Constant light exposure alters gut microbiota and promotes the progression of steatohepatitis in high fat diet rats Lin Wei1, Fangzhi Yue1, Lin Xing1, Shanyu Wu1, Ying Shi1, Jinchen Li2, Xingwei Xiang1, Lam S. Man3, Guanghou Shui3, Ryan Russell4, Dongmei Zhang1* 1Department of Endocrinology, Xiangya Hospital, Central South University, China, 2National Clinical Research Center for Geriatric Disorders, Xiangya Hospital, Central South University, China, 3State Key Laboratory of Molecular Developmental Biology, Institute of Genetics and Developmental Biology (CAS), China, 4Department of Health \& ProvisionalHuman Performance, College of Health Professions, University of Texas Rio Grande Valley Brownsville, United States Submitted to Journal: Frontiers in Microbiology Specialty Section: Microbial Physiology and Metabolism ISSN: 1664-302X Article type: Original Research Article Received on: 28 Dec 2019 Accepted on: 27 Jul 2020 Provisional PDF published on: 27 Jul 2020 Frontiers website link: www.frontiersin.org Citation: Wei L, Yue F, Xing L, Wu S, Shi Y, Li J, Xiang X, Man LS, Shui G, Russell R and Zhang D(2020) Constant light exposure alters gut microbiota and promotes the progression of steatohepatitis in high fat diet rats. Front. Microbiol. 11:1975. doi:10.3389/fmicb.2020.01975 Copyright statement: © 2020 Wei, Yue, Xing, Wu, Shi, Li, Xiang, Man, Shui, Russell and Zhang. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution and reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

This Provisional PDF corresponds to the article as it appeared upon acceptance, after peer-review. Fully formatted PDF and full text (HTML) versions will be made available soon. Frontiers in Microbiology | www.frontiersin.org Provisional

1 Journal: Frontier in Microbiology 2 Manuscript Category: Original Investigation 3 Title: Constant light exposure alters gut microbiota and promotes 4 the progression of steatohepatitis in high fat diet rats 5 Authors: Lin Wei1, Fangzhi Yue1, Lin Xing1, Shanyu Wu1, Ying Shi1, 6 Jinchen Li2, Xingwei Xiang1, Sin Man Lam3, Guanghou 7 Shui3, Ryan Russell4, Dongmei Zhang1* 8 Institutions: 1Department of Endocrinology, Xiangya Hospital, Central 9 South University, Changsha. 2National Clinical Research 10 Center for Geriatric Disorders, Department of Geriatrics, 11 Xiangya Hospital, Central South University, Changsha. 12 3State key laboratory of molecular developmental biology, 13 Institute of genetics and developmental biology, Chinese Provisional14 Academy of Sciences, No 1 west beichen road, Beijing. 15 4Cardiomatabolic Exercise lab Director, Department of 16 Health and Human Performance, College of Health 17 Professions, University of Texas Rio Grande Valley, 18 Brownsville, Texas. 19 Corresponding to: *Dongmei Zhang, PhD, MD, Department of Endocrinology, 20 Xiangya Hospital, Central South University, 87 Xiangya 21 Road, Changsha 410008, Hunan, China. Tel: (86) 731- 22 89753727. Fax: (86) 731-84327166. E-mail: 23 drdmzhang@csu.edu.cn 24 Short Title: Constant light exposure alters microbiota and promotes 25 NASH 26 Word Count: 3781 27 Conflict of interest: The authors do not have any disclosures to report. 28 1

29 Abstract 30 Background: Non-alcoholic fatty liver disease (NAFLD) poses a significant health 31 concern worldwide. With the progression of urbanization, light pollution may be a 32 previously unrecognized risk factor for NAFLD/NASH development. However, the 33 role of light pollution on NAFLD is insufficiently understood, and the underlying 34 mechanism remains unclear. Interestingly, recent studies indicate the gut microbiota 35 affects NAFLD/NASH development. Therefore, the present study explored effects of 36 constant light exposure on NAFLD and its related microbiotic mechanisms. 37 Material and method: Twenty-eight SD male rats were divided into four groups (n=7 38 each): rats fed a normal chow diet, and exposed to standard light-dark cycle (ND-LD); 39 rats fed a normal chow diet, and exposed to constant light (ND-LL); rats fed a high fat 40 diet, and exposed to standard light-dark cycle (HFD-LD); and rats on a high fat diet, 41 and exposed to constant light (HFD-LL). Body weight, hepatic pathophysiology, gut 42 microbiota, and short/medium chain fatty acids in colon contents, serum Provisional43 lipopolysaccharide (LPS) and liver LPS-binding protein (LBP) mRNA expression were 44 documented post intervention and compared among groups. 45 Result: In normal chow fed groups, rats exposed to constant light displayed glucose 46 abnormalities and dyslipidemia. In HFD-fed rats, constant light exposure exacerbated 47 glucose abnormalities, insulin resistance, inflammation and liver steatohepatitis. 48 Constant light exposure altered composition of gut microbiota in both normal chow and 49 HFD fed rats. Compared with HFD-LD group, HFD-LL rats displayed less 50 Butyricicoccus, Clostridium and Turicibacter, butyrate levels in colon contents, 51 decreased colon expression of occludin-1 and zonula occluden-1 (ZO-1) , and increased 52 serum LPS and liver LBP mRNA expression. 53 Conclusion: Constant light exposure impacts gut microbiota and its metabolic 54 products, impairs gut barrier function and gut-liver axis, promotes NAFLD/NASH 55 progression in HFD rats. 56 Key Words: Non-alcoholic fatty liver disease, light pollution, gut microbiota, short 57 chain fatty acids, gut-liver axis 2

58 Graphical abstract 59 constant light exposure normal chew fed rats HFD fed rats gut microbiota gut microbiota metabolic products imbalance dysbiosis dysbiosis (e.g. butyrate↓) (e.g.Proteobacteria↑) (e.g. Clostridium↓, gut barrier dysfunction ↑ Turicibacter↓) e.g. endogenous alcohol ↑LPS translocation↑ Provisional60 61 3

62 Funding sources 63 This work was supported by the National Science and Foundation of China (NO. 64 81670788) and Bethune-Merck Diabetes Research Fund (G2018030). 65 Ethics review and approval 66 The experiments were conducted according to the Animal Use and Care 67 Committee of Central South University and were conducted according to the 68 regulations set by Central South University (No.2018sydw184). 69 Declaration of interests 70 The authors have no known competing financial interests or personal relationships 71 that could have influenced the work reported in this paper. Provisional72 4

73 Highlights 74 • Constant light exposure promotes NAFLD/NASH progression in HFD rats 75 • Constant light exposure alters composition of gut microbiota 76 • Constant light exposure impairs gut barrier function and gut-liver axis in HFD 77 rats 78 Provisional 5

79 Abbreviations 80 NAFLD, non-alcoholic fatty liver disease; NASH, non-alcoholic steatohepatitis; 81 SCN, suprachiasmatic nuclei; ND, normal chow diet; HFD, high fat diet; IPGTT, 82 intraperitoneal glucose tolerance test; ITT, insulin tolerance test; AUC, area under the 83 curve; TG, triglyceride; TC, total cholesterol; LDL-C, low-density lipoprotein 84 cholesterol; HDL-C, high-density lipoprotein cholesterol; ALT, alanine 85 aminotransferase; AST, aspartate aminotransferase; NAS, NAFLD activity score; 86 OTUs, operational taxonomic units; PCA, principal component analysis; LDA, linear 87 discriminant analysis; LEfSe, LDA effect size; GEE, generalized estimated equation; 88 PICRUSt, phylogenetic investigation of communities by reconstruction of unobserved 89 states; KEGG, Kyoto Encyclopedia of Genes and Genomes; SCFAs, short chain fatty 90 acids; MCFAs, medium chain fatty acids; IL-6, interleukin-6; TNF-α, tumor necrosis 91 factor-alpha; LPS, lipopolysaccharide; LBP, LPS-binding protein; RT-PCR, reverse Provisional92 transcription polymerase chain reaction; ZO-1, zonula occluden-1; IHC, 93 immunohistochemistry. 94 6

