1
The intestinal circadian clock drives microbial rhythmicity to maintain
1
gastrointestinal homeostasis
2
3
Marjolein Heddes
1, 2,
*, Baraa Altaha
1, 2,
*, Yunhui Niu
1, 2
, Sandra Reitmeier
1, 2
, Karin
4
Kleigrewe
3
, Dirk Haller
1, 2,
, Silke Kiessling
1,2,4,†
5
6
†
Corresponding Author: Dr. Silke Kiessling, silke.kiessling@tum.de
7
*These authors contributed equally
8
1
ZIEL - Institute for Food & Health, Technical University of Munich, 85354 Freising,
9
Germany
10
2
Chair of Nutrition and Immunology, Technical University of Munich, Gregor-Mendel-Str. 2,
11
85354 Freising, Germany
12
3
Bavarian Center for Biomolecular Mass Spectrometry, Technical University of Munich,
13
Gregor-Mendel-Str. 4, 85354 Freising, Germany
14
4
Lead Contact
15
preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
The copyright holder for thisthis version posted October 18, 2021. ; https://doi.org/10.1101/2021.10.18.464061doi: bioRxiv preprint
2
Summary
16
Diurnal (i.e., 24-hour) oscillations of the gut microbiome have been described in various
17
species including mice and humans. However, the driving force behind these rhythms
18
remains less clear. In this study, we differentiate between endogenous and exogenous time
19
cues driving microbial rhythms. Our results demonstrate that fecal microbial oscillations
20
are maintained in mice kept in the absence of light, supporting a role of the host’s
21
circadian system rather than representing a diurnal response to environmental changes.
22
Intestinal epithelial cell-specific ablation of the core clock gene Bmal1 disrupts
23
rhythmicity of microbiota. Targeted metabolomics functionally link intestinal clock-
24
controlled bacteria to microbial-derived products, in particular branched-chain fatty
25
acids and secondary bile acids. Microbiota transfer from intestinal clock-deficient mice
26
into germ-free mice altered intestinal gene expression, enhanced lymphoid organ weights
27
and suppressed immune cell recruitment. These results highlight the importance of
28
functional intestinal clocks for circadian microbiota composition and function, which is
29
required to balance the host’s gastrointestinal homeostasis.
30
31
Introduction
32
Various physiological processes show 24-hour fluctuations. These rhythms are the expression
33
of endogenous circadian (Lat. circa = about; dies = day) clocks that have evolved in most
34
species [1] to facilitate anticipation of daily recurring environmental changes. In mammals, the
35
circadian system includes a central pacemaker regulating sleep-wake behavior and
36
orchestrating subordinated tissue clocks by humoral and neuronal pathways [2]. At the cellular
37
level, these clocks consist of inter-regulated core clock genes [3] that drive tissue-specific
38
transcriptional programs of clock-controlled genes (CCGs) [4]. Through these CCGs, circadian
39
clocks regulate various aspects of physiology including metabolism, gastrointestinal transit
40
preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
The copyright holder for thisthis version posted October 18, 2021. ; https://doi.org/10.1101/2021.10.18.464061doi: bioRxiv preprint
3
time (GITT), mucus secretion, antimicrobial peptide secretion, immune defense and intestinal
41
barrier function (reviewed by [5-8]).
42
In mice, 10-15 % of gut bacteria undergo diurnal oscillations in their abundance influenced by
43
meal timing, diet type and other environmental conditions [9-11]. Recently, we found similar
44
rhythms in microbiota composition and function in population-based human cohorts [12].
45
Although functionality of the host’s circadian system impacts microbial rhythmicity [10, 13,
46
14], it is unclear which tissue clocks contribute to this effect.
47
A balanced gut microbiome promotes health and microbial dysbiosis has been linked to
48
metabolic diseases, colorectal cancer and gastrointestinal inflammation [12, 15-17]. Similar
49
pathological consequences are associated with circadian rhythm disruption (reviewed by [18-
50
20]), which also induces microbiota dysbiosis [10, 11, 13, 14, 21]. Consequently, we
51
hypothesized that circadian regulation of microbiota composition and function may contribute
52
to the host’s GI health.
