Human Genetics (2021) 140:747–760

https://doi.org/10.1007/s00439-020-02237-0

REVIEW

The relationship between the gut microbiome and host gene

expression: a review
Robert G. Nichols1 · Emily R. Davenport1,2

Received: 3 September 2020 / Accepted: 6 November 2020 / Published online: 22 November 2020
© The Author(s) 2020

Abstract
Despite the growing knowledge surrounding host–microbiome interactions, we are just beginning to understand how the gut
microbiome influences—and is influenced by—host gene expression. Here, we review recent literature that intersects these
two fields, summarizing themes across studies. Work in model organisms, human biopsies, and cell culture demonstrate that
the gut microbiome is an important regulator of several host pathways relevant for disease, including immune development
and energy metabolism, and vice versa. The gut microbiome remodels host chromatin, causes differential splicing, alters the
epigenetic landscape, and directly interrupts host signaling cascades. Emerging techniques like single-cell RNA sequencing
and organoid generation have the potential to refine our understanding of the relationship between the gut microbiome and
host gene expression in the future. By intersecting microbiome and host gene expression, we gain a window into the physi-
ological processes important for fostering the extensive cross-kingdom interactions and ultimately our health.

Introduction of disease remains largely unknown. Although, it has been

demonstrated that the microbiota play causal roles in several
The human body plays host to large numbers of bacteria, diseases, including obesity (Ridaura et al. 2014; Turnbaugh
fungi and other microorganisms—commonly referred to et al. 2006) and diabetes (Wen et al. 2008). Given the impor-
as the microbiota (Shreiner et al. 2015). These microbes tance for our health, it is necessary to characterize the physi-
are important to our health. They assist in establishing host ological relationships between the microbiota and human
immunity (Kamada et al. 2013), strengthen the gut bar- host to understand disease etiology and design therapeutics
rier (Leclercq et al. 2014), and provide beneficial metabo- involving the microbiome.
lites (Lukovac et al. 2014). Consequently, the microbiota There is much to be learned about human–microbiome
is associated with a large number of complex diseases in interactions by studying the genetic components of each.
humans, including inflammatory bowel disease (IBD) (Alek- The human genome contains approximately 20,000 protein
sandrova et al. 2017), cardiovascular disease (CVD) (Tang coding genes (Salzberg 2018). These genes are regulated
et al. 2017), and colon cancer (Nelson and Chia 2019). in a tissue-specific manner by both intrinsic host factors as
Whether shifts in the microbiota lead to or are the cause well as sensing environmental cues. Collectively, the micro-
bial genomes within each of our microbiomes contain an
estimated 100 times the gene content as our own genome
Electronic supplementary material The online version of this (Nelson et al. 2010). This genetic material is sometimes
article (https​://doi.org/10.1007/s0043​9-020-02237​-0) contains referred to as our ‘second genome’, as the coding potential
supplementary material, which is available to authorized users.
of the microbiome greatly expands upon the coding poten-
* Emily R. Davenport tial our own genome (Grice and Segre 2012). For example,
[email protected] genes unique to the microbiota create metabolites required
Robert G. Nichols by the human host (such as vitamin B12, biotin and folic
[email protected] acid) (Hooper et al. 2002) and allow for microbial survival
(such as adhesion factors and transport systems) (Reidl et al.
1
Department of Biology, The Pennsylvania State University, 2009). Understanding how the gene products of the host and
University Park, PA 16802, USA
microbiome interact can offer clues into what physiological
2
Huck Institutes of the Life Sciences, The Pennsylvania State
University, University Park, PA 16802, USA

13
Vol.:(0123456789)
748 Human Genetics (2021) 140:747–760

processes are necessary for maintaining these complex To explore microbial genomes and functional capac-
cross-kingdom relationships. ity of the gut microbiome, shotgun metagenomics is used.
In this review, we offer insights about the physiology of This involves sequencing the total DNA composition of a
host–microbiota relationships gained by studying host gene sample. Several aspects of the microbial community can be
expression jointly with the microbiome, specifically in the assayed using metagenomics, including characterizing taxo-
gut. We briefly describe the microbiome and how it is stud- nomic composition [through programs like Kraken (Wood
ied, synthesize themes gained through studies performed in and Salzberg 2014) and MetaPhlAn2 (Truong et al. 2015)],
powerful model systems and in humans, and highlight poten- functional capabilities [through the program HUMAnN2
tial mechanisms through which host transcription–micro- (Franzosa et al. 2018)], and assembling individual micro-
biota cross-talk occurs. Finally, we propose future direc- bial genomes [metagenome assembled genomes, or MAGs
tions, both experimental and analytical, that will further our (Albertsen et al. 2013; Allen and Banfield 2005; Nielsen
understanding of how host transcription and the microbiome et al. 2014; Parks et al. 2017)]. While metagenomics has
interact. the potential to reveal additional functional information over
phylogenetic marker gene sequencing, it is expensive and
can be cost prohibitive in samples with a high host to bacte-
The human gut microbiome: a primer rial biomass ratio.
Finally, to explore functional activity of the gut microbi-
The relationship between the host and the gut microbi- ome, RNA within a microbiome can be sequenced via RNA-
ome starts at birth when the microbiome of the newborn seq (referred to as metatranscriptomics). Typically, research-
is seeded. Delivery mode (cesarean section or vaginal ers deplete ribosomal and transfer RNA experimentally to
delivery) plays an important role in the establishment of enrich for bacterial transcripts prior to sequencing, as ribo-
the microbiome (Bäckhed et al. 2015; Dominguez-Bello somal and transfer RNA make up approximately 95–97%
et al. 2010; Montoya-Williams et al. 2018; Papathoma et al. of total RNA in a bacterial cell (Rosenow et al. 2001).
2016). Proper nutrition and the transition from breast feed- Metatranscriptomics investigates bacterial gene expression
ing to more solid foods results in maturation of the infant (transcription) changes between conditions or individuals
microbiome (Bäckhed et al. 2015). The establishment of the [analyzed with the program SAMSA2, for example (Westre-
microbiome in neonates works in concert with the establish- ich et al. 2018)]. 16S rRNA gene sequencing, metagenomics,
ment of innate mucosal immunity, but can be disrupted by and metatranscriptomic sequencing can be coupled with host
early use of antibiotics (Russell et al. 2012). The compo- transcriptomics to investigate how different aspects of the
sition of the microbiome rapidly diversifies up to the age gut microbiome (composition, functional capacity, or func-
of three, steadily increases until around the age of 40, and tional activity) affects host gene expression and vice versa.
then remains fairly stable (de la Cuesta-Zuluaga et al. 2019;
Yadav et al. 2016; Yatsunenko et al. 2012). Though highly
unique between individuals, the microbiome is relatively Relationship between microbiome and gene
resistant to long-term changes (Bäckhed et al. 2012). The expression in model organisms
microbiome will typically return to a state of equilibrium
after a stress like a dietary change, a short term adult anti- A powerful way to investigate both the microbiome and
biotic treatment, or an acute invasion by a pathogenic bac- host transcriptomics is with model organisms, such as mice
terium (Bäckhed et al. 2012). However, short-term modula- (Fig. 1), zebrafish, C. elegans or Drosophila melanogaster.
tions of the gut microbiome can interrupt normal metabolite With model organisms, researchers have control over the
production (Yoon and Yoon 2018). This in turn may cause environment, complete control over diet, and, importantly,
changes in host gene expression that could lead to more the ability to study all tissues. These aspects are either
long-lasting effects in the host. impossible or extremely difficult to do with human subjects.
To investigate the microbiome, researchers employ a Organisms bred without a microbiome (e.g., germ-free mice)
plethora of techniques, many involving next-generation can be used to identify gut microbiome-mediated effects in
sequencing technology. To explore the taxonomic makeup the host organism. For example, if the effects of a treatment
of the gut microbiome, sequencing a phylogenetic marker are lost in germ-free mice when compared to wild-type or
gene is easy and efficient. Most often the 16S rRNA gene conventionalized mice, then it can be inferred that gut micro-
is assayed, as it is universally present in archaea and bac- biome plays a role in the specific effect. While powerful,
teria. The gene contains alternating segments of high and there are caveats to consider when using germ-free mice.
low conservation, which can be used for PCR priming and With no microbiome present, there is much thinner mucous
taxonomic identification, respectively (Davidson and Epper- layer in the gastrointestinal (GI) tract of a germ-free mouse
son 2018). compared to a conventional mouse (Miyakawa et al. 1971).