95 1. Introduction 96 Non-alcoholic fatty liver disease (NAFLD) is defined as the presence of hepatic 97 steatosis in at least 5% of hepatocytes which is not attributed to alcohol consumption 98 or other secondary causes of steatosis (Cairns and Peters, 1983). Although steatosis has 99 long been considered as a benign liver disease, it may progress into a more aggressive 100 form of nonalcoholic steatohepatitis (NASH), which in turn may lead to cirrhosis and, 101 sometimes, to hepatocellular carcinoma (Xanthakos et al., 2006). NAFLD is becoming 102 the most common chronic liver disease worldwide, affecting about 20-30% of the 103 general population (Sheka et al., 2020). 104 Although fatty liver and steatohepatitis most commonly stem from overnutrition 105 and lack of exercise, additional mediators, such as environmental factors, have recently 106 been postulated (Eslam et al., 2018). One novel environmental risk factor for NAFLD 107 is light pollution. Light pollution, defined as the alteration of natural light levels due to 108 the introduction of artificial light at night, is a major side-effect of urbanization (Falchi Provisional109 et al., 2016). Artificial light allows people to extend daytime activities into the night 110 and engage in countercyclical nighttime shift work. As such, it may adversely affect 111 health via circadian rhythm disruption (Lunn et al., 2017). 112 Circadian rhythms refer to physiological processes that occur with a repeating period 113 of approximately 24 h, and ensure that internal physiology is synchronized with the 114 external environment (Gachon et al., 2004). In mammals, suprachiasmatic nuclei (SCN) 115 acts as the master circadian clock. SCN communicates time-of-day information by 116 synaptic and diffusible signals to clocks in various brain regions and peripheral organs 117 (i.e. peripheral clock). Light information is transmitted by way of the retino- 118 hypothalamic tract connecting the eye to the SCN, and is the most potent synchronizing 119 factor and the main zeitgeber that synchronizes the clock with the external environment 120 (Panda et al., 2002). Thus, the SCN serves to synchronize the timing of rhythmic 121 activities throughout the body to the light/dark cycle, and responds to light more rapidly 7

122 than peripheral tissues (Welsh et al., 2010). Taken together, the endogenous biological 123 clocks can be disrupted by nighttime light exposure (light pollution). 124 It is reasonable to suspect light pollution affects physiological function due to the 125 importance of circadian system in regulating homeostatic functions. Data show that 126 circadian disruptions from light pollution causes significant disruptions in 127 physiological function. Epidemiological evidence from shift workers exposed to high 128 levels of light at night suggest that prolonged exposure to light at night increases the 129 risk for cancer, mood disorders, and metabolic dysfunction (Kubo et al., 2011;Xiao et 130 al., 2019). Additionally, the increase in exposure to light at night parallels the global 131 increase in the prevalence of obesity and metabolic disorders (Koo et al., 2016), well- 132 known risk factors for NAFLD. However, the effects of environmental light pollution 133 on the development of NAFLD remains unclear. 134 The gut microbiota is recognized as an “external\" organ playing an important role 135 in host physiology and metabolism (Lozupone et al., 2012;Raman et al., 2013). Both Provisional136 animal and observational studies in NAFLD patients suggest links between gut 137 microbiota changes and NAFLD (Chiu et al., 2017;Tripathi et al., 2018;Zhou et al., 138 2018). The changes of gut microbiota may disrupt the gut tight junctions, leading to 139 increased gut permeability and LPS translocation. Increased LPS translocation induces 140 “metabolic endotoxemia,” which triggers inflammatory reactions, insulin resistance, 141 and promotes the development of NAFLD (Leung et al., 2016;Aron-Wisnewsky et al., 142 2020). 143 It has been reported that circadian disruption by constant light exposure changes 144 gut microbiome taxa and their functional gene composition (Wu et al., 2018). However, 145 data on the effects of light pollution on gut microbiome in NAFLD subjects is limited. 146 High fat diet (HFD) feeding is extensively used models of NAFLD in rodents. 147 Accumulating evidence shows that HFD reduces microbial diversity and alters 148 gut microbiota composition (Cani et al., 2008;Zhang et al., 2012). In the present 149 animal study, we observed the effects of constant light exposure on NAFLD, and 8

150 explored changes of gut microbiota in the colon content in HFD-fed rats model of 151 NAFLD. 152 2. Materials and methods 153 2.1 Animal experiment 154 Twenty-eight male Sprague-Dawley (SD) rats (6 weeks old) were purchased from 155 Hunan Slac-Jingda Laboratory Animal Co. (Changsha, China). All rats were housed 156 under specific pathogen-free conditions in a temperature-controlled room with free 157 access to water and food. All rats were fed with normal chow diet (ND, fat 12%, 158 carbohydrate 66%, protein 22%, 3.50 kcal/g) and under 12:12 h light/dark cycle for one 159 week to adapt to the environment. The rats were then randomly divided into 4 160 experimental groups and placed in two separate rooms (n=7 each): (1) ND-LD group: 161 rats on a normal diet (ND), and exposed to standard light/dark (LD) cycle; (2) ND-LL 162 group: rats on a ND, and exposed to constant light (LL); (3) HFD-LD group: rats fed Provisional163 on a HFD ( fat 37%, protein 17.5%, carbohydrate 45.5%, 4.50 kcal/g), and exposed to 164 standard LD cycle; and (4) HFD-LL group: rats on HFD, and exposed to constant light 165 (LL). The LD rats were placed in a room with a standard 12 h light/dark (LD) cycle: 166 lights on between 8am and 8pm, with the light intensity in the cage set at 200 lux. The 167 LL rats were placed in a constant-light room where they were exposed to 200 lux of 168 continuous light at cage level. The light source were natural white fluorescent light 169 tubes with a wavelength range of 400nm~560nm. 170 Food intake and body weight were recorded weekly for all animals throughout the 171 experiment for 16 weeks. Animal protocols were approved by the Animal Use and Care 172 Committee of Central South University and were conducted according to the 173 regulations set by Central South University (No.2018sydw184). 174 2.2 Intraperitoneal glucose tolerance test (IPGTT) and insulin tolerance test (ITT) 175 An IPGTT was performed in the 14th week of experiment after an overnight fast. 176 The rats were given an intraperitoneal injection of 50% D-glucose (2.0 g/kg) after 177 fasting. Blood samples were measured from the tip of the tail at 0, 15, 30, 60, and 9

