F1000Research F1000Research 2046-1402 F1000 Research Limited London, UK 10.12688/f1000research.125628.1 Research Article Articles Genetic and environmental perturbations alter the rhythmic expression pattern of a circadian long non-coding RNA, Per2AS, in mouse liver

[version 1; peer review: 2 approved]

 Miao Lin Data Curation Formal Analysis Investigation Project Administration Supervision Validation Visualization Writing – Original Draft Preparation https://orcid.org/0000-0002-3866-3104 1 2 Batty Kyle R. Formal Analysis Investigation 2 3 Jackson Ayana N. Formal Analysis Investigation 1 2 Pieno Heather A. Formal Analysis Investigation 2 Rhoades Maisy W. Formal Analysis Investigation 2 Kojima Shihoko Conceptualization Data Curation Funding Acquisition Project Administration Resources Supervision Validation Writing – Review & Editing https://orcid.org/0000-0001-9579-7830 a 1 2 3 Department of Biological Sciences, Virginia Tech, Blacksburg, VA, 24061, USA Fralin Life Sciences Institute, Virginia Tech, Blacksburg, VA, 24061, USA Division of Systems Biology, Academy of Integrated Science, Virginia Tech, Blacksburg, VA, 24061, USA skojima@vt.edu

No competing interests were disclosed.

 20 9 2022 2022 11 1073 12 9 2022 Copyright: © 2022 Miao L et al. 2022 This is an open access article distributed under the terms of the Creative Commons Attribution Licence, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Background: Long non-coding RNAs (lncRNAs) play a wide variety of biological roles without encoding a protein. Although the functions of many lncRNAs have been uncovered in recent years, the regulatory mechanism of lncRNA expression is still poorly understood despite that the expression patterns of lncRNAs are much more specific compared to mRNAs. Here, we investigated the rhythmic expression of Per2AS, a novel lncRNA that regulates circadian rhythms. Given that Per2AS expression is antiphasic to Period2 ( Per2), a core circadian clock gene, and transcribed from the antisense strand of Per2, we hypothesized that the rhythmic Per2AS expression is driven either by its own promoter or by the rhythmic Per2 transcription via transcriptional interference.

Methods: We leveraged existing circadian RNA-seq datasets and analyzed the expression patterns of Per2AS and Per2 in response to the genetic or environmental disruption of the circadian rhythm in mouse liver. We tested our hypotheses by comparing the changes in the expression patterns of Per2AS and Per2.

Conclusions: We found that, in some cases, Per2AS expression is independently controlled by other circadian transcription factors. In other cases, the pattern of expression change is consistent with both transcriptional interference and independent regulation hypotheses. Although additional experiments will be necessary to distinguish these possibilities, findings from this work contribute to a deeper understanding of the mechanism of how the expression of lncRNA is regulated.

 LncRNAs circadian rhythm antisense rhythmicity feeding regimen National Institute of General Medical Sciences R01GM12622 This work was supported by R01-GM126223 from the National Institutes of Health (to S.K.). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Introduction

Long non-coding RNAs (lncRNAs) are a subgroup of RNAs longer than 200 nucleotides that do not produce proteins and play a variety of roles in a number of biological processes, including innate immune response, 1 cell cycle control, 2 cell differentiation, 3 , 4 X-inactivation, 5 , 6 and neuronal activity. 7 , 8 Because lncRNAs do not produce proteins, it is important for lncRNAs to interact with other molecules, such as other nucleic acids and RNA-binding proteins, to exert their functions. For example, nuclear lncRNAs interact with DNA, chromatin, and proteins and modulate nuclear processes, like chromatin organization, RNA transcription, splicing, and lncRNA nuclear export and retention. In contrast, cytoplasmic lncRNAs interact with proteins, other RNAs, or different organelles to alter mRNA stability and localization, protein translation, post-translational modification, mitochondrial functions, or protein trafficking. 9 – 15 In some cases, the act of transcription, rather than the lncRNA transcripts, is the functional entity to regulate target gene expression locally. 16

Although significant progress has been made in understanding the functions of lncRNAs in recent years, their transcription regulatory mechanism has been poorly understood. Similar to mRNAs which are transcribed by RNA Polymerase II, a considerable number of lncRNAs are also transcribed by RNA Polymerase II, 5′ capped, 3′ polyadenylated, and multi-exonic. 17 Interestingly, however, lncRNAs exhibit more cell type-, tissue-, developmental stage- or disease state-specific expression patterns compared to mRNAs. 18 – 21 This has raised many interesting questions: What are the regulatory mechanisms of lncRNA transcription to achieve highly specific expression patterns? Is the transcription of lncRNAs also regulated by transcription factors which bind to its promoter/enhancer, similar to mRNAs? If so, why is lncRNA expression more specific? Is there a universal mechanism to regulate the transcription of all lncRNAs, or is the transcription of different lncRNAs regulated by different mechanisms? Does the expression of lncRNAs respond to external inputs similar to mRNAs?

We have recently shown that Per2AS, a lncRNA, plays an important role in regulating circadian rhythms, 22 an internal timing mechanism to anticipate and respond to daily environmental rhythms driven by the rotation of the Earth. Interestingly, Per2AS is transcribed from the antisense strand of Period 2 ( Per2), one of the core clock genes essential for generating circadian rhythmicity, and its expression is rhythmic and antiphasic to Per2 mRNA. 23 – 26 Most strikingly, we further demonstrated that the transcription of Per2AS, rather than its transcripts, is important for regulating circadian rhythms. 22 These data prompted us to interrogate how rhythmic Per2AS transcription is regulated.

In this study, we analyzed the expression patterns of Per2AS under conditions in which the circadian clock was perturbed either genetically or environmentally. We used publicly available circadian transcriptomic datasets from mouse liver, where Per2AS is abundantly and rhythmically expressed. 23 – 26 Our first hypothesis is that the rhythmic Per2AS transcription is regulated by rhythmic antiphasic transcription of Per2 by means of transcriptional interference. 27 – 29 Our alternative hypothesis is that rhythmic Per2AS expression is driven by its own promoter, similar to mRNAs. 30 – 33 If the former is true, we anticipate that changes in Per2 and Per2AS expression would always be antiphasic. If the latter, then we anticipate that a change in Per2AS expression would be independent of that of Per2. We also take into account the possibility that the two hypotheses are not mutually exclusive. Results from this study contribute to our mechanistic understanding of how circadian rhythm is regulated by Per2AS and, more broadly, how the transcription of antisense lncRNAs is regulated.

 Methods RNA-seq data acquisition and processing

All the fastq files were obtained from NCBI SRA (GSE135898, GSE135875, GSE107787, GSE102072, GSE143528, 34 – 37 except for PRJDB7789, which was obtained from DDBJ DRA (PRJDB7789). 38 Fastq reads were mapped to the Ensembl mouse genome release 38 (mm10) using STAR 2.7.7a 39 with outFilterScoreMinOverLRead = 0.3 and outFilterMatchNMinOverLRead = 0.3 options. The ‘condenseGenes’ option was also used to select the most abundant isoform of each gene. The mapped reads were quantified by HOMER (v 4.11.1) 40 and normalized by the transcripts per million (TPM) option. We used the -sspe option for paired-end reads and the -strand – or -strand + option to quantify mapped reads in a strand-specific manner. The option -bp10 was used for GSE102072 and PRJDB7789 to filter alignments with mapQ smaller than 10. Per2AS expression was calculated from the antisense strand of the Per2 genomic region as it is not annotated in Mus musculus GRCm38.95 GTF or NCBI RefSeq mm10 GTF files.