13
Human Genetics (2021) 140:747–760 749

Fig. 1  Using mice as a model organism. Mice are popular model ome-derived samples and techniques. Both microbiome-derived and
organism for characterizing gut microbiome–host relationships. host-derived techniques can be categorized in three groups; sequenc-
Researchers compare gnotobiotic (germ-free), mono-colonized, ing techniques, other -omic techniques and gross tissue techniques.
and/or conventionalized/wild-type mice to determine the role of the SHIME stands for the Simulator of the Human Intestinal Microbial
microbiome in their phenotype of interest. Orange elements represent Ecosystem (Molly et al. 1993)
host-derived tissues and techniques. Blue elements represent microbi-

Barrier function is worse, and there is likely innate inflam- not observed in a germ-free ­HDAC3ΔIEC mouse model, dem-
mation present in the gut (Miyakawa et al. 1971). Germ-free onstrating that the effects were mediated by the microbiome.
mice also have reduced metabolic rates and enlarged ceca These results highlight one mechanism by which the micro-
(Miyakawa et al. 1971). These innate functional differences biome is sensed and affects gene expression in the host: by
must be noted when drawing conclusions with germ-free influencing the expression of a host epigenetic modifying
mice. enzyme.
Regardless, germ-free mouse experiments demonstrate Modulation of the microbiome (either by probiotics
that the microbiome plays a role in regulating host gene or creating germ free mice) can have dramatic effects on
expression. For example, a murine model with a tissue-spe- host gene expression. In a study focusing on 303 xeno-
cific deletion of histone deacetylase 3 (HDAC3) results in biotic processing genes expressed in the intestine, 116
the dysregulation of intestinal epithelial cell gene regulation genes were significantly differentially expressed between
dependent on the microbiome (Alenghant et al. 2013). Spe- mice raised conventionally vs. germ-free (Fu et al. 2017).
cifically, an intestinal epithelial cell specific HDAC3 knock- These genes include Phase I enzymes, Phase II enzymes,
out mouse line was bred (­ HDAC3ΔIEC) to investigate inflam- transporters, and transcription factors. Additionally,
matory bowel disease (IBD) progression. Conventionalized expression of drug-metabolizing enzymes in the liver are
­HDAC3ΔIEC mice show dysregulated host gene expression significantly altered between groups of mice raised conven-
and disrupted homeostasis. Conversely, this dysregulation is tionally, conventionally + probiotics (VSL3), germ-free, and

13
750 Human Genetics (2021) 140:747–760

germ-free + probiotics (Selwyn et al. 2016). These studies reveal insights into the relationship between the microbi-
demonstrate the causal role the microbiome plays in altering ome and host transcription. Studies conducted in zebrafish
host gene expression across the body. (Murdoch and Rawls 2019), C. elegans (Dirksen et al. 2020;
Host energy metabolism is affected by the microbiome, Yang et al. 2019), and Drosophila melanogaster (Broderick
as evidenced by changes in host gene expression. Short- et al. 2014; Douglas 2018; Elya et al. 2016) for example, all
chain fatty acids (SCFAs) are small molecules made solely have revealed the impact of the microbiome on immunity,
by the microbiome via microbial fermentation. In animal metabolism, and developmental gene regulation.
guts, these molecules are generated by the breakdown of
host-indigestible foods and serve as a major energy source
for the host. For example, an abundant SCFA of major Gastric pathogens influence host
health importance is butyrate, which colonocytes use as an transcription
energy source (Donohoe et al. 2011). In germ-free mice,
the TCA cycle is dysregulated in colonocytes, due to the Given the difficulties of studying human gene expression,
lack of butyrate (Donohoe et al. 2011). However, upon sup- far less is known about the relationship between the human
plementation of butyrate, isolated colonocytes from a germ- microbiome and the human transcriptome. The area with
free mouse dramatically increased mitochondrial respira- the most research to date is the study of how gut pathogens
tion. Additionally, the microbiome influences host hepatic influence host gene expression, either directly or indirectly
lipogenesis in wild-type mice. Increased monosaccharide through immune stimulation. One of the ways pathogens
absorption in the mouse gut suppresses the expression of modulate host gene expression is indirectly through the acti-
fasting-induced adipocyte factor (FAIF), which results in vation of host microRNAs (miRNA), which are small RNA
the deposit of triglycerides in adipocytes (Bäckhed et al. molecules (~ 20 nucleotides long) that repress transcribed
2004). These are just two examples of how microbes in the mRNAs in human cells by targeting specific RNAs for deg-
gut regulate host energy metabolism, even in distant tissues. radation or inhibiting their translation (Maudet et al. 2014).
Host immunity is also affected by the microbiome. In a Pathogenic bacteria like Listeria monocytogenes (Schnitger
mouse model comparing wild-type mice to mice lacking et al. 2011), Salmonella Typhimurium (Schulte et al. 2011)
innate immunity (Myd88 knockout model) in both con- and Helicobacter pylori (Zhang et al. 2008) all stimulate
ventionally raised and germ-free conditions, over half of host miRNAs that dampen the immune response, repress
the expressed genes in the GI tract were regulated by the apoptotic signals, and increase autophagy to avoid host
microbiome (2844 of 5652 genes) (Larsson et al. 2012). clearance (Maudet et al. 2014).
Gene Ontology (GO) analysis, showed the top differentially In addition to activating host miRNAs, pathogenic bacte-
expressed genes were involved with host immune responses rial species release specific virulence factors called effector
and host energy metabolism. Interestingly, the microbiomes proteins which can either bind to host proteins to inhibit host
involvement in both patterns of gene expression and depend- cellular pathways or can act as enhancers or repressors for
ency on MyD88 shifted along the GI tract. The composition, host genes (Shames and Finlay 2012). Pathogenic bacteria
gene expression, and epigenetic profile of innate lymphoid like Salmonella Typhimurium utilize their specific effector
cells (ILCs) are also shaped by the microbiome (Gury- proteins to repress host innate immunity, leading to a suc-
BenAri et al. 2016). Specifically, single-cell RNA-seq, cessful invasion (Hausmann and Hardt 2019). In cases of
ATAC-seq, and iChIP-IVT on the intestinal lamina propria Clostridium difficile infection (CDI), the two major toxins
of conventional and antibiotic treated mice revealed hun- produced (TcdA and TcdB) increase expression of vascular
dreds of transcripts that were differentially regulated by the endothelial growth factor A (VEGF-A) in gut epithelial cells
microbiome in several ILC clusters. These included genes (Huang et al. 2019). VEGF-A promotes angiogenesis and
related to cellular adhesion and interaction with the extracel- vasodilation and is upregulated in cases of IBD (Danese
lular matrix, chemokine signaling, and MAPK signaling. In et al. 2006). Vasodilation exacerbates inflammation asso-
general, the transcriptomic profile of two ILC subsets (ILC1 ciated with IBD and promotes CDI pathogenesis (Huang
and ILC2) shifted towards a third (ILC3). These results et al. 2019). The toxins associated with CDI also downregu-
demonstrate the degree of plasticity host cells display in late the expression of aquaporins (specifically AQP1) in a
response to microbial stimuli and how that can affect down- human intestinal microvascular endothelial cell line (Hui
stream immune function. The relationship between the gut et al. 2018). Irregular aquaporin activity results in diarrhea,
microbiome and host innate immunity has been extensively due to disrupted osmosis (Hui et al. 2018).
reviewed by Pott and Hornef (2012), Honda and Littman Gut pathogens affect the host gene expression in a cell-
(2016), and Shi et al. (2017). type specific manner, and new techniques like dual RNA-
While we highlight mouse models above, many addi- seq and single cell RNAseq (scRNA-seq) can be used
tional organisms serve as powerful laboratory systems to to investigate that relationship. Dual RNA-seq involves

13
Human Genetics (2021) 140:747–760 751

simultaneously sequencing both the pathogenic bacteria Relationship between the gut microbiome
and the afflicted host cell (Westermann et al. 2012). This and host transcription in humans
technique allows researchers to see the exact interactions
that are occurring between the pathogen and the host cell When investigating microbiome–host interactions in
(Westermann et al. 2016). The second technique mentioned, humans, researchers employ various approaches, such as
scRNA-seq, is a modified RNA-seq technique that involves collecting biopsies (taken during colonoscopies, gastric
sorting individual cells followed by RNA-seq to characterize surgeries, or colonic surgeries), generating organoids, the
the transcriptome on a cell-by-cell basis. ScRNA-seq allows gut on a chip model (Kim et al. 2016) or using human cell
for detection of gene expression differences both within and cultures (Fig. 2, Supplemental Table 1). Biopsies provide
between different cell-types and characterization of cell-type the most biologically faithful representation of host gene
proportions in a sample. expression–microbiome relationships at the interface of
Currently, most scRNA-seq studies involving some the intestinal lumen, but suffer from the drawback that
microbiome component investigate the mechanisms under-
lying viral infection of host cells. For example, scRNA-seq
revealed that there are highly heterogeneous infectivity
and host transcriptional responses in a primary fibroblast
model infected with herpes simplex virus 1 (HSV-1) (Dray-
man et al. 2019), identifiable transcriptomic signatures of
SARS-CoV-2 infection in patients with severe disease (Bost
et al. 2020), and no discernable differences between lytic
and latent human cytomegalovirus (HCMV) transcriptomes
in CD14 + monocytes and CD34 + HPCs (Shnayder et al.
2018). Though bacteria have different pathogenesis than
viruses, scRNA-seq could be used to show how different
microbial species of the microbiome affect the composition
and transcriptional profiles of different host cells.
While exciting, technical limitations exist for scRNA-seq
on host cells as well as for using scRNA-seq on a micro-
biome sample. The first hurdle of scRNA-seq is that the
equipment needed for the experiments are expensive (Liu
and Trapnell 2016). Additionally, the procedures used to
isolate cells and their respective RNA are complicated and
could introduce bias based on the enzyme treatments used.
ScRNA-seq methods are most criticized for their low capture
rates, meaning that sparse transcripts may be missed. When
applying scRNA-seq to the microbiome, the above issues
are compounded with the fact that there are thousands of
species of bacteria in our gut. There has not been a method
developed for using scRNA-seq specially on a microbi-
ome sample. However, one group took existing scRNA-seq
methods that address multiple species (Butler et al. 2017), Fig. 2  Techniques for examining microbiome—gene expression rela-
zero inflation issues (Van den Berge et al. 2018) and issues tionships in humans. Several different approaches are used to investi-
gate the relationship between the microbiome and host transcriptome
with technical noise (Kharchenko et al. 2014), tested them in humans. a Human subjects are split into two groups: one group is
on simulated and manually curated 16S and metagenomic placed on antibiotics to temporally reduce the gut microbiome, while
microbiome data, and compared the results with existing the other acts as a control. Researchers collect fecal and blood or tis-
metagenomic analysis techniques (Calgaro et al. 2020). sue samples to study the microbiome and host transcriptome, respec-
tively. b Tissue biopsies are collected during medical procedures
This group concluded that there was no perfect method from both inflamed and non-inflamed tissue, for example during colo-
but existing microbiome methods like DESeq2, edgeR and noscopies. RNAseq data are collected from the host tissue, while the
corncob preformed the best when analyzing the data. The microbiome is characterized from either the associated mucosal layer
prospect of combining scRNA-seq with the microbiome is or a separately collected fecal sample. c Human cell lines are estab-
lished and co-cultured in the presence of a microbiome, individual
attractive, new analytical techniques will need to be imple- bacterium, or vehicle control. Differences in gene expression between
mented to deal with the complexity of both the host and the conditions can then be attributed to the microbiome
microbiome.