178 120 min after glucose injection using a portable glucose monitor (ACCU-CHEK; 179 Roche Diagnostics, Mannheim, Germany). 180 ITT was carried out 5 days after IPGTT. Insulin (0.75 IU/kg, Novolin R, Novo 181 Nordisk, Denmark) was injected intraperitoneally after an overnight fast, and blood 182 glucose measurement procedure was the same as IPGTT. 183 2.3 Body composition assessment 184 Dual Energy X-ray Absorptiometry (DXA, GE Lunar Corp., USA) was utilized to 185 assess body fat mass using small animal software (GE Medical Systems Lunar, 186 Madison, WI, USA). 187 2.4 Sample collection 188 After 16 weeks of intervention, rats were sacrificed in 2 consecutive days from 8 am 189 -12 pm after fasting for 12 h. They were sacrificed while under anesthesia (inhaled) 190 with 1.5%–3.0% isoflurane (RWD Life Since Co., China). The abdominal cavity was 191 then rapidly opened, and blood samples were collected from the superior vena cava Provisional192 and centrifuged at 2000 × g for 20 min to isolate serum (stored at -20℃). 193 Liver tissues were collected and weighed. Liver tissues were then either fixed with 194 4% paraformaldehyde solution, or frozen in liquid nitrogen and stored at -80°C until 195 analysis. 196 Colon content samples were collected under a sterile fume to prevent miscellaneous 197 bacterium contamination and then frozen in liquid nitrogen and stored at -80°C until 198 they were analyzed for gut microbiota or short/medium chain fatty acids measurements. 199 Proximal colon segments were collected and then fixed with 4% paraformaldehyde 200 solution for immunohistochemical analysis. 201 2.5 Relative visceral fat weight 202 Epididymal fat were quickly excised and weighted. The epididymal fat relative 203 body weight was represented for relative visceral fat weight. 204 2.6 Serum Biochemical Analysis 10

205 Serum levels of triglycerides (TG), total cholesterol (TC), low-density lipoprotein 206 cholesterol (LDL-C), high-density lipoprotein cholesterol (HDL-C), alanine 207 aminotransferase (ALT) and aspartate aminotransferase (AST) were measured using 208 commercial reagents (Serotec Co., Sapporo, Japan) according to the manufacturer’s 209 recommendations. 210 2.7 Liver pathology 211 Paraffin-embedded liver sections (5μm) were stained by haematoxylin and eosin 212 (H\&E). The NAFLD activity score (NAS) was assessed by an experienced physiologist 213 using indices of inflammation, steatosis and hepatocyte ballooning as previously 214 published (Kleiner et al., 2005). 215 2.8 Gut microbiota analysis 216 Microbial DNA was extracted by FastDNA™ Spin Kit for Soil (MP bio, USA), and 217 quantified using NanoDrop 2000. The V3-V4 region of the 16S rRNA gene was 218 intensified by PCR (94℃ for 2min, followed by 25 cycles at 94℃ for 30s, 55℃ for Provisional219 30s, and 72℃ for 1min and a final extension at 72℃ for 10min) using primers 5′- 220 CCTACGGGNGGCWGCAG -3′ for 341F and 5′-GACTACHVGGGTATCTAATCC 221 -3′ for 805R. The Agencourt AMPure XP PCR Purification Beads (Bechman Coulter, 222 USA) was used to purify the amplicons. The solution was checked for integrity using 223 Agilent 2100 Bioanalyzer (Agilent Technologies, USA) and quantified using 224 Invitrogen Qubit3.0 spectrophotometer (Thermo Fisher Scientific, USA). The Miseq 225 Reagnent Kit V3 was used for normalization and sequencing on an Illumina MiSeq 226 (Illumina, USA). 227 Adapter sequences and low-quality ends were removed using Trimmomatic v0.33. 228 Fastq files were demultiplexed and quality-filtered with FLASH2. The operational 229 taxonomic units (OTUs) which reached a 97% sequence identity were subjected to 230 alpha-diversity analyses to evaluate samples biodiversity using mothur software 231 (version 1.30.1). The R package (R 3.6.0.) was used for the visualization of bacterial 232 community classification and distribution. Linear discriminant analysis (LDA) effect 11

233 size (LEfSe) was performed to identify taxonomic biomarkers that characterize the 234 differences between groups with the logarithmic LDA score threshold set at 2.5. A 235 generalized estimated equation (GEE) analysis was performed using R package to 236 investigate whether changes of microbiota were associated with NAS. Functional 237 capacity for each sequence was analyzed with phylogenetic investigation of 238 communities by reconstruction of unobserved states (PICRUSt). The Kyoto 239 Encyclopedia of Genes and Genomes (KEGG) analysis was used to identify credible 240 biological functions. 241 2.9 Short and medium chain fatty acids measurement 242 Short chain fatty acids (SCFAs) and medium chain fatty acids (MCFAs) in colon 243 contents were measured using HPLC-MS/MS method as previously described (Li et al., 244 2019). Briefly，samples were extracted with solvent mixtures containing acetonitrile 245 and double distilled water and analyzed on a Thermo Fisher DGLC-3000 coupled to 246 Sciex QTRAP 6500 Plus system. Octanoic acid-1-13C1 was used as an internal standard. Provisional247 2.10SerumIL-6andTNF-α 248 Serum IL-6 and TNF-α were measured using respective commercial rat-specific 249 enzyme-linked immunosorbent assay (ELISA) kit (Cusabio, Wuhan, China). 250 2. 11 Immunohistochemical analysis 251 Paraffin-embedded colon tissues (4μm) were stained with occludin (1:200; Abcam, 252 Hong Kong) or zonula occluden-1 (ZO-1, 1:100; Affinity Bioscience, USA) primary 253 antibodies and incubated in a humidified chamber overnight at 4℃. The sections were 254 then incubated with biotinylated goat anti-rabbit secondary antibodies (Boster 255 Biological Technology, Wuhan, China) at room temperature for 20min. Finally, color 256 was developed in diaminobenzidine (DAB; ZSGB Biotechnology, Beijing, China) 257 substrate solution. The average optical density (AOD) value of immunohistochemical 258 intensity was analyzed by the Image J software (version 1.53a; national institutes of 259 Health, USA). 12

260 2.12 Serum lipopolysaccharide (LPS) and liver LPS-binding protein (LBP) 261 quantitative real time-PCR analysis 262 LPS, a component of the outer membrane of gram-negative bacteria, is an indicator 263 of intestinal leakiness (Summa et al., 2013). The concentration of serum LPS was 264 measured with commercially available limulus amebocyte lysate chromogenic kit 265 (Dynamiker Biotechnology, Tianjin, China) according to the manufacturer’s protocol. 266 LBP is a type 1 acute phase protein that is constitutively produced by the liver and 267 rapidly upregulated during acute phase responses. LBP binds LPS to facilitate immune 268 responses in conjunction with cell-surface pattern recognition receptors and is used as 269 an indicator of LPS exposure(Summa et al., 2013). Liver expression of LBP mRNA 270 was assessed by quantitative real time -PCR. Primers sequence of LBP were as follows: 271 F: 5’-TTACCGCCTGACTCCAACAT-3’, R: 5’-CAAGCCGGAAGACAGATTCG- 272 3’). Quantification of LBP gene expression was performed using a ΔΔCt method with 273 glyceraldehyde 3- phosphate dehydrogenase (GADPH) as an internal control. Provisional274 2.12Statisticalanalysis 275 All data were analyzed using SPSS 23.0 (IBM, Armok, U.S.). A one-way analysis 276 of variance (ANOVA) followed by Tukey’s post hoc test was used to identify statistical 277 differences between groups. Data were shown as mean ± standard error of the mean. 278 Statistical significance was defined as p<0.05. 279 3. Results 280 3.1 Constant light exposure aggravated obesity and visceral adiposity in HFD rats 281 There were no significant differences in food intakes, body weights, body fat mass, 282 and relative visceral fat weights between ND-LD and ND-LL group (Figure 1). In HFD 283 rats, body weights in HFD-LL group were significantly higher than those in HFD-LD 284 group from the 11th week of HFD feeding, with no difference in caloric intake was 285 noted between HFD-LD and HFD-LL groups (Figure 1A-D). Compared with the HFD- 286 LD group, body weight and visceral fat weights in the HFD-LL group were 287 significantly higher after 16 weeks (Figure 1E-F). 13

