PLoS Genetics
Public Library of Science
image
Long noncoding RNA PAHAL modulates locust behavioural plasticity through the feedback regulation of dopamine biosynthesis
DOI 10.1371/journal.pgen.1008771 , Volume: 16 , Issue: 4

Table of Contents

Highlights

Notes

Abstract

The neurotransmitter dopamine is crucial for the neuronal and behavioral response in animals. Phenylalanine hydroxylase (PAH) is involved in dopamine biosynthesis and behavioral regulation in the migratory locust. However, the molecular mechanism for the fine tuning of PAH expression in behavioral response remains ambiguous. Here we discovered a nuclear-enriched lncRNA PAHAL that is transcribed from the coding strand of the PAH gene in the locust (i.e., sense lncRNA). PAHAL positively regulated PAH expression and dopamine production in the brain. In addition, PAHAL modulated behavioral aggregation of the locust. Mechanistically, PAHAL mediated the transcriptional activation of PAH by recruiting SRSF2, a transcription/splicing factor, to the promoter-associated nascent RNA of PAH. These data support a model of feedback modulation of dopamine biosynthesis and behavioral plasticity via a sense lncRNA in the catecholamine metabolic pathway.

Zhang, Xu, Chen, Kang, and Ewer: Long noncoding RNA PAHAL modulates locust behavioural plasticity through the feedback regulation of dopamine biosynthesis

Introduction

Long noncoding RNAs (lncRNAs) are transcripts comprising > 200 nucleotides and possessing minimal or non-existent protein-coding capacity [1] and are increasingly recognised as key players in numerous cellular processes [2,3]. LncRNAs are the main products of RNA polymerase II, often polyadenylated and processed through splicing [3]. Growing evidence shows that most lncRNAs may be functionally relevant. LncRNAs engage in various biological processes, such as X-chromosome inactivation [46], DNA damage response [7,8], differentiation and development [5,9,10], metabolism [11,12] and immunity and disease response [1214].

The brain and neuronal specificity of lncRNA expression has prompted the exploration of the potential roles of lncRNAs in neuronal development and cognitive and behavioural regulation [10,1521]. In Drosophila species, cytoplasmic lncRNA yar regulates sleep behaviour [22]. The lncRNA CRG exhibits spatiotemporal specific expression patterns within the central nervous system; in addition, CRG affects the locomotor behaviour of Drosophila by positively regulating a neighbouring gene that encodes a Ca2+ /calmodulin-dependent protein kinase [23]. LncRNAs can regulate gene expression through distinct modes [19,2426], such as their association with transcription factors and/or chromatin modification factors [3,5,2729]. In mice, Gomafu mediates anxiety-like behaviour by maintaining the polycomb repressive complex 1 at the promoter of the schizophrenia-related gene beta crystallin [30]. Many more findings implicate the widespread involvement of lncRNAs in neuronal response and neurological diseases [10,17,18,29,31]. Despite the recognised roles of lncRNAs in the neuronal system and behaviour, how neuronal and behavioural responses to environmental stimuli are modulated at the cellular and organismal levels by lncRNAs remains incompletely understood.

The metabolic and signal transduction pathways of dopamine (DA) are important for the behavioural responses of animals. DA is responsible for motor control, learning and memory and is associated with several important neural diseases [32,33]. Phenylalanine hydroxylase (PAH, also known as Henna in Drosophila and Locusta ) catalyses the synthesis of tyrosine from phenylalanine and affects the production of DA and other bioamines in brain [3437] (Fig 1A). We previously elucidated the metabolic and signalling pathways of DA-regulated aggregation behaviour in the migratory locust Locusta migratoria , which exhibits density-dependent behavioural plasticity [3840]. Gregarious (G) locusts display high mobility and conspecific attraction. By contrast, solitarious (S) locusts are tardy and repulsive to other conspecific individuals [41]. Locusts can reversibly and rapidly shift their behaviours between the S and G phases in response to population density changes. Specifically, G nymphs exhibit significant behavioural solitarization upon 1 h isolation and full solitarization upon 16 h isolation. S nymphs display significant behavioural gregarization upon 32 h crowding [42]. Therefore, locusts are considered ideal models for studying animal behavioural plasticity [38,41,43]. PAH expression affects the catecholamine metabolic pathway (e.g., DA synthesis) and controls the behavioural phase shift of locusts [39,40]. Specifically, miR-133 targeting PAH coding region repressed PAH expression, reduced DA production and drove locust soliterization while activated, and vice versa while inhibited [36]. However, the molecular mechanism underlying the fine-tuning of PAH expression for the mediation of brain DA dynamics in behavioural shifts remains ambiguous.

PAHAL is expressed as a sense lncRNA transcript within the PAH locus.
Fig 1
(A) Schematic representation of the catecholamine metabolic pathway. PAH catalyses the reaction from phenylalanine to tyrosine. PAH, phenylalanine hydroxylase; TH, tyrosine hydroxylase; TYDC, tyrosine decarboxylase; TβH, tyramine β-hydroxylase; DDC, decarboxylase. (B) Gene and transcript structure of the locust PAH and PAHAL. The bent arrows indicate the transcriptional orientation of PAH and PAHAL, respectively. “E 1–8” represents the eight exons of PAH transcript. The half-arrows indicate the following strand-specific primers for quantitative PCR (qPCR): primers 1 and 2 for PAH, primers 3 and 4 for PAHAL, primers 5 and 6 for PAHAL dsRNA, and primers 7 and 8 for PAH dsRNA. The broken line represents siRNA sequences used for PAHAL knockdown. The entire PAH locus is drawn proportional to its length. The black scale bar for PAH locus represent 1 kb. The gene structure of PAHAL is scaled up in red. The red scale bar for PAHAL represent 0.1 kb. (C) PAHAL and PAH were different transcripts. The half-arrows indicate the following strand-specific primers for PCR: Primer 13 is the specific reverse primer for cDNA synthesis of PAH and PAHAL, primers 10 and 12 are for PAH (289 bp), primers 9 and 10 are for PAHAL (476 bp), and primers 11 and 12 yield no PCR product. The red shading indicates the location of PAHAL in PAH locus. (D) 5′-triphosphate cap of PAHAL validated through the 5′ exonuclease digestion assay. GAPDH and 18S rRNA acted as the positive and negative controls, respectively. “Total” means total RNA in reverse transcription (RT), and “capped” indicates capped RNA that was 5′ exonuclease digested in the cDNA synthesis. (E) Detection of PAHAL poly(A) tail by RT-PCR in the brain RNA of the fourth-stadium nymphs. Total RNA was G/I-tailed, reverse-transcribed and amplified with gene-specific primers F and R. U6 and GAPDH were used as negative and positive control, respectively. (F) Coding potential of the transcript in the PAHAL and reference transcripts according to the CPAT algorithm. In the references, MALAT1 and HOTAIR are lncRNAs, and the others are protein-coding genes. (G) In vitro translation assay for PAHAL. Rluc, biotinylated luciferase; NC, no DNA-template control.PAHAL is expressed as a sense lncRNA transcript within the PAH locus.

In this study, we explored a new regulatory mechanism, which involves an lncRNA in the DA metabolic pathway and underlies the elaborate control of locust behavioural plasticity. We discovered a novel PAH-activating lncRNA (PAHAL), which is sense to the ancestral PAH gene. PAHAL tightly controls the metabolic regulation of DA biosynthesis in the brain and mediates reversible behavioural changes in response to population density. Mechanistic analysis revealed that PAHAL can enhance PAH expression by recruiting the transcription activator serine/arginine-rich splicing factor 2 (SRSF2) to the PAH promoter-proximal region. These results demonstrate that the lncRNA orchestrates gene expression for DA biosynthesis by regulating the local feedback transcription of PAH. Thus, our findings provide new insights into the role of lncRNAs in the fine tuning of neuronal and behavioural responses cued by environmental stimuli.

Results

Sense lncRNA PAHAL is expressed from the intron/exon of the PAH gene locus

We identified PAHAL, a novel lncRNA that overlaps with the PAH gene, in the genome and transcriptomes of the locust. We validated the transcriptional origin of PAHAL relative to that of PAH through 5′ and 3′ RACE. The locust PAH gene encodes PAH by a 1,652 bp-long transcript comprising eight exons (Fig 1B). PAHAL is unspliced and is 2,612- nucleotide (nt) long (S1 Fig). The PAHAL sequence covers the sixth exon and part of the two introns flanking the exon in the PAH gene (Fig 1B and S1 Fig). Sense-specific reverse transcription polymerase chain reactions (RT-PCRs) confirmed that PAH and PAHAL are different transcripts (Fig 1C). The 5′ exonuclease digestion assay indicated that PAHAL is 5′ capped (Fig 1D), and the RT-PCR from the 3′-end tailed-RNA demonstrated that PAHAL is a polyadenylated transcript (Fig 1E). These results indicate that the transcription direction of PAHAL is the same as that of the canonical PAH, and PAHAL transcription is sense to PAH gene (Fig 1B).