13
752 Human Genetics (2021) 140:747–760

they are invasive to obtain and are difficult to maintain the of host gene expression changes, and that the relationship
features of ‘normal’ human tissue. Consequently, many between the two is highly dependent on disease context.
profiled biopsies originate from diseased tissue, and do Biopsies are highly invasive and expensive, and alterna-
not offer insight into host–microbiome interactions in tive strategies exist for studying the relationship between
healthy states. In some instances, matching control sam- the human microbiome and transcriptome. Cell cultures are
ples are collected from patients with diseased or inflamed cheap, relatively easy to maintain, and can be completely
tissue (Häsler et al. 2017; Lloyd-Price et al. 2019). This controlled by the researcher. To study how the host cells and
matched study design allows direct comparison between microbes interact, co-culture techniques can be employed
“normal” and inflamed tissue, which can be a more power- where intestinal cells are co-cultured with fecal isolates to
ful approach that takes into account host-specific factors simulate a host–microbiome environment (Richards et al.
like diet and genetics versus comparing case and control 2016). Although an imperfect proxy, colonocyte-microbi-
samples collected in different individuals. ome co-culture experiments identify thousands of genes dif-
One such study used biopsies taken from the terminal ferentially regulated in the presence of a microbiome com-
ileum and sigmoidal colon from healthy individuals and pared to control. The differentially expressed genes in the
those with IBD (inflamed and healthy intestinal tissue) co-culture model are significantly enriched for genes identi-
to study IBD pathophysiology in humans (Häsler et al. fied as differentially expressed in murine colonic epithelial
2017). Specifically, when comparing inflamed and non- cells from conventional vs. germ-free mice, demonstrating
inflamed tissues in the same individual, 13 pathways were the biological utility of the model.
differentially expressed in host tissue, including repres-
sion of citric acid cycle, bile acid synthesis and fatty acid
oxidation, and induction of tryptophan, glycine, serine, The microbiome and host gene expression:
alanine and threonine metabolism. Additionally, ~ 90% of a two‑way conversation
the genes were differentially expressed between healthy
and IBD individuals correlated with microbial taxa present It is clear from the many studies in model organisms and
in the healthy state and had almost no correlation to the humans that there is an association between the gut micro-
microbial taxa present in the IBD gut. This “uncoupling” biome and gene expression in the host. In many cases, it is
of mucosal gene regulation is hypothesized to be an impor- unclear what the direction of causality is with these associa-
tant component of the environmental-host axis underlying tions, however. Does a change in microbiome composition
IBD etiology. cause changes in host gene expression, or does a change
A second study, also examining IBD progression used in host gene expression change microbiome composition
biopsies taken from patients during routine colonosco- (Fig. 3)? Deciphering this relationship is important for
pies with both normal and inflamed intestinal tissue (4–14 understanding disease etiology and ultimately designing
biopsies per patient) coupled with stool and blood sam- therapeutics that target the microbiome.
ples (Lloyd-Price et al. 2019). Interestingly, antimicrobial Comparing microbiome-containing to germ-free system
genes, including CXCL6, LCN2, DUOX2 and SAA2, showed is one way to assess whether the microbiome plays a causa-
increased expression in patients with IBD compared with tive role in regulating gene expression. This type of approach
controls. A separate study also saw increases in antimicro- can be done by comparing either germ-free to conventional
bial host genes in IBD [specifically those with ulcerative organisms (Bäckhed et al. 2004; Fu et al. 2017; Larsson et al.
colitis (UC)] patients when compared to control (Bennet 2012; Sayin et al. 2013; Selwyn et al. 2016) or by comparing
et al. 2018). The increase of antimicrobial genes in IBD cell cultures co-cultured with a microbiome or un-inocu-
patients could be one of the many reasons for the micro- lated media (Richards et al. 2016). For example, expres-
biome dysbiosis observed in the disease, due to the direct sion of intestinal Cytochrome P450 (Cyp) 3a sub-family and
impact host antimicrobial genes would have on the members transporter genes are significantly decreased in germ-free
of the gut microbiome. This might also suggest that host compared to conventionalized mice (Fu et al. 2017). Lower
gene expression plays a larger role in altering microbiome expression of these genes drastically reduces the detoxifica-
composition than the microbiome plays on altering host gene tion capability of the host. The decrease in expression of
expression in cases of IBD. However, during outbreaks of these genes occurs in the germ-free host, demonstrating the
pouchitis in patients who underwent ileal pouch-anal anas- causal role the microbiome plays in regulating detoxification
tomosis as a treatment for IBD, the host transcriptome of pathways in the host.
ileal tissue drastically changed within the ileal pouch, while In population samples, where germ-free conditions are
the taxonomic makeup of the microbiome remained stable unavailable, Mendelian Randomization (MR) offers a tool
between the ileal pouch and pre-pouch ileum (Morgan et al. to assess whether the microbiome leads to a phenotype in
2015). This suggests that microbiome can act independently the host or vice versa (Wade and Hall 2020). For example,

13
Human Genetics (2021) 140:747–760 753

Fig. 3  Direction of causality between microbiome and host gene or vice versa. b Here, differential expression of a host gene expres-
expression associations. a When an association between the micro- sion leads to changes in the microbiome. c Here, changes in the
biome and host gene expression is identified, an open question is microbiome cause a change in host gene expression
whether the microbiome is leading to the changes in gene expression

MR analysis revealed that the gut microbiome casually these studies demonstrate that host genetics, likely via gene
effects metabolic traits related to type 2 diabetes and obesity regulation, modulates aspects of the gut microbiome.
(Sanna et al. 2019). Although not applied to studies of gene
expression and the microbiome yet, this would be a power-
ful framework to examine direction of causality between the Molecular mechanisms linking
host transcriptome and microbiome. the microbiome to host gene expression
Conversely, genome-wide association studies (GWAS) of
the gut microbiota provide evidence that gene expression Many molecular mechanisms foster the cross-talk between
likely influences the abundance of certain bacteria in the gut the microbiome and host gene expression (Fig. 4). Transcrip-
(Blekhman et al. 2015; Bonder et al. 2016; Davenport 2016; tion factors are host proteins that bind to DNA and regu-
Goodrich et al. 2016; Goodrich et al. 2017; Turpin et al. late the transcription of genes. Elements of the microbiome
2016; Wang et al. 2017). In particular, taxa such as Chris- bind directly to transcription factors (Davison et al. 2017;
tensenellaceae, Akkermansia, and Bifidobacterium are either Krautkramer et al. 2017). In zebrafish, the microbiome sup-
heritable or associated with genetic variation in the human presses the transcription factor hepatocyte nuclear factor
genome. As it is expected that an individual’s genome 4A (HNF4α), preventing the regulation of host inflamma-
sequence will not change in the presence of the microbiome, tory pathways, potentially leading to an inflammatory state

13
754 Human Genetics (2021) 140:747–760

Fig. 4  Potential mechanisms underlying host gene expression– tin availability. These changes lead to differential expression of host
microbiome associations Interactions between the microbiome and genes. c Microbiome-derived molecules can also lead to changes in
microbiome-derived molecules and host can occur in either cellular transcription and alternative splicing, resulting in different host gene
or extra-cellular compartments and involve a variety of processes. a products. Both b and c result in differential protein expression, that
Microbiome derived stimuli are recognized by host cells, either via are then released back into the gut lumen. d Host-derived proteins
interaction with extracellular receptors or entering the cell. Stimuli can also modulate the bacteria present in the gut microbiome, causing
include microbiome-derived transcripts, small molecules, proteins differential microbial growth and transcription. Microbiome-derived
and enzymes, or pH changes. b Microbiome-derived molecules cause molecules illustrated in blue. Host-derived molecules illustrated in
differential transcription factor binding, methylation, or chroma- orange

