Combinational conditions of low air pressure and noise impact on locomotor rhythms
In this study, 6 free-moving mice in the capsule and 6 free-moving mice in the animal room were raised for recording of locomotion rhytmicity. For hypothalamus and feces sampling, 3 mice were raised for each time point either in the capsule or animal room (Fig. 1a). Uniform field noise of 85 dbA and ~ 0.9 atmospheric pressure were loaded in the simulated space capsule. We recorded the locomotion rhythms of control mice but not those of mice undergoing hind limb unloading (HU) to mimic microgravity, in both the simulated space capsule and animal room. Compared to the mice in the animal room, the mice in the simulated space capsule showed a decrease in amplitude of locomotion during the days after the initiation of exposure to noise and low atmospheric pressure (Fig. 1b-d). Furthermore, it took the control mice in the capsule took 5.8 ± 0.5 d to adjust to the 6-h advance of the LD cycles, which was significantly faster than that of the control mice in the animal room (7.7 ± 0.8 d) (Fig. 1b,c,e,f). These data suggest that noise and low atmospheric pressure may affect the function of the circadian system.
Simulated space environmental factors differentially affect global gene expression
To address the effects of different simulated space environmental factors (noise, low air pressure and microgravity) on global gene expression, the hypothalamus tissues containing suprachiasmatic nuclei (SCN) from HU and control mice, in the capsule and animal room in the control (day 6) and experiment (day 36) periods, were collected for RNA sequencing (RNA-seq), respectively. Global changes in gene expression were observed (Fig. 2a,b). Notably, the quantities of differentially expressed genes (DEGs) in the capsule were much lower than those in the animal room (Fig. 2b,c). For instance, of the detected genes (Table S1), 22 were upregulated at the early stage and 151 were upregulated at the late stage of the control mice in the capsule while 884 were upregulated at the early stage and 4350 were upregulated at the late stage of the control mice in the animal room. Similarly, the number of DEGs was much higher in the HU mice (Fig. 2b-d; Figure S1b-g). It is worth noting that the number of DEGs in animals in SSC was much lower than those in AR, for both HU or control mice.
Gene enrichment and functional annotation by Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis indicated that a number of important biological processes associated with diseases were altered in the capsule mice, such as viral protein infection in HU mice and prion diseases in the control mice in the capsule. Pathways implicated in metabolism, neural systems, cell physiology, and diseases and cancers were also found. The pathway of circadian entrainment was enriched in the HU and control mice in the animal room (late vs. early) but not in the mice in the simulated space capsule. (Fig. 2e-f; Figure S1h-k; Table S3).
Simulated space environmental factors differentially modify the rhythmicity of global gene expression
JTK_Cycle analysis revealed differential reprogramming patterns of global circadian expression of hypothalamus transcripts between early and late stages under different conditions. For instance, 828 and 1558 genes of HU mice in the simulated space capsule showed rhythmicity in the early and late stages, respectively, and 78 genes showed rhymicity at both stages (Fig. 3a,b; Figure S2a-f; Table S2). Compared to HU mice in the animal room, the phase of global gene expression of HU mice in the capsule was limited in much narrower ranges at either early or late stages (Fig. 3b).
We next predicted the enriched pathways of those genes with significant changes in their rhythmicities under different conditions by KEGG pathway analysis. The pathways implicated in hypertrophic cardiomyopathy (HCM) and dilate cardiomyopathy (DCM) were identified between HU and control mice at the early stage in the animal room and between the early and late stages of HU mice in the animal room. Alterations in the rhythmicity of genes implicated in the pathway of circadian rhythm were identified between HU and control mice in the animal room at the late stage. Furthermore, the pathway associated with human papillomavirus infection was found in HU and noise and low atmospheric conditions. Additionally, pathways associated with cancers, cell physiology, immune defense, digestion, and neurodegenerative diseases were also found under the respective conditions (Fig. 3g-j; Figure S3; Table S4).
Simulated space environmental factors differentially affect the expression of core circadian clock genes
Together, the changes in locomotion rhythms (Fig. 1b-e) and pathways associated with circadian entrainment in the animal room (Fig. 3a-d) suggest that circadian rhythms may be affected at the molecular level. RNA-seq analysis confirmed that the average levels of circadian clock genes showed differentially altered gene expression patterns (Fig. 4a,b).
The average levels of many core circadian clock genes showed significant changes in mice in the animal room but not simulated space capsule (Fig. 4a-c). In the animal room, the levels of Clock, Per1, Per2, Per3, Cry1, Cry2, Nr1d1, Arntl and Npas2 were upregulated at the late stage in control mice, and Clock, Per2, Per3, Cry1, Cry2, Nr1d1 and Npas2 were upregulated in the HU mice. In contrast, none of these genes showed significant changes in the mice in the capsule. Arntl2 showed no significant change in either HU or control mice (Fig. 4a,b). Relative levels of some circadian clock genes (late levels normalized to early levels), including Clock, Per1, Per2, Cry1, Cry2 and Nr1d1, were decreased in HU compared to control mice in the animal room. In contrast, the levels of these genes were comparable between HU and control mice in the simulated space capsule (Fig. 4c). Together, these data suggest that microgravity and noise and low atmospheric pressure may repress the expression and function of circadian clock genes. Changes in circadian rhythmicity (amplitude or phase) of these clock genes were also observed in most of these clock genes (Fig. 4d-k; Figure S4a-l).