 Statistical analyses

We used the two-way analysis of variance (ANOVA) in Microsoft Excel to test for differences in RNA levels between the experimental groups (i.e., genotype, diet), except for the dataset GSE102072 in which some samples had only one biological replica. The rhythmicity of each RNA expression was assessed by MetaCycle, 41 which integrates three algorithms, ARSER, JTK CYCLE, and Lomb-Scargle, to determine the p-value, Benjamini-Hochberg q-value (BH.Q value), period, phase, baseline value, amplitude (AMP), and relative amplitude (rAMP). We defined the expression of an RNA as rhythmic when meta2d p<0.05.

 Results Core clock genes <italic toggle="yes">Bmal1</italic>, <italic toggle="yes">Cry1/2</italic>, and <italic toggle="yes">Nr1d1/2</italic> affect the expression patterns of <italic toggle="yes">Per2AS</italic> and <italic toggle="yes">Per2</italic>

To test whether any of the core clock genes have an effect on the expression of Per2AS and Per2, we analyzed the circadian transcriptomic datasets from mice livers lacking one or more “core” clock genes ( Figure 1). Circadian rhythms in each cell are driven by a cell-autonomous molecular clock, composed of a group of core clock genes that form a network of transcriptional–translational feedback loops (TTFLs) 42 and regulate daily rhythms in biochemistry, physiology, and behavior. In the first loop, transcription activators BMAL1 (gene name: Arntl) and CLOCK form a heterodimeric complex and bind to the E-box sequence in the promoter regions to regulate the rhythmic expression of their target genes, including Per1-3 and Cryptochrome ( Cry)1-2. High levels of PER and CRY proteins form a protein complex in the cytoplasm, then translocate back to the nucleus to inhibit its own transcription by interacting with CLOCK/BMAL1. This cycle takes approximately 24 hours, and this is the molecular basis of generating circadian rhythms. The secondary feedback loop is comprised of the transcriptional repressors REV-ERBs (gene name: Nr1d) and the activator RORs that are regulated by BMAL1/CLOCK. REV-ERB and ROR proteins both recognize and bind to the RORE sequence in the promoter region and compete with each other to drive the rhythmicity of the target gene expression, including Bmal1. The last loop consists of transcription activators of proline and acidic amino acid-rich basic leucine zipper (PAR bZip) proteins: DBP, TEF, and HLF, and the repressor NFIL3, all of which target genes containing D-box element within their promoters, including Rev-erbs, Rors, and Pers. 42

The expression patterns of <italic toggle="yes">Per2AS</italic> and <italic toggle="yes">Per2</italic> in the knockout mice of core clock genes.

Mice were kept under an L:D=12:12 cycle and fed ad libitum. Liver RNAs were extracted every four hours in (A) Bmal1, (B) Nfil3, (C) Cry1/2 double, and (E) Dbp/Tef/Hlf triple knockout mice (n=2), or every three hours in (D) Nr1d1/2 double knockout mice (n=3). Y-axis represents strand-specific TPM. Points and error bars represent mean±SE. (**) p<0.01, (***), p<0.001, N.S. indicates no significant difference between genotypes (two-way ANOVA). Data derived from Weger et al., Proc Natl Acad Sci U S A, 2020 34 ; Yoshitane et al., Commun Biol., 2019 38 ; Guan et al., Science, 2020. 37

Removal of Bmal1 from the TTFL results in arrhythmic locomotor activity and disrupts circadian expression of hepatic core clock genes. 43 , 44 Similarly, removal of both Cry1 and 2 also results in the loss of the rhythm in locomotor activities and core clock gene expression. 45 , 46 In both Bmal1 knock-out (KO) and Cry1/2 double knock-out (DKO) animals, our analysis demonstrated that the rhythmicity of Per2 was abolished but expression was maintained at intermediate levels, as was reported previously 34 , 47 , 48 ( Figure 1A, C, Table S1 84 ). Whereas the expression of Per2AS was markedly low and arrhythmic, indicating that Bmal1 and Cry1/2 activate the expression of Per2AS ( Figure 1A, C, Table S1 84 ).

In contrast, the removal of Nfil3 or PAR bZip genes ( Dbp/Tef/Hlf) did not change the expression pattern of Per2 ( Figure 1B, E). This is in line with previous findings that the rhythmic expression of core clock genes was nearly identical between wild-type (WT) and Nfil3 or PAR bZip-deficient mice in liver. 38 , 49 Interestingly, however, the amplitude of Per2AS increased in both Nfil3 KO and Dbp/Tef/Hlf triple knock-out (TKO) mice, indicating that the NFIL3 and PAR bZip proteins repress the expression of Per2AS without affecting Per2 ( Figure 1B, E, Table S1 84 ).

Removal of both Nr1d1 and Nr1d2 abolishes rhythmic locomotor activity and interferes with the circadian expression of many hepatic core clock genes, including dampening the rhythm of Per2. 50 Consistent with this, the amplitude of Per2 expression was decreased in Nr1d1/2 DKO mice while the relative amplitude of Per2AS expression was increased ( Figure 1D, Table S1 84 ). These data suggest that Nr1d1/2 have an opposing effect on Per2 and Per2AS, activating Per2 while repressing Per2AS. We also confirmed the genotypes of each dataset by checking the mRNA levels of the knock-out genes, all of which were significantly decreased ( Figures 1, 3, 4).

 24-hr fasting decreased the amplitude of the rhythmic expression of <italic toggle="yes">Per2</italic>, but not that of <italic toggle="yes">Per2AS</italic>

We next tested the effect of environmental perturbation on the expression patterns of Per2AS and Per2. In mammals, light is the most potent environmental cue that entrains the circadian clock through the suprachiasmatic nucleus (SCN) of the hypothalamus. However, food intake also serves as a strong ‘Zeitgeber’ (time giver) to entrain the circadian clock of peripheral organs in an SCN-independent manner. 51 – 54 Previous studies have demonstrated that fasting directly affects a large number of physiological parameters, such as body temperature, 55 , 56 body weight, 57 , 58 hormone levels, 59 , 60 and hepatic glucose levels, 61 , 62 in addition to the expression patterns of the core clock genes in mouse liver.

In particular, the expression levels of BMAL1-target genes, including Per2, are lower in the liver of fasting mice. 35 , 63 – 67 In line with this, the amplitude of Per2 expression was considerably lower in the fasted mice ( Figure 2, Table S1 84 ) whose liver samples were collected after 24 hours of fasting for each time point, compared to the mice fed under the ad libitum (ad lib) condition. In contrast, the expression patterns of Per2AS show little or no difference between the ad lib fed and 24-hr fasting conditions ( Figure 2, Table S1 84 ). These data indicate that the 24-hr fasting alters the expression pattern of Per2, but not that of Per2AS.

The expression patterns of <italic toggle="yes">Per2AS</italic> and <italic toggle="yes">Per2</italic> under <italic toggle="yes">ad libitum</italic> or 24-hr fasting conditions.