(Davison et al. 2017). Studies on the skin microbiome of have major effects on host histone post-translational modi-
mice showed that colonization of the skin microbiome regu- fications, which then could change host gene expression for
lates the expression of several key transcription factors (Klf4, a variety of genes.
AP-1 and SP-1), albeit via an unknown mechanism (Meisel The microbiome also remodels the chromatin in host
et al. 2018). intestinal epithelial cells (Fig. 4b), albeit with conflicting
The human microbiome also affects epigenetic modifica- evidence. ATAC-seq in a colonic epithelial cell co-culture
tions like DNA methylation and histone acetylation (Krau- model demonstrates that specific microbes regulate chro-
tkramer et al. 2017; Yu et al. 2015) (Fig. 4b). For example, matin accessibility and transcription factor binding in host
germ-free mice have lower genome-wide DNA methylation tissues (Richards et al. 2019). In mouse models, the presence
in colonic tissue compared to conventionally raised animals of a microbiome results in more highly accessible chromatin
(Yu et al. 2015). A fecal transplant, however, drastically in intestinal epithelial cells as compared to a gnotobiotic
increases global DNA methylation in previously germ-free mouse (Semenkovich et al. 2016). The authors even specu-
mice. Additionally, short-chain fatty acids (SCFAs) influ- late that the intestinal epithelium could have evolved to have
ence histone acetylation (Krautkramer et al. 2017). Even a a chromatin structure that requires a microbiome to activate
small change in diet can cause major shifts in SCFAs and appropriately. However, a separate study showed there are
other bacterial metabolite levels in the host (David et al. not significant changes in chromatin accessibility in intes-
2014). Taken together, small dietary changes can potentially tinal epithelial cells between wild-type mice and germ-free

13
Human Genetics (2021) 140:747–760 755

mice (Camp et al. 2014). Instead microbial regulation of host of the gut microbiome changes, in addition to shifts in
gene expression most likely came from different transcrip- taxonomy (Blakeley-Ruiz et al. 2019; Schirmer et al.
tion factor binding regions in the available chromatin. 2019). Bacterial metagenomics, metatranscriptomics,
Conversely, in a diet-induced obesity mouse experiment, and metaproteomics can be used to investigate the func-
the microbiome remodels chromatin in colonic cells (Qin tional potential of the gut microbiome. Future studies that
et al. 2018). This results in an increase of accessible HNF4α combine metatranscriptomics, metaproteomics, and 16S
binding sites, subsequently leading to downregulation of sequencing with metabolomics and RNAseq of host tissue
genes near those sites. One important gene that gains an may be able to reveal more insights about how microbiome
HNF4α-binding site and is downregulated is Scd1 (Qin et al. functions interact with host gene expression.
2018). Scd1 is an important regulator of the amount of free As briefly mentioned above, scRNA-seq of both the
floating fatty acids and is responsible for combining them microbiome and host tissue has the potential to reveal how
into triglyceride storage (Miyazaki et al. 2000). Downregu- the microbiome associates with cell-type composition and
lation of Scd1 has been seen in patients with nonalcoholic cell-specific gene expression. For example, using scRNA-
fatty liver disease (NAFLD) (Gornicka et al. 2011). Taken seq, the transcriptome of only the first layer of intestinal
together it is possible that the microbiome in an obese state epithelial cells that have direct contact with the microbiome
indirectly promotes the formation of NAFLD through modu- could be assayed. This would limit the noise in gene expres-
lation the structure of the chromatin in colonic cells, result- sion measurements collected from multiple cell types and
ing in more binding sites for HNF4α. instead focus analysis on only the interactions happening
Finally, the gut microbiome also has a role in the alterna- directly at the interface of the host–gut axis.
tive splicing of host genes (Martínez-Montiel et al. 2016) Additionally, organoid systems can be utilized to create
(Fig. 4c). Products from Pseudomonas sp (FR901464) and a better in vitro model of the human gut. Organoid systems
Streptomyces sp (pladienolide) inhibit host splicing machin- for human colon and small intestinal tissue contain crypts
ery (Fan et al. 2011). Additionally, 320 differential splicing and villi, which more accurately model an intestinal system
events in intestinal tissue occurred between patients with versus cells adherent to a dish (Sato et al. 2011). Recently,
IBD and healthy controls, after controlling for tissue type, intestinal organoids were used to model Clostridium diffi-
inflammation status, and diagnosis. (Häsler et al. 2017). The cile infections (CDI) (Leslie et al. 2015). The CDI organoid
alternative splicing events upregulated in IBD patients were model more closely represented in vivo CDI when compared
mapped to the KEGG pathways for ‘bacterial invasion of to CDI modeled with a cell culture (Leslie et al. 2015). The
epithelial cells’, ‘pathogenic E. coli infection’ and ‘allograft next step for these types of studies would be to move on to
rejection’. It should be noted that there was only a weak organoid studies and co-culture organoids with microbiome
correlation between host gene expression and the shared isolates. Recent methods involving microinjection could
alternative splicing events. Research into the relationship be employed to recapitulate the gut microbiome inside the
between alternative splicing and the microbiome remains organoids to provide researchers with a better model to study
scarce and should be considered in future studies, consider- host–gut interactions (Williamson et al. 2018). Additionally,
ing the major role splicing plays in complex disease etiology organoids created from patients with Crohn’s disease and
(Li et al. 2016). healthy controls showed that the ex vivo organoid model
could recapitulate the DNA methylation profiles seen in both
host tissue (Howell et al. 2018). Organoid models have also
Future directions been used to investigate intestinal cytokine secretion, inde-
pendent of host immune functionality, to show that cytokine
The complexity of the interactions between the microbes secretion is dependent on differentiation state (Lyons et al.
in the gut and the host is immense. Although recent studies 2018). Therefore, while we cannot make germ-free humans
in model organisms and humans demonstrate that trans- or knockout humans (like we can with mice), researchers
kingdom crosstalk occurs, there are still many avenues can utilize 3D organoids with and without the microbiome
to explore to gain further insight into host–microbiome instead.
interactions. Many existing studies identify individual Finally, novel computational methods need to be devel-
bacteria associated with host gene expression. While this oped that integrate multiple highly complex datasets together
can occur, other effects are likely driven by a variety of to reveal biological insight, such as gene expression and tax-
taxa, due to functional redundancy. There are thousands onomic composition. Each type of high dimensional data
of bacterial species present in the gut microbiome. Many has its own quirks, such as sparsity, compositionality, or
different species share genes and pathways to produce the overdispersion (Anders et al. 2010; Tsilimigras and Fodor
same metabolites (Moya and Ferrer 2016). Therefore, it 2016). Designing tools that specifically take these issues
is important to understand how the functional capability into account will allow researchers to identify novel and