Open reading frame (ORF) analysis showed that 19 short ORFs are present in PAHAL. The longest ORF (201 nt) is located at most 3′ end and annotated as the long terminal repeat region of a Gypsy transposable element. No match to any known genomic sequence of insect species, except the contained PAH exon six (262 nt), was identified. Therefore, the lncRNA sequence is not conserved among insect species. The coding capacity assessment by using the Coding Potential Assessment Tool (CPAT) [44] showed that PAHAL lacks protein-coding capacity (Fig 1F). An in vitro translation assay demonstrated that no protein was produced by PAHAL expression (Fig 1G). Therefore, PAHAL is a sense noncoding RNA generated from the PAH locus.

PAHAL controls dopamine biosynthesis by regulating the PAH expression

We knocked down PAHAL through dsRNA (double-stranded RNA) and siRNA (small interfering RNA) interferences to test the regulatory relationship between PAHAL and PAH in the G locust brain. The dsRNA and siRNAs of PAHAL were targeted to the right intron flanking the sixth exon in the PAH gene and ought to be specific for PAHAL not influence the pre-mRNA of PAH . Primer sequences of these dsRNAs and siRNAs can be found in S3 Table. PAHAL dsRNA injection in the brain significantly repressed the PAHAL expression (t test: P = 0.002, N = 9). The PAHAL knockdown reduced the PAH mRNA level by 91% (t test: P = 0.006, N = 9) and PAH protein level by 96% (t test: P < 0.001, N = 4; Fig 2A). The PAHAL -specific knockdown through siRNA interference provided similar results (Fig 2B). However, PAH knockdown had no effect on PAHAL level (Fig 2C). These findings imply that PAHAL is involved in the positive regulation of PAH in locust brains.

PAHAL regulates the PAH expression and controls the DA biosynthesis in the locust brain.
Fig 2
The PAHAL and PAH expression levels in the locust brain at 72 h after dsRNA (A) and siRNA (B) knockdown of PAHAL. dsGFP and siGFP represent GFP dsRNA and siRNA injection and acts as the control. dsPAHAL and siPAHAL represent PAHAL dsRNA and siRNA injection. The expression levels in dsPAHAL and siPAHAL are normalised against those in dsGFP and siGFP. Rp49 is the internal control. Nine biological replicates of 10 brains were measured. (C) PAH knockdown through dsPAH injection did not alter the level of PAHAL. Five biological replicates of eight brains were measured. (D) Median-centered, hierarchical clustering of annotated genes that were altered in PAHAL-silenced brains. Three biological replicates of 12 brains were sequenced. (E) Functional clustering of differentially expressed genes using KEGG enrichment analysis. Only P < 0.05 terms are shown (Fisher’s exact test). (F) Effects of PAHAL knockdown on the bioamine metabolite level in the catecholamine metabolic pathway in the brain. Bioamine metabolite levels were measured through high performance liquid chromatography–mass spectrometry. Ten biological replicates of 12 brains were measured. The data in this figure, as well as those in the following figures, are represented as the mean ± SEM with the fourth-stadium nymphs, unless stated otherwise. Student’s t-test: *P < 0.05; **P < 0.01; ***P < 0.001.PAHAL regulates the PAH expression and controls the DA biosynthesis in the locust brain.

Thereafter, we applied brain transcriptome sequencing to investigate the downstream pathways affected by the PAHAL expression. The RNA-seq data showed that 52 genes were significantly altered by PAHAL knockdown (Fisher’s exact test: P < 0.05, fold change > 2, N = 3; Fig 2D and S1 Table, Supplemental File 1). PAH was the most differentially expressed gene (Fisher’s exact test: P = 2.5 × 10−37, N = 3). The functional clustering of the 52 genes revealed that the phenylalanine metabolism pathway to include the PAH gene was the most significantly enriched pathway (Fisher’s exact test: P = 4.9 × 10−4 , N = 3; Fig 2E).

PAH encodes a dual function enzyme, that hydroxylates both phenylalanine for tyrosine synthesis and tryptophan for the synthesis of 5-hydrotryptophan (5-HTP), the precursor of serotonin in insects [34,45,46]. Therefore, we then quantified the levels of tyrosine, a direct product of PAH catalysis, and the six downstream metabolites in the catecholamine metabolic pathways, namely, l -dopa, DA, 5-hydroxytryptophan, serotonin, tyramine and octopamine by liquid chromatography–mass spectrometry (LC–MS) (Figs 1A and 2F). The PAHAL knockdown significantly decreased the concentrations of tyrosine (t test: P < 0.001, N = 10), and metabolites (i.e. l-dopa, t test: P = 0.001, N = 8) and DA (t test: P < 0.001, N = 7), in one downstream branch pathway but not those (i.e. tyramine and octopamine) in another one in the brain. Such knockdown also reduced the content of 5-hydroxytryptophan (t test: P < 0.001, N = 10) but not that of the downstream product, such as serotonin (Fig 2F). These results indicate that PAHAL affects the DA synthesis in the catecholamine metabolic pathway.

PAHAL is a positive regulator of PAH expression in locust behaviour

We tested the functional relationship between PAHAL and PAH by monitoring their expression levels in five tissues from G nymphs (Fig 3A). PAHAL and PAH were highly expressed in the locust brain. G locusts display high mobility and sociable. By contrast, S locusts are sedentary and live at very low densities [41]. The G and S individuals possess distinct, population density-dependent behavioural features [38]. Thus, we compared the expression levels of the two transcripts in the brains of G and S locusts (Fig 3B). The expression of PAHAL and PAH in the G locusts were 17.7- (t test: P = 0.004, N = 5) and 2.2-fold higher (t test: P < 0.001, N = 7), respectively, than those in the S locusts (Fig 3B). Accordingly, PAHAL and PAH were upregulated in the G locust brain. The brains of G locusts were subjected to fluorescence in situ hybridisation (FISH). The result showed that PAH and PAHAL transcripts are both localised to neuronal cell bodies, but as expected PAH is cytoplasmic whereas PAHAL is nuclear (Fig 3C and S2 Fig). Furthermore, PAH and PAHAL have broad expression patterns rather than being localised only in a subset of locust neurons.

PAHAL expression is positively correlated with the PAH expression in the brain during the behavioural transition of locusts.
Fig 3
(A) Tissue expression of PAHAL and PAH. The mRNA level was quantified through qPCR. Means labelled with different letters within each gene are significantly different (P < 0.05). (B) PAHAL and PAH expression in the brains of gregarious (G) and solitarious (S) locusts. (C) Localisation of PAHAL and PAH in the G brains as revealed through double fluorescence in situ hybridization (FISH). Images are shown at 10× (left images) and 63× (right images) magnification. The "a" squares specifically indicate the brain region of the right images in the locust protocerebrum. The brain region of the Fig 5B is indicated by the "b" squares in the locust deutocerebrum. (D) Profiles of the PAHAL and PAH expression in the brain during locust isolation and crowding. Eight biological replicates of eight brains were measured. Asterisks indicate significant differences between each time point and at 0 h (P < 0.05). Student’s t-test: *P < 0.05; **P < 0.01; ***P < 0.001.PAHAL expression is positively correlated with the PAH expression in the brain during the behavioural transition of locusts.

We tested the time-course expression dynamics of PAHAL and PAH in the brain to determine their dynamic correlations during locust isolation and crowding treatments (Fig 3D). PAHAL was significantly upregulated at 4 h upon aggregation (t test: P = 0.002, N = 8) compared with that at 0 h. The upregulation of the PAHAL expression was sustained at 8 h (t test: P = 0.040, N = 8). By contrast, the PAHAL expression was significantly downregulated at 4 h after isolation (t test: P = 0.013, N = 8). Such expression continued to decrease at 8 h after isolation (t test: P= 0.001, N = 8). PAH exhibited the same time-course expression pattern. Thus, the expression levels of PAHAL and PAH are positively correlated during locust aggregation and isolation.

PAHAL controls phase-related behavioural transition

We investigated the role of PAHAL in behavioural regulation in the locusts using dsRNA and siRNA interference of PAHAL , respectively. Locust aggregation in an arena was measured by using an established automatic behavioural assay [40,42] (Fig 4A). Aggregation behaviour was quantified by using Pgreg , a summary statistic provided by this behavioural assay and a reliable indicator of locust aggregation propensity [40]. The shift in the median Pgreg value of G locusts from 0.89 to 0.23 when PAHAL was knocked down through dsRNA interference indicated a significant behavioural shift from the G to the S states (Mann–Whitney U test: P < 0.001, NdsGFP = 30, NdsPAHAL = 57, Fig 4B). The results from siRNA interference lead to a change of Pgreg from 0.90 to 0.20 (Mann–Whitney U test: P < 0.001, NsiGFP = 33, NsiPAHAL = 35) (Fig 4C). This outcome from siRNA demonstrates an effect of PAHAL on locust aggregation similar to that from dsRNA.

PAHAL modulates locust behavioural aggregation.
Fig 4
(A) Movement trajectories of the G and S nymphs in the behavioural assay arena. A stimulus group of 30 G nymphs were placed behind a transparent porous partition on the left. (B) dsRNA interference of the PAHAL expression resulting in the behavioural transition from G to S phase. Pgreg is a measure of the predisposition of locusts to swarming, that is, 0 for typical solitarious behaviour and 1 for typical gregarious behaviour. Median Pgreg values are indicated by red arrows. (C) Behavioural changes resulting from siRNA interference of PAHAL. (D) dsRNA knockdown of the PAHAL expression altering behavioural traits, including the total distance moved (TDM), the total duration of movement (TDMV), and the attraction index (AI). (E) Behavioural traits affected by the siRNA knockdown of PAHAL. Asterisks indicate the significance level by Student’s t-test for TDM and TDMV and by Mann–Whitney U test for AI: **P < 0.01; ***P < 0.001.PAHAL modulates locust behavioural aggregation.