Mice were kept under an L:D=12:12 cycle and either fed ad libitum or food was removed 24 hours prior to tissue sampling for the fasting group of mice. Liver RNAs were extracted every four hours. Points and error bars represent mean±SE (n=3). (***) p<0.001, N.S. indicates no significant difference between feeding conditions (two-way ANOVA). Data derived from Kinouchi et al., Cell Rep., 2018. 35

<italic toggle="yes">Bmal1</italic> and <italic toggle="yes">Cry1/2</italic> regulate <italic toggle="yes">Per2AS</italic> expression even under the time-restricted feeding condition

Time-restricted feeding (TRF) is a form of intermittent fasting in which food consumption is restricted to a certain time window of the day. The TRF regimen can protect mice from excessive body weight gain and liver damage depending on the timing of the food availability, and also improve metabolic and physiological rhythms when food access is restricted to the active phase. 68 – 71 Here, we also looked at the differences in Per2AS and Per2 expression patterns when food access was restricted to nighttime (i.e., the active phase of mice) to understand whether core clock genes still have the same effect on the expression of Per2AS and Per2. Similar to what was observed with the ad lib feeding condition ( Figure 1A, C), the expression of Per2AS was very low and arrhythmic in Bmal1 KO and Cry1/2 DKO animals experiencing TRF compared to WT ( Figure 3, Table S1 84 ). The expression of Per2 was also arrhythmic but still maintained intermediate levels both in Bmal1 KO and Cry1/2 DKO animals even under TRF ( Figure 3, Table S1 84 ). Compared to ad lib feeding, TRF did not affect the expression patterns of Per2AS and Per2, showing that the core clock gene KO is the primary factor to alter the expressions of Per2AS and Per2. Additionally, these findings further support that BMAL1 and CRY are crucial transcription factors to activate Per2AS expression, regardless of the feeding patterns.

The expression patterns of <italic toggle="yes">Per2AS</italic> and <italic toggle="yes">Per2</italic> with time-restricted feeding (TRF).

Mice were kept under an L:D=12:12 cycle and access to food were restricted between ZT12 to ZT24. Liver RNAs were extracted every four hours in (A) Bmal1 and (B) Cry1/2 double knockout mice. Y-axis represents strand-specific TPM. Points and error bars represent mean±SE (n=2). (**) p<0.01, (***) p<0.001, N.S. indicates no significant difference between genotypes (two-way ANOVA). Data derived from Weger et al., Proc Natl Acad Sci U S A, 2020. 34

Core clock genes <italic toggle="yes">Bmal1</italic>, <italic toggle="yes">Cry1/2</italic>, and <italic toggle="yes">Nr1d1/</italic>2 knockout abolished the expression of <italic toggle="yes">Per2AS</italic> and dampen the rhythm of <italic toggle="yes">Per2</italic> when feeding with HFD

In addition to the timing of diet, the composition of the diet, such as high-fat or ketogenic, also affects circadian rhythms and alters core clock gene expression. 68 , 72 – 75 To understand whether different diet compositions affect the expression of Per2AS and Per2 differently, we also examined the expression levels of Per2AS and Per2 when mice were fed with a 60% high-fat diet (HFD) either ad lib or TRF during the active phase (ZT13-22 or 23). 36

The expression of Per2 was arrhythmic in WT fed with HFD under the ad lib condition ( Figure 4A, Table S1 84 ), as was reported previously. 73 However, it was rhythmic under the TRF condition ( Figure 4D, Table S1 84 ), supporting the idea that TRF restores peripheral oscillations of core clock gene expressions. 76 , 77 Similar to the results observed with a regular chow diet under the ad lib condition ( Figure 1), Per2 expression was arrhythmic but maintained intermediate expression in both ad lib and TRF conditions in Bmal1 KO and Cry1/2 DKO ( Figure 4, Table S1 84 ), but not in Nr1d1/2 DKO mice with HFD ( Figure 4C, F, Table S1 84 ). In contrast, Per2AS levels were low and arrhythmic in both ad lib and TRF conditions in Bmal1 KO and Cry1/2 DKO mice compared to WT mice with HFD ( Figure 4A- B, D- E), similar to what was observed in regular chow mice under both ad lib and TRF conditions ( Figure 1A, C; Figure 3). This further supports the idea that Per2AS expression is regulated by Bmal1 and Cry1/2, regardless of the composition of the diet. On the other hand, Per2AS expression was arrhythmic in Nr1d1/2 DKO mice with HFD in contrast to the regular chow diet under ad lib condition ( Figure 1D), suggesting that the effect of Nr1d1/2 on Per2AS is different between regular chow and HFD ( Figure 4C, F; Table S1 84 ).

The expression patterns of <italic toggle="yes">Per2AS</italic> and <italic toggle="yes">Per2</italic> with high-fat diet (HFD).

Mice were kept under an L:D=12:12 cycle and fed with a 60% high-fat diet either ad lib (A-C) or with TRF (D-F) condition. Liver RNAs were extracted every four hours in wild type and (A) Bmal1 knockout mice, or every three hours in (B) Cry1/2 double and (C) Nr1d1/2 double knockout mice. Y-axis represents strand-specific TPM. The same WT mice were used in each group. Points and error bars represent mean±SE (n=2). Two-way ANOVA analysis was not performed for this dataset, because the KO samples consisted of a single biological replicate. Data derived from Chaix et al., Cell Metab., 2019. 36

Discussion

In this study, we focused on the circadian antisense lncRNA, Per2AS, and asked whether the rhythmic expression of Per2AS is regulated independently by its own promoter like many other circadian mRNAs, or transcriptional interference driven by the antiphasic expression of its sense-strand gene Per2. By using circadian transcriptomic datasets from mouse liver, in which the molecular clock machinery was genetically and environmentally disrupted, we examined how these perturbations affect the expression patterns of Per2AS and Per2. Our data demonstrated that the expression of Per2AS can be altered by both genetic and environmental perturbations of the circadian clock.

We were able to test the effect of Bmal1, Nfil3, Cry1/2, Nr1d1/2, and Dbp/Tef/Hlf, but not that of Rora/b/c as the circadian transcriptome dataset for Rora/c KO mouse liver is only available with a microarray platform and this does not allow us to quantify Per2AS levels. 78 We also tested the effect of fasting, TRF, and an HFD on the expression patterns of Per2AS and Per2. Even though the expression of many core clock genes, including Per2, is affected by fasting ( Figure 2) or HFD ( Figure 4), 35 , 36 , 63 – 67 Per2AS appears to be less sensitive to these changes and its expression remained rhythmic under fasting, TRF, or HFD conditions ( Figures 2- 4). These data suggest that the timing of food intake, the composition of the diet, and fasting are not the primary factor that regulates the rhythmic expression of Per2AS. By contrast, the effect of genetic perturbation within the circadian rhythm system is stronger than environmental perturbation on the expression Per2AS.

We tested our hypotheses for Per2AS transcription regulation by comparing the changes in the expression pattern between Per2AS and Per2. We found that, in some cases, the Per2AS expression pattern was altered even when that of Per2 was unaltered ( Figure 1B, E). We also found, in other cases, that the Per2AS expression pattern was unaltered even when that of Per2 was altered ( Figure 2). These data strongly support the independent hypothesis that the expression of Per2AS is regulated by its own promoter, and its transcription is independent from that of Per2. Indeed, the expression of many lncRNAs is controlled by their promoter or other DNA elements, such as enhancers. 30 , 79 The majority of lncRNAs also contain highly conserved core promoter sequences and can be regulated by different transcription factors. 80 , 81 In addition, the promoters of lncRNAs are evolutionarily conserved as much as that of mRNAs at least between humans and mice, even though their nucleotide sequences are less conserved than mRNAs. These data support the significance of promoter sequences in regulating lncRNA expression patterns. 82 Our data also indicate that BMAL1 and CRY1/2 are the activators, and NFIL3 and PAR bZip proteins are the repressors of Per2AS ( Figure 1, 3). However, the removal of one particular core clock gene may alter the expression of other core clock genes or their downstream genes. 34 , 83 Thus, we cannot eliminate the possibility that the Per2AS expression is indirectly impacted by the change of the core clock gene circuit.