13
756 Human Genetics (2021) 140:747–760

complex associations between the gut microbiome and host Reyes A, Shannon P, Smyth GK, Tenenbaum D, Waldron L,
gene expression. Morgan M, Liao Y, Smyth GK, Shi W, Love MI, Huber W,
Anders S, Narayana A, Mathew M, Tam M, Kannan R, Madden
The study of the relationship between the microbiome KM, Golfinos JG, Parker EC, Ott PA, Pavlick AC, Saeidipour B,
and host gene expression has already yielded great insights Bakhshi S (2010) Differential expression analysis for sequence
into the physiological mechanisms that link a host with the count data via mixtures of negative binomials. Adv Environ Biol
microbes that live within it. With the exponential growth of 7:2803–2809
Bäckhed F, Ding H, Wang T, Hooper LV, Gou YK, Nagy A, Semen-
sequencing techniques and technologies, new cell culture kovich CF, Gordon JI (2004) The gut microbiota as an environ-
techniques, the growing popularity of scRNA-seq, and the mental factor that regulates fat storage. PNAS 101:15718–15723.
advent of more advanced statistical methods, we are cur- https​://doi.org/10.1073/pnas.04070​76101​
rently poised to gain even greater insight into the trans-king- Bäckhed F, Fraser CM, Ringel Y, Sanders ME, Sartor RB, Sherman
PM, Versalovic J, Young V, Finlay BB (2012) Defining a healthy
dom relationship we have with our microbiome. human gut microbiome: Current concepts, future directions, and
clinical applications. Cell Host Microbe 12:611–622. https​://doi.
org/10.1016/j.chom.2012.10.012
Author contributions Performed literature search (RGN, ERD), wrote Bäckhed F, Roswall J, Peng Y, Feng Q, Jia H, Kovatcheva-Datchary P,
manuscript (RGN), revised manuscript (ERD). Li Y, Xia Y, Xie H, Zhong H, Khan MT, Zhang J, Li J, Xiao L,
Al-Aama J, Zhang D, Lee YS, Kotowska D, Colding C, Tremaroli
Funding The Pennsylvania State University. V, Yin Y, Bergman S, Xu X, Madsen L, Kristiansen K, Dahl-
gren J, Jun W (2015) Dynamics and stabilization of the human
gut microbiome during the first year of life. Cell Host Microbe
Compliance with ethical standards 17:690–703. https​://doi.org/10.1016/j.chom.2015.04.004
Bennet SMP, Sundin J, Magnusson MK, Strid H, Tap J, Derrien M,
Conflict of interest None to declare. Le Nevé B, Doré J, Törnblom H, Simrén M, Öhman L (2018)
Altered intestinal antibacterial gene expression response profile
Open Access This article is licensed under a Creative Commons Attri- in irritable bowel syndrome is linked to bacterial composition
bution 4.0 International License, which permits use, sharing, adapta- and immune activation. Neurogastroenterol Motil 30:1–15. https​
tion, distribution and reproduction in any medium or format, as long ://doi.org/10.1111/nmo.13468​
as you give appropriate credit to the original author(s) and the source, Blakeley-Ruiz JA, Erickson AR, Cantarel BL, Xiong W, Adams R,
provide a link to the Creative Commons licence, and indicate if changes Jansson JK, Fraser CM, Hettich RL (2019) Metaproteomics
were made. The images or other third party material in this article are reveals persistent and phylum-redundant metabolic functional
included in the article’s Creative Commons licence, unless indicated stability in adult human gut microbiomes of Crohn’s remission
otherwise in a credit line to the material. If material is not included in patients despite temporal variations in microbial taxa, genomes,
the article’s Creative Commons licence and your intended use is not and proteomes. Microbiome 7:1–15. https​://doi.org/10.1186/
permitted by statutory regulation or exceeds the permitted use, you will s4016​8-019-0631-8
need to obtain permission directly from the copyright holder. To view a Blekhman R, Goodrich JK, Huang K, Sun Q, Bukowski R, Bell JT,
copy of this licence, visit http://creat​iveco​mmons​.org/licen​ses/by/4.0/. Spector TD, Keinan A, Ley RE, Gevers D, Clark AG (2015) Host
genetic variation impacts microbiome composition across human
body sites. Genome Biol 16:1–12. https​://doi.org/10.1186/s1305​
9-015-0759-1
References Bonder MJ, Kurilshikov A, Tigchelaar EF, Mujagic Z, Imhann F, Vila
AV, Deelen P, Vatanen T, Schirmer M, Smeekens SP, Zherna-
Albertsen M, Hugenholtz P, Skarshewsk A (2013) Genome sequences kova DV, Jankipersadsing SA, Jaeger M, Oosting M, Cenit MC,
of rare, uncultured bacteria obtained by differential coverage bin- Masclee AAM, Swertz MA, Li Y, Kumar V, Joosten L, Harmsen
ning of multiple metagenomes. Nat Biotechnol 31 H, Weersma RK, Franke L, Hofker MH, Xavier RJ, Jonkers D,
Aleksandrova K, Romero-Mosquera B, Hernandez V (2017) Diet, gut Netea MG, Wijmenga C, Fu J, Zhernakova A (2016) The effect of
microbiome and epigenetics: Emerging links with inflammatory host genetics on the gut microbiome. Nat Genet 48:1407–1412.
bowel diseases and prospects for management and prevention. https​://doi.org/10.1038/ng.3663
Nutrients 9:1–13. https​://doi.org/10.3390/nu909​0962 Bost P, Giladi A, Liu Y, Bendjelal Y, Xu G, David E, Blecher-Gonen
Alenghant T, Osborne L, Saenz S, Kobuley D, Ziegler C, Mullican R, Cohen M, Medaglia C, Li H, Deczkowska A, Zhang S,
SE, Choi I, Grunberg S, Sinha R, Wynosky-Dolfi M, Snyder A, Schwikowski B, Zhang Z, Amit I (2020) Host-viral infection
Gaicomin P, Joyce K, Hoang T, Bewtra M, Brodsky I, Sonnen- maps reveal signatures of severe COVID-19 patients. Cell. https​
berg G, Bushman F, Won K-J, Lazar M, Artis D (2013) Histone ://doi.org/10.1016/j.cell.2020.05.006
Deacetylase 3 orchestrates commensal bacteria- dependent intes- Broderick NA, Buchon N, Lemaitre B (2014) Microbiota-induced
tinal homeostasis. Nature 504:153–157. https​://doi.org/10.1038/ changes in drosophila melanogaster host gene expression and
natur​e1268​7.Histo​ne gut morphology. mBio 5:1–13
Allen EE, Banfield JF (2005) Community genomics in microbial ecol- Butler A, Hoffman P, Smibert P, Papalexi E, Satija R (2017) Integrat-
ogy and evolution. Nat Rev Microbiol 3:489–498 ing single-cell transcriptomic data across different conditions,
Anders S, Huber W, Dobin A, Davis CA, Schlesinger F, Drenkow J, technologies, and species. Physiol Behav 176:139–148
Zaleski C, Jha S, Batut P, Chaisson M, Gingeras TR, Gu Z, Eils Calgaro M, Romualdi C, Waldron L, Risso D, Vitulo N (2020) Assess-
R, Schlesner M, Huber W, Carey VJ, Gentleman R, Anders S, ment of statistical methods from single cell, bulk RNA-seq,
Carlson M, Carvalho BS, Bravo HC, Davis S, Gatto L, Girke and metagenomics applied to microbiome data. Genome Biol
T, Gottardo R, Hahne F, Hansen KD, Irizarry RA, Lawrence 21:1–31
M, Love MI, MaCdonald J, Obenchain V, Oles̈ AK, Pagès H, Camp JG, Frank CL, Lickwar CR, Guturu H, Rube T, Wenger AM,
Chen J, Bejerano G, Crawford GE, Rawls JF (2014) Microbiota