Furthermore, we examined the specific parameters of the behavioural traits of locusts after the gene expression interference. In comparison with the dsGFP control treatment, the PAHAL knockdown reduced the total distance of movement by 71% (t test: P < 0.001, NdsGFP = 30, NdsPAHAL = 57), decreased the total duration of movement by 95% (t test: P < 0.001, NdsGFP = 30, NdsPAHAL = 57) and induced the reversion from attraction to repulsion, as shown by the attraction index (Mann–Whitney U test: P = 0.008, NdsGFP = 30, NdsPAHAL = 57; Fig 4D). We also examined the specific behavioural effect from siRNA knock down. The PAHAL interference by siRNA has behavioural effects similar to that by dsRNA in terms of mobility and conspecific attraction (Fig 4E). Overall, PAHAL effectively modulates the behavioural aggregation of the locusts.

PAHAL facilitates the transcription activation of PAH by interacting with the PAH promoter-proximal region

We determined the mechanism by which PAHAL mediates PAH transcription. The sequence analysis revealed that one nuclear localisation signal (NLS) [47] is present at the 3′ end of PAHAL but not in PAH (S3 Fig). The nuclear fractionation experiment showed that 94% of PAHAL mRNAs (possibly together with trace amount of PAH pre-mRNA) localises in the nucleus relative to nuclear RNA U6 (positive control) and cytoskeleton actin (negative control). By contrast, 97% of PAH mRNAs localised in the cytoplasm (Fig 5A). FISH results also showed that PAHAL and PAH localised in the nucleus and cytoplasm, respectively (Figs 3C and 5B). This finding indicates that these two transcripts are distinct in the subcellular location.

Nuclear-enriched PAHAL promotes the PAH expression by interacting with the promoter-proximal region of PAH.
Fig 5
(A) Subcellular localisation of PAHAL and PAH in the brain. Five biological replicates of eight brains were examined for each treatment. (B) FISH detection of the PAHAL probe (red), PAH probe (green), and Hoechst 33342 (blue) in the brain. The arrows indicate the subcellular location of PAHAL (in the nucleus) and PAH (in the cytoplasm). Scale bars, 50 μM. The brain region in these images was indicated by the "b" squares of Fig 3C. (C) PAHAL-enhanced PAH promoter activity. The PAH promoter construct (P+5′-UTR) contained −1,168 to +89 bp relative to the TSS. “P” represents −1,168 to +1 bp, and “5′-UTR” denotes +1 to +89 bp. Reverse PAHAL is a reverse full-length fragment of PAHAL. Lacz is a fragment of Lacz gene containing frame shift and acted as a negative control. PAHAL, reverse PAHAL and lacz were constructed into pAC5.1/V5-His A vectors to their high-level transient expression in Drosophila S2 cells. (D) PAHAL interacts with the proximal promoter of PAH. (E) 5′-UTR of PAH is required for transcriptional activation by PAHAL. Error bars indicate ± SEM. Student’s t-test: *P < 0.05, **P < 0.01; ***P < 0.001; n.s., insignificant.Nuclear-enriched PAHAL promotes the PAH expression by interacting with the promoter-proximal region of PAH.

Whether PAHAL regulates PAH expression through a cis interaction with a PAH promoter was determined. The sequence from −1,169 nt to +89 nt relative to the PAH initiator was selected as the regulatory segment of the PAH promoter (labelled as P+5′-UTR). The luciferase assay showed that PAHAL substantially enhanced the luciferase expression driven by the PAH promoter in S2 cells (t test: P < 0.001, N = 4). By contrast, PAHAL expressed in the reverse orientation (i.e., reverse PAHAL) had no effect on PAH promoter activity (Fig 5C). Thus, PAHAL enhances the promoter activity of PAH not by acting as an enhancer element.

Moreover, the specific regulatory region in the PAH promoter targeted by PAHAL was determined by testing the transcription activities of truncated promoters coexpressing PAHAL. PAHAL significantly enhanced the activity of these promoter regions, even that of the shortest promoter (from −48 nt to +89 nt; t test: P < 0.001, N = 4; Fig 5D). Thus, the active PAHAL interacting region is close to the transcription start site (TSS) of PAH. The removal of the 5′-UTR abrogated the PAHAL -mediated transcriptional activation (Fig 5E). Therefore, PAHAL promotes the PAH expression by actively interacting with the promoter-proximal region, including the 5′-UTR, of the PAH gene.

PAHAL regulates PAH expression by binding with SRSF2

To determine proteins that interact with PAHAL , we performed RNA pulldown with brain tissue extracts (Fig 6A) and identified 11 protein candidates by mass spectrometry (S2 Table). Among these proteins, SRSF2 bound with the sense, but not antisense, strand of PAHAL (Fig 6B) and was associated with behavioural phase changes (Fig 6C and S4 Fig). Specifically, SRSF2 expression in the brain of a G locust significantly differed from that of an S locusts (t test: P < 0.001, N = 8). SRSF2 was continuously upregulated during locust crowding treatments and downregulated during isolation treatments (Fig 6C). Thus, the SRSF2 expression is positively related to the PAHAL and PAH expression during locust behavioural transition.

PAHAL activates PAH transcription by interacting with SRSF2.
Fig 6
(A) RNA pulldown identifying PAHAL-associated proteins in the brain. Red arrows represent different protein bands between PAHAL and its antisense. (B) Western blot detects SRSF2 in protein pools from PAHAL RNA pulldown (N = 4). (C) SRSF2 expression in the brain during locust isolation and crowding. Eight replicates of eight brains were measured. (D) RNA immunoprecipitation (RIP) enrichment verifies that SRSF2 bound with PAHAL in vitro and in vivo. Five replicates of eight brains were measured. (E, F) SRSF2 facilitates the PAHAL-mediated transcription activation of PAH. Luciferase assays of P+5′-UTR were performed by using SRSF2 protein-depleted mouse embryonic fibroblasts (SRSF2-MEFs, SRSF2-KO). “DOX +” represents SRSF2 knockout, whereas “DOX–” indicates the normal expression of SRSF2 in SRSF2-MEFs or in either S2 cells overexpressed SRSF2 (S2, SRSF2-ov). Student’s t-test: *P < 0.05; **P < 0.01; ***P < 0.001; n.s., insignificant.PAHAL activates PAH transcription by interacting with SRSF2.

To determine the physical interaction between PAHAL and SRSF2 that binds to the promoter-associated nascent RNA during transcription regulation, we conducted several in vitro and in vivo RNA immunoprecipitation (RIP). In vitro RIP using an antibody against V5 tag in SRSF2-V5 protein-overexpressed cell lysis revealed significant PAHAL enrichments with SRSF2 (S5A Fig and Fig 6D, t test: P = 0.010, N = 6). In vivo RIP with brain tissues showed that endogenous PAHAL mRNA and SRSF2 were significantly enriched by the SRSF2 antibody (t test: P = 0.004, N = 5 for enriched PAHAL , Fig 6D; t test: P = 0.029, N = 5 for SRSF2, S5B Fig). Thus, the SRSF2 protein physically binds with PAHAL RNA.

Thereafter, we examined the regulatory effects of SRSF2 on PAHAL -mediated transcriptional activation through the SRSF2 knockout and overexpression. We first used SRSF2 protein-depleted cells generated from mouse embryonic fibroblasts (SRSF2-MEFs); an HA-tagged SRSF2 gene replaced the endogenous gene and expressed from a tetracycline (tet)-off promoter, thereby enabling the elimination of the protein by adding the tet analogue DOX [48,49]. The PAH promoter activity was significantly inhibited by 67% by SRSF2 knockout (One-Way ANOVA: P < 0.001, N = 6). The enhanced activity from co-transfected PAHAL was abrogated (One-Way ANOVA: P < 0.001, N = 6) by SRSF2 knockout (Fig 6E). By contrast, the SRSF2 overexpression in S2 cells significantly increased the PAH promoter activity by 56% (One-Way ANOVA: P< 0.001, N = 5). The elevated promoter activity was further enhanced by 91% (One-Way ANOVA: P < 0.001, N = 5) with PAHAL co-transfection. The effects of the SRSF2 and PAHAL expression exhibited a significant interaction in the PAH promoter activation (Mann–Whitney U test: P < 0.001, N = 5, Fig 6F). Therefore, SRSF2 substantially promoted the PAHAL-mediated cis-activation effects.

PAHAL–SRSF2 complex binds with sequence specificity for transcriptional regulation

We determined the specific sites for SRSF2 interaction in the PAHAL sequence. The domain mapping of sequence fragments containing the 3′ end of PAHAL revealed that all the fragments interacted with SRSF2 (Fig 7A, left panel). The mapping analysis of sequence fragments covering the 5′ end of PAHAL revealed that only the fragment that simultaneously covered the 3′ end (i.e., the last 245 nt) of PAHAL bound with SRSF2 (Fig 7A, middle panel). Two overlapping fragments at the 3′ terminal of PAHAL spanning 1,968–2,613 exhibited SRSF2 binding (Fig 7A, right panel). Thus, the sequence at the most 3′ end of PAHAL is necessary for PAHAL–SRSF2 binding.