Although these data strongly support the independent hypothesis, we cannot completely reject the alternative hypothesis that Per2AS expression is regulated by transcriptional interference from Per2, since there are some instances where the changes in the expression pattern of Per2AS and Per2 can still be explained by the transcriptional interference hypothesis. For example, in Bmal1 KO and Cry1/2 DKO animals under any dietary conditions (i.e., ad lib, TRF, and HFD feeding), the Per2AS expression was completely abolished, and the Per2 expression pattern was also arrhythmic but stayed at the intermediate levels ( Figures 1A, C, 3, 4D, E, Table S1 84 ). It is possible that the constant Per2 expression prevents the transcription of Per2AS on the other strand, leading to the decreased and arhythmic Per2AS expression. Additionally, in Nr1d1/2 DKO mice, Per2AS expression increased and its rhythmicity became more robust, while Per2 expression decreased and its rhythmicity was dampened ( Figure 1D). This could be due to decreased Per2 transcription leading to the increased Per2AS transcription on the other strand. Therefore, the two alternative hypotheses can both be viable and coordinated together to regulate the rhythmic transcription of Per2AS. At the same time, these data can also be explained solely by the independent hypothesis, and additional experimental evidence will be required to distinguish these possibilities. For example, it would be helpful to understand whether these core clock proteins are indeed recruited to the Per2AS promoter or enhancer sequences. We can also modify the transcription of Per2 directly and test whether this would lead to changes in Per2AS expression patterns.

Regardless, these results help us better understand not only how the transcription of Per2AS is regulated, but also how Per2AS interacts with other core clock proteins in transcriptional-translational feedback loops to regulate circadian rhythms because the rhythmic transcription of Per2AS is important for its functions in regulating circadian rhythms. More broadly, our results also help us understand the transcription regulation mechanism of antisense lncRNAs.

 Data availability Underlying data

NCBI GEO: Temporal profiles of gene expression in Cry1/2 KO, Bmal1 KO under night restricted feeding and ad libitum feeding regimen. Accession number: GSE135898, https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE135898.

NCBI GEO: Temporal profiles of hepatic gene expression in PAR bZip triple knockout mice. Accession number: GSE135875, https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE135875.

NCBI GEO: Fasting Imparts a Switch to Alternative Circadian Transcriptional Pathways in Liver and Muscle. Accession number: GSE107787, https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE107787.

NCBI GEO: Hepatic transcriptome by Next Generation Sequencing of WT and clock mutant mice fed a HFD ad libitum or time-restricted feeding. Accession number: GSE102072, https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE102072.

NCBI GEO: The Hepatocyte Clock and Feeding Interdependently Control Chrono-Homeostasis of Multiple Liver Cell Types (RNA-seq). Accession number: GSE143524, https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE143524.

NCBI BioProject: Transcriptome of mouse liver in Per2::Luc KI and E4bp4 KO/ Per2::Luc KI mice. Accession number: PRJDB7789, https://www.ncbi.nlm.nih.gov/bioproject?term=PRJDB7789&cmd=DetailsSearch.

 Extended data

figshare: Table_S1.xlsx. https://doi.org/10.6084/m9.figshare.21067537. 84

This project contains the following underlying data:

Table S1. xlsx (Metacycle analysis to determine whether the gene expression is rhythmic)

Data are available under the terms of the Creative Commons Attribution 4.0 International license (CC-BY 4.0).

 Acknowledgments

The authors thank Dr. Nobuya Koike from the Kyoto Prefectural University of Medicine for technical assistance and the members of the Kojima laboratory for critical proofreading of the manuscript.