288 3.2 Effects of constant light exposure on glucose homeostasis and serum lipid 289 profiles. 290 In normal chow fed rats, serum TC in the ND-LL group was significantly higher 291 than that of the ND-LD group (Table 1). During the IPGTT, blood glucose levels at 292 30min and 60min in the ND-LL group were higher than those noted in the ND-LD 293 group. The area under the curve (AUC) of the IPGTT in the ND-LL group was higher 294 than that in the ND-LD group (Figure 2 A-B). ITT suggested an increased tendency of 295 AUCITT in the ND-LL group vs the ND-LD group (p=0.0844) (Figure 2D). 296 In HFD rats, serum TC and LDL-C in the HFD-LL group was significantly higher 297 than those in the HFD-LD group (Table 1). During IPGTT, the HFD-LL group 298 displayed increased levels of blood glucose starting at 15min post-glucose injection vs 299 the HFD-LD group (Figure 2A). ITT demonstrated greater insulin resistance in the 300 HFD-LL group vs the HFD-LD group (Figure 2D). 301 3.3 Constant light exposure promote the progress of NAFLD in HFD rats. Provisional302 Compared with the ND-LD group, the ND-LL group had increased serum IL-6 303 concentration and AST/ALT ratio (Table 1). Liver histopathological examination 304 showed more inflammatory cell infiltration in the ND-LL group as well (Figure 3A-B). 305 In HFD rats, hepatic steatosis and inflammatory infiltrates were significantly 306 elevated in the HFD-LL group (Figure 3A). Also, the HFD-LL group had higher NAS 307 and inflammation scores than the HFD-LD group (p < 0.05, Figure 3B). Serum TNF- 308 α, IL-6, and AST/ALT ratio were also higher in the HFD-LL group (Table 1). 309 3.4 Constant light exposure altered intestinal microbial communities. 310 There were no significant differences in alpha diversity and beta diversity between 311 the ND-LD and ND-LL groups (Figure 4A-B). LEfSe analysis showed the relative 312 higher abundance of Protobacteria, Firmicutes, Spirochaetes phylum, Spirochaetes 313 class, Spirochaetales order, Spirochaetaceae and Odoribacteraceae families, and 314 Treponema genus in the ND-LL group compared to the ND-LD group. (Figure 4F). 14

315 Compared with the ND-LD group, HFD-LD rats had decreased alpha diversity, 316 increased abundance of Firmicutes, Proteobacteria and Verrucomicrobia and 317 decreased abundance of Bacteroides (Figure 4C-E). 318 The alpha diversity did not reveal significant differences between the HFD-LL and 319 HFD-LD groups. The PCA represented structural change by the second principal 320 component (PC2) with no statistic differences in PC1 (Figure 4A-B). The LEfSe 321 analysis demonstrated constant light exposure restrained the growth of genus 322 Butyricicoccus, Clostridium, Turicibacter and class Bacilli in HFD fed rats. Compared 323 with the HFD-LD group, the amount of Christensenella and Dehalobacterium were 324 higher in the HFD-LL group (Figure 4G). Consistent with the notion that constant light 325 exposure causes increased severity of NASH, GEE analysis demonstrated that NAS 326 was inversely correlated with genus Turicibacter (p < 0.01) and genus Clostridium (p 327 < 0.05) (Figure 4H-I). 328 Pathways related to type 2 diabetes mellitus were elevated in the ND-LL group vs Provisional329 those of the ND-LD group. Compared with the HFD-LD group, LPS biosynthesis 330 phagosome, LPS biosynthesis proteins, Vitamin B6 metabolism, and antimicrobial 331 drug resistance were significantly upregulated in the HFD-LL group (Figure 4J). 332 3.5 Effects of constant light exposure on SCFAs/MCFAs levels of colon content 333 No significant differences in SCFAs and MCFAs levels of colon contents between 334 ND-LD and ND-LL group. Compared with ND-LD group, an increase in acetate acid, 335 propionate acid, and a decrease in butyrate acid, valerate acid and MCFAs in colon 336 contents were detected in the HFD-LD group. Among HFD groups, reduced butyrate 337 acid was observed in the HFD-LL rats (Figure 5A-D). 338 3.6 Constant light exposure caused gut barrier dysfunction 339 Gut barrier dysfunction plays an important role in the progression of NASH (Xue 340 et al., 2017). Imbalances of gut microbiota and decreased SCFAs (e.g. butyrate acid) 341 are associated with gut barrier dysfunction (Hamer et al., 2008;Lee et al., 2019). 342 Compared to the control (ND-LD) group, HFD groups had lower expression of 15

343 occludin and ZO-1 in proximal colon, and significantly higher circulating LPS levels 344 and liver LBP mRNA expression. In HFD rats, constant light exposure caused a further 345 increase in serum LPS and liver LBP mRNA expression, with further decrement of 346 occludin and ZO-1 expression (all p < 0.05) (Figure 6A-F). 347 4. Discussion 348 With the wide-spread prevalence of electric lights, particularly at night, light 349 pollution is rising by approximately 6% per year worldwide (Holker et al., 2010). Light 350 pollution is expected to rise dramatically in the next several decades through more 351 urban development, such as street lightening, vehicles lightening, and security 352 lightening. However, consequences of light pollution remain largely unknown. In this 353 report, we demonstrated that constant light exposure predisposed HFD rats, a widely 354 used animal model of obesity and NAFLD, to increased obesity and NAFLD/NASH 355 progression. This suggests that light pollution is a novel risk factor for NAFLD/NASH 356 progression. In normal chow fed rats, ND-LL rats showed increased levels of TC and Provisional357 blood glucose, and higher levels of IL-6, also suggesting deleterious effects of constant 358 light exposure on metabolism. 359 Gut dysbiosis is associated with the development of NAFLD (Tripathi et al., 2018). 360 Studies suggest that the development of NAFLD is associated with marked changes of 361 fecal microbiota composition (Aron-Wisnewsky et al., 2013;Wang et al., 2018). In 362 addition, results of both animal and human studies suggest that targeting intestinal 363 microbiota may have protective effects on the development of NAFLD (Tripathi et al., 364 2018). However, changes in gut microbiota were not entirely consistent among studies 365 (Kolodziejczyk et al., 2019). For example, among NAFLD individuals, an increase in 366 Proteobacteria, and Bacteroidetes has been reported (Boursier et al., 2016;Shen et al., 367 2017), while a reduction in Proteobacteria and Bacteroidetes were also observed (Shin 368 et al., 2017;Koopman et al., 2019). Different ethnicities and living environments may 369 help explain these discrepancies. Our experiment showed an increase of Firmicutes, 16