SRSF2 is recruited by the 3′ terminal of PAHAL to PAH 5′-UTR for the transcriptional activation of PAH.
Fig 7
(A) Mapping analysis of the biotin-labelled fragments of PAHAL incubated with SRSF2-MEF lysates that overexpress the V5-tagged SRSF2 (SRSF2-V5) of locusts. (B) Mutational analysis of the activity of ESEs in PAH gene with the overexpressed PAHAL by using the luciferase assay. WT indicates wild type. MT-ESE1, MT-ESE2 or MT-ESE3 indicate mutated ESE1, ESE2 or ESE3 in the pGL4.10 vector, respectively. MT-ESEs indicate the mutation of all the three ESEs. (C) The interaction of PAHAL, SRSF2 and ESE promotes the transcription activation of PAH. Means labelled with different letters are significantly different (P < 0.05). The following vectors were transfected in SRSF2-MEFs: “ESE+”, P+5′-UTR (containing −554/+89 fragment without ESE mutation) in pGL4.10; “ESE−”, P+5′-UTR containing the mutated tandem ESEs; “PAHAL+”, the pcDNA3.1(+)/PAHAL; “PAHAL−”, the pcDNA 3.1(+)/lacz; “SRSF2+”, the pcDNA 3.1/V5-His/SRSF2; “SRSF2−”, the pcDNA3.1/V5-His /lacz. (D) SRSF2 knockdown by dsRNA injection reduces the PAH expression in the brain. Six replicates of eight brains were measured. Student’s t-test: *P < 0.05; **P < 0.01; ***P < 0.001; n.s., insignificant.SRSF2 is recruited by the 3′ terminal of PAHAL to PAH 5′-UTR for the transcriptional activation of PAH.

Furthermore, we analysed the mechanism by which PAHAL and SRSF2 interaction facilitated the transcription activation of the PAH gene. SRSF2 facilitates transcription by binding to a high-affinity binding site (exonic-splicing enhancer, i.e., ESE) of nascent RNA [49]. We found three ESE motifs in the 5′-UTR of PAH (S5C Fig). The PAHAL-mediated activation of the PAH promoter activity was disrupted by the mutation of the three ESEs; by contrast, a single ESE mutation did not eliminate the activation effect of PAHAL (Fig 7B). Therefore, the regulatory function of PAHAL involves interaction with the ESEs of the PAH 5′-UTR. The luciferase assay further demonstrated that the three factors, namely, PAHAL, SRSF2 and tandem ESEs, had significant interactions in the effects of transcriptional regulation (One-Way ANOVA: P < 0.001, N = 6). The tandem ESEs exhibited the greatest effects among the three factors (Fig 7C). These results indicate that PAHAL directs SRSF2 to the ESEs of PAH 5′-UTR to promote PAH transcription.

Finally, we validated the role of SRSF2 in the transcriptional activation of PAH through RNAi. SRSF2 knockdown significantly reduced the mRNA level of PAH (t test: P = 0.006, N = 6) and protein level of PAH by 89% (t test: P = 0.037, N = 3, Fig 7D). Thus, SRSF2 regulates PAH transcription in vivo.

Discussion

In our study, we have identified a novel sense lncRNA named PAHAL in the PAH locus in the catecholamine metabolic pathways, and have revealed its specific role in regulating DA biosynthesis and locust behavioral aggregation. Moreover, we have also characterized the mode of transcriptional activation orchestrated by this lncRNA in the biosynthetic regulation. PAHAL exhibits a distinct mode of cis-regulation transcription, which is verified in locust, fruit fly and mice in vivo and in vitro . DA has been confirmed as a crucial modulator for locust behavioural phase transition [36,40]. Thus, our present study established the novel linkage between genetic and epigenetic modulation of behavioural plasticity of locust regulated by DA.

PAHAL knockdown attenuated the PAH expression and caused a behavioural shift toward S traits. Furthermore, PAH catalyses the reaction from phenylalanine to tyrosine and affects the downstream metabolic pathways of several bioamine neurotransmitters. Immunocytochemical studies show that aminergic neurons present wide distribution in the locust brain [5052]. Specifically, most dopaminergic neurons are in the upper division of the central body [51]. Serotoninrgic neurons distribute 6 groups that innervate the central complex (groups l-5) [50]. Our study indicates that PAH and PAHAL are expressed in almost all neurons in the locust brain (Figs 3C and 5B). However, the transcriptome sequencing and LC–MS measurements illustrate that PAHAL predominantly affects the DA synthetic pathway in the locust brain (Fig 2), although the results do not exclude non-PAH players in DA metabolism and PAH’s role in other processes. Thus, this study establishes the regulatory role of PAHAL in the orchestration of the gene expression and DA metabolism to modulate animal behavioural transition in response to environmental stimuli.

This study also reveals the regulatory relevance of lncRNAs in strictly controlled genetic programs. Animal behaviours are generally deliberately tuned in response to a specific environmental cue. This response requires triggering the precise spatial–temporal regulation of the expression of specific genes or genetic pathways and the production of specific neurotransmitters [33,40]. The mode of cis-regulation of PAHAL lncRNA exhibits common transcription mechanism, which is verified in locust, fruit fly and mice. Numerous lncRNA loci act as local regulators and influence the expression of nearby genes through cis -regulation [24,25]. These lncRNAs with cis -regulatory roles may arise from diverse ways, such as from promoters or enhancers, antisense transcripts or from within introns or near TSSs of other host genes [25]. For example, about half fraction of antisense transcripts are noncoding RNAs [25]. In another study, genetic manipulation in mouse cell lines revealed that, among 12 genomic loci that produce lncRNAs, five loci influence the expression of a neighboring gene in cis [24]. These studies have revealed the prevalence of lncRNA-mediated cis regulation [53,54]. Cis-interference or silencing by intragenic lncRNAs is also a common phenomenon in numerous organisms. For example, the antisense Tsix RNA that originates from the Xist loci inhibits the expression of the maternal Xist allele [6]. By contrast, COLDAIR, which is a sense intronic lncRNA, antagonises FLC in Arabidopsis [55]. Nevertheless, PAHAL is mainly derived from the intronic region of PAH but also shares its partial sequence with the sixth exon of the coding PAH transcript. Such transcription mode does not solely occur to PAHAL. For example, the sense lncRNA GClnc1 is transcribed in large part from the intronic region of the SOD2 gene, but it also overlaps with the last exon of two SOD2 transcripts [56]. PAHAL and PAH are positively regulated in the behavioural change of the locusts. In vivo and in vitro analyses demonstrated that nuclear-enriched PAHAL promoted the PAH expression. PAHAL is engaged in a regulatory feedback loop that is stimulated by population density changes, by interacting with PAH (Fig 8). Thus, PAHAL acts as a sense transcriptional activator and represents an lncRNA subclass that participates in local cis-regulation.

Working model for the feedback activation of PAH via PAHAL–SRSF2 interaction in the DA metabolic pathway.
Fig 8
The sense nuclear lncRNA PAHAL is transcribed from the exon–intron region of PAH gene locus cued by population crowding. PAHAL then binds with SRSF2 at its 3′ end to form a RNA–protein complex. PAHAL guides the PAHAL–SRSF2 complex to bind with the ESE elements of the nascent PAH RNA and activates the PAH transcription. PAH catalyses phenylalanine into DA. The red genomic region in the PAH loci represents the PAHAL emanative fragment. The half arrowhead indicates the transcriptional direction of PAHAL.Working model for the feedback activation of PAH via PAHAL–SRSF2 interaction in the DA metabolic pathway.

In line with the behavioural phase changes, PAHAL is generated at transcription levels highly contingent upon locust population density. Thus, PAHAL transcription is precisely controlled commensurate with the crowding state and phase-related behaviour of locusts. The PAHAL transcription then triggers the fine tuning of the PAH expression and in turn the downstream biosynthesis of specific neurotransmitters such as the DA that controls the aggregation behaviour of locusts (Fig 8). Therefore, PAHAL is dedicated to orchestrating the PAH expression and downstream DA production during the population density-dependent behavioural aggregation of locusts. The density-contingent transcription of PAHAL could be also crucial for adjusting the DA dosage. A large proportion of lncRNAs in animal models present high spatial–temporal specificity of expression in various biological processes, such as development and behavioural responses [20,5759]. Overall, these findings highlight the important role of lncRNAs as a regulator of tightly controlled genetic networks.

Interestingly, two distinct regulatory mechanisms for PAH expression rendered by two noncoding RNAs have been employed in the locust. On one hand, lncRNA PAHAL is induced by locust crowding, positively regulates PAH expression and DA production in the brain at the transcriptional level. On the other hand, miR-133 is upregulated by locust isolation, evokes degradation of PAH mRNA and inhibits DA production, thus representing a mechanism for post-transcriptional regulation of PAH [36]. Both PAHAL and miR-133 are involved in the modulation of behavioral aggregation but in contrasting directions. Therefore, the two mechanisms at different regulatory levels may be orchestrated in response to population density to ensure the fine tuning of PAH expression and DA biosynthesis and thus guarantee the deliberate behavioral response of locusts.