53
Here, we functionally dissect the circadian origin of microbiota oscillations in mice. We provide
54
evidence that intestinal epithelial cell (IEC) clocks generate the majority of gut microbial
55
rhythms and their metabolic products, particularly short-chain fatty acids (SCFAs) and bile
56
acids (BAs). Transfer of microbiota from IEC clock-deficient mice in germ-free (GF) wild type
57
hosts directly indicate the consequences of microbiota arrhythmicity on the gastrointestinal
58
homeostasis. Thus, we identify a mechanistic link between IEC clocks, gut bacteria rhythms
59
and their functions through transfer experiments, showing the importance of rhythmic
60
microbiota for host physiology.
61
Results
62
Rhythms in microbiota are generated endogenously by the circadian system
63
Diurnal rhythms in microbiota composition and function have been demonstrated in animal
64
models and in humans [9-12]. However, it has not been demonstrated whether these rhythms
65
preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
The copyright holder for thisthis version posted October 18, 2021. ; https://doi.org/10.1101/2021.10.18.464061doi: bioRxiv preprint
4
are a response to rhythmic external cues (Zeitgebers), such as the light-dark cycle, or are
66
generated by endogenous clocks [22] and, thus, persist when the organism is placed in an
67
environment devoid of timing cues. To address this question, we compared fecal microbiota
68
rhythms of the same wild-type mice kept in a rhythmic 12-hour light/12-hour dark (LD) cycle
69
and constant darkness (DD) for two weeks (Fig. 1A). Host-driven rhythmic factors, which
70
might influence microbiota composition, such as locomotor activity, food intake as well as total
71
GI transit time (GITT), did not differ between light conditions (Suppl. Fig 1A-D). 16S rRNA
72
profiling of fecal samples revealed clustering based on sampling time points in both light
73
conditions (Fig. 1B). Generalized UniFrac distances (GUniFrac) quantification identified
74
rhythmicity in both light conditions, although with a 40% reduced amplitude in DD (p = 0.0033,
75
Fig. 1C). Importantly, 24-hour rhythms of species richness and Shannon effective number of
76
species found in LD persisted in DD, supporting their circadian origin (Fig. 1D). Rhythms were
77
also preserved in DD on phylum level with the two most dominant phyla Bacteroidetes and
78
Firmicutes, oscillating in antiphase (Fig. 1E, F). Although microbiota composition is
79
commonly analyzed by relative abundance, rhythmicity of highly abundant taxa may mask
80
oscillations of small microbial communities, as previously demonstrated in fecal samples
81
collected in LD [13]. Therefore, using synthetic DNA spikes, we performed relative
82
quantification of the copy number of 16S rRNA genes according to Tourlousee et al. [23], from
83
here on referred to as ‘quantitative analysis’. With both approaches, highly abundant phyla and
84
families, including Lachnospiraceae and Muribaculaceae, showed comparable circadian
85
rhythmicity in LD and DD (Fig. 1F; Suppl. Fig. 1E). Few families, such as Prevotellaceae
86
showed significant diurnal (LD), but no circadian (DD) rhythmicity, suggesting that their
87
rhythms are regulated by the environmental LD cycle (Suppl. Fig. 1E). After removal of low-
88
abundant taxa (mean relative abundance > 0.1%; prevalence > 10%), the remaining 580 zOTUs
89
displayed robust and comparable 24-hour oscillations in both light conditions, independent of
90
the analysis (Fig. 1G, H; Suppl. Fig. 1F-I). The wide distribution of peak abundances, e.g.,
91
preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
The copyright holder for thisthis version posted October 18, 2021. ; https://doi.org/10.1101/2021.10.18.464061doi: bioRxiv preprint
5
bacteria peaking during the day (mainly belonging to Muribaculaceae) and during the night
92
(Lachnospiraceae) (Fig. 1G, H; Suppl. Fig. 1F-H), suggests that different microbial taxa
93
dominate different daytimes. Importantly, about ¾ of all identified zOTUs were found to
94
significantly oscillate in LD and about 80 % of these diurnal zOTUs remained rhythmic in DD
95
(Fig. 1G, H, Suppl. Fig. 1F-H, Suppl. table 1), suggesting that their rhythmicity is generated
96
by the circadian system. Rhythmicity of the majority of zOTUs (e.g. the genera Alistipes,
97
Oscillibacter and Fusimonas) identified by relative analysis was further validated by
98
quantitative analysis (73 % in LD, 58 % in DD) (Suppl. Fig. 1 I, J). Although most zOTUs
99
were found to be controlled by the circadian system, less than 20 % of zOTUs (e.g. the genera
100
Alloprevotella and Aneaeroplasma) lost rhythmicity in DD (Fig. 1G; Suppl. Fig. 1K),
101
indicating the rhythmicity of these taxa depends on rhythmic environmental cues. Altogether,
102
these data suggest that the diurnal rhythmicity found in ¾ of all zOTUs examined is generated
103
by the endogenous circadian system.