13
Human Genetics (2021) 140:747–760 757

modulate transcription in the intestinal epithelium without the microbiome Julia. Science 352:532–535. https ​ : //doi.
remodeling the accessible chromatin landscape. Genome Res org/10.1016/j.physb​eh.2017.03.040
24:1504–1516. https​://doi.org/10.1101/gr.16584​5.113 Goodrich JK, Davenport ER, Clark AG, Ley RE (2017) The relation-
Danese S, Sans M, de la Motte C, Graziani C, West G, Phillips MH, ship between the human genome and microbiome comes into
Pola R, Rutella S, Willis J, Gasbarrini A, Fiocchi C (2006) Angi- view. Annu Rev Genet 51:413–433. https​://doi.org/10.1146/
ogenesis as a novel component of inflammatory bowel disease annur​ev-genet​-11071​1-15553​2
pathogenesis. Gastroenterology 130:2060–2073. https​://doi. Gornicka A, Morris-Stiff G, Thapaliya S, Papouchado BG, Berk M,
org/10.1053/j.gastr​o.2006.03.054 Feldstein AE (2011) Transcriptional profile of genes involved
Davenport ER (2016) Elucidating the role of the host genome in shap- in oxidative stress and antioxidant defense in a dietary murine
ing microbiome composition. Gut Microbes 7:178–184. https​:// model of steatohepatitis. Antioxid Redox Signal 15:437–445.
doi.org/10.1080/19490​976.2016.11550​22 https​://doi.org/10.1089/ars.2010.3815
David LA, Maurice CF, Carmody RN, Gootenberg DB, Button JE, Grice E, Segre J (2012) The human microbiome: our second genome.
Wolfe BE, Ling AV, Devlin AS, Varma Y, Fischbach MA, Annu Rev Genomics Hum Genet 13:151–170. https​: //doi.
Biddinger SB, Dutton RJ, Turnbaugh PJ (2014) Diet rapidly org/10.1146/annur​ev-genom​-09071​1-16381​4.The
and reproducibly alters the human gut microbiome. Nature Gury-BenAri M, Thaiss CA, Serafini N, Winter DR, Giladi A, Lara-
505:559–563. https​://doi.org/10.1038/natur​e1282​0 Astiaso D, Levy M, Salame TM, Weiner A, David E, Shapiro H,
Davidson RM, Epperson LE (2018) Microbiome sequencing methods Dori-Bachash M, Pevsner-Fischer M, Lorenzo-Vivas E, Keren-
for studying human diseases Rebecca. Dis Gene Identification: Shaul H, Paul F, Harmelin A, Eberl G, Itzkovitz S, Tanay A, Di
Methods Protocols 1706:77–90 Santo JP, Elinav E, Amit I (2016) The spectrum and regulatory
Davison JM, Lickwar CR, Song L, Breton G, Crawford GE, Rawls landscape of intestinal innate lymphoid cells are shaped by the
JF (2017) Microbiota regulate intestinal epithelial gene microbiome. Cell 166:1231-1246.e13. https​://doi.org/10.1016/j.
expression by suppressing the transcription factor Hepatocyte cell.2016.07.043
nuclear factor 4 alpha. Genome Res 27:1195–1206. https​://doi. Häsler R, Sheibani-Tezerji R, Sinha A, Barann M, Rehman A, Esser D,
org/10.1101/gr.22011​1.116 Aden K, Knecht C, Brandt B, Nikolaus S, Schäuble S, Kaleta C,
de la Cuesta-Zuluaga J, Kelley ST, Chen Y, Escobar JS, Mueller Franke A, Fretter C, Müller W, Hütt MT, Krawczak M, Schreiber
NT, Ley RE, McDonald D, Huang S, Swafford AD, Knight R, S, Rosenstiel P (2017) Uncoupling of mucosal gene regulation,
Thackray VG (2019) Age- and sex-dependent patterns of gut mRNA splicing and adherent microbiota signatures in inflamma-
microbial diversity in human adults. mSystems 4:1–12. https​ tory bowel disease. Gut 66:2087–2097. https​://doi.org/10.1136/
://doi.org/10.1128/msyst​ems.00261​-19 gutjn​l-2016-31165​1
Dirksen P, Assié A, Zimmermann J, Zhang F, Tietje A-M, Marsh Hausmann A, Hardt W-D (2019) The interplay between salmonella
SA, Félix M-A, Shapira M, Kaleta C, Schulenburg H, Samuel enterica serovar typhimurium and the intestinal mucosa dur-
B (2020) CeMbio—the Caenorhabditis elegans microbiome ing oral infection. Microbiol Spectrum 7:1–16. https​://doi.
resource. bioRxiv, pp 1–48 org/10.1128/micro​biols​pec.bai-0004-2019
Dominguez-Bello MG, Costello EK, Contreras M, Magris M, Honda K, Littman DR (2016) The microbiota in adaptive immune
Hidalgo G, Fierer N, Knight R (2010) Delivery mode shapes homeostasis and disease. Nature 535:75–84. https ​ : //doi.
the acquisition and structure of the initial microbiota across org/10.1038/natur​e1884​8
multiple body habitats in newborns. PNAS 107:11971–11975 Hooper LV, Midtvedt T, Gordon JI (2002) How host-microbial interac-
Donohoe DR, Garge N, Zhang X, Sun W, O’Connell TM, Bunger tions shape the nutrient environment of the mammalian intestine.
MK, Bultman SJ (2011) The microbiome and butyrate regu- Annu Rev Nutr 22:283–307
late energy metabolism and autophagy in the mammalian Howell KJ, Kraiczy J, Nayak KM, Gasparetto M, Ross A, Lee C,
colon. Cell Metab 13:517–526. https​: //doi.org/10.1016/j. Mak TN, Koo BK, Kumar N, Lawley T, Sinha A, Rosenstiel
cmet.2011.02.018 P, Heuschkel R, Stegle O, Zilbauer M (2018) DNA methyla-
Douglas A (2018) The Drosophila model for microbiome research. tion and transcription patterns in intestinal epithelial cells from
Lab Anim 47:157–164 pediatric patients with inflammatory bowel diseases differentiate
Drayman N, Patel P, Vistain L (2019) HSV-1 single-cell analysis disease subtypes and associate with outcome. Gastroenterology
reveals the activation of anti-viral and developmental programs 154:585–598. https​://doi.org/10.1053/j.gastr​o.2017.10.007
in distinct sub-populations. eLife 8:1–25. https:​ //doi.org/10.7554/ Huang J, Kelly C, Bakirtzi K, Villafuerte J, Lyras D, Mileto S, Lar-
eLife​.46339​ combe S, Xu H, Yang X, Shields K, Zhu W, Zhang Y, Gold-
Elya C, Zhang V, Ludington WB, Eisen MB (2016) Stable host gene smith J, Patel I, Hansen J, Huang M, Yla-Herttuala S, Moss A,
expression in the gut of adult Drosophila melanogaster with dif- Paredes-Sabja D, Pothoulakis C, Shah Y, Wang J, Chen X (2019)
ferent bacterial mono-associations. PLoS ONE 11:1–23 Clostridium difficile toxins induce VEGF-A and vascular perme-
Fan L, Lagisettia C, Edwards C, Webb T, Potter P (2011) Sudemycins, ability to promote disease pathogenesis. Nat Microbiol 4:269–
novel small molecule analogues of FR901464, induce alternative 279. https:​ //doi.org/10.1097/CCM.0b013e​ 31823​ da96d​ .Hydrog​ en
gene splicing Liying. ACS Chem Biol 6:582–589. https​://doi. Hui L, Zang K, Wang M, Shang F, Zhang G (2018) Coculture with
org/10.1038/jid.2014.371 clostridium difficile promotes apoptosis of human intestinal
Franzosa EA, McIver LJ, Rahnavard G, Thompson LR, Schirmer M, microvascular endothelial cells. J Int Med Res 46:4631–4739.
Weingart G, Lipson KS, Knight R, Caporaso JG, Segata N, Hut- https​://doi.org/10.1177/03000​60518​79926​7
tenhower C (2018) Species-level functional profiling of metage- Kamada N, Seo S-U, Chen GY, Núñez G (2013) Role of the gut micro-
nomes and metatranscriptomes. Nat Methods 15:962–968. https​ biota in immunity and inflammatory disease. Nat Rev Immunol
://doi.org/10.1038/s4159​2-018-0176-y 13:321–335. https​://doi.org/10.1038/nri34​30
Fu ZD, Selwyn FP, Cui JY, Klaassen CD (2017) RNA-seq profil- Kharchenko PV, Silberstein L, Scadden DT (2014) Bayesian approach
ing of intestinal expression of xenobiotic processing genes in to single-cell differential expression analysis. Nat Methods
germ-free mice. Drug Metab Dispos 45:1225–1238. https​://doi. 11:740–742
org/10.1124/dmd.117.07731​3 Kim HJ, Li H, Collins JJ, Ingber DE (2016) Contributions of micro-
Goodrich J, Davenport E, Waters J, Clark A, Ley R (2016) Cross- biome and mechanical deformation to intestinal bacterial
species comparisons of host genetic associations with