PAHAL mediates regulatory activation by recruiting the transcription activator SRSF2 to the promoter-proximal region of PAH . In this regulatory network (Fig 8), PAHAL involves a distinct regulatory element in the lncRNAs–SRSF2 interaction complex that is destined to the transcriptional machinery of the focal gene. Previous studies showed that some lncRNAs participate in regulatory complexes through different mechanisms, such as interaction with transcription factors and chromatin modifiers [44,60]. SRSF2 was first characterised as a splicing factor [49,6163]. The nuclear-retained lncRNA MALAT1 is also involved in the regulation of alternative splicing by interacting with SRSF2 [64]. In contrast with MALAT1, PAHAL functions in transcriptional activation by binding with SRSF2. Notably, SRSF2 also plays an active role in transcription elongation and activation [49,62]. In this study, PAHAL RNA recruits SRSF2 to the promoter-proximal region of the target gene to promote local transcription (Fig 8). Therefore, SRSF2 possibly exerts different functions by interacting with various lncRNAs. SRSF2 bound to the ESE sites of the promoter-associated nascent RNA. Three tandem ESE motifs in the 5′-UTR of PAH significantly increases the PAHAL-mediated activation of the PAH- promoter activity in the presence of SRSF2. The ESE sites near the 5′ end of nascent RNA may act as critical signals for transcription activation by triggering the progressive release of SRSF2 and P-TEFb from the transcription pause complex [49]. Thus, PAHAL may enhance the transcription activation of the PAH gene by facilitating SRSF2 binding to the nascent PAH RNA. However, the activity assay with reverse PAHAL demonstrated that the enhancer effect rendered by PAHAL is not due to a genomic enhancer element residing in its sequence (Fig 5C).

The local lncRNA-interwired transcription machinery could orchestrate locus-specific transcription regulation in response to an environmental stimulus. SRSF2 is a general transcription activator in the transcription elongation complex and targets a large set of genes [62]. The locus-specific sense lncRNA PAHAL is generated in response to the environmental signal and destined to the transcriptional machinery of the ancestral gene in collaboration with nascent RNA-bound SR proteins (e.g. SRSF2). The lncRNA–SRSF2 interaction complex may have also delivered conformation modifiers to its target genomic loci while still attached to the elongating RNAP II [28,60]. Hence, PAHAL may facilitate the regulatory role of SRSF2 in a gene locus-specific manner. The sequence at the 3′ end of PAHAL is crucial for the binding of SRSF2 with PAHAL, although the underlying mechanism needs further investigation. This region also harbours one NLS that determines the nuclear retention of PAHAL . SRSF2 is the only SR protein retained in the nucleus [65]. MALAT1 to bind with SRSF2 is required for the correct localisation of SRSF2 to nuclear speckles [64,66]. The sequence at the 3′ end of PAHAL may confer a structural component or binding site that promotes the nuclear retention of SRSF2. In summary, these data suggest that PAHAL could govern the spatial specification of the regulatory components of transcriptional machinery in cells and control the target specificity of interacting transcriptional factors in animal behavioural response. Such a precise regulatory mechanism derived from interactions between PAHAL and SRSF2 enlightens us to develop the novel approaches for behavioural manipulation of animals.

Materials and methods

Animals

Locusts were derived from one locust population (Hebei, China) and maintained under standard conditions at the Institute of Zoology, Chinese Academy of Sciences, Beijing, China [67]. The locust colonies were reared under a photoperiod regime of 14 h of light:10 h of darkness at 30°C ± 2°C and fed on fresh wheat seedlings and bran. The G locusts were reared in large cages (40 cm × 40 cm × 40 cm) at a density of approximately 500 insects per cage. The S locusts were obtained from the G colony and individually cultured in white metal boxes (10 cm × 10 cm × 25 cm) supplied with fresh air. The S locusts were reared in isolation for at least three generations prior to experimentation. One-day-old fourth-stadium nymphs, which are most prominent for the behavioural phase plasticity [40,67,68], were used in all experiments.

Cells

Drosophila S2 cells (Gibco, NY, USA, R69007) were maintained in SFX-insect medium (HyClone, Logan, UT, SH30278.02) at 28°C. Human HEK 293T cells (ATCC, Manassas, USA, CRL-3216) and SRSF2 protein-depleted SRSF2-MEFs were maintained in DMEM (GIBCO, NY, USA, C11965500BT) supplemented with 10% FBS (BI, Beit-Haemek, Israel, 04-001-1A) or tet-free FBS (Clontech, CA, USA, 631106) under humidified 5% CO2/95% air at 37°C. SRSF2-MEFs were treated with 10 μg/mL doxycycline (DOX, Sigma, MO, USA, D9891-1G) for 1 day to deplete SRSF2 transcription (DOX+) [48]. We used HEK 293T cells to express the locust SRSF2-V5 fusion protein because this cell line represents a rapid and high-yield protein production system and express more locust SRSF2 than Drosophila S2 cells. Lastly, we adopted the SRSF2-MEF cell line to verify the interaction of SRSF2 with PAHAL by using an endogenous SRSF2-KO system constructed in the cell line.

RNA isolation, 5′ and 3′ RACE and qPCR

Nymphal tissues were rapidly dissected and stored in liquid nitrogen for ≤ 2 months before RNA preparation. The total RNA was extracted by using TRIzol reagent (Invitrogen, CA, USA, 15596018) and incubated for 15 min at room temperature (RT) with 15 kunitzK units of RNase-free DNase I (Qiagen, Hilden, Germany, 79254) to remove genomic DNA. The RNA quality, integrity and quantity were detected using NanoDrop 2000c (Thermo, CA, USA) and 1% agarose gel electrophoresis. Analyses using 5′ and 3′ RACE were performed with a SMARTer RACE cDNA amplification kit in accordance with the manufacturer’s instructions (Clontech, CA, USA, 634858). The PCR products were purified and cloned into a pGEM-T vector (Promega, WI, USA, A1360) for sequencing. The cDNA was reverse-transcribed with a FastQuant RT Kit (Tiangen, Beijing, China, KR106-02). Briefly, 2 μg of DNase-treated total RNA was mixed with 1 μL of reverse transcriptase mix, 2 μL of FQ-RT primer mix and 2 μL of Fast RT buffer into RNase-free water to obtain a final volume of 20 μL. The mixture was incubated for 30 min at 42°C. The cDNA was attenuated to 100 μL and stored at −20°C. Each 10 μL qPCR reaction contained 2 μL of cDNA template, 5 μL of SYBR Green I Master and 0.5 μM of upstream and downstream primers. The qPCR was performed with LightCycler 480 SYBR Green I Master kit (Roche, Mannheim, Germany, 4887352001) on a LightCycler 480 instrument (Roche, Switzerland). Predenaturation was performed at 95°C for 10 min, followed by 45 cycles of PCR at 95°C, 58°C and 68°C for 20 s. The amplification specificity of the target genes was assessed on the basis of a melting curve. All PCR products were verified through sequencing before qPCR. The housekeeping gene ribosomal protein 49 (RP49 ) was used as the endogenous control in qPCR analysis [67]. The relative expression levels of the specific genes were quantified by using the 2−ΔCt method, where ΔCt is the Cp value of RP49 subtracted from that of the gene of interest. Four to eight biological replicates and three technical replicates for every replicate were prepared for each treatment. The outliers were removed from the qPCR standard curves on the basis of the results of Grubbs’ test [69]. S3 Table illustrates all primers for RACE and qPCR.

Sense-specific RT-PCR

The sense-specific PCR was performed as previously described with slight modifications [56]. Briefly, 2 μg of DNase-treated total RNA was mixed with 2 pmol of a reverse primer (primer 13 in Fig 1C) for reverse transcription by using SuperScript IV Reverse Transcriptase (Thermo Fisher; Vilnius, Lithuania, 18090050). The PCR was performed with Taq polymerase (Takara, Tokyo, R001A). An initial denaturation of 94°C for 3 min was followed by 94°C for 30 s, 58°C for 30 s, 72°C for 40 s for 30 cycles. Next, primers 10 and 12 were used to specifically amplify PAH transcript, and primers 9 and 10 were used to specifically amplify the PAHAL transcript. The PCR with primers 11 and 12 is expected to yield no product if PAH and PAHAL are different transcripts. S3 Table present these primers.

5′ exonuclease digestion assay

5′-Phosphate-dependent exonuclease digestion was performed with a terminator 5′-phosphate-dependent exonuclease kit in accordance with the manufacturer’s instructions (Epicentre, IL, US, TER51020). Briefly, total RNA isolated from locust brains was treated with the exonuclease, which specifically digests RNA species with a 5′-monophosphate end, at 30°C for 60 min. The 5′-capped mRNA was isolated through phenol extraction and ethanol precipitation, and cDNA was generated using high-capacity RNA-to-cDNA kit (ABI, CA, USA, 4387406). The mRNA fraction of PAHAL, GAPDH and 18S rRNA was analysed through semi-quantitative PCR. GAPDH and 18S rRNA were used as positive and negative controls, respectively.