 References Ouyang J Hu J Chen JL : lncRNAs regulate the innate immune response to viral infection. Wiley Interdiscip. Rev. RNA. 2016 Jan-Feb;7(1):129–143. Epub 20151215. eng. 26667656 10.1002/wrna.1321 PMC7169827 Zhou Z Tan F Pei Q : lncRNA SNHG4 modulates colorectal cancer cell cycle and cell proliferation through regulating miR-590-3p/CDK1 axis. Aging (Albany NY). 2021 Mar 19;13(7):9838–9858. Epub 20210319. eng. 33744866 10.18632/aging.202737 PMC8064176 Feng J Bi C Clark BS : The Evf-2 noncoding RNA is transcribed from the Dlx-5/6 ultraconserved region and functions as a Dlx-2 transcriptional coactivator. Genes Dev. 2006 Jun 1;20(11):1470–1484. Epub 20060516. eng. 16705037 10.1101/gad.1416106 PMC1475760 Omote N Sakamoto K Li Q : Long noncoding RNA TINCR is a novel regulator of human bronchial epithelial cell differentiation state. Physiol. Rep. 2021 Feb;9(3):e14727. eng. 33527707 10.14814/phy2.14727 PMC7851438 Plath K Fang J Mlynarczyk-Evans SK : Role of histone H3 lysine 27 methylation in X inactivation. Science. 2003 Apr 4;300(5616):131–135. Epub 20030320. eng. 12649488 10.1126/science.1084274 Ohhata T Hoki Y Sasaki H : Crucial role of antisense transcription across the Xist promoter in Tsix-mediated Xist chromatin modification. Development. 2008 Jan;135(2):227–235. Epub 20071205. eng. 18057104 10.1242/dev.008490 Barry G Briggs JA Hwang DW : The long non-coding RNA NEAT1 is responsive to neuronal activity and is associated with hyperexcitability states. Sci. Rep. 2017 Jan 5;7:40127. Epub 20170105. eng. 28054653 10.1038/srep40127 PMC5214838 Mercer TR Qureshi IA Gokhan S : Long noncoding RNAs in neuronal-glial fate specification and oligodendrocyte lineage maturation. BMC Neurosci. 2010 Feb 5;11:14. Epub 20100205. eng. 20137068 10.1186/1471-2202-11-14 PMC2829031 Bridges MC Daulagala AC Kourtidis A : LNCcation: lncRNA localization and function. J. Cell Biol. 2021 Feb 1;220(2). eng. 33464299 10.1083/jcb.202009045 PMC7816648 Carlevaro-Fita J Polidori T Das M : Ancient exapted transposable elements promote nuclear enrichment of human long noncoding RNAs. Genome Res. 2019 Feb;29(2):208–222. Epub 20181226. eng. 30587508 10.1101/gr.229922.117 PMC6360812 Fazal FM Han S Parker KR : Atlas of Subcellular RNA Localization Revealed by APEX-Seq. Cell. 2019 Jul 11;178(2):473–90.e26. Epub 20190620. eng. 31230715 10.1016/j.cell.2019.05.027 PMC6786773 Herman AB Tsitsipatis D Gorospe M : Integrated lncRNA function upon genomic and epigenomic regulation. Mol. Cell. 2022 Jun 16;82(12):2252–2266. eng. 35714586 10.1016/j.molcel.2022.05.027 PMC9219586 Rackham O Shearwood AM Mercer TR : Long noncoding RNAs are generated from the mitochondrial genome and regulated by nuclear-encoded proteins. RNA. 2011 Dec;17(12):2085–2093. Epub 20111025. eng. 22028365 10.1261/rna.029405.111 PMC3222122 Zhang Y Li B Bai Q : The lncRNA Snhg1-Vps13D vesicle trafficking system promotes memory CD8 T cell establishment via regulating the dual effects of IL-7 signaling. Signal Transduct. Target. Ther. 2021 Mar 24;6(1):126. Epub 20210324. eng. 33758164 10.1038/s41392-021-00492-9 PMC7987995 Lyu Q Xu S Lyu Y : SENCR stabilizes vascular endothelial cell adherens junctions through interaction with CKAP4. Proc. Natl. Acad. Sci. U. S. A. 2019 Jan 8;116(2):546–555. Epub 20181224. eng. 30584103 10.1073/pnas.1810729116 PMC6329947 Statello L Guo CJ Chen LL : Gene regulation by long non-coding RNAs and its biological functions. Nat. Rev. Mol. Cell Biol. 2021 Feb;22(2):96–118. Epub 20201222. eng. 33353982 10.1038/s41580-020-00315-9 PMC7754182 Guttman M Amit I Garber M : Chromatin signature reveals over a thousand highly conserved large non-coding RNAs in mammals. Nature. 2009 Mar 12;458(7235):223–227. 19182780 10.1038/nature07672 PMC2754849 Quinn JJ Chang HY : Unique features of long non-coding RNA biogenesis and function. Nat. Rev. Genet. 2016 Jan;17(1):47–62. 26666209eng 10.1038/nrg.2015.10 Derrien T Johnson R Bussotti G : The GENCODE v7 catalog of human long noncoding RNAs: analysis of their gene structure, evolution, and expression. Genome Res. 2012 Sep;22(9):1775–1789. eng. 22955988 10.1101/gr.132159.111 PMC3431493 Cabili MN Trapnell C Goff L : Integrative annotation of human large intergenic noncoding RNAs reveals global properties and specific subclasses. Genes Dev. 2011 Sep 15;25(18):1915–1927. Epub 20110902. eng. 21890647 10.1101/gad.17446611 PMC3185964 Mercer TR Dinger ME Sunkin SM : Specific expression of long noncoding RNAs in the mouse brain. Proc. Natl. Acad. Sci. U. S. A. 2008 Jan 15;105(2):716–721. Epub 20080109. eng. 18184812 10.1073/pnas.0706729105 PMC2206602 Mosig RA Castaneda AN Deslauriers JC : Natural antisense transcript of Period2, Per2AS, regulates the amplitude of the mouse circadian clock. Genes Dev. 2021 Jun;35(11-12):899–913. Epub 20210520. eng. 34016691 10.1101/gad.343541.120 PMC8168560 Koike N Yoo SH Huang HC : Transcriptional architecture and chromatin landscape of the core circadian clock in mammals. Science. 2012 Oct 19;338(6105):349–354. Epub 20120830. eng. 22936566 10.1126/science.1226339 PMC3694775 Vollmers C Schmitz RJ Nathanson J : Circadian oscillations of protein-coding and regulatory RNAs in a highly dynamic mammalian liver epigenome. Cell Metab. 2012 Dec 5;16(6):833–845. eng. 23217262 10.1016/j.cmet.2012.11.004 PMC3541940 Menet JS Rodriguez J Abruzzi KC : Nascent-Seq reveals novel features of mouse circadian transcriptional regulation. elife. 2012 Nov 13;1:e00011. Epub 20121113. eng. 23150795 10.7554/eLife.00011 PMC3492862 Zhang R Lahens NF Ballance HI : A circadian gene expression atlas in mammals: implications for biology and medicine. Proc. Natl. Acad. Sci. U. S. A. 2014 Nov 11;111(45):16219–16224. Epub 20141027. eng. 25349387 10.1073/pnas.1408886111 PMC4234565 Shearwin KE Callen BP Egan JB : Transcriptional interference--a crash course. Trends Genet. 2005 Jun;21(6):339–345. eng. 15922833 10.1016/j.tig.2005.04.009 PMC2941638 Xue Z Ye Q Anson SR : Transcriptional interference by antisense RNA is required for circadian clock function. Nature. 2014 Oct 30;514(7524):650–653. Epub 20140817. eng. 25132551 10.1038/nature13671 PMC4214883 Georg J Hess WR : cis-antisense RNA, another level of gene regulation in bacteria. Microbiol Mol Biol Rev. 2011 Jun;75(2):286–300. eng. 21646430 10.1128/MMBR.00032-10 PMC3122628 Mattioli K Volders PJ Gerhardinger C : High-throughput functional analysis of lncRNA core promoters elucidates rules governing tissue specificity. Genome Res. 2019 Mar;29(3):344–355. Epub 20190125. eng. 30683753 10.1101/gr.242222.118 PMC6396428 Brantl S Wagner EG : An antisense RNA-mediated transcriptional attenuation mechanism functions in Escherichia coli. J. Bacteriol. 2002 May;184(10):2740–2747. eng. 11976303 10.1128/JB.184.10.2740-2747.2002 PMC135013 Brantl S : Regulatory mechanisms employed by cis-encoded antisense RNAs. Curr. Opin. Microbiol. 2007 Apr;10(2):102–109. Epub 20070326. eng. 17387036 10.1016/j.mib.2007.03.012 Brophy JA Voigt CA : Antisense transcription as a tool to tune gene expression. Mol. Syst. Biol. 2016 Jan 14;12(1):854. Epub 20160114. eng. 26769567 10.15252/msb.20156540 PMC4731013 Weger BD Gobet C David FPA : Systematic analysis of differential rhythmic liver gene expression mediated by the circadian clock and feeding rhythms. Proc. Natl. Acad. Sci. U. S. A. 2021 Jan 19;118(3). pii: e2015803118. eng. 33452134 doi: 10.1073/pnas.2015803118 PMC7826335 Kinouchi K Magnan C Ceglia N : Fasting Imparts a Switch to Alternative Daily Pathways in Liver and Muscle. Cell Rep. 2018 Dec 18;25(12):3299–314.e6. eng. 30566858 10.1016/j.celrep.2018.11.077 PMC6433478 Chaix A Lin T Le HD : Time-Restricted Feeding Prevents Obesity and Metabolic Syndrome in Mice Lacking a Circadian Clock. Cell Metab. 2019 Feb 5;29(2):303–19.e4. Epub 20180830. eng. 30174302 10.1016/j.cmet.2018.08.004 PMC7751278 Guan D Xiong Y Trinh TM : The hepatocyte clock and feeding control chronophysiology of multiple liver cell types. Science. 2020 Sep 11;369(6509):1388–1394. Epub 20200730. eng. 32732282 10.1126/science.aba8984 PMC7849028 Yoshitane H Asano Y Sagami A : Functional D-box sequences reset the circadian clock and drive mRNA rhythms. Commun. Biol. 2019;2:300. Epub 20190808. eng. 31428688 10.1038/s42003-019-0522-3 PMC6687812 Dobin A Davis CA Schlesinger F : STAR: ultrafast universal RNA-seq aligner. Bioinformatics. 2013 Jan 1;29(1):15–21. Epub 20121025. eng. 23104886 10.1093/bioinformatics/bts635 PMC3530905 Heinz S Benner C Spann N : Simple combinations of lineage-determining transcription factors prime cis-regulatory elements required for macrophage and B cell identities. Mol. Cell. 2010 May 28;38(4):576–589. eng. 20513432 10.1016/j.molcel.2010.05.004 PMC2898526 Wu YW Simmons BA Singer SW : MaxBin 2.0: an automated binning algorithm to recover genomes from multiple metagenomic datasets. (1367-4811 (Electronic)). eng. Takahashi JS : Transcriptional architecture of the mammalian circadian clock. Nat. Rev. Genet. 