Stichwort Mikrobiom:
Fördern Darmbakterien
das Stammzellwachstum
bei Darmkrebs?
Prof. Dr. med. Sebastian Zeißig
Medizinische Klinik 1, Universitätsklinikum Dresden
Zentrum für Regenerative Therapien Dresden
13.10.2016
40 Billion (4x1013)
Mikroorganismen
(Entspricht der Zahl
körpereigener Zellen)
0
1000000
2000000
3000000
4000000
An
za
hl d
er
Ge
ne
~23.000
menschliche
Gene
~3.3 Millionen
mikrobielle
Gene
Intestinale Mikrobiota
Qin et al., Nature 2010
Arumugam et al., Nature 2012 Sender et al., Cell, 2016
Intestinale Mikrobiota
Ley et al., Cell 2006
• circa 1200 mikrobielle Spezies
• Davon ca. 160 pro Individuum
• 99% Bakterien
• 0.1% Viren (primär Bakteriophagen)
Qin et al., Nature 2010
Helicobacter pylori und Magenkarzinom
Amieva, Peek, Gastroenterology 2016
Helicobacter, Entzündung, Karzinogenese
Polk and Peek, Nat Rev Cancer 2010
Terzic et al., Gastroenterology 2010
Fearon, Vogelstein, Cell 1990
Sequenzielle Mutagenese als Grundlage des
kolorektalen Karzinoms
Vermeulen, Snippert, Nat. Rev. Cancer 2014
Kolorektale Karzinogenese & Stammzellen
Die Mikrobiota als pathogener Faktor?
Rakoff-Nahoum et al., Science 2007
Dove et al., Cancer Res 2007
Lee et al., Nat Med 2010
Li et al., Carcinogenesis 2012
Arthur et al., Science 2012
Grivennikov et al., Nature 2014
Belcheva et al., Cell 2014
Song et al., Immunity 2014
Mikrobiota und entzündliche Signalwege im
kolorektalen Karzinom
Irrazabal, Mol Cell 2014
Greten et al., Cell 2004
Bollrath et al., Cancer Cell 2009
Grivennikov et al., Cancer Cell 2009
Vlantis et al., J Clin Invest 2011
Schwitalla et al., Cell 2013
Schwitalla et al., Cancer Cell 2013
Grivennikov et al., Nature 2014
Mikrobielle Barrieren im Darm
Johansson, PNAS 2008
Barrierestörung in intestinalen Adenomen und
Karzinomen
Normalgewebe Adenom
Barrierestörung, Entzündung, Proliferation
Dejea et al., PNAS 2015
Garrett et al., Science 2015
Dejea et al., PNAS 2014
Grivennikov et al., Nature 2014
Kostic, Cell Host Microbe 2013
Castellarin et al., Genome Res 2012
Kostic et al., Genome Res 2012
Arthur et al., Science 2011
Wu et al., Nat Med 2009
Nougayrede et al., Science 2006
Bakterielle Promotoren der Karzinogenese?
Fusobacterium nucleatum als Promoter des
Tumorwachstums
Kostic,
Cell Host Microbe
2013
INTESTINAL TUMOR NORMAL IEC
mic
robio
ta
MA
MP
s
str
atificatio
n
(mu
cus, A
MP
s)
Calmodulin
Calcineurin
Dclk1
tumor cell proliferation
tumor growth
Lgr5 Olfm4
NFAT
P P P
NFAT
Nucleus
NFAT
CRAC PLCγ
TLR
IP3
ER
STIM1?
Ca2+
Ca2+
Calmodulin
Calcineurin
NFAT
P P P
CRAC PLCγ
TLR
ER
Ca2+
(Dclk1)
(Lgr5) (Olfm4)
alte
red
mic
robio
ta
alte
red
stra
tificatio
n
Nucleus
non-proliferating IEC
Calcineurin, NFAT und Darmkrebs
Peuker et al.,
Nat. Med. 2016
Zusammenfassung
• Bakterien und ihre Erkennung durch das Immunsystem
sind mit Entzündung auf zellulärer Ebene verbunden
• Diese Entzündungsprozesse tragen durch Regulation der
epithelialen Zellteilung, insbesondere in
Tumorstammzellen, zur Karzinogenese bei
• Keimfreie Mäuse sowie Mäuse mit Deletion von
Rezeptoren der bakteriellen Erkennung weisen eine
reduzierte kolorektale Karzinogenese auf
Nature Reviews | Microbiology
a b
Epithelium
Lamina propria
Intestinal lumen
Infla
m
ed epithelium
Secondary
colonizer
Tertiary colonizer
IL-17
CPT
ToxinLCN2
b-def
TH17 cell
IL-23DC
Adenoma
Primary
colonizer
their temporal association with the colonic
mucosa, and that these different associations
are dictated by alterations in the micro
environment as the disease progresses.
Our model does not exclude the active
involvement of bacterial passengers, such
as Fusobacterium spp., in CRC develop
ment. Instead, our analysis indicates that
the tumour microenvironment provides a
preferred niche for these bacteria and that
an involvement in CRC progression, if any,
will be more pronounced during the later
stages of the disease. Finally, it should be
realized that the microbiome ‘snapshots’ that
are provided by the currently available next
generation sequencing data do not reflect
the whole continuum of microbiome shifts
that take place during CRC development.