The pressure was 0.9 atm in the capsule from day 7 to the end in the present study. The RNA-seq data showed no significant difference in the hypothalamus Hif1a level in control mice between the early and late stages in the capsule (Figure S4m,n). We further measured the HIF1A protein levels in liver samples, and revealed no significant change between these two stages; together, these data suggest that the low atmospheric pressure in the capsule is not sufficient to elicit hypoxia stress. However, a significant increase in HIF1A levels was observed in control mice at the late stage in the animal room and a significant decrease was observed in HU mice at the late stage in the capsule (Figure S4o).
Simulated space environmental factors differentially affect the abundance of the gut microbiome
Stool samples from the transverse colon were subjected to 16S rRNA sequencing. In total, the sequenced microbiome consisted of 25 phyla, 58 classes, 82 orders, 129 families, and 172 genera of microorganisms, of which Bacteroidetes and Firmicutes were the most dominant. Together with Proteobacteria, Verrucomicrobia, Cyanobacteria, Actinobacteria, Deferribacteres, and Tenericutes, these phyla constituted over 99% of the known phylogenetic categories (Fig. 5a,b; Table S5).
The α-diversity between the data sets showed no significant difference (Fig. 5a-c). The PCoA results based on the OTU abundance revealed different separation under the indicated conditions (Figure S5a-h). Interestingly, a clear separation was visualized between the early and late stages of the HU mice gut microbiome in the animal room but not the HU mice in the simulated space capsule (Fig. 5d-g; Figure S5f,h), suggesting that microgravity imposes long-term impacts on the structure of the gut microbial community, which could be repressed by superimposition with noise and low atmospheric pressure.
Among the detected phyla, changes in the abundance of many bacteria were observed (Fig. 6a-d; Figure S6). For instance, interestingly, Adlercreutzia was increased at the late stage of HU mice in the animal room but not in the capsule(Figure S6a,c). Verrucomicrobia abundance was increased at the late stage of control mice in the capsule and the animal room but was decreased at the late stage of HU mice in the capsule and the animal room (Fig. 6a-d; Figure S6a-g).
PICRUSt analysis revealed that microbial pathways were significantly correlated with host cardiovascular functions. For instance, pathways of cardiac muscle contraction and HCM were identified between the early and late stages of control mice in both the animal room and the simulated space capsule, such as the renin-angiotensin system and HCM between the HU and control mice at the early stage in the simulated space capsule and cardiac muscle contraction, viral myocarditis, renin-angiotensin system and HCM pathways between the HU and control mice at the late stage in the simulated space capsule, and between HU and control mice at the late stage in the simulated space capsule. In addition, the pathway of amyotrophic lateral sclerosis was identified between the HU and control mice at the late stage in the simulated space capsule (Fig. 6e,f; Figure S7; Table S7). Other important pathways implicated in immune defense, metabolism, cancers, cell physiology, and other processes were also found under different conditions (Fig. 6e,f; Figure S7; Table S7).
Simulated space environmental factors differentially affect the diel oscillation of gut microbiome
Most of prokaryotic organisms possess no endogenous circadian systems; however, they exhibit diurnal rhythms owing to daily cycling environmental factors or the host internal milieu[31]. It has been revealed that mammalian oral and gut microbiomes undergo diurnal changes in the abundance of many microorganisms [32, 33].
The composition of some taxa displayed overt diurnal patterns over a day, including Akkermansia, Allobaculum, Bacteroides, Prevotella, Sutterella and so on (Fig. 7a). We next looked at the compositional changes of some bacteria at the genus level and found that the rhythmicities were changed under some conditions, such as Cetobacterium, Kaistobacter, Rhodoplanes, Acinetobacter, Corynebulcterium, Paraprevotella, Enterobacter, Devosia, etc (Fig. 7b-o; Figure S8; Table S6). Changes in phase or amplitude were found in some of these bacteria. For instance, Prevotella showed a higher amplitude in the HU mice in the capsule at the late stage than at the early, and a higher amplitude of Corynebacterium was seen in the control mice in the capsule. Rhodoplanes showed a higher amplitude in the HU mice in the capsule at the late stage. And Paraprevotella showed a dramatic increase in the oscillation in the HU mice at the late stage while it showed induction in abundance only at night in the control mice at the late stage in the animal room (Fig. 7b-o; Figure S8; Table S6).