2017 Mar;18(3):164–179. Epub 20161219. eng. 27990019 10.1038/nrg.2016.150 PMC5501165 Bunger MK Wilsbacher LD Moran SM : Mop3 is an essential component of the master circadian pacemaker in mammals. Cell. 2000 Dec 22;103(7):1009–1017. eng. 11163178 10.1016/S0092-8674(00)00205-1 PMC3779439 Kondratov RV Shamanna RK Kondratova AA : Dual role of the CLOCK/BMAL1 circadian complex in transcriptional regulation. FASEB J. 2006 Mar;20(3):530–532. Epub 20060125. eng. 16507766 10.1096/fj.05-5321fje Horst GT van der Muijtjens M Kobayashi K : Mammalian Cry1 and Cry2 are essential for maintenance of circadian rhythms. Nature. 1999 Apr 15;398(6728):627–630. eng. 10217146 10.1038/19323 Oishi K Miyazaki K Kadota K : Genome-wide expression analysis of mouse liver reveals CLOCK-regulated circadian output genes. J. Biol. Chem. 2003 Oct 17;278(42):41519–41527. Epub 20030715. eng. 12865428 10.1074/jbc.M304564200 Zhou B Zhang Y Zhang F : CLOCK/BMAL1 regulates circadian change of mouse hepatic insulin sensitivity by SIRT1. Hepatology. 2014 Jun;59(6):2196–2206. Epub 20140425. eng. 24442997 10.1002/hep.26992 Koronowski KB Kinouchi K Welz PS : Defining the Independence of the Liver Circadian Clock. Cell. 2019 May 30;177(6):1448–62.e14. eng. 31150621 10.1016/j.cell.2019.04.025 PMC6813833 Gachon F Fonjallaz P Damiola F : The loss of circadian PAR bZip transcription factors results in epilepsy. Genes Dev. 2004 Jun 15;18(12):1397–1412. Epub 20040602. eng. 15175240 10.1101/gad.301404 PMC423191 Cho H Zhao X Hatori M : Regulation of circadian behaviour and metabolism by REV-ERB-α and REV-ERB-β. Nature. 2012 Mar 29;485(7396):123–127. Epub 20120329. eng. 22460952 10.1038/nature11048 PMC3367514 Stokkan KA Yamazaki S Tei H : Entrainment of the circadian clock in the liver by feeding. Science. 2001 Jan 19;291(5503):490–493. eng. 11161204 10.1126/science.291.5503.490 Challet E Caldelas I Graff C : Synchronization of the molecular clockwork by light- and food-related cues in mammals. Biol. Chem. 2003 May;384(5):711–719. eng. 12817467 10.1515/BC.2003.079 Damiola F Le Minh N Preitner N : Restricted feeding uncouples circadian oscillators in peripheral tissues from the central pacemaker in the suprachiasmatic nucleus. Genes Dev. 2000 Dec 1;14(23):2950–2961. eng. 11114885 10.1101/gad.183500 PMC317100 Honma S : The mammalian circadian system: a hierarchical multi-oscillator structure for generating circadian rhythm. J. Physiol. Sci. 2018 May;68(3):207–219. Epub 20180219. eng. 29460036 10.1007/s12576-018-0597-5 Wankhade UD Vella KR Fox DL : Deletion of Nhlh2 results in a defective torpor response and reduced Beta adrenergic receptor expression in adipose tissue. PLoS One. 2010 Aug 23;5(8):e12324. Epub 20100823. eng. 20808804 10.1371/journal.pone.0012324 PMC2925948 Kanizsai P Garami A Solymár M : Energetics of fasting heterothermia in TRPV1-KO and wild type mice. Physiol. Behav. 2009 Jan 8;96(1):149–154. Epub 20081008. eng. 18938188 10.1016/j.physbeh.2008.09.023 Heijboer AC Donga E Voshol PJ : Sixteen hours of fasting differentially affects hepatic and muscle insulin sensitivity in mice. J. Lipid Res. 2005 Mar;46(3):582–588. Epub 20041201. eng 15576835 10.1194/jlr.M400440-JLR200 Furner RL Feller DD : The influence of starvation upon hepatic drug metabolism in rats, mice, and guinea pigs. Proc. Soc. Exp. Biol. Med. 1971 Jul;137(3):816–819. eng. 5560669 10.3181/00379727-137-35674 Moitra J Mason MM Olive M : Life without white fat: a transgenic mouse. Genes Dev. 1998 Oct 15;12(20):3168–3181. eng. 9784492 10.1101/gad.12.20.3168 PMC317213 Gavrilova O Leon LR Marcus-Samuels B : Torpor in mice is induced by both leptin-dependent and -independent mechanisms. Proc. Natl. Acad. Sci. U. S. A. 1999 Dec 7;96(25):14623–14628. eng. 10588755 10.1073/pnas.96.25.14623 PMC24486 Ayala JE Bracy DP McGuinness OP : Considerations in the design of hyperinsulinemic-euglycemic clamps in the conscious mouse. Diabetes. 2006 Feb;55(2):390–397. eng. 16443772 10.2337/diabetes.55.02.06.db05-0686 Strubelt O Dost-Kempf E Siegers CP : The influence of fasting on the susceptibility of mice to hepatotoxic injury. Toxicol. Appl. Pharmacol. 1981 Aug;60(1):66–77. eng. 7281177 10.1016/0041-008X(81)90136-8 Kawamoto T Noshiro M Furukawa M : Effects of fasting and re-feeding on the expression of Dec1, Per1, and other clock-related genes. J. Biochem. 2006 Sep;140(3):401–408. Epub 20060727. eng. 16873396 10.1093/jb/mvj165 Shavlakadze T Anwari T Soffe Z : Impact of fasting on the rhythmic expression of myogenic and metabolic factors in skeletal muscle of adult mice. Am. J. Physiol. Cell Physiol. 2013 Jul 1;305(1):C26–C35. Epub 20130417. eng. 23596176 10.1152/ajpcell.00027.2013 Sun X Dang F Zhang D : Glucagon-CREB/CRTC2 signaling cascade regulates hepatic BMAL1 protein. J. Biol. Chem. 2015 Jan 23;290(4):2189–2197. Epub 20141205. eng. 25480789 10.1074/jbc.M114.612358 PMC4303670 Xie Z Zhang D Chung D : Metabolic Regulation of Gene Expression by Histone Lysine β-Hydroxybutyrylation. Mol. Cell. 2016 Apr 21;62(2):194–206. eng. 27105115 10.1016/j.molcel.2016.03.036 PMC5540445 Pickel L Lee JH Maughan H : Circadian rhythms in metabolic organs and the microbiota during acute fasting in mice. Physiol. Rep. 2022 Jul;10(14):e15393. eng. 35851583 10.14814/phy2.15393 PMC9295129 Hatori M Vollmers C Zarrinpar A : Time-restricted feeding without reducing caloric intake prevents metabolic diseases in mice fed a high-fat diet. Cell Metab. 2012 Jun 6;15(6):848–860. Epub 20120517. eng. 22608008 10.1016/j.cmet.2012.04.019 PMC3491655 Longo VD Panda S : Fasting, Circadian Rhythms, and Time-Restricted Feeding in Healthy Lifespan. Cell Metab. 2016 Jun 14;23(6):1048–1059. eng. 27304506 10.1016/j.cmet.2016.06.001 PMC5388543 Duncan MJ Smith JT Narbaiza J : Restricting feeding to the active phase in middle-aged mice attenuates adverse metabolic effects of a high-fat diet. Physiol. Behav. 2016 Dec;167(167):1–9. Epub 20160830. eng. 27586251 10.1016/j.physbeh.2016.08.027 Jamshed H Beyl RA Della Manna DL : Early Time-Restricted Feeding Improves 24-Hour Glucose Levels and Affects Markers of the Circadian Clock, Aging, and Autophagy in Humans. Nutrients. 2019 May 30;11(6). Epub 20190530. eng. pii: E1234. 31151228 10.3390/nu11061234 PMC6627766 Yanagihara H Ando H Hayashi Y : High-fat feeding exerts minimal effects on rhythmic mRNA expression of clock genes in mouse peripheral tissues. Chronobiol. Int. 2006;23(5):905–914. eng. 17050208 10.1080/07420520600827103 Kohsaka A Laposky AD Ramsey KM : High-fat diet disrupts behavioral and molecular circadian rhythms in mice. Cell Metab. 2007 Nov;6(5):414–421. eng. 17983587 10.1016/j.cmet.2007.09.006 Eckel-Mahan KL Patel VR Mateo S de : Reprogramming of the circadian clock by nutritional challenge. Cell. 2013 Dec 19;155(7):1464–1478. eng. 24360271 10.1016/j.cell.2013.11.034 PMC4573395 Lizier M Bomba L Minuti A : The nutrigenomic investigation of C57BL/6N mice fed a short-term high-fat diet highlights early changes in clock genes expression. Genes Nutr. 2013 Sep;8(5):465–474. Epub 20130416. eng. 23588623 10.1007/s12263-013-0344-8 PMC3755136 Sherman H Genzer Y Cohen R : Timed high-fat diet resets circadian metabolism and prevents obesity. FASEB J. 2012 Aug;26(8):3493–3502. Epub 20120516. eng. 22593546 10.1096/fj.12-208868 Vollmers C Gill S DiTacchio L : Time of feeding and the intrinsic circadian clock drive rhythms in hepatic gene expression. Proc. Natl. Acad. Sci. U. S. A. 2009 Dec 15;106(50):21453–21458. Epub 20091125. eng. 19940241 10.1073/pnas.0909591106 PMC2795502 Zhang Y Papazyan R Damle M : The hepatic circadian clock fine-tunes the lipogenic response to feeding through RORα/γ. Genes Dev. 2017 Jun 15;31(12):1202–1211. Epub 20170726. eng. 28747429 10.1101/gad.302323.117 PMC5558923 Batut PJ Gingeras TR : Conserved noncoding transcription and core promoter regulatory code in early Drosophila development. elife. 2017 Dec 20;6. Epub 20171220. eng. 29260710 10.7554/eLife.29005 PMC5754203 Ali T Grote P : Beyond the RNA-dependent function of LncRNA genes. elife. 2020 Oct 23;9. Epub 20201023. eng. 33095159 10.7554/eLife.60583 PMC7584451 Ransohoff JD Wei Y Khavari PA : The functions and unique features of long intergenic non-coding RNA. Nat. Rev. Mol. Cell Biol. 2018 Mar;19(3):143–157. Epub 20171115. eng. 29138516 10.1038/nrm.2017.104 PMC5889127 Ponjavic J Ponting CP Lunter G : Functionality or transcriptional noise? Evidence for selection within long noncoding RNAs. Genome Res. 2007 May;17(5):556–565. Epub 20070326. eng. 17387145 10.1101/gr.6036807 PMC1855172 Yang G Chen L Grant GR : Timing of expression of the core clock gene Bmal1 influences its effects on aging and survival. Sci. Transl. Med. 2016 Feb 3;8(324):324ra16. Epub 20160203. eng. 26843191 10.1126/scitranslmed.aad3305 PMC4870001 Miao L Batty K Jackson A : Table_S1.xlsx. figshare. [Dataset]. 2022. 10.6084/m9.figshare.21067537.v1 10.5256/f1000research.137958.r151597 Reviewer response for version 1 Cai Yao 1 Referee https://orcid.org/0000-0002-3571-2978 Department of Entomology and Nematology, College of Agricultural and Environmental Sciences, University of California Davis, Davis, CA, USA