Factors influencing the identification of bac-
terial drivers and passengers. As mentioned
above, technical limitations (attributable
to the complex and diverse composition of
the colonic microbiota), combined with the
limited depth of the microbiome analyses,
may impede the identification of bacterial
drivers and passengers of CRC. Moreover,
the lack of consistent colonization pat
terns in the offtumour samples from the
recent CRC microbiome studies (that is,
the inconsistency of CRC driver bacteria)
may also reflect biological factors. First,
CRC may be triggered by multifactorial
signals, and distinct bacteria may confer a
similar CRC risk (for example, ETBF and
members of the family Enterobacteriaceae, as discussed above). Increased coloniza
tion by any one of these bacterial drivers
may initiate carcinogenesis, and this would
preclude consistent findings on the species
(or strain) level. Second, although meta
genomic27 and metatranscriptomic28 data
sets were obtained by the recent studies,
these investigations focused mainly on iden
tifying the taxonomic clades associated with
CRC, whereas the encoded functions were
neglected. Thus, there is no discrimination
between, for example, Bacteroides strains
that do and do not produce BFT, a toxin that
has been implicated in CRC initiation77,78.
Third, the presence or absence of certain
drivers may depend on their interaction
potential with other members of the intes
tinal ecosystem. Multiple mechanisms can
be envisioned for such a dependency. For
instance, colonization by a maladapted spe
cies may be aided by the ability of primary
colonizers to modulate the host response, as
put forward by the alphabug hypothesis29.
In this model, ETBF induces the production
of IL22 and IL23 by dendritic cells48. IL23
stimulates several subsets of T cells to secrete
IL17, which promotes amplification of the
host response by stimulating the intestinal
epithelium to secrete CXC-chemokines (which
are neutrophil chemoattractants) and anti
microbial peptides (such as βdefensins,
lipocalin 2 (LCN2; also known as NGAL)
and calprotectin)79,80. The primary role of
this TH17 response is to prevent bacterial
dissemination from the gut, but it also
promotes colonization of the mucosa by
bacteria (including pathogens) that are
resistant to some of the induced antimicro
bial responses81 (FIG. 3a). For instance, LCN2
binds to and sequesters bacterial sidero
phores, which are required by a number of
species to survive in ironlimiting environ
ments81. Certain pathogenic members of
the family Enterobacteriaceae, such as
Salmonella, Escherichia and Klebsiella spp.,
have evolved ironsequestering siderophores
that are resistant to LCN2 binding81, giving
them a competitive advantage in the
environ ment of the inflamed gut. Similarly,
ETBF may facilitate colonization of species
that are resistant to βdefensins. One could
speculate that if these secondary colonizers
become established at the inflamed site, they
have the opportunity to become drivers of
CRC, especially if they produce genotoxins,
which would augment intestinal carcino
genesis in the long term. Thus, when deciding
whether an individual is at high risk for
the development of CRC, it may be neces
sary not only to identify the organisms that
are present in the indigenous (that is, off
tumour) microbiota, but also to determine
their functional repertoire.
Interactiondependent colonization may
also be one of the factors that explain the
large variation in the ontumour microbi
omes82,83. For instance, certain bacteria may
be poor colonizers of developing tumours
but may initially adhere to other species
Figure 3 | Interaction-dependent colonization of the intestinal epithe -
lium. Two models for interaction-dependent bacterial colonization of
inflamed and adenomatous intestinal tissue. Bacteria that ar e unable to colo-
nize the colonic mucosa in the absence of ‘helper’ bacteria are referred to as
secondary colonizers, and the helper bacteria are referred to as primary
colonizers. a | Secondary colonizers may be indirectly stimulated to colonize
an inflamed gut following induction of the host immune r esponse by primary
colonizers. Bacterial toxins released from the primary colonizers may induce
interleukin-23 (IL-23) and IL-17 responses from the dendritic cells (DCs) and
T helper 17 (TH17) cells, respectively, in the lamina propria, and this in turn
would cause intestinal epithelial cells to secr ete anti bacterial compounds
such as β-defensins (β‑def), calprotectin (CPT) and lipocalin 2 (LCN2) into the
gut lumen79–81. Increased levels of these compounds provide a selective pres-
sure in favour of resistant secondary colonizers (see main text for examples).
When they are established, secondary colonizers have the potential to
become drivers of colorectal cancer (CRC). b | Primary colonizers of adeno-
matous tissue may directly facilitate seeding of secondary colonizers by , for
example, providing an adherent surface for their attachment to the CRC
microenvironment. In addition, these secondary colonizers may in turn
function as ‘bridging organisms’ for tertiary colonizers.
