause the insect or animal to stop consuming it.
In most cases, in humans, we actually don't consume cruciferous vegetables raw, and that
means that the myrosinase is denatured as a result of cooking. But our gut bacteria,
many of which contain thioglucosidases, can also carry out this reaction and allow for
release of isothiocyanates.
And I think this is a nice example of the difference in exposure to isothiocyanates
across a group of individuals. This is work of Tom Kensler's from a number of years ago.
But they were conducting an intervention trial with glucosinolate supplement, glucoraphanin,
which is the glucosinolate precursor of sulforaphane. And we're looking at the impact on reducing
aflatoxin DNA adducts in a population in China. As you can see here, the urinary isothiocyanate
recovery ranged from 1 to 45 percent of the dose administered when they measured the metabolites
in urine, suggesting that the availability of these compounds varies considerably across
individuals. And this -- as you increase recovery in the urine, you see a decrease in DNA adduct
formation, suggesting that those exposed to more may be benefited more from the impact
of the isothiocyanate.
We were interested in whether -- to what extent we could understand the relationship between
these urinary measures of phenotype, as far as recovery of isothiocyanates in relation
to gut microbial activity. And Fay Li in our group recruited a group of individuals, challenged
them with a dose of cooked broccoli, so, i.e., glucosinolate-rich but no myrosinase to cleave
the isothiocyanate, and could recover between 1 to 28 percent of the isothiocyanate in urine.
You can see there is quite a range in the distribution. And then took the high and low
urinary ITC excreters and incubated fecal bacteria from their stool samples with glucoraphanin.
And looked at rate of decrease in relation -- of glucoraphanin in relation to either
a high excreter in yellow or a low excreter in blue. And suggesting that those individuals
who are excreting more isothiocyanates are typically more likely to break down the glucoraphanin,
which allows for release and absorption.
Another example of phytochemicals that are heavily metabolized by gut microbes are the
soy isoflavones. These have been of interest for a number of decades now in relation to
their weak estrogenic properties, capacity to bind to the estrogen receptors, both alpha
and beta, and illicit either weak estrogenic responses or to have anti-estrogenic effect.
In the context of this story, the soy isoflavone daidzein is handled by gut microbes to convert
two major metabolites, equol and O-desmethylangolensin, or ODMA. What we find in most populations,
that most individuals have got microbes that can carry out this reaction. And if you measure
urinary ODMA levels in a population, 80 to 90 percent of individuals will have substantial
ODMA excretion. And whereas the equol really varies, one, across populations, and certainly
is a lot lower in production. And it's not necessarily that either you're an ODMA producer
or an equol producer; typically, ODMA producers can be equol producers or vice versa.
This just shows that if you give individuals a soy challenge containing daidzein, you can
see a range of excretion; this is 24-hour urinary excretion. But this is a tenfold jump
between those individuals who really aren't excreting any. In this case, what we're measuring
here is likely what's coming from other animal sources of equol coming from milk and meat
sources of ruminants compared to what you seek produced by the humans.
The reason people have been interested in this is that in a number of the soy intervention
studies, there are have been identified differences in response between those of an equol-producing
capacity. And many of these relate to sex hormone effects, either in relation to menstrual
cycle or steroid hormone concentrations, bone mineral densities, as well as some differential
gene expression that is estrogen receptor driven in peripheral lymphocytes. And then
also similarly, some observation studies that show differences. But these are certainly
not always shown in all studies. There are -- certainly are in all studies too, but...
So what gut microbes are producing equol? There have been a number of attempts to identify
these, and there's a number of specific bacteria that have been shown to be able to convert
daidzein to equol listed here. But often, when you go looking by QPCR for some of these
specifically in human stool samples, you don't necessarily find them. There are also some
suggestions that it may -- some individuals may rely on a consortium of bacteria: some
that can carry out the first part of the reaction, and some that carry out the second. So I think
this sort of is an indication of the importance of being able to come to a more functional
approach to be able to characterize some of these activities, because we're going to find,
for many of these reactions, that there are a whole host of bacteria that could do this.
But -- and they're not necessarily all within the same group, so I think that's an important
piece.
So just to summarize the presentation. I think -- I hope I've shown that microbial metabolism
can modify a variety of different dietary components; that the differences in the gut
microbial community capacity to handle substrates can be detected using metabolic phenotypes,
even on a more gross level of monitoring urinary metabolites, and certainly as we move into
using the metabolomics approaches, can get a wide range of ways to evaluate this.
I think one of the things that is always a good reminder is that diet, as consumed, it
is not necessarily that experienced by the host. So from the standpoint of the components,
the metabolites produced by the bacteria, these may be, in some cases, quite different
from those that went into your mouth. And lastly, I think the gut microbiome really
needs to be considered in the context of the host diet in order for us to best understand
its impact on both host metabolism as well as its own, and on disease risk.
In relation to some of the gaps, needs, and challenges, coming at this from a very -- sort
of the nutritional angle, and it's interesting to see how these are sort of recapitulating
across many of our presentations in the last two days, I think one of our big challenges
certainly is trying to test the causality of gut microbiome and contribution to health
and disease, and certainly in the context of diet. And in that regard, I think we have
a strong need for a number of different approaches to this, both from the standpoint of the prospective
cohorts that allow us to follow individuals for extended periods of time with good measure
of exposure, whether it be diet or otherwise, as well as capacity to sample and characterize
the gut microbiome. I think there's a place for well-controlled dietary interventions
to try to get at some of the relationships for inter-individual variation. And finding
ways to harness these and make the most of them, I think, is important. And certainly,
the model systems are going to be critical as there are some many things that we certainly
can't do in humans.
And lastly, on a more broader level, I think, to reiterate Vince's comment this morning,
really trying to facilitate transdisciplinary research to allow for the breadth and depth
of knowledge that we need in this area, not only from the standpoint of a funding situation,
but certainly from the support of our institutions to be able to allow for this cross-fertilization.
And big plug for methods for assessing composite functionality of the gut microbial community,
and being able to integrate the structure and functional relationships. And lastly,
having the computational methods to integrate multiple omics platforms, and I think we've
heard that one before.
So lastly, just want to thank the members of my group, as well as collaborators at the
Hutch and University of Washington, Texas A&M University with collaboration with Rob
Chapkin project, and long-term collaborations. And lastly, to thank NCI for the support over
the years. Thank you.
[applause]
Female Speaker: We have time for a few questions.
Johanna Lampe: Everybody's just ready for dinner. [laughs]
Female Speaker: Well, if people aren't rushing for questions
now, it'll probably give us the opportunity to -- oh, someone's --
Female Speaker: I do have one question.
Female Speaker: Yeah.
Female Speaker: You mentioned lecithin, and you were looking
at meat and egg sources. I was wondering, have you looked at all at soy lecithin? I
know it's very commonly and almost ubiquitously put into processed foods.
Johanna Lampe: I would expect it probably is going to be
similar. I mean, as phosphatidylcholine, it's going to be similarly broken doing, and the
choline is going to be available as a substrate for the bacteria. So whether or not it's a
pure source, or whether it's coming from different food products, it's likely to have that effect.
Female Speaker: Well, please join me in showing appreciation
for all the speakers in this session.
[applause]
And we'll now start the discussion session for this afternoon.