Competing interests: No competing interests were disclosed.

 10 10 2022 Copyright: © 2022 Cai Y 2022 This is an open access peer review report distributed under the terms of the Creative Commons Attribution Licence, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. recommendation approve

Miao et al. investigated how genetic and environmental manipulations influence the expression of Per2AS, a circadian long non-coding RNA. Despite the identification of  Per2AS in several circadian transcriptome studies and publication of a functional studies (Mosig et al., 2021, G&D), how Per2AS transcription is regulated remains a major gap in knowledge. A lncRNA transcription can be regulated via transcriptional interference. Alternatively, the promoter of a lncRNA is regulated independent of its sense RNA transcript. To gain insights on the transcriptional regulation of Per2AS, the authors analyzed the expression pattern of Per2AS and Per2 in response to genetic and/or environmental manipulation from publicly available RNA-seq database. Per2AS expression is altered in most core clock gene mutant mice. Whereas Per2AS expression is less sensitive to environmental manipulations including fasting, time-restricted feeding and high-fat diet. They authors suggest both models regarding transcription of the Per2AS lncRNA may be true. (1) In some conditions (e.g. Cry1/2 double knock-out) where Per2 expression is in intermediate level throughout the day, Per2AS level remains in a trough level, supporting the transcriptional interference hypothesis. (2) When Per2 expression remains in the trough level under 24-hour fasting condition, Per2AS level is unaltered, favoring the independent regulation hypothesis. Taken together, this manuscript provides new insights on the transcriptional regulation of Per2AS. Below are some minor suggestions for the authors to improve the manuscript.

Clarify what transcriptional interference (TI) means in the introduction section. Is there a specific TI mechanism you refer to, such as promoter competition, RNAPII collision? I believe this helps readers who are unfamiliar with TI to understand the TI hypothesis.

Transcripts of Per2AS seems to be a few fold higher than Per2 in any given conditions (e.g., Figure 1A). Is this relevant to the two hypotheses? Can the authors comment on this phenomenon?

Can the authors speculate whether Per2AS is more sensitive to signals from the central clock than the peripheral liver clock? It is interesting that Per2AS is unaltered under 24-hour fasting condition (Figure 2), despite the dampening of Per2 transcripts. Maybe analyzing Per2AS level in the liver of SCN lesioned mice or mice kept under constant light will be informative.

To date, there are no data that can solidly differentiate whether BMAL1 and CRY1/2 activates or depresses Per2AS transcription. For this reason, I think the word “promote” is better than “activate” when describing upregulation of Per2AS expression (e.g., the sentence that cites Figure 1A, C, Table S1). This will include both possibilities.

