Exploring host-microbiota interactions

The role of our intestinal microbiota
reaches far beyond fermentation of
indigestible food components. Apart
from immunological functions,
they have a major impact on our
metabolic and perhaps even mental
health. Dutch research organization
TNO is exploring this exciting
field, known as host-microbiota
interactions, and is developing and
integrating a range of advanced
models, techniques and trials. The
insights generated will support
manufacturers in the development
of functional foods targeted at, for
example, obesity.
The intestinal microbiota is frequently referred to
as “an extra organ in the human body”. This is not
surprising, given that our intestines contain up to
100 trillion bacteria, about 10 times more than there
are cells in the body. The intestinal microbiota exert
a wide range of functions: from playing a key role
in the fermentation of indigestible food leftovers,
to producing components that benefit intestinal
epithelial cells and help inhibit attachment of
pathogenic bacteria to the intestinal wall.
‘The intestinal microbiota affect not only our
immune system, but also our metabolic and mental
health’, says Guus Roeselers, Senior Research
Scientist at TNO in Zeist, the Netherlands. ‘People
with obesity appear to have different microbiota
compositions compared to people of “normal”
body weight. The hypothesis is that this affects
dietary energy harvest’, he continues. ‘Moreover,
experiments with germ-free mice have shown that
mice display different behaviors depending on
the composition of their intestinal microbiota. For
example when showing anxiety.’
Obvious link
The link between intestinal microbiota, and
metabolic and mental health is, actually, an obvious
one, says Roeselers: ‘The intestines have their own
local nervous system, with over 100 million neurons
and sensors in the intestinal wall. These neurons
interact with the microbiota and communicate,
through signaling molecules, neurotransmitters
and hormones, with the central nervous system in
the brain. This communication is responsible for,
amongst others, feelings of appetite and satiety, and
liking and wanting of certain foods.
‘Understanding how our gut microbiota interact
with environmental factors and our own genes.

allows us to assess whether actively manipulating
our microbial balances could help stem the
worldwide increase in metabolic diseases, and
improve mental health’, says Roeselers. The
mechanisms behind host-microbe interactions
are, however, complex. The intestinal microbiota
contain a variety of micro-organisms that not only
interact with the host but also with each other. These
interactions are, in their turn, affected by the host’s
nutrition and lifestyle.
Causal relationships
‘In the past few years the field of host-microbiota
interactions has, worldwide, attracted increasing
attention’, says Roeselers. ‘Progress in the areas of
DNA sequencing and bioinformatics has facilitated
detailed exploration of this promising new area.’
Though spectacular findings have been published,
until now most research findings remain at the
descriptive level. ‘However, we need to establish
causal relationships in order to enable the food
industry to make well-substantiated health claims in
the future’, he stresses.
Establishing causal relationships is only possible
with an interdisciplinary approach and state-of-theart
methods for data integration, stresses Marijana
Radonjic, Research Scientist Systems Biology at
TNO: ‘For instance, the advanced Network Biology
Approach we use allows for mapping the host’s
system components and the causal interactions
between those components at different levels, for
example at a molecular or physiological level.’
The Network Biology Approach can also be
focused on the ‘cross-talks’ between different
organs, in addition to their interaction with the
microbiome. ‘We can then relate these findings to
the health status of the host’, says Radonjic. ‘This
gives us understanding of processes that play a role
in achieving and maintaining optimal health. It also
helps us to identify “hotspots” and biomarkers that
reflect health effects of interventions’.
In addition to a Network Biology Approach, noninvasive,
predictive markers are vital in answering
questions on causal relationships. ‘It is virtually
impossible to study host-microbe interactions
in detail in human volunteers. Studies are timeconsuming
and you cannot easily get inside the
different compartments of the intestines of your
subjects to observe what is happening’, Roeselers
explains. ‘In addition, for an increasing number of
manufacturers, animal experiments are no longer an
ethical option.’
Complementary models
TNO has developed and applied a range of
advanced non-invasive models that closely simulate
processes taking place in the gastrointestinal tract.
These include the TIM-2 system for investigation
of fermentation processes in the large intestine,
the InTESTine model which uses intestinal wall
segments from pigs obtained from the veterinary
faculty of a local university, and the organoid
system.
The “organoids” originate from a spectacular
advance in stem cell research by the Hubrecht
Laboratory in Utrecht. ‘We have used insights from
this research to develop a complex host–microbe
model that very-well represents in vivo situations.
Fig 1
Organoid model Figure caption: A) Schematic
representation of a gut organoid with the lumen
corresponding to the intestinal lumen. Stem cells,
in the crypt base, give rise to transit amplifying cells
that differentiate into the four major epithelial cell
types: enterocytes, goblet cells, enteroendocrine
cells, and Paneth cells. B) Darkfield photograph of
a mouse gut organoid six days post- isolation of the
crypts.