Poly(A) tail identification

Poly(A) tails were detected with the Poly(A) tail length assay kit (USB Corporation, Vilnius, Lithuania, 76455). Briefly, a limited number of guanosine and inosine residues were added to the 3′ ends of poly(A)-containing RNAs that were prepared from locust brains. The tailed-RNA was converted to cDNA through reverse transcription by using the newly added G/I tails as the priming sites. PAHAL U6 and GAPDH were amplified from the cDNA with gene-specific primers (S3 Table). The RT-PCR of PAHAL, U6 and GAPDH gave a fragment of 242, 71 and 180 bp, respectively.

In vitro translation assay

TNT Quick for the PCR DNA kit (Promega, WI, USA, L5540) was used for the in vitro translation assay to assess the coding capability of PAHAL. The assay was performed in accordance with the manufacturer’s instructions. Sense primers containing T7 promoters were used to amplify the Renilla luciferase gene (coding negative control) and the full-length PAHAL sequence. The PCR products (1 μg) were mixed with 0.75 μg of transcend biotin–lysyl–tRNA and 1 μL of methionine (1 mM) in 40 μL of TNT T7 Quick Master Mix to a final volume of 50 μL. The mixture was incubated for 1 h at 37°C. Thereafter, 15 μL of the reaction mix was added into 50 μL of the rehydration/sample lysis buffer (8 M urea, 2 M thiourea, 1% SDS and 0.02% [wt/vol] β-mercaptoethanol) in a tube and mixed by inverting the tube until the mixture became homogeneous. After incubation overnight at RT, 10 μL of the homogenised samples was mixed with 1 μL of 1% bromophenol blue buffer (1% bromophenol blue and 50 mM Tris base) and subjected to polyacrylamide gel (15%) electrophoresis. The samples were then transferred to polyvinylidene difluoride (PVDF) membranes (Millipore, CA, USA, ISEQ00010). The membranes were blocked with 5% (wt/vol) skimmed milk at RT for 1 h. The newly synthesised protein containing biotin was detected by using a chemiluminescent nucleic acid detection module (Pierce, CA, USA, 89880).

RNA interference and RNA-seq

dsPAHAL, dsPAH and dsSRSF2 were used to knock down the PAHAL, PAH and SRSF2 expression, respectively. This method has been widely used for specific and efficient knockdown in the locust [40,7072]. The dsRNAs of GFP (dsGFP) were used as the negative control. The dsRNAs were synthesised by using the T7 RiboMAX express RNAi system (Promega, WI, USA, P1700). Three PAHAL -specific siRNAs (Fig 1B) were designed for siRNA interference. The siRNAs were synthesised by RiboBio (Guangzhou, China) and used as a siRNA pool to interfere with the PAHAL expression. The short interfering RNAs of GFP (siGFP) were used as the negative control. The brains of G locusts were injected with 69 nL of dsRNAs (2 μg/μL) or siRNAs (2 μg/μL) with a glass micropipette tip mounted on a nanoliter injector (World Precision Instruments, FL, USA) under an anatomical lens. The injected locusts were returned to normal rearing conditions and reared for 3 days before their brains were harvested for RNA or protein extraction. Three independent biological replicates were prepared for the RNA-seq analysis of dsGFP and dsPAHAL. The integrity of total RNA was quantified with an Agilent 2100 Bioanalyser (Agilent). The cDNA libraries were prepared and sequenced in accordance with the manufacturer’s protocol of Illumina Nova-seq 6000 (150 PE) at Berry Genomic Corporation, Ltd., Beijing, China. HISAT2 software was used to acquire clean reads from the raw data and map them with the locust genome sequence. The gene-expression level was calculated using the clean reads per kb million mapped reads by using the HTseq tool. Genes with corrected P < 0.05 were considered differentially expressed. Differentially expressed genes were defined as corrected P < 0.05 and log2|FoldChange|> 1. The expression levels were shown by a heat map signal that indicates log2 fold-change values relative to the median expression level within the group. The high and low expression levels relative to the median expression level within the group are represented by yellow and blue signals, respectively. KEGG enrichment was performed by using KOBAS software. Significance was analysed through Fisher’s exact test.

Liquid chromatography–mass spectrometry (LC–MS)

Brains were immediately dissected and stored in liquid nitrogen before being assayed. Ten brains were collected for each of the ten biological replicates. Each sample was homogenised with 100 μL of ice-cold PBS. Thereafter, 10 μL of 1 M perchloric acid was added to 90 μL of the homogenate for protein precipitation. The homogenate was centrifuged at 5200 ×g for 30 min at 4°C. The supernatants were filtered through 0.22 μm filters (Millipore, MA, USA). An equal volume of methanol (Fisher, NJ, USA) was added to the filtered supernatant of each sample for immediate LC–MS assay. The total protein of the remaining homogenate was applied as the loading control and quantified through the BCA assay (Fisher, NJ, USA, 23227).

LC–MS was performed with a rapid resolution liquid chromatography system (ACQUITY UPLC I-Class, Waters, USA). Hydrophilic interaction chromatography separation was performed with an ACQUITY UPLCHSS PFP column (100 mm × 2.1 mm, 1.8 μm). The autosampler was set at 10°C. The mobile phase A of the system was gradient elution with 0.1% formic acid acetonitrile (Pierce, CA, USA; Fisher, NJ, USA), whereas phase B was 0.1% formic acid water. The linear gradient for A was as follows: 10%, 0–3 min; 10%–15%, 3–5 min; 15%–100%, 5–7 min; 100%, 7–8 min; 100%–10%, 8–8.1 min; and 10%, 8.1–11 min. The flow rate was adjusted to 0.2 mL/min, and the injection volume was 10 μL. The total run time for each sample was 11 min.

LC/MS/MS analysis was performed on AB SCIEX Triple Quad 4500 (Applied Biosystems, CA, USA) with an electrospray ionisation source (Turbo Ion spray). Mass spectrometry detection was performed in positive electrospray ionisation mode. The [M + H] of the analyte was selected as the precursor ion. The quantification mode was the multiple-reaction monitoring (MRM) mode using mass transitions (precursor/product ions).

The ESI ion source temperature was 500°C. The other mass spectrometric parameters were as follows: curtain gas flow: 10 psi; collisionally activated dissociation gas setting: medium; ion spray voltage: 5500 V; and ion gas 1 and 2: 50 psi. Acquisition and processing were performed by using AB SCIEX Analyst 1.6 Software (Applied Biosystems, CA, USA). The collision energies for different MRM pairs were individually optimised with octopamine (m/z 154 > 136, CE 11V), DA (m/z 154 > 137, CE 14 V), serotonin (m/z 177 > 160, CE 17 V), tyramine (m/z 138 > 77, CE 38 V), l-dopa (m/z 198.1 > 152.1, CE 18 V), 5-hydroxytryptophan (m/z 221.1 > 204.1, CE 14 V) and tyrosine (m/z 182.1 > 136.1, CE 18 V).

Fluorescence in situ hybridization (FISH)

A double FISH experiment was performed as previously described with slight modifications [73]. The RNA probes for PAHAL (digoxigenin [DIG]–PAHAL) and PAH (biotin–PAH) were synthesised by using a T7/SP6 RNA polymerase kit (Promega, WI, USA, P2075, P1085) and DIG RNA labelling mixture/biotin RNA labelling mix (Roche, Mannheim, Germany, 11277073910, 11685597910). Nymphal brains were fixed in 4% (wt/vol) paraformaldehyde for 30 min at RT or overnight at 4°C. The brain tissue was embedded in 5% agarose. Hardened brains were trimmed and dissected into 40 μm slices with a Leica VT1200 S Vibrating Microtome (Leica, Bensheim, Germany). Brain slices were washed twice with PBST (0.5% Triton X-100 in 1× PBS) and then soaked in PBST for 10 min at RT to permeabilise the brain tissue. Brains were digested with proteinase K (160 μg/mL; Invitrogen, CA, USA, AM2548) at 37°C for 20 min and then washed thrice with PBST. Prehybridisation was performed by using a prehybridisation buffer (Wuhan Boster, Wuhan, China, AR0152) at 37°C for 30 min. Brain slices were hybridised with DIG–PAHAL and biotin–PAH probes (5 ng/μL) at 37°C overnight and then blocked at 4°C in 2% BSA (2% BSA in 0.2× SSC) for 20 min. Next, brain slices were incubated in anti-DIG alkaline phosphatase-conjugated antibody (1:500; Roche, Mannheim, Germany, 11093274910) and streptavidin–HRP (1:100) for 1 h at RT and then washed thrice with PBS. The fluorescent signal of DIG for PAHAL or that of biotin for PAH was detected with an HNPP Fluorescent Detection Set (Roche, Mannheim, Germany, 11758888001) or TSA Fluorescein System (Perkin-Elmer, MA, USA, NEL701A001KT). Images were captured under a LSM 710 confocal fluorescence microscope (Zeiss, Oberkochen, Germany) at 10× and 40× magnifications. S3 Table lists the primers used for the probe synthesis of PAHAL and PAH.