I have difficulty understanding the first sentence of the last paragraph in the Discussion section. What do the authors mean by “interacts with other core clock proteins”? To my understanding, this manuscript investigates how Per2AS transcription is influenced upon genetic manipulation on core clock genes, instead of how Per2AS regulates core clock proteins.

Is the work clearly and accurately presented and does it cite the current literature?

Yes

If applicable, is the statistical analysis and its interpretation appropriate?

Yes

Are all the source data underlying the results available to ensure full reproducibility?

Yes

Is the study design appropriate and is the work technically sound?

Yes

Are the conclusions drawn adequately supported by the results?

Yes

Are sufficient details of methods and analysis provided to allow replication by others?

Yes

Reviewer Expertise:

Circadian clock, phosphorylation, Drosophila genetics

I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard.

Kojima Shihoko

Competing interests: NA

 26 10 2022

1. We agree that the explanation of transcriptional interference (TI) will help readers better understand our hypothesis. We supplemented this information, and it now reads: “Our first hypothesis is that the rhythmic Per2AS transcription is regulated by rhythmic antiphasic transcription of Per2 by means of transcriptional interference, in which the transcription process on one strand suppresses the transcription process of the other strand. 27-29” We currently do not have any specific mechanism in mind and did not mention any.

2. The reviewer is correct that the strand-specific TPM (transcripts per million) for Per2AS is generally higher than that of Per2. This is because we had to separately calculate the strand-specific TPM on each strand. Since the total number of raw read counts is different in each strand, the expression level of Per2 and Per2AS cannot be directly compared with each other. This is not relevant to the two hypotheses.

3. We have shown that Per2AS is sensitive to the local liver clock as well as environmental perturbation triggered by changes in diet (Figs 1-4). It is unclear whether Per2AS is sensitive to the central clock (i.e., SCN). To our knowledge, there is no circadian transcriptome dataset in liver from SCN-lesioned mice.

4. We agree with the reviewer that the word “promote” is more accurate than “activate”. We have now changed these, as suggested. 1) “Whereas the expression of Per2AS was markedly low and arrhythmic, indicating that Bmal1 and Cry1/2 promote the expression of Per2AS (Figure 1A, C, Table S1 84).” 2) “Additionally, these findings further support that BMAL1 and CRY are crucial transcription factors to promote Per2AS expression, regardless of the feeding patterns. ”

5. We apologize for the confusion. We revised this sentence, which now reads:

“Regardless, these results help us better understand not only how the transcription of Per2AS is regulated, but also how Per2AS is wired with other core clock proteins in transcriptional-translational feedback loops to regulate circadian rhythms because the rhythmic transcription of Per2AS is important for its functions in regulating circadian rhythms.”

 10.5256/f1000research.137958.r150976 Reviewer response for version 1 Yoshitane Hikari 1 2 Referee https://orcid.org/0000-0001-6319-3354 Circadian Clock Project, Tokyo Metropolitan Institute of Medical Science, Tokyo, Japan Department of Biological Sciences, School of Science, The University of Tokyo, Tokyo, Japan

Competing interests: No competing interests were disclosed.

 21 9 2022 Copyright: © 2022 Yoshitane H 2022 This is an open access peer review report distributed under the terms of the Creative Commons Attribution Licence, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. recommendation approve

Dr. Shihoko Kojima and their colleagues focused on the molecular basis of the rhythmic expression pattern of a circadian long non-coding RNA, PerAS, in the mouse liver. PerAS gene is located in the genomic locus of one of the core clock gene, Per2, and is transcribed as an antisense strand RNA. Intriguingly, PerAS shows a clear circadian rhythm of its RNA expression in the mouse liver, and the rhythmic pattern is almost anti-phasic to that of Per2 mRNA. Many chronobiologists, including me, predict that these genes probably repress each other's transcription, resulting in their anti-phasic RNA rhythms. In this study, however, the authors clearly demonstrated that Per2AS expression and Per2 mRNA expression are independently controlled by unknown mechanisms that will be revealed in the near future. They obtained a series of RNA-Seq datasets in this field to analyze RNA expression patterns of Per2AS in the mouse livers that lack core clock genes or under altered food conditions. The results are clear and the conclusion is so simple, and I strongly encourage to Approve this research in F1000Research paper with the following minor modifications.

The authors cite Cho et al., (2012) in which conditional DKO mice of Nr1d1/2 (REV-ERBs) genes did not show clear circadian rhythms in the locomotor activity. However, this arrhythmic phenotype was probably due to the severe unhealthiness of the mutant mice, because canonical DKO mice are lethal. In Adlanmerini et al., (2021) 1 (from the same Dr. Lazar Lab), they showed clear locomotor rhythms with significant short period in the SCN-specific DKO of REV-ERBs. Therefore, REV-ERBs are not essential genes for the circadian oscillation. This will be informative for the readers to understand the results in this paper.

The authors re-analyzed a series of RNA-Seq datasets in this field to analyze RNA expression patterns of Per2AS in the mouse livers. Since the bioinformatic methods for analyzing RNA-Seq data are advancing day by day and the obtained results will change depending on the method, the results analyzed by the same authors using the same method are very useful for many chronobiologists. For example in Figure 1, they showed only expression patterns of Per2AS, Per2, and the deleted genes in a series of KO mice, but I suggest them to show expression patterns of core clock genes in all the case probably in supplemental figures; especially E-box genes (Dbp and Rev-erb) expression in Bmal1-KO and Cry-DKO mice and RRE genes (Bmal1 and E4bp4) expression in Rev-erb-DKO. These results will support the quality of their bioinformatic method in this paper and also will show the quality of the raw data in the previous studies.

 Minor:

I find it easier to compare the gene expression patterns if panels are displayed in the same size even between different figures. For example, panels in figure 2 is too big and figure 4 is too small.

Is the work clearly and accurately presented and does it cite the current literature?

Yes

If applicable, is the statistical analysis and its interpretation appropriate?

Yes

Are all the source data underlying the results available to ensure full reproducibility?

Yes

Is the study design appropriate and is the work technically sound?

Yes

Are the conclusions drawn adequately supported by the results?

Yes

Are sufficient details of methods and analysis provided to allow replication by others?

Yes

Reviewer Expertise:

Circadian clock, post-translational modifications, phosphorylation, LC-MS/MS

I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard.

References : REV-ERB nuclear receptors in the suprachiasmatic nucleus control circadian period and restrict diet-induced obesity. Sci Adv .2021;7(44) : 10.1126/sciadv.abh2007 eabh2007 34705514 10.1126/sciadv.abh2007 Kojima Shihoko

Competing interests: NA

 26 10 2022

1. We thank the reviewer for the suggestion. We have added the paper that the reviewer mentioned, and edited the last paragraph of the section “Core clock genes Bmal1, Cry1/2, and Nr1d1/2 affect the expression patterns of Per2AS and Per2”.

2. We agree with the reviewer. We have now provided an additional extended data file as Table S2, which includes RNA expression levels (i.e., strand-specific TPM) of the core clock genes as well as the rhythmicity analysis of the expression patterns with MetaCycle (doi: 10.6084/m9.figshare.21375783). We also mentioned this extended data by the end of the session “Core clock genes Bmal1, Cry1/2, and Nr1d1/2 affect the expression patterns of Per2AS and Per2”.

3. Unfortunately, we, the authors, have no control over the size of the figures and are not able to change their sizes.