We cultivate and stimulate single intestinal
stem cells to create crypt-like structures with
a recognizable organization of differentiated
cell types’, explains Roeselers. ‘We currently
use organoids from pigs, mice and humans in
studying host-microbiota interactions.’ Using the
organoid model, TNO scientists have been able
to demonstrate that butyrate – a short chain fatty
acid produced by intestinal bacteria – stimulates
the expression of genes in gut epithelial cells that
prevent fat storage and stimulate fat mobilization.
With all three models, data can be obtained within
a relatively short time span compared to human and
animal studies. The models complement each other
and often precede human intervention trials, for
example to optimize study designs.
The TIM-2 model has been an interesting point
of departure for several TNO studies. ‘This system
mimics processes that take place in the lumen of
the large intestine, without the presence of intestinal
cells. It can represent a variety of situations, for
example comparing the intestinal microbiota of
healthy people versus people with intestinal disease
or obesity. TIM-2 allows the taking of samples from.

places that would be unthinkable in humans or
animals’, explains Koen Venema, Project Manager
Gastrointestinal Health at TNO. ‘The samples we
take using TIM-2 are incubated in InTESTine cell
cultures or organoids for further research’, says
Venema.
TIM also enables the investigation of the link
between gut microbiota and mental health. ‘Although
at first this seems far-fetched, the validated in vitro
model allows one to study production of neuroactive
components by the microbiota, such as serotonin,
melatonin, biogenic amines and gamma-amino
butyrate (GABA)’, Venema explains.
Fig 2
Network biology of host-microbiome interactions
Human trial
Recently, TNO conducted a human trial to establish
correlations between the presence of certain gut
bacteria and specific metabolic effects. In this study,
10 male subjects received a high-fat, high-caloric
diet for 4 weeks, resulting in an average weight
gain of 3 kg. At the beginning of the study, and
after 4 weeks, DNA material was isolated from the
gut microbiota in feces and sequenced. Metabolic
health parameters, such as BMI and fatty acid
levels in the blood, were also determined. Univariate
correlation networks were used to examine the
relationship between microbiota composition and
physiological parameters.
The researchers observed diet-induced changes
in Bacteroidetes levels that could be related
to changes in carbohydrate oxidation in the
volunteers’ metabolic system. Changes in levels of
Firmicutes bacteria correlated with changes found
in fat oxidation. The correlations were statistically
significant.
Similar correlations have been observed in studies
with mice. ‘Using the outcomes from the human
trial and the mouse experiments, we have used
TIM-2 to screen for dietary substrates – primarily
prebiotics – to modulate the composition and the
activity of the microbiota from both lean and obese
individuals’ says Venema. The ratio of Firmicutes
and Bacteroidetes levels could be changed by
selecting the appropriate substrates. ‘Surprisingly,
the magnitude of the effect of some of the tested
substrates was different in a microbiota originating
from lean or obese volunteers. For example, apple
pectin and sugar beet pectin lead to a lower energy
extraction by a obese microbiota compared to a lean
microbiota’, he explains.
TIM-2 samples were also incubated with pig
intestinal tissue in the InTESTine model. The
experiments showed that incubations with
samples from the obese microbiota led to an
increased secretion of PYY – a hormone that
suppresses appetite – compared to those of the
“lean” microbiota. ‘Some substrates might have an
appetite suppressing effect in obese people, via
modulation of the microbiota”, says Venema.
European research programs
TNO is involved in a number of European research
initiatives in which host-microbe interactions are
studied. One of them is the NutriTech program
(www.nugo.org/nutritech) that aims to use cuttingedge
analytical technologies and methods to
evaluate the diet-health relationship, quantify the
effect of diet on “phenotypic flexibility” and find
biomarkers for the quantification of health effects of
diet and dietary ingredients.
Within the Nutritech program currently a human
caloric-restriction intervention study is carried
out. Radonjic is coordinator of the program’s
“Integration” work package. ‘The project will provide
an opportunity to further benchmark the Network
Biology Approach for studying host-microbiome
interactions as a powerful mean to infer host health
status, effects of nutrient interventions and find
biomarkers for health’, she says.
Another interesting international research program
is MyNewGut that focuses at the influence of the
microbiome on the hosts’ energy balance and
brain function. Roeselers is coordinating the work
package “Data Integration”. The program will start
by the end of 2013 and will run for five years.
According to the three TNO scientists, the next
five years will be both challenging and rewarding
for research on host-microbe interactions. ‘We have
generated many insights but there remains much
uncharted territory’, says Roeselers. ‘The main
challenge will be to correlate in vivo and in vitro
results and predict what will happen in the host,
following specific change(s) in diet or lifestyle. This
is essential for the development and optimization of
food and pharmaceutical products’, says Venema.
nutraceuticals now 23
‘Using TIM 2, we have been able to predict the
survival rates of probiotics in milk. These outcomes
support our idea to use a similar approach
for predicting survival and growth of intestinal
microbiota.’