PERSPECTIVES
580 | AUGUST 2012 | VOLUME 10 www.nature.com/reviews/micro
© 2012 Macmillan Publishers Limited. All rights reserved
Cancer is defined as uncontrolled, malignant
cell proliferation caused by accumulated
genetic and epigenetic mutations1,2. The
triggers for these mutations can be multi
factorial in origin and remain elusive in
many cases. Several types of cancer are
associated with infectious agents, and many
of these cancers occur in tissues with a high
exposure to the microbiota3. It has been esti
mated that about 20% of the global cancer
burden can be linked to infectious agents4.
Wellknown examples include cervical
and gastric cancer, which can be caused by
human papilloma viruses and the bacterium
Helicobacter pylori, respectively4,5.
Bacteria constitute about 90% of all cells
in the human body, and it has been esti
mated that the total number of microbial
genes exceeds the number of human genes
by two orders of magnitude or more6. The
majority of these bacteria, an estimated
1014 cells comprising >103 different species,
colonize the large intestine6–9. Interestingly,
the bacterial density in the large intestine
(~1012 cells per ml) is much greater than that
in the small intestine (~102 cells per ml), and
this is paralleled by an approximately 12fold
increase in cancer risk for the large intes
tine compared with the small intestine10,11.
Moreover, mutant mice that are genetically
susceptible to colorectal cancer (CRC)
develop significantly fewer tumours under
germfree conditions than when they have
a conventional microbiota12–14. Despite this
knowledge, the possibility that intestinal
microorganisms have a direct effect on the
initiation and progression of sporadic CRC
has been largely ignored since Fearon and
Vogelstein formulated a genetic model for
this disease more than 15 years ago15.
In this Opinion article, we present a
brief overview of the intestinal microbiome
of humans, paying special attention to the
changes in bacterial composition during
CRC progression. We highlight recent data
that support a role for the gut microbiota in
this disease, and we propose that CRC can
be initiated by ‘driver’ bacteria, which are
eventually replaced by ‘passenger’ bacteria
that either promote or stall tumorigenesis.
On the basis of these predictions, we present
a bacterial driver–passenger model to clarify
how intestinal bacteria can directly or
indirectly mediate CRC development.
Colorectal cancer
CRC is the fourth most commonly diag
nosed cancer in the world, with more
than one million new cases and more than
600,000 deaths annually16. Many of the
risk factors for CRC are associated with
a ‘Western’ lifestyle. In particular, a high
consumption of animal fat, processed meat
and red meat combined with a low intake
of vitamin D, fibre and fish is thought to
increase the risk of disease development17.
The underlying genetic basis of the disease
is described by the ‘adenoma–carcinoma
sequence’ model that was developed by
Fearon and Vogelstein, which posits that
accumulating genetic and epigenetic muta
tions (genomic instability) drive epithelial
dysplasia and hyperplasia in the colon, ulti
mately resulting in CRC15,18. Specifically,
CRC is initiated when the stem cells at the
base of the villus crypt develop a mutation
that renders them immortal and prone to the
accumulation of additional mutations19,20.
The most commonly mutated genes include
tumour suppressors (adenoma tous polyposis
coli (APC), the βcatenin gene (CTNNB1),
deleted in colorectal cancer (DCC) and
P53 (also known as TP53)) and oncogenes
(Kirsten rat sarcoma (KRAS) and myelo
cytomatosis oncogene (MYC))21–24. These
mutations are termed driver mutations
and are associated with several disease
hallmarks, including cell growth without
external growth signals, insensitivity to
antigrowth signals, evasion of apoptosis
and immune destruction, limitless replica
tive potential, reprogramming of energy
metabolism, increased angiogenesis, and
tissue invasion and metastasis25. Because the
transition from an adenoma to a carcinoma
requires a mutation in a tumour suppressor
gene or an oncogene, the process can be slow
and may take more than 10 years depending
on the mutation frequency. Meanwhile, the
genomes of the adenoma cells accumulate
numerous passenger mutations that have
no direct effect on tumour progression. The
triggers leading to this accumulation of
mutations remain ill defined.
Recently, deepsequencing technology
has allowed us to explore the microbial
compo sition of both healthy and diseased
body sites, and we now have experimental
O PI NI O N
A bacterial driver–passenger model for colorectal cancer: beyond the usual suspects
Harold Tjalsma, Annemarie Boleij, Julian R. Marchesi and Bas E. Dutilh
Abstract | Cancer has long been consider ed a genetic disease. However,
accumulating evidence supports the involvement of infectious agents in the
development of cancer, especially in those organs that are continuously exposed
to microorganisms, such as the large intestine. Recent next-generation sequencing
studies of the intestinal microbiota now offer an unprecedented view of the
aetiology of sporadic colorectal cancer and have revealed that the microbiota
associated with colorectal cancer contains bacterial species that differ in their
temporal associations with developing tumours. Her e, we propose a bacterial
driver–passenger model for microbial involvement in the development of
colorectal cancer and suggest that this model be incorpor ated into the genetic
paradigm of cancer progression.
PERSPECTIVES
NATURE REVIEWS | MICROBIOLOGY VOLUME 10 | AUGUST 2012 | 575
© 2012 Macmillan Publishers Limited. All rights reserved
Nat Rev Microbiol, 2012
normal
tumor
Bakterien als therapeutische Vehikel?
Vivit Vivit
Vivit
Vielen Dank!