13
758 Human Genetics (2021) 140:747–760

overgrowth and inflammation in a human gut-on-a-chip. Proc modulate gene expression in the skin. Microbiome 6:1–15. https​
Natl Acad Sci USA 113:E7–E15 ://doi.org/10.1186/s4016​8-018-0404-9
Krautkramer KA, Kreznar JH, Romano KA, Vivas EI, Barrett-wilt GA, Miyakawa M, Gordon HA, Wostmann BS (1971) The gnotobiotic ani-
Rabaglia ME, Keller MP, Attie AD, Federico E, Denu JM (2017) mal as a tool in the study of inflammation. Science 173:171–173.
Diet-microbiota interactions mediate global epigenetic program- https​://doi.org/10.1126/scien​ce.173.3992.171
ming in multiple host tissues. Mol Cell 64:982–992. https​://doi. Miyazaki M, Kim YC, Gray-Keller MP, Attie AD, Ntambi JM (2000)
org/10.1016/j.molce​l.2016.10.025.Diet-micro​biota​ The biosynthesis of hepatic cholesterol esters and triglycerides
Larsson E, Tremaroli V, Lee YS, Koren O, Nookaew I, Fricker A, is impaired in mice with a disruption of the gene for stearoyl-
Nielsen J, Ley RE, Bäckhed F (2012) Analysis of gut microbial CoA desaturase 1. J Biol Chem 275:30132–30138. https​://doi.
regulation of host gene expression along the length of the gut org/10.1074/jbc.M0054​88200​
and regulation of gut microbial ecology through MyD88. Gut Molly K, Vande Woestyne M, Verstraete W (1993) Development of
61:1124–1131. https​://doi.org/10.1136/gutjn​l-2011-30110​4 a 5-step multi-chamber reactor as a simulation of the human
Leclercq S, Matamoros S, Cani PD, Neyrinck AM, Jamar F, Starkel intestinal microbial ecosystem. Appl Microbiol Biotechnol
P, Windey K, Tremaroli V, Backhed F, Verbeke K, de Timary 39:254–258. https​://doi.org/10.1007/BF002​28615​
P, Delzenne NM (2014) Intestinal permeability, gut-bacterial Montoya-Williams D, Lemas DJ, Spiryda L, Patel K, Carney ONO,
dysbiosis, and behavioral markers of alcohol-dependence sever- Neu J, Carson TL (2018) The neonatal microbiome and its partial
ity. Proc Natl Acad Sci U S A 111:E4485–E4493. https​://doi. role in mediating the association between birth by cesarean sec-
org/10.1073/pnas.14151​74111​ tion and adverse pediatric outcomes. Neonatology 114:103–111.
Leslie JL, Huang S, Opp JS, Nagy MS, Kobayashi M, Young VB, https​://doi.org/10.1159/00048​7102
Spence JR (2015) Persistence and toxin production by Clostrid- Morgan XC, Kabakchiev B, Waldron L, Tyler AD, Tickle TL, Milgrom
ium difficile within human intestinal organoids result in disrup- R, Stempak JM, Gevers D, Xavier RJ, Silverberg MS, Hutten-
tion of epithelial paracellular barrier function. Infect Immun hower C (2015) Associations between host gene expression, the
83:138–145. https​://doi.org/10.1128/IAI.02561​-14 mucosal microbiome, and clinical outcome in the pelvic pouch of
Li YI, Van De Geijn B, Raj A, Knowles DA, Petti AA, Golan D, Gilad patients with inflammatory bowel disease. Genome Biol 16:67.
Y, Pritchard JK (2016) RNA splicing is a primary link between https​://doi.org/10.1186/s1305​9-015-0637-x
genetic variation and disease. Science 352:600–604 Moya A, Ferrer M (2016) Functional redundancy-induced stability
Liu S, Trapnell C (2016) Single-cell transcriptome sequencing: recent of gut microbiota subjected to disturbance. Trends Microbiol
advances and remaining challenges. F1000Research. https​://doi. 24:402–413. https​://doi.org/10.1016/j.tim.2016.02.002
org/10.12688​/f1000​resea​rch.7223.1 Murdoch CC, Rawls JF (2019) Commensal microbiota regulate verte-
Lloyd-Price J, Arze C, Ananthakrishnan AN, Schirmer M, Avila- brate innate immunity-insights from the zebrafish. Front Immu-
Pacheco J, Poon TW, Andrews E, Ajami NJ, Bonham KS, nol 10:1–14
Brislawn CJ, Casero D, Courtney H, Gonzalez A, Graeber TG, Nelson H, Chia N (2019) Gut microbiome and colon cancer: a plausible
Hall AB, Lake K, Landers CJ, Mallick H, Plichta DR, Prasad explanation for dietary contributions to cancer. J Am Coll Surg
M, Rahnavard G, Sauk J, Shungin D, Vázquez-Baeza Y, White 229:231–235. https:​ //doi.org/10.1016/j.jamcol​ lsurg​ .2019.05.003
RA, Bishai J, Bullock K, Deik A, Dennis C, Kaplan JL, Khalili Nelson K, Weinstock G, Highlander S, Worley K, Creasy H, Wortman
H, McIver LJ, Moran CJ, Nguyen L, Pierce KA, Schwager R, J, Rusch D, Mitreva M, Sodergren E, Chinwalla A, Feldgarden
Sirota-Madi A, Stevens BW, Tan W, ten Hoeve JJ, Weingart G, M, Gevers D, Haas B, Madupu R, Ward D, Birren B, Gibbs R,
Wilson RG, Yajnik V, Braun J, Denson LA, Jansson JK, Knight Methe B, Petrosino J, Strausberg R, Sutton G, White O (2010) A
R, Kugathasan S, McGovern DPB, Petrosino JF, Stappenbeck catalog of reference genomes from the human microbiome. Sci-
TS, Winter HS, Clish CB, Franzosa EA, Vlamakis H, Xavier RJ, ence 328:994–999. https​://doi.org/10.1126/scien​ce.11836​05.A
Huttenhower C (2019) Multi-omics of the gut microbial ecosys- Nielsen B, Almeida M, Juncker A, Rasmussen S, Li J, Sunagawa S,
tem in inflammatory bowel diseases. Nature 569:655–662. https​ Plichta D, Gautier L, Pedersen A, Chatelier E, Pelletier E, Bonde
://doi.org/10.1038/s4158​6-019-1237-9 I, Nielsen T, Manichanah C, Arumugam M, Batto J-M, Quinta-
Lukovac S, Belzer C, Pellis L, Keijser BJ, de Vos WM, Montijn RC, nilha dos Santos M, Blom N, Borruel N, Burgdof K, Boumez-
Roeselers G (2014) Differential modulation by akkermansia beur F, Casellas F, Dore J, Dworzynski P, Guarner F, Hansen T,
muciniphila and faecalibacterium prausnitzii of host peripheral Hildebrand F, Kass R, Kennedy S, Kristiansen K, Kultima JR,
lipid metabolism and histone acetylation in mouse gut organoids. Leonard P, Levenez F, Lund O, Moumen B, Paslier D, Pons N,
mBio 5:1–10. https​://doi.org/10.1128/mBio.01438​-14 Pederson O, Prifti E, Qin J, Raes J, Sorensen S, Tap J, Tims S,
Lyons J, Ghazi PC, Starchenko A, Tovaglieri A, Baldwin KR, Poulin Ussey D, Yamada T, Consortium M, Renault P, Sicheritz-Ponten
EJ, Gierut JJ, Genetti C, Yajnik V, Breault DT, Lauffenburger T, Bork P, Wang J, Brunak S, Ehrlich D (2014) Identification and
DA, Haigis KM (2018) The colonic epithelium plays an active assembly of genomes and genetic elements in complex metagen-
role in promoting colitis by shaping the tissue cytokine profile. omic samples without using reference genomes. Nat Biotechnol
PLoS Biol 16:1–24. https​://doi.org/10.1371/journ​al.pbio.20024​ 32:822–828
17 Papathoma E, Triga M, Fouzas S, Dimitriou G (2016) Cesarean section
Martínez-Montiel N, Rosas-Murrieta NH, Martínez-Montiel M, Gas- delivery and development of food allergy and atopic dermatitis
pariano-Cholula MP, Martínez-Contreras RD (2016) Micro- in early childhood. Pediatr Allergy Immunol 27:419–424. https​
bial and natural metabolites that inhibit splicing: a powerful ://doi.org/10.1111/pai.12552​
alternative for cancer treatment. Biomed Res Int. https​://doi. Parks DH, Rinke C, Chuvochina M, Chaumeil PA, Woodcroft BJ,
org/10.1155/2016/36810​94 Evans PN, Hugenholtz P, Tyson GW (2017) Recovery of nearly
Maudet C, Mano M, Eulalio A (2014) MicroRNAs in the interaction 8000 metagenome-assembled genomes substantially expands the
between host and bacterial pathogens. FEBS Lett 588:4140– tree of life. Nat Microbiol 2:1533–1542. https​://doi.org/10.1038/
4147. https​://doi.org/10.1016/j.febsl​et.2014.08.002 s4156​4-017-0012-7
Meisel JS, Sfyroera G, Bartow-McKenney C, Gimblet C, Bugayev J, Pott J, Hornef M (2012) Innate immune signalling at the intestinal
Horwinski J, Kim B, Brestoff JR, Tyldsley AS, Zheng Q, Hod- epithelium in homeostasis and disease. EMBO Rep 13:684–698.
kinson BP, Artis D, Grice EA (2018) Commensal microbiota https​://doi.org/10.1038/embor​.2012.96