Pre- and probiotics
At present, well-substantiated health claims
for food products that modulate the intestinal
microbiota in order to improve health do not exist.
Roeselers expects them to be in evidence within
five years. ‘Prebiotics and probiotics will become
an accepted intervention in a multidisciplinary
approach to countering obesity, via modulating
energy harvest and through control of appetite, and
liking and wanting of foods’, he says. ‘ln addition,
we will know much more about the relationship
between the microbiota and mental health.’
Venema agrees with Roeselers: ‘Our initial focus
will be on modulating brain development and
behavior, but in five years from now, it is likely that
we will be able to influence psychological wellbeing
as well. I believe that the gut microbiota plays
a role in every disease and disorder that you can
think of’. Venema expects major advances will be
made in all aspects of the analysis and detection
of samples, and in correlating the presence of
certain bacterial strains to host responses: ‘This will
definitely contribute to the substantiation of health
claims.’
Microbiota development
The development of the microbiota starts at
birth, when the sterile gastrointestinal tract of the
infant is colonized by vaginal and anal bacteria
from the mother and other micro-organisms in
the environment and the infant’s nutrition. The
colonization process persists during infancy, until,
at age four, children have developed their own
unique, stable microbiota. The composition of this
microbiota can only be changed temporarily, for
example by nutrition. This implies that development
of the microbiota, in the early years of life, can have
a major impact later in life, for example in terms of
being sensitive for developing obesity, or impacting
brain development.
TNO research program
TNO’s work on host-microbe interactions forms
part of the organization’s research strategy for
Food and Health. One of the program’s goals
is to develop generic methods for the shortterm
assessment of the health effects of food
products. The ambition is to establish groundbreaking
methodologies for the substantiation and
assessment of health claims that will be adopted
in the Netherlands and abroad. The organization
collaborates with respected universities and
research institutes in the Netherlands, Europe
and the United States. TNO is acknowledged,
worldwide, as a leader in the fields of systems
biology – with focus on metabolic and inflammatory
health – and in microbiological research.
For more information, please contact edwin.abeln@
tno.nl
References
• Den Besten G, van Eunen K, Groen AK, Venema
K, Reijngoud DJ, Bakker BM. The role of shortchain
fatty acids in the interplay between diet, gut
microbiota and host energy metabolism. J Lipid
Res. 2013 Jul:. [Epub ahead of print].
• Radonjic M, Wielinga PY, Wopereis S, Kelder
T, Goelela VS, Verschuren L, Toet K, Van
Duyvenvoorde W, Van der Werff van der Vat
B, Stroeve JHM, Cnubben N, Kooistra T, Van
Ommen B, Kleemann R. Differential effects of drug
interventions and dietary lifestyle in developing type
2 diabetes and complications: a systems biology
analysis in LDLR-/- mice. PLoS ONE. 2013. Feb.
8(2): e56122. doi:10.1371.
• Kelder T, Conklin BR, Evelo CT, Pico AR. Finding
the right questions: exploratory pathway analysis
to enhance biological discovery in large datasets.
PLoS Biol. 2010. Aug. 8(8): e1000472. doi:10.1371.
• Roeselers G, Bouwman J, Venema K, Montijn
R. The human gastrointestinal microbiota – an
unexplored frontier for pharmaceutical discovery.
Pharmacological Research. 2012. Dec. 66(6):443–
447.
• Roeselers G, Mittge EK, Stephens WZ, Parichy
DM, Cavanaugh CM, Guillemin K, Rawls JF
Evidence for a core gut microbiota in the zebrafish.
The ISME journal, 2011. Oct. 5(10):1595–608.
• Roeselers G, Ponomarenko M, Lukovac S,
Wortelboer HM Ex vivo systems to study host–
microbiota interactions in the gastrointestinal tract.
Best Practice & Research Clinical Gastroenterology,
2013. Apr 27(1):101–113.
• Venema K, van den Abbeele P Experimental
models of the gut microbiome. Best Pract Res Clin
Gastroenterol. 2013. Feb. 27(1):115-26.
• Venema K Role of gut microbiota in the control
of energy and carbohydrate metabolism. Curr Opin
Clin Nutr Metab Care. 2010. Jul. 13(4):432-8
• Wopereis S, Radonjic M, Rubingh CM, van Erk
M, Smilde AK, van Duyvenvoorde W, Cnubben
NH, Kooistra T, van Ommen B, Kleemann R.
Identification of prognostic and diagnostic
biomarkers of glucose intolerance in ApoE3Leiden
mice. PhysiGenomics. 2012. Mar.44(5):293-304.