Isolation and crowding of locusts

The locusts were isolated by introducing typical G nymphs into metal boxes and individually rearing them under standard conditions. The locusts were crowded by introducing 10 labelled S nymphs and 20 G nymphs into an optic Perspex box (10 cm × 10 cm × 10 cm). Adequate fresh food was provided. After 0, 4 or 8 h of treatment, the locust brains were dissected and immediately frozen in liquid nitrogen for RNA preparation. All insects were sampled at the same time of the day to avoid the effects of circadian rhythm on the locust phenotypes. Equal numbers of male and female insects were sampled for each biological replicate.

Behavioural assays

Behavioural assays were performed as previously described [40,42]. A rectangular Perspex arena (40 cm × 30 cm × 10 cm) was used in the assay. One of the separated side chambers (7.5 cm × 30 cm × 10 cm) contained 30 G locusts as a stimulus group. Another chamber with the same dimensions was left empty. A locust was released into the centre of the arena and monitored for 300 s. Individual behavioural data were automatically recorded and analysed with the EthoVision video tracking system (v.3.1.16, Noldus Inc., Wageningen, Netherlands). The three behavioural variables were as follows: total distance moved (TDM) and total duration of movement (TDMV), which represent motor activity levels, and attraction index (AI, i.e., total duration in stimulus area minus total duration in opposite area), which represents the attraction or repulsion to the stimulus group. The behavioural phase state of each locust was assessed by applying a single probabilistic metric of gregariousness Pgreg, Pgreg = eη/(1 + eη), where η = −2.11 + 0.005 × AI + 0.012 × TDM + 0.015 × TDMV [36,42]. Pgreg indicates the probability of a locust regarded as G. Pgreg = 1 indicates fully G behaviour, whereas Pgreg = 0 means fully S behaviour.

Nuclear fractionation

Nuclear and cytoplasmic fractionation was performed as previously described [73]. Twenty nymphal brains were harvested and homogenised in a cold lysis buffer [1× PBS containing 0.2% IGEPAL CA-630 (Sigma, MO, USA, I8896-50ml) and 1× proteinase inhibitor (Pierce MA, USA, 88266) and RNase inhibitor (Promega, WI, USA, N2111S)] for nuclear fractionation. The homogenate was then centrifuged at 30 ×g for 2 min at 4°C. The supernatants were transferred to a fresh tube and centrifuged at 425 ×g for 15 min at 4°C to obtain the nuclear pellet. The cytoplasmic fraction in the supernatant was centrifuged at 2000 ×g for 10 min at 4°C to remove residual nuclei. The nuclear pellet and cytoplasmic supernatant were maintained at −80°C prior to RNA extraction.

Reporter and expression plasmid construction, luciferase assay and antibodies

Luciferase assays were performed to verify whether PAHAL regulates PAH in cis. Plasmids containing different promoter regions of PAH and their deletions were constructed into pGL4.10 (Promega, WI, USA, E665A) by using the KpnI and XhoI restriction sites. Promoter regions were amplified from locust genomic DNA. A series of ESEs with mutations in the PAH 5′-UTR (containing −554/+89 fragment) were inserted into pGL4.10 obtained from PolePolar Biotechnology Co., Ltd., Beijing, China.

The full-length PAHAL sequence, the PAHAL sequence in the reverse orientation (reverse PAHAL), the PAHAL sequence with deletions, the SRSF2 ORF and a 3 kb lacz ORF (as the negative control) were cloned into the pcDNA3.1 (+) vector (Invitrogen, CA, USA, V79020) and/or pAc5.10/V5-His A vector (Invitrogen, CA, USA, V4110-20) by using KpnI and XhoI. These vectors were transfected with Drosophila S2 cells, HEK 293T cells and SRSF2-MEFs.

Cells were plated on 48-well plates and transfected by using Lipofectamine 3000 in accordance with the manufacturer’s instructions (Invitrogen, CA, USA, L3000015). Reporter plasmids (10 ng) with 200 ng of the expression plasmid or negative control vector were co-transfected with 5 ng of the internal control plasmid pRL-TK/pGL4.73 (Promega, WI, USA, E2241/ E691A). Luciferase activity was measured with Dual-Luciferase Reporter Assay System (Promega, WI, USA, E1960) at 30 h after incubation.

The polyclonal antibodies for PAH and monoclonal SRSF2 antibody were produced by immunising mice with the prokaryotic expression peptide (Beijing Protein Innovation, Beijing, China). The tubulin (rabbit) and GAPDH (rabbit) antibodies [74] were provided by Dr. Yun-Dan Wang.

RNA pulldown and Western blot analysis

Twenty brains in one biological duplicate were lysed to acquire total protein for RNA pulldown by using 200 μL of tissue lysis buffer [200 μL of T-PER tissue protein extraction reagent (Pierce, CA, USA, 78510) and 2 μL of Halt Protease Inhibitor Cocktail, EDTA-free (Pierce, CA, USA, 87785)]. Such an undertaking was carried out to identify the interactive proteins with PAHAL in vivo. RNA pulldown was performed with a magnetic RNA–protein pull-down kit in accordance with the manufacturer’s instructions (Pierce, CA, USA, 20164). The RNA probes for full-length PAHAL and their deletions (biotin–PAHAL) were synthesised with a T7 RNA polymerase kit. The RNA-associated proteins were separated on 15% SDS-PAGE gel and then subjected to protein silver staining. Single silver-stained bands that were present in PAHAL pulldown but absent from antisense PAHAL pulldown were excised and then bleached by using a destaining buffer (50% acetonitrile and 25 mM NaHCO3). Disulphide bonds of protein samples were disrupted by adding 10 mM DTT for 1 h at 56°C and 55 mM Iodoacetamide for 45 min at RT. Protein was hydrolysed by incubation in trypsin buffer (62.5 ng/μL trypsin and 25 mM NaHCO3) at 37°C overnight. The protein samples were subjected to mass spectrometry (Beijing Protein Innovation, Beijing, China). MASCOT software (Multiple Alignment System developed by iCOT) [75] was used to identify and quantity proteins from the mass spectrometric data. The protein functions were annotated by the protein scores > 85, which is a criterion of the significant high expression level of proteins (P < 0.05) in this single silver-stained band.

RNA pulldown and western blot analysis were performed on SRSF2-MEFs (2 × 107) that co-transfected with pcDNA3.1/V5-His/SRSF2 ORF and a series of PAHAL-pcDNA3.1+ deletions. This task was carried out to identify the specific sites for SRSF2 interaction in the PAHAL sequence.

The total proteins for Western blot analysis were first extracted by using TRIzol reagent. The proteins were dissolved in rehydration/sample lysis buffer (in vitro translation assay) to a concentration of 10 μg/μL overnight at RT and then mixed with 1 μL of 1% bromophenol blue buffer. The proteins were then separated through SDS-PAGE on 10% NuPAGER Bis–Tris gel (Invitrogen, CA, USA, NP0315BOX) by using 1× NuPAGE MOPS SDS running buffer (Invitrogen, CA, USA, NP0050) in accordance with the manufacturer’s instructions. The separated proteins were transferred to PVDF membranes by using 1× transfer buffer and blocked with 5% (wt/vol) skimmed milk for 1 h at RT. The membranes were incubated in a blocking buffer at 4°C overnight with a primary antibody in the following concentration: anti-PAH (mouse), 1:500; anti-SRSF2 (mouse), 1:500; anti-SmD1 (rabbit), 1:2,000; anti-GAPDH (rabbit), 1:5,000; and anti-Tubulin (rabbit), 1:5,000. The secondary antibody (1:5,000; Easybio, Beijing, China, BE0101-100, BE0102-100) in the blocking buffer was incubated for 1 h at RT. The immunological blot was detected with SuperSignal West Femto Substrate Trial Kit (Pierce, CA, USA, 34094).

RNA immunoprecipitation (RIP) assay

The SRSF2 ORF was first cloned in frame with the V5 epitope of the pcDNA3.1/V5-His vector for the construction of pcDNA3.1/V5-His/SRSF2 ORF to test the binding of SRSF2 with PAHAL in vitro. HEK 293T cells were then co-transfected with this vector and pcDNA3.1 (+)/PAHAL for the following RIP assay. After 3 days, 2 × 107 of HEK 293T cells were harvested for the RIP experiment.

The RIP assay was performed with Magna RIP Quad RNA-Binding Protein Immunoprecipitation Kit (Millipore, CA, USA, 17–704). The cell pellet was homogenised in ice-cold RIP lysis buffer containing 1× proteinase inhibitor and RNase inhibitor and stored at −80°C overnight. Magnetic beads were pre-incubated with 5 μg of V5 antibody (Invitrogen, CA, USA, R96025) or normal mouse IgG (Millipore, CA, USA, CS200621) for 30 min at RT with rotation. The supernatants of the lysate from the centrifugation were co-incubated with the bead–antibody complex overnight at 4°C with rotation. Thereafter, 10 μL of the supernatants was stored as the input. The RNA in the immunoprecipitates and input was extracted by using TRIzol reagent. Reverse transcription was performed with a high-capacity RNA-to-cDNA kit. The target gene expression was analysed through qPCR.