370 Proteobacteria, Verrucomicrobia and a decrease of Bacteroides in the HFD-LD vs ND- 371 LD group. 372 Evidences indicate that circadian clock has a profound impact on gut 373 microbiota(Liang et al., 2015;Voigt et al., 2016). Circadian disruption by constant 374 darkness or ClockΔ19 mutation, changes the composition and diurnal oscillation of gut 375 microbiome in normal chow fed mice (Voigt et al., 2014;Voigt et al., 2016;Wu et al., 376 2018). In normal chow fed rats, we found that constant light exposure increased 377 Proteobacteria and Firmicutes. It has been reported that Proteobacteria ferment to 378 produce alcohol, and higher concentrations of endogenous alcohols are thought to 379 contribute to liver injury and inflammation (Leung et al., 2016). Thus, the increased 380 abundance of Proteobacteria may explain the increased AST/ALT and liver 381 inflammatory cells infiltration in the ND-LL group. 382 Very few experiments have been performed to study the effects of constant light 383 exposure on gut microbiome in HFD-fed animals. The present study was the first to Provisional384 note increased Christensenella and Dehalobacterium, and a decrease in genus 385 Butyricicoccus, Clostridium, Turicibacter and class Bacilli in HFD rats exposed to 386 constant light. Clostridium has been associated with improved lipid regulation, reduced 387 risk of hyperlipidemia, intestinal permeability, inhibition of harmful pathogens and 388 normalization of lipid metabolism (Ling et al., 2016). In NAFLD patients, a negative 389 correlation between severity of NASH and abundance of Clostridium has been reported 390 (Pataky et al., 2016;Zhou et al., 2019). The present animal study also demonstrated a 391 negative correlation between NAS and abundance of genus Clostridium. 392 Another interesting discovery of our study was that constant light exposure caused 393 decreased butyrate acid levels in colon content of HFD rats. The impact of gut 394 microbiota on NAFLD is considered to be mediated by its metabolic products, such as 395 SCFAs. Among SCFAs, butyrate has been reported to upregulate the expression of tight 396 junction proteins, and consequently can enhance the gut barrier function (Lu et al., 397 2016). Butyrate supplementation to animals has been demonstrated to reduce hepatic 17

398 fat accumulation and hepatic inflammation (Canfora et al., 2019). The decreased 399 butyrate level in the HFD-LL group is consistent with the reduction in Clostridium 400 genus, as the Clostridium genus has butyrate-producing effects (Vital et al., 401 2014;Ganesan et al., 2018). Targeting gut microbiota therapy has protective effects on 402 the development of NAFLD (Tripathi et al., 2018). Light pollution is sometimes 403 inevitable, e.g. shift workers, medical staff, et cetera. It may possible that 404 supplementation of some bacteria (e.g. Clostridium genus) or butyrate acid can be used 405 as an intervention therapy for early-stage NASH caused by light pollution. 406 NAFLD is associated with increased gut permeability and impaired gut-liver axis 407 (Jiang et al., 2015;Szabo, 2015). Gut hyperpermeability permits the translocation of 408 proinflammatory bacterial products (most prominently, LPS) from the lumen of the gut 409 into systemic circulation and promotes the progression of NASH (Okubo et al., 2016). 410 We found that HFD-LL rats had decreased colon expression of tight junction proteins 411 ZO-1 and occludin, and increased serum LPS level and liver LBP mRNA expression Provisional412 vs the HFD-LD group, suggesting increased gut barrier dysfunction and impaired gut- 413 liver axis exists in HFD rats exposed to constant light. 414 Taken together, the results presented herein demonstrate that constant light 415 exposure alters gut microbiota and its metabolic products, impairs gut barrier function 416 and gut-liver axis, and promotes the progression of NAFLD/NASH in HFD rats. 417 5. Limitations 418 This study was the first to explore the impact of constant light exposure on 419 NAFLD. However, there are some limitations to note. First, the light intensity was only 420 set at 200lux. Different results may occur when light intensity changes. Secondly, the 421 illumination wavelength in our study was 400nm~560nm. It has been reported that 422 circadian responses have different spectral sensitivity, peaking at wavelengths between 423 450 and 490 nm (Dominoni et al., 2016). This suggests that illumination with different 424 wavelengths may have varying effects. Our experiment used a circadian clock- 425 disrupting light source, though specifying which wavelength has the most impact 18

426 cannot determine. Lastly, we carried out the experiment only in male SD rats, though 427 gender is thought to be an important factor in metabolic syndrome and its outcomes. 428 There are gender differences in prevalence, risk factors and mechanisms of NAFLD. It 429 is still unknown the effects of constant light exposure on female subjects. Moreover, 430 the number of animals was small. 431 6. Conclusion 432 Constant light exposure changes gut microbiota and its metabolic products, 433 impairs gut barrier function and gut-liver axis, promotes the progression of 434 NAFLD/NASH in HFD rats. 435 436 Acknowledgements 437 We thank Dr. Zhijun Zhou, the director of department of zoology, Xiangya medical 438 college, Central South university, for the support and assistance in this study. 439 Provisional440 CRediTauthorstatement 441 Lin Wei: Conceptualization, Methodology, Data curation, Writing-Original draft 442 preparation Fangzhi Yue: Investigation, Data curation, Formal analysis Lin Xing: 443 Investigation, Data curation Shanyu Wu: Investigation, Data curation Ying Shi: 444 Resources, Investigation Jinchen Li: Visualization, Data curation Xingwei Xiang: 445 Investigation Sin Man Lam: Resources Guanghou Shui: Resources Ryan Russell: 446 Data curation, Writing-Reviewing and Editing Dongmei Zhang: Supervision, 447 Conceptualization, Formal analysis, Writing- Reviewing and Editing. 19

448 Reference 449 Aron-Wisnewsky, J., Gaborit, B., Dutour, A., and Clement, K. (2013). 450 Gut microbiota and non-alcoholic fatty liver disease: new 451 insights. Clin Microbiol Infect 19, 338-348.doi:10.1111/1469- 452 0691.12140 453 Aron-Wisnewsky, J., Vigliotti, C., Witjes, J., Le, P., Holleboom, 454 A.G., Verheij, J., Nieuwdorp, M., and Clement, K. (2020). Gut 455 microbiota and human NAFLD: disentangling microbial signatures 456 from metabolic disorders. Nat Rev Gastroenterol Hepatol 17, 457 279-297.doi:10.1038/s41575-020-0269-9 458 Boursier, J., Mueller, O., Barret, M., Machado, M., Fizanne, L., 459 Araujo-Perez, F., Guy, C.D., Seed, P.C., Rawls, J.F., David, 460 L.A., Hunault, G., Oberti, F., Cales, P., and Diehl, A.M. 461 (2016). The severity of nonalcoholic fatty liver disease is 462 associated with gut dysbiosis and shift in the metabolic 463 function of the gut microbiota. Hepatology 63, 764- 464 775.doi:10.1002/hep.28356 465 Cairns, S.R., and Peters, T.J. (1983). Biochemical analysis of 466 hepatic lipid in alcoholic and diabetic and control subjects. 467 Clin Sci (Lond) 65, 645-652.doi:10.1042/cs0650645 468 Canfora, E.E., Meex, R.C.R., Venema, K., and Blaak, E.E. (2019). Gut Provisional469 microbial metabolites in obesity, NAFLD and T2DM. Nat Rev 470 Endocrinol 15, 261-273.doi:10.1038/s41574-019-0156-z 471 Cani, P.D., Bibiloni, R., Knauf, C., Waget, A., Neyrinck, A.M., 472 Delzenne, N.M., and Burcelin, R. (2008). Changes in gut 473 microbiota control metabolic endotoxemia-induced inflammation 474 in high-fat diet-induced obesity and diabetes in mice. Diabetes 475 57, 1470-1481.doi:10.2337/db07-1403 476 Chiu, C.C., Ching, Y.H., Li, Y.P., Liu, J.Y., Huang, Y.T., Huang, 477 Y.W., Yang, S.S., Huang, W.C., and Chuang, H.L. (2017). 478 Nonalcoholic Fatty Liver Disease Is Exacerbated in High-Fat 479 Diet-Fed Gnotobiotic Mice by Colonization with the Gut 480 Microbiota from Patients with Nonalcoholic Steatohepatitis. 481 Nutrients 9.doi:ARTN 1220 482 10.3390/nu9111220 483 Dominoni, D.M., Borniger, J.C., and Nelson, R.J. (2016). Light at 484 night, clocks and health: from humans to wild organisms. Biol 485 Lett 12, 20160015.doi:10.1098/rsbl.2016.0015 486 Eslam, M., Valenti, L., and Romeo, S. (2018). Genetics and 487 epigenetics of NAFLD and NASH: Clinical impact. J Hepatol 68, 488 268-279.doi:10.1016/j.jhep.2017.09.003 20