13
Human Genetics (2021) 140:747–760 759

Qin Y, Roberts JD, Grimm SA, Lih FB, Deterding LJ, Li R, Chrysover- the control of major cytokines by the let-7 family. EMBO J
gis K, Wade PA (2018) An obesity-associated gut microbiome 30:1977–1989. https​://doi.org/10.1038/emboj​.2011.94
reprograms the intestinal epigenome and leads to altered colonic Selwyn FP, Cheng SL, Klaassen CD, Cui JY (2016) Regulation of
gene expression. Genome Biol 19:1–14. https​://doi.org/10.1186/ hepatic drug-metabolizing enzymes in germ-free mice by con-
s1305​9-018-1389-1 ventionalization and probiotics. Drug Metab Dispos 44:262–274.
Reidl S, Lehmann A, Schiller R, Salam Khan A, Dobrindt U (2009) https​://doi.org/10.1124/dmd.115.06750​4
Impact of O-glycosylation on the molecular and cellular adhesion Semenkovich NP, Planer JD, Ahern PP, Griffin NW, Lin CY, Gordon
properties of the Escherichia coli autotransporter protein Ag43. JI (2016) Impact of the gut microbiota on enhancer accessibil-
Int J Med Microbiol 299:389–401. https​://doi.org/10.1016/j. ity in gut intraepithelial lymphocytes. Proc Natl Acad Sci USA
ijmm.2009.01.001 113:14805–14810. https​://doi.org/10.1073/pnas.16177​93113​
Richards AL, Burns MB, Alazizi A, Barreiro LB, Pique-Regi R, Ble- Shames SR, Finlay BB (2012) Bacterial effector interplay: a new way
khman R, Luca F (2016) Genetic and transcriptional analysis to view effector function. Trends Microbiol 20:214–219. https​://
of human host response to healthy gut microbiota. mSystems doi.org/10.1016/j.tim.2012.02.007
1:1–16. https​://doi.org/10.1128/msyst​ems.00067​-16 Shi N, Li N, Duan X, Niu H (2017) Interaction between the gut micro-
Richards AL, Muehlbauer AL, Alazizi A, Burns MB, Findley A, biome and mucosal immune system. Milit Med Res 4:1–7. https​
Messina F, Gould TJ, Cascardo C, Pique-Regi R, Blekhman R, ://doi.org/10.1186/s4077​9-017-0122-9
Luca F (2019) Gut Microbiota Has a Widespread and Modi- Shnayder M, Nachshon A, Krishna B, Poole E, Boshkov A, Binyamin
fiable Effect on Host Gene Regulation. mSystems. https​://doi. A, Maza I (2018) Defining the transcriptional landscape dur-
org/10.1128/msyst​ems.00323​-18 ing cytomegalovirus latency with single-cell RNA sequencing.
Ridaura VK, Faith JJ, Rey FE, Cheng J, Alexis E, Kau AL, Griffin NW, MBio 9:1–17
Lombard V, Henrissat B, Bain JR, Muehlbauer MJ, Ilkayeva O, Shreiner A, Kao J, Young V (2015) The gut microbiome in health
Semenkovich CF, Funai K, Hayashi DK, Lyle BJ, Martini MC, and in disease. Curr Opin Gastroenterol 31:69–75. https​://doi.
Luke K, Clemente JC, Treuren WV, Wa W (2014) Cultured gut org/10.1097/MOG.00000​00000​00013​9.The
microbiota from twins discordant for obesity modulate adiposity Tang W, Kitai T, Hazen S (2017) Gut microbiota in cardiovascular
and metabolic phenotypes in mice. Science 341:1–22. https:​ //doi. health and disease W.H. Circ Res 120:139–148. https​://doi.
org/10.1126/scien​ce.12412​14.Cultu​red org/10.1016/j.physb​eh.2017.03.040
Rosenow C, Saxena R, Durst M, Gingeras T (2001) Prokaryotic RNA Truong DT, Franzosa EA, Tickle TL, Scholz M, Weingart G, Pasolli
preparation methods useful for high density array analysis: com- E, Tett A, Huttenhower C, Segata N (2015) MetaPhlAn2 for
parison of two approaches. Nucleic Acids Res 29:112e–1112. enhanced metagenomic taxonomic profiling. Nat Methods
https​://doi.org/10.1093/nar/29.22.e112 12:902–903. https​://doi.org/10.1038/nmeth​.3589
Russell SL, Gold MJ, Hartmann M, Willing BP, Thorson L, Wlodarska Tsilimigras MCB, Fodor AA (2016) Compositional data analysis of the
M, Gill N, Blanchet MR, Mohn WW, McNagny KM, Finlay BB microbiome: fundamentals, tools, and challenges. Ann Epidemiol
(2012) Early life antibiotic-driven changes in microbiota enhance 26. Elsevier Inc, pp 330–335
susceptibility to allergic asthma. EMBO Rep 13:440–447. https​ Turnbaugh PJ, Ley RE, Mahowald MA, Magrini V, Mardis ER, Gordon
://doi.org/10.1038/embor​.2012.32 JI (2006) An obesity-associated gut microbiome with increased
Salzberg SL (2018) Open questions: how many genes do we have? capacity for energy harvest. Nature 444:1027–1031. https​://doi.
BMC Biol 16:10–12. https:​ //doi.org/10.1186/s12915​ -018-0564-x org/10.1038/natur​e0541​4
Sanna S, Zuydam N, Mahajan A, Kurilshikov A, Vich Vila A, Vosa U, Turpin W, Espin-Garcia O, Xu W, Silverberg M, Kevans D, Smith M,
Mujagic Z, Masclee A, Jonkers D, Oosting M, Joosten L, Netea Guttman D, Griffiths A, Panaccione R, Otley A, Xu L, Shesto-
M, Franke L, Zhernakova A, Fu J, Wijmenga C, McCarthy M paloff K, Moreno-Hagelsieb G, Consortium GPR, Paterson A,
(2019) Causal relationships between gut microbiome, short-chain Croitoru K (2016) Association of host genome with intestinal
fatty acids and metabolic diseases. Nat Genet 51:549–562. https​ microbial composition in a large healthy cohort. Nat Genet
://doi.org/10.1097/CCM.0b013​e3182​3da96​d.Hydro​gen 48:1413–1417
Sato T, Stange DE, Ferrante M, Vries RGJ, Van Es JH, Van Den Brink Van den Berge KVD, Perraudeau F, Soneson C, Love MI, Risso D,
S, Van Houdt WJ, Pronk A, Van Gorp J, Siersema PD, Clevers H Vert J-p, Robinson MD, Dudoit S, Clement L (2018) Observa-
(2011) Long-term expansion of epithelial organoids from human tion weights unlock bulk RNA-seq tools for zero inflation and
colon, adenoma, adenocarcinoma, and Barrett’s epithelium. Gas- single-cell applications. Genome Biology 1–17
troenterology 141:1762–1772. https​://doi.org/10.1053/j.gastr​ Wade KH, Hall LJ (2020) Improving causality in microbiome research:
o.2011.07.050 Can human genetic epidemiology help? Wellcome Open
Sayin SI, Wahlström A, Felin J, Jäntti S, Marschall HU, Bamberg K, Research 4
Angelin B, Hyötyläinen T, Orešič M, Bäckhed F (2013) Gut Wang J, Thingholm L, Skieceviciene J, Rausch P, Kummen M, Hov J,
microbiota regulates bile acid metabolism by reducing the lev- Degenhardt F, Heinsen F, Ruhlemann M, Szymczak S, Holm K,
els of tauro-beta-muricholic acid, a naturally occurring FXR Esko T, Sun J, Pricop-Jeckstadt M, Al-Dury S, Bohov P, Bethune
antagonist. Cell Metab 17:225–235. https​://doi.org/10.1016/j. J, Sommer F, Ellinghaus D, Berge R, Hubenthal M, Koch M,
cmet.2013.01.003 Schwarz K, Rimbach G, Hubbe P, Oan W, Sheibani-Tezerji R,
Schirmer M, Garner A, Vlamakis H, Xavier RJ (2019) Microbial genes Hasler R, Rosensteil P, D’Amatto M, Cloppenborg-Schmidt K,
and pathways in inflammatory bowel disease. Nat Rev Microbiol Kunzel S, Laudes M, Marschall H, Leib W, Nothlings U, Karles
17:497–511. https​://doi.org/10.1038/s4157​9-019-0213-6 T, Baines J, Franke A (2017) Genome-wide association analysis
Schnitger AKD, Machova A, Mueller RU, Androulidaki A, Schermer identifies variation in vitamin D receptor and other host factors
B, Pasparakis M, Krönke M, Papadopoulou N (2011) Listeria influencing the gut microbiota. Nat Genet 546:651–655. https​://
monocytogenes infection in macrophages induces vacuolar- doi.org/10.1038/natur​e2281​4.Trans​-kingd​om
dependent host miRNA response. PLoS ONE. https​: //doi. Wen L, Ley RE, Volchkov PY, Stranges PB, Avanesyan L, Stone-
org/10.1371/journ​al.pone.00274​35 braker AC, Hu C, Wong FS, Szot GL, Bluestone JA, Gordon JI,
Schulte LN, Eulalio A, Mollenkopf HJ, Reinhardt R, Vogel J (2011) Chervonsky AV (2008) Innate immunity and intestinal micro-
Analysis of the host microRNA response to Salmonella uncovers biota in the development of Type 1 diabetes. Nature 455:1109–
1113. https​://doi.org/10.1016/j.physb​eh.2017.03.040

13
760 Human Genetics (2021) 140:747–760

Westermann AJ, Gorski SA, Vogel J (2012) Dual RNA-seq of path- caenorhabditis elegans to members of its natural microbiota
ogen and host. Nat Rev Microbiol 10:618–630. https​://doi. across development and adult life. Front Microbiol 10:1–16
org/10.1038/nrmic​ro285​2 Yatsunenko T, Rey F, Manary M, Trehan I, Dominguez-Bello M, Con-
Westermann AJ, Förstner KU, Amman F, Barquist L, Chao Y, Schulte treras M, Magris M, Hidalgo G, Baldassano R, Anokhin A, Heath
LN, Müller L, Reinhardt R, Stadler PF, Vogel J (2016) Dual A, Warner B, Reeder J, Kuczynski J, Caporaso J, Lozupone C,
RNA-seq unveils noncoding RNA functions in host-pathogen Lauber C, Clemente J, Knights D, Knight R, Gordon J (2012)
interactions. Nature 529:496–501. https​://doi.org/10.1038/natur​ Human gut microbiome viewed across age and geography. Nature
e1654​7 486:222–227. https​://doi.org/10.1038/natur​e1105​3.Human​
Westreich ST, Treiber ML, Mills DA, Korf I, Lemay DG (2018) Yoon MY, Yoon SS (2018) Disruption of the gut ecosystem by antibiot-
SAMSA2: a standalone metatranscriptome analysis pipe- ics. Yonsei Med J 59:4–12
line. BMC Bioinform 19:175. https​://doi.org/10.1186/s1285​ Yu DH, Gadkari M, Zhou Q, Yu S, Gao N, Guan Y, Schady D, Roshan
9-018-2189-z TN, Chen MH, Laritsky E, Ge Z, Wang H, Chen R, Westwater
Williamson IA, Arnold JW, Samsa LA, Gaynor L, DiSalvo M, Coc- C, Bry L, Waterland RA, Moriarty C, Hwang C, Swennes AG,
chiaro JL, Carroll I, Azcarate-Peril MA, Rawls JF, Allbritton Moore SR, Shen L (2015) Postnatal epigenetic regulation of
NL, Magness ST (2018) A high-throughput organoid microin- intestinal stem cells requires DNA methylation and is guided by
jection platform to study gastrointestinal microbiota and luminal the microbiome. Genome Biol 16:211. https​://doi.org/10.1186/
physiology. Cmgh 6:301–319. https​://doi.org/10.1016/j.jcmgh​ s1305​9-015-0763-5
.2018.05.004 Zhang Z, Li Z, Gao C, Chen P, Chen J, Liu W, Xiao S, Lu H (2008)
Wood DE, Salzberg SL (2014) Kraken: ultrafast metagenomic sequence miR-21 plays a pivotal role in gastric cancer pathogenesis and
classification using exact alignments. Genome Biol 15:R46. https​ progression. Lab Invest 88:1358–1366. https​://doi.org/10.1038/
://doi.org/10.1186/gb-2014-15-3-r46 labin​vest.2008.94
Yadav D, Ghosh TS, Mande SS (2016) Global investigation of compo-
sition and interaction networks in gut microbiomes of individuals Publisher’s Note Springer Nature remains neutral with regard to
belonging to diverse geographies and age-groups. Gut Pathogens jurisdictional claims in published maps and institutional affiliations.
8:17. https​://doi.org/10.1186/s1309​9-016-0099-z
Yang W, Petersen C, Pees B, Zimmermann J, Waschina S, Dirksen
P, Rosenstiel P, Tholey A, Leippe M, Dierking K, Kaleta C,
Schulenburg H (2019) The inducible response of the nematode

13