104
Robust microbial rhythms during simulated shift work
105
Alterations in gut microbiota communities as well as disrupted diurnal oscillation in specific
106
taxa, such as Faecalibacterium, have been reported in shift work and jetlag conditions in human
107
and mouse studies [10, 24]. In our study, external light conditions only mildly effect microbial
108
rhythms (Fig. 1), thus prompting us to determine microbial rhythmicity under simulated shift
109
work (SSW) conditions. SSW mice were exposed to 8-hour shifts of the LD cycle every 5
th
day
110
for a minimum of 6 weeks and compared to littermates kept in stable LD conditions (Fig. 2A).
111
Total activity, food intake and GITT were comparable between both cohorts (Suppl. Fig. 2A-
112
C). However, altered activity profiles and advanced activity onsets were observed in SSW mice
113
(Fig. 2A, Suppl. Fig. 2A) reflecting the transient state of chronodisruption observed after such
114
rapid LD phase shifts [25]. Microbial profiles oscillated in both light conditions and rhythms
115
of major phyla were preserved in SSW (Fig. 2B,-E). Heatmaps of bacterial abundances over
116
preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
The copyright holder for thisthis version posted October 18, 2021. ; https://doi.org/10.1101/2021.10.18.464061doi: bioRxiv preprint
![Fig. 1C). Importantly, 24-hour rhythms of species richness and Shannon effective number of 76 species found in LD persisted in DD, supporting their circadian origin (Fig. 1D). Rhythms were 77 also preserved in DD on phylum level with the two most dominant phyla Bacteroidetes and 78 Firmicutes, oscillating in antiphase (Fig. 1E, F). Although microbiota composition is 79 commonly analyzed by relative abundance, rhythmicity of highly abundant taxa may mask 80 oscillations of small microbial communities, as previously demonstrated in fecal samples 81 collected in LD [13]. Therefore, using synthetic DNA spikes, we performed relative 82 quantification of the copy number of 16S rRNA genes according to Tourlousee et al. [23], from 83 here on referred to as ‘quantitative analysis’. With both approaches, highly abundant phyla and 84 families, including Lachnospiraceae and Muribaculaceae, showed comparable circadian 85 rhythmicity in LD and DD (Fig. 1F; Suppl. Fig. 1E). Few families, such as Prevotellaceae 86 showed significant diurnal (LD), but no circadian (DD) rhythmicity, suggesting that their 87 rhythms are regulated by the environmental LD cycle (Suppl. Fig. 1E). After removal of low-88 abundant taxa (mean relative abundance > 0.1%; prevalence > 10%), the remaining 580 zOTUs 89 displayed robust and comparable 24-hour oscillations in both light conditions, independent of 90 the analysis (Fig. 1G, H; Suppl. Fig. 1F-I). The wide distribution of peak abundances, e.g., 91](/figures/fig-1c-importantly-24-hour-rhythms-of-species-richness-and-1l6vqsw0.webp)