489 Falchi, F., Cinzano, P., Duriscoe, D., Kyba, C.C., Elvidge, C.D., 490 Baugh, K., Portnov, B.A., Rybnikova, N.A., and Furgoni, R. 491 (2016). The new world atlas of artificial night sky brightness. 492 Sci Adv 2, e1600377.doi:10.1126/sciadv.1600377 493 Gachon, F., Nagoshi, E., Brown, S.A., Ripperger, J., and Schibler, U. 494 (2004). The mammalian circadian timing system: from gene 495 expression to physiology. Chromosoma 113, 103- 496 112.doi:10.1007/s00412-004-0296-2 497 Ganesan, K., Chung, S.K., Vanamala, J., and Xu, B. (2018). Causal 498 Relationship between Diet-Induced Gut Microbiota Changes and 499 Diabetes: A Novel Strategy to Transplant Faecalibacterium 500 prausnitzii in Preventing Diabetes. Int J Mol Sci 501 19.doi:10.3390/ijms19123720 502 Hamer, H.M., Jonkers, D., Venema, K., Vanhoutvin, S., Troost, F.J., 503 and Brummer, R.J. (2008). Review article: the role of butyrate 504 on colonic function. Aliment Pharmacol Ther 27, 104- 505 119.doi:10.1111/j.1365-2036.2007.03562.x 506 Holker, F., Moss, T., Griefahn, B., Kloas, W., Voigt, C.C., Henckel, 507 D., Hanel, A., Kappeler, P.M., Volker, S., Schwope, A., Franke, 508 S., Uhrlandt, D., Fischer, J., Klenke, R., Wolter, C., and 509 Tockner, K. (2010). The Dark Side of Light: A Transdisciplinary Provisional510 Research Agenda for Light Pollution Policy. Ecology and Society 511 15 512 Jiang, W., Wu, N., Wang, X., Chi, Y., Zhang, Y., Qiu, X., Hu, Y., Li, 513 J., and Liu, Y. (2015). Dysbiosis gut microbiota associated 514 with inflammation and impaired mucosal immune function in 515 intestine of humans with non-alcoholic fatty liver disease. Sci 516 Rep 5, 8096.doi:10.1038/srep08096 517 Kleiner, D.E., Brunt, E.M., Van Natta, M., Behling, C., Contos, M.J., 518 Cummings, O.W., Ferrell, L.D., Liu, Y.C., Torbenson, M.S., 519 Unalp-Arida, A., Yeh, M., Mccullough, A.J., Sanyal, A.J., and 520 Clin, N.S. (2005). Design and validation of a histological 521 scoring system for nonalcoholic fatty liver disease. Hepatology 522 41, 1313-1321.doi:10.1002/hep.20701 523 Kolodziejczyk, A.A., Zheng, D., Shibolet, O., and Elinav, E. (2019). 524 The role of the microbiome in NAFLD and NASH. EMBO Mol Med 525 11.doi:10.15252/emmm.201809302 526 Koo, Y.S., Song, J.Y., Joo, E.Y., Lee, H.J., Lee, E., Lee, S.K., and 527 Jung, K.Y. (2016). Outdoor artificial light at night, obesity, 528 and sleep health: Cross-sectional analysis in the KoGES study. 529 Chronobiology International 33, 301- 530 314.doi:10.3109/07420528.2016.1143480 21

531 Koopman, N., Molinaro, A., Nieuwdorp, M., and Holleboom, A.G. (2019). 532 Review article: can bugs be drugs? The potential of probiotics 533 and prebiotics as treatment for non-alcoholic fatty liver 534 disease. Aliment Pharmacol Ther.doi:10.1111/apt.15416 535 Kubo, T., Oyama, I., Nakamura, T., Shirane, K., Otsuka, H., Kunimoto, 536 M., Kadowaki, K., Maruyama, T., Otomo, H., Fujino, Y., 537 Matsumoto, T., and Matsuda, S. (2011). Retrospective cohort 538 study of the risk of obesity among shift workers: findings from 539 the Industry-based Shift Workers' Health study, Japan. Occup 540 Environ Med 68, 327-331.doi:10.1136/oem.2009.054445 541 Lee, S., Kirkland, R., Grunewald, Z.I., Sun, Q., Wicker, L., and De 542 La Serre, C.B. (2019). Beneficial Effects of Non-Encapsulated 543 or Encapsulated Probiotic Supplementation on Microbiota 544 Composition, Intestinal Barrier Functions, Inflammatory 545 Profiles, and Glucose Tolerance in High Fat Fed Rats. Nutrients 546 11.doi:10.3390/nu11091975 547 Leung, C., Rivera, L., Furness, J.B., and Angus, P.W. (2016). The 548 role of the gut microbiota in NAFLD. Nat Rev Gastroenterol 549 Hepatol 13, 412-425.doi:10.1038/nrgastro.2016.85 550 Li, B., Li, L., Li, M., Lam, S.M., Wang, G., Wu, Y., Zhang, H., Niu, 551 C., Zhang, X., Liu, X., Hambly, C., Jin, W., Shui, G., and Provisional552 Speakman, J.R. (2019). Microbiota Depletion Impairs 553 Thermogenesis of Brown Adipose Tissue and Browning of White 554 Adipose Tissue. Cell Rep 26, 2720-2737 555 e2725.doi:10.1016/j.celrep.2019.02.015 556 Liang, X., Bushman, F.D., and Fitzgerald, G.A. (2015). Rhythmicity of 557 the intestinal microbiota is regulated by gender and the host 558 circadian clock. Proc Natl Acad Sci U S A 112, 10479- 559 10484.doi:10.1073/pnas.1501305112 560 Ling, Z., Jin, C., Xie, T., Cheng, Y., Li, L., and Wu, N. (2016). 561 Alterations in the Fecal Microbiota of Patients with HIV-1 562 Infection: An Observational Study in A Chinese Population. Sci 563 Rep 6, 30673.doi:10.1038/srep30673 564 Lozupone, C.A., Stombaugh, J.I., Gordon, J.I., Jansson, J.K., and 565 Knight, R. (2012). Diversity, stability and resilience of the 566 human gut microbiota. Nature 489, 220- 567 230.doi:10.1038/nature11550 568 Lu, Y.Y., Fan, C.N., Li, P., Lu, Y.F., Chang, X.L., and Qi, K.M. 569 (2016). Short Chain Fatty Acids Prevent High-fat-diet-induced 570 Obesity in Mice by Regulating G Protein-coupled Receptors and 571 Gut Microbiota. Scientific Reports 6.doi:ARTN 37589 572 10.1038/srep37589 22

573 Lunn, R.M., Blask, D.E., Coogan, A.N., Figueiro, M., Gorman, M.R., 574 Hall, J.E., Hansen, J., Nelson, R.J., Panda, S., Smolensky, 575 M.H., Stevens, R.G., Turek, F.W., Vermeulen, R., Carreon, T., 576 Caruso, C.C., Lawson, C.C., Thayer, K.A., Twery, M.J., Ewens, 577 A.D., Garner, S.C., Schwingl, P.J., and Boyd, W.A. (2017). 578 Health consequences of electric lighting practices in the 579 modern world: A report on the National Toxicology Program's 580 workshop on shift work at night, artificial light at night, and 581 circadian disruption. Science of the Total Environment 607, 582 1073-1084.doi:10.1016/j.scitotenv.2017.07.056 583 Okubo, H., Kushiyama, A., Sakoda, H., Nakatsu, Y., Iizuka, M., Taki, 584 N., Fujishiro, M., Fukushima, T., Kamata, H., Nagamachi, A., 585 Inaba, T., Nishimura, F., Katagiri, H., Asahara, T., Yoshida, 586 Y., Chonan, O., Encinas, J., and Asano, T. (2016). Involvement 587 of resistin-like molecule beta in the development of 588 methionine-choline deficient diet-induced non-alcoholic 589 steatohepatitis in mice. Sci Rep 6, 20157.doi:10.1038/srep20157 590 Panda, S., Antoch, M.P., Miller, B.H., Su, A.I., Schook, A.B., 591 Straume, M., Schultz, P.G., Kay, S.A., Takahashi, J.S., and 592 Hogenesch, J.B. (2002). Coordinated transcription of key 593 pathways in the mouse by the circadian clock. Cell 109, 307- Provisional594 320.doi:10.1016/s0092-8674(02)00722-5 595 Pataky, Z., Genton, L., Spahr, L., Lazarevic, V., Terraz, S., Gaia, 596 N., Rubbia-Brandt, L., Golay, A., Schrenzel, J., and Pichard, 597 C. (2016). Impact of Hypocaloric Hyperproteic Diet on Gut 598 Microbiota in Overweight or Obese Patients with Nonalcoholic 599 Fatty Liver Disease: A Pilot Study. Dig Dis Sci 61, 2721- 600 2731.doi:10.1007/s10620-016-4179-1 601 Raman, M., Ahmed, I., Gillevet, P.M., Probert, C.S., Ratcliffe, N.M., 602 Smith, S., Greenwood, R., Sikaroodi, M., Lam, V., Crotty, P., 603 Bailey, J., Myers, R.P., and Rioux, K.P. (2013). Fecal 604 microbiome and volatile organic compound metabolome in obese 605 humans with nonalcoholic fatty liver disease. Clin 606 Gastroenterol Hepatol 11, 868-875 e861- 607 863.doi:10.1016/j.cgh.2013.02.015 608 Sheka, A.C., Adeyi, O., Thompson, J., Hameed, B., Crawford, P.A., and 609 Ikramuddin, S. (2020). Nonalcoholic Steatohepatitis: A Review. 610 JAMA 323, 1175-1183.doi:10.1001/jama.2020.2298 611 Shen, F., Zheng, R.D., Sun, X.Q., Ding, W.J., Wang, X.Y., and Fan, 612 J.G. (2017). Gut microbiota dysbiosis in patients with non- 613 alcoholic fatty liver disease. Hepatobiliary \& Pancreatic 23

614 Diseases International 16, 375-381.doi:10.1016/S1499- 615 3872(17)60019-5 616 Shin, N.R., Bose, S., Wang, J.H., Ansari, A., Lim, S.K., Chin, Y.W., 617 Choi, H.S., and Kim, H. (2017). Flos Lonicera Combined with 618 Metformin Ameliorates Hepatosteatosis and Glucose Intolerance 619 in Association with Gut Microbiota Modulation. Frontiers in 620 Microbiology 8.doi:ARTN 2271 621 10.3389/fmicb.2017.02271 622 Summa, K.C., Voigt, R.M., Forsyth, C.B., Shaikh, M., Cavanaugh, K., 623 Tang, Y., Vitaterna, M.H., Song, S., Turek, F.W., and 624 Keshavarzian, A. (2013). Disruption of the Circadian Clock in 625 Mice Increases Intestinal Permeability and Promotes Alcohol- 626 Induced Hepatic Pathology and Inflammation. PLoS One 8, 627 e67102.doi:10.1371/journal.pone.0067102 628 Szabo, G. (2015). Gut-liver axis in alcoholic liver disease. 629 Gastroenterology 148, 30-36.doi:10.1053/j.gastro.2014.10.042 630 Tripathi, A., Debelius, J., Brenner, D.A., Karin, M., Loomba, R., 631 Schnabl, B., and Knight, R. (2018). The gut-liver axis and the 632 intersection with the microbiome. Nat Rev Gastroenterol Hepatol 633 15, 397-411.doi:10.1038/s41575-018-0011-z 634 Vital, M., Howe, A.C., and Tiedje, J.M. (2014). Revealing the Provisional635 bacterial butyrate synthesis pathways by analyzing 636 (meta)genomic data. MBio 5, e00889.doi:10.1128/mBio.00889-14 637 Voigt, R.M., Forsyth, C.B., Green, S.J., Mutlu, E., Engen, P., 638 Vitaterna, M.H., Turek, F.W., and Keshavarzian, A. (2014). 639 Circadian disorganization alters intestinal microbiota. PLoS 640 One 9, e97500.doi:10.1371/journal.pone.0097500 641 Voigt, R.M., Summa, K.C., Forsyth, C.B., Green, S.J., Engen, P., 642 Naqib, A., Vitaterna, M.H., Turek, F.W., and Keshavarzian, A. 643 (2016). The Circadian Clock Mutation Promotes Intestinal 644 Dysbiosis. Alcohol Clin Exp Res 40, 335- 645 347.doi:10.1111/acer.12943 646 Wang, W., Zhao, J., Gui, W., Sun, D., Dai, H., Xiao, L., Chu, H., Du, 647 F., Zhu, Q., Schnabl, B., Huang, K., Yang, L., and Hou, X. 648 (2018). Tauroursodeoxycholic acid inhibits intestinal 649 inflammation and barrier disruption in mice with non-alcoholic 650 fatty liver disease. Br J Pharmacol 175, 469- 651 484.doi:10.1111/bph.14095 652 Welsh, D.K., Takahashi, J.S., and Kay, S.A. (2010). Suprachiasmatic 653 nucleus: cell autonomy and network properties. Annu Rev Physiol 654 72, 551-577.doi:10.1146/annurev-physiol-021909-135919 24

655 Wu, G., Tang, W., He, Y., Hu, J., Gong, S., He, Z., Wei, G., Lv, L., 656 Jiang, Y., Zhou, H., and Chen, P. (2018). Light exposure 657 influences the diurnal oscillation of gut microbiota in mice. 658 Biochem Biophys Res Commun 501, 16- 659 23.doi:10.1016/j.bbrc.2018.04.095 660 Xanthakos, S., Miles, L., Bucuvalas, J., Daniels, S., Garcia, V., and 661 Inge, T. (2006). Histologic spectrum of nonalcoholic fatty 662 liver disease in morbidly obese adolescents. Clin Gastroenterol 663 Hepatol 4, 226-232 664 Xiao, Q., Gee, G., Jones, R.R., Jia, P., James, P., and Hale, L. 665 (2019). Cross-sectional association between outdoor artificial 666 light at night and sleep duration in middle-to-older aged 667 adults: The NIH-AARP Diet and Health Study. Environ Res 180, 668 108823.doi:10.1016/j.envres.2019.108823 669 Xue, L., He, J., Gao, N., Lu, X., Li, M., Wu, X., Liu, Z., Jin, Y., 670 Liu, J., Xu, J., and Geng, Y. (2017). Probiotics may delay the 671 progression of nonalcoholic fatty liver disease by restoring 672 the gut microbiota structure and improving intestinal 673 endotoxemia. Sci Rep 7, 45176.doi:10.1038/srep45176 674 Zhang, C., Zhang, M., Pang, X., Zhao, Y., Wang, L., and Zhao, L. 675 (2012). Structural resilience of the gut microbiota in adult Provisional676 mice under high-fat dietary perturbations. ISME J 6, 1848- 677 1857.doi:10.1038/ismej.2012.27 678 Zhou, W., Guo, R., Guo, W., Hong, J., Li, L., Ni, L., Sun, J., Liu, 679 B., Rao, P., and Lv, X. (2019). Monascus yellow, red and orange 680 pigments from red yeast rice ameliorate lipid metabolic 681 disorders and gut microbiota dysbiosis in Wistar rats fed on a 682 high-fat diet. Food Funct 10, 1073-1084.doi:10.1039/c8fo02192a 683 Zhou, Y., Zheng, T., Chen, H., Li, Y., Huang, H., Chen, W., Du, Y., 684 He, J., Li, Y., Cao, J., Nie, Y., and Zhou, Y. (2018). 685 Microbial Intervention as a Novel Target in Treatment of Non- 686 Alcoholic Fatty Liver Disease Progression. Cell Physiol Biochem 687 51, 2123-2135.doi:10.1159/000495830 688 689 25

690 Table 1. Comparison of serum lipids, AST/ALT and serum IL-6, TNF-α among groups at the end 691 of the experiment. Group ND-LD ND-LL HFD-LD HFD-LL n 7777 TG (mmol/L) 0.74±0.10 0.80±0.27 0.86±0.07 0.82±0.17 TC (mmol/L) 1.41±0.06 1.71±0.08a 1.71±0.09a 2.05±0.12a, b, c HDL-C (mmol/L) 0.87±0.07 0.86±0.05 0.65±0.03a, b 0.63±0.04a, b LDL-C (mmol/L) 0.48±0.05 0.54±0.03 0.74±0.05a, b 0.93±0.07a, b, c AST(U/L) 92.56±6.95 106.63±10.08 129.89±21.61 209.74±33.07 a, b ALT(U/L) 36.24±3.09 33.31±2.95 41.11±9.03 45.59±8.33 AST/ALT 2.56±0.16 3.24±0.19 a 3.47±0.30 a 4.79±0.22a, b, c IL-6 (pg/mL) 26.82±1.65 40.26±2.75a 46.26±2.21a, b 70.01±2.10a, b, c TNF-α (pg/mL) 8.38±0.51 8.86±0.15 11.84±0.61a, b 14.92±0.59a, b, c 692 Results were expressed as Mean ± SEM. ap < 0.05, compared with ND-LD group; 693 bp < 0.01, compared with ND-LL group; cp < 0.01, compared with HFD-LD group. 694 Abbreviations: TG, triglyceride; TC, total cholesterol; HDL-C, high-density Provisional695 lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; AST, aspartate 696 aminotransferase; ALT, alanine aminotransferase; IL-6, interleukin-6; TNF-α, tumor 697 necrosis factor-alpha. 698 26

699 Figure Legends: 700 701 Figure1. Constant light exposure aggravated obesity and visceral adiposity in HFD rats. 702 (A) Calorie intakes, (B) average calorie intakes, (C) body weights, (D) final body 703 weights, (E) body fat mass, and (F) relative visceral fat weight. Data were expressed as 704 Mean ± SEM, n = 7 per group, a1p < 0.05, a2p < 0.01, a3p < 0.001, vs ND-LD group; 705 b1p < 0.05, b2p < 0.01, vs ND-LL group; c1p < 0.05, vs HFD-LD group. 706 707 Figure2. Effects of constant light exposure on glucose homeostasis. (A) IPGTT, (B) 708 AUC of IPGTT, (C) ITT (% basal), and (D) AUC of ITT. Data were expressed as 709 Mean ± SEM, n = 7 per group, a1p < 0.05, a2p < 0.01, a3p < 0.001, vs ND-LD group; 710 b1p < 0.05, b2p < 0.01, vs ND-LL group; c1p < 0.05, c2p < 0.01 vs HFD-LD group. 711 712 Figure3. Constant light exposure aggravated the progression of NAFLD/NASH. 713 (A)Representative pictures of H\&E staining (200× magnifications) in liver tissue; (B) Provisional714 NAFLD activity score. Data were expressed as Median, a1p < 0.05, a3p < 0.001, vs 715 ND-LD group; b3p < 0.001, vs ND-LL group; c1p < 0.05, c2p < 0.01, vs HFD-LD group. 716 717 Figure 4. Constant light exposure altered colon microbiota. (A)Alpha diversity of colon 718 microbiota estimated by Chao 1, Shannon and Simpson indexes, (B) principle 719 component analysis (PCA), (C-D) heatmap of colon microbiota at phylum (C) and 720 genus (D) level, (E) cladogram, (F-G) scores of taxonomic biomarkers identified by 721 linear discriminant analysis (LDA) using LEfSe in normal chow-fed rats (F) and HFD- 722 fed rats(G), LDA value > 2.5 were showed in the figure, (H-I) correlation between 723 NAS and abundance of Clostridium (H), Turicibacter (I) revealed by generalized 724 estimated equation (GEE) analysis, (J) KEGG pathways changed by constant light 725 exposure. 726 a1p < 0.05, a2p < 0.01, a3p < 0.001, vs ND-LD group; b1p < 0.05, b2p < 0.01, b3p < 0.001, 727 vs ND-LL group. 27

728 729 Figure5. Effects of constant light exposure on short chain fatty acids (SCFAs) and 730 medium chain fatty acids (MCFAs) of colon contents. (A) Total amount of SCFAs, 731 (B)SCFAs species, (C) total amount of MCFAs, (D) MCFAs species. Data were 732 expressed as Mean ± SEM, a1p < 0.05, a2p < 0.01, vs ND-LD group; b1p < 0.05, vs 733 ND-LL group; c1p < 0.05, vs HFD-LD group. 734 735 Figure6. Constant light exposure impaired gut barrier function and gut-liver axis in 736 HFD rats. (A) Serum LPS concentration, (B) hepatic LBP mRNA expression by RT- 737 PCR, (C) expression of occludin in colon by IHC (200×), (D) expression of ZO-1 in 738 colon by IHC (200×). Data were expressed as Mean ± SEM, a1p < 0.05, a2p < 0.01, a3p 739 < 0.001, vs ND-LD group; b1p < 0.05, b2p < 0.01, b3p < 0.001, vs ND-LL group; c1p < Provisional740 0.05,c2p < 0.01, c3p < 0.001 vs HFD-LD group. 28

Figure 01.TIFF Provisional

Figure 02.TIFF Provisional

Figure 03.TIFF Provisional

Figure 04.TIFF Provisional

Figure 05.TIFF Provisional

Figure 06.TIFF Provisional