The binding of SRSF2 with PAHAL in vivo was tested by performing the RIP assay on brain tissues. A Dynabeads Protein G immunoprecipitation kit (Thermo Fisher, CA, USA, 10007D) was used. S4 Table shows the antibody epitopes of PAH and SRSF2. Fifty brains in one biological replicate were lysed with 700 μL of T-PER Tissue Protein Extraction Reagent containing RNasin Plus RNase Inhibitor (Promega, WI, USA, N2611) and EDTA-free 1× Halt Protease Inhibitor Cocktail. After the lysates were centrifuged at 10,000 ×g for 10 min, the supernatants were collected and subjected to RIP by following the protocol included with the kit. A total of 100 μL of Dynabeads was pre-incubated with 10 μg of antibody for 10 min at RT with rotation. The supernatants (10 μL) of the lysate were stored as the input. A total of 300 μL of supernatants were co-incubated with the bead-specific antibody complex with rotation for 10 min at RT. The RNA and proteins were extracted using a TRIzol reagent before quantitative measurement.

Bioinformatics and statistical analysis

Transcriptome data subjected to poly(A) lncRNA screening were obtained from a previous publication [76]. Locust lncRNA sequences were predicted and identified through an integrative method as previously described [44,57]. Sequence conservation of PAHAL was analysed by blasting in the genome assembly of all species in flybase (www.flybase.org). The sequence motifs were WNNNNSNNAGCCC (W = A/T, S = G/C) for the NLS [47] and WSSNGYY (W = A/T, S = G/C Y = C/T) for the SRSF2-responsive ESE [49,77]. Data from the tissue expression and truncated luciferase assay were analysed through ANOVA, then by post-hoc Tukey’s b-test for multiple comparisons. Independent sample Student’s t-tests were performed for comparing differences in gene expression and other values between treatments. The frequency data of behavioural features, namely, Pgreg and AI, were analysed through Mann–Whitney U test. Two-sided P-values were provided. Data are presented as the mean ± SEM unless stated otherwise. All statistical data were analysed by using SPSS 21.0 (SPSS Inc., IL, USA). The locust genome data are available at the following website: http://www.locustmine.org. All sequences for PAHAL, PAH and SRSF2 have been deposited in GenBank under accession numbers KX962170, KX951493 and KX951494, respectively. The RNA-seq data have been uploaded to NCBI with the accession number PRJNA522953. Numerical data that underlies graphs or summary statistics and sample image data in support of all reported results have been uploaded to Harvard Dataverse Network with the websit https://dataverse.harvard.edu/dataset.xhtml?persistentId=doi:10.7910/DVN/LXRBKJ and https://dataverse.harvard.edu/dataset.xhtml?persistentId=doi:10.7910/DVN/KSVYWY

Acknowledgements

We are grateful to Xiang-Dong Fu for providing SRSF2-MEFs, Yun-Dan Wang for providing tubulin and GAPDH antibodies, Yi Ding and Weihua Wang for assisting in LC-MS, and Fangqing Zhao for the useful discussion.

References

1 

    TR Mercer, ME Dinger, JS Mattick. . Long non-coding RNAs: insights into functions. Nat Rev Genet. 2009; 10: , pp.155–159. , doi: 10.1038/nrg2521

2 

    S Geisler, J Coller. . RNA in unexpected places: long non-coding RNA functions in diverse cellular contexts. Nat Rev Mol Cell Biol. 2013; 14: , pp.699–712. , doi: 10.1038/nrm3679

3 

    JJ Quinn, HY Chang. . Unique features of long non-coding RNA biogenesis and function. Nat Rev Genet. 2016; 17: , pp.47–62. , doi: 10.1038/nrg.2015.10

4 

    JT Lee. . The X as Model for RNA’s Niche in Epigenomic Regulation.Cold Spring Harbor Perspect Biol.2010; 2: , pp.a003749, doi: 10.1101/cshperspect.a003749

5 

6 

    T Ohhata, CE Senner, M Hemberger, A Wutz. . Lineage-specific function of the noncoding Tsix RNA for Xist repression and Xi reactivation in mice. Gene Dev. 2011; 25: , pp.1702–1715. , doi: 10.1101/gad.16997911

7 

    F Michelini, S Pitchiaya, V Vitelli, S Sharma, U Gioia, F Pessina, et al. Damage-induced lncRNAs control the DNA damage response through interaction with DDRNAs at individual double-strand breaks. Nat Cell Biol. 2017; 19: , pp.1400–1411. , doi: 10.1038/ncb3643

8 

    L Williamson, M Saponaro, S Boeing, P East, R Mitter, T Kantidakis, et al. UV irradiation induces a non-coding RNA that functionally opposes the protein encoded by the same gene. Cell. 2017; 168: , pp.1–13. , doi: 10.1016/j.cell.2016.12.043

9 

    A Fatica, I Bozzoni. . Long non-coding RNAs: new players in cell differentiation and development. Nat Rev Genet. 2014; 15: , pp.7–21. , doi: 10.1038/nrg3606

10 

    S Bian, T Sun. . Functions of noncoding RNAs in neural development and neurological diseases. Mol Neurobiol. 2011; 44: , pp.359–373. , doi: 10.1007/s12035-011-8211-3

11 

    T Sallam, MC Jones, T Gilliland, L Zhang, X Wu, A Eskin, et al. Feedback modulation of cholesterol metabolism by the lipid-responsive non-coding RNA LeXis. Nature. 2016; 534: , pp.124–128. , doi: 10.1038/nature17674

12 

    X Liu, Z-D Xiao, L Han, J Zhang, S-W Lee, W Wang, et al. LncRNA NBR2 engages a metabolic checkpoint by regulating AMPK under energy stress. Nat Cell Biol. 2016; 18: , pp.431–442. , doi: 10.1038/ncb3328

13 

    YJ Tang, T Zhou, X Yu, ZX Xue, N Shen. . The role of long non-coding RNAs in rheumatic diseases.Nat Rev Rheumatol.2017; 13: , pp.657–669. , doi: 10.1038/nrrheum.2017.162

14 

    L Soreq, A Guffanti, N Salomonis, A Simchovitz, Z Israel, H Bergman, et al. Long non-coding RNA and alternative splicing modulations in Parkinson's leukocytes identified by RNA sequencing. PLoS Comput Biol. 2014; 10: , pp.e1003517, doi: 10.1371/journal.pcbi.1003517

15 

    IA Qureshi, MF Mehler. . Emerging roles of non-coding RNAs in brain evolution, development, plasticity and disease.Nat Rev Neurosci.2012; 13: , pp.528–541. , doi: 10.1038/nrn3234

16 

    VJ Peschansky, C Pastori, Z Zeier, K Wentzel, D Velmeshev, M Magistri, et al. The long non-coding RNA FMR4 promotes proliferation of human neural precursor cells and epigenetic regulation of gene expression in trans. Mol Cell Neurosci. 2016; 74: , pp.49–57. , doi: 10.1016/j.mcn.2016.03.008

17 

18 

    LR Earls, JJ Westmoreland, SS Zakharenko. . Non-coding RNA regulation of synaptic plasticity and memory: Implications for aging. Ageing Res Rev. 2014; 17: , pp.34–42. , doi: 10.1016/j.arr.2014.03.004

19 

    N Rani, TJ Nowakowski, H Zhou, SE Godshalk, V Lisi, AR Kriegstein, et al. A primate lncRNA mediates Notch signaling during neuronal development by sequestering miRNA. Neuron. 2016; 90: , pp.1174–1188. , doi: 10.1016/j.neuron.2016.05.005

20 

    EJ Shields, L Sheng, AK Weiner, BA Garcia, R Bonasio. . High-quality genome assemblies reveal long non-coding RNAs expressed in Ant brains. Cell Rep. 2018; 23: , pp.3078–3090. , doi: 10.1016/j.celrep.2018.05.014

21 

    M Sauvageau, LA Goff, S Lodato, B Bonev, AF Groff, C Gerhardinger, et al. Multiple knockout mouse models reveal lincRNAs are required for life and brain development. Elife. 2013; 2: , pp.e01749, doi: 10.7554/eLife.01749

22 

    AA Soshnev, H Ishimoto, BF McAllister, X Li, MD Wehling, T Kitamoto, et al. A conserved long noncoding RNA affects sleep behavior in Drosophila. Genetics. 2011; 189: , pp.455–468. , doi: 10.1534/genetics.111.131706

23 

    M Li, S Wen, X Guo, B Bai, Z Gong, X Liu, et al. The novel long non-coding RNA CRG regulates Drosophila locomotor behavior. Nucleic Acids Res. 2012; 40: , pp.11714–11727. , doi: 10.1093/nar/gks943

24 

    JM Engreitz, JE Haines, EM Perez, G Munson, J Chen, M Kane, et al. Local regulation of gene expression by lncRNA promoters, transcription and splicing. Nature. 2016; 539: , pp.452–455. , doi: 10.1038/nature20149

25 

    S Guil, M Esteller. . Cis-acting noncoding RNAs: friends and foes. Nat Struct Mol Biol. 2012; 19: , pp.1068–1075. , doi: 10.1038/nsmb.2428

26 

    FP Marchese, I Raimondi, M Huarte. . The multidimensional mechanisms of long noncoding RNA function. Genome Biol. 2017; 18: , pp.206, doi: 10.1186/s13059-017-1348-2

27 

    LL Chen. . Linking long noncoding RNA localization and function. Trends Biochem Sci. 2016; 41: , pp.761–772. , doi: 10.1016/j.tibs.2016.07.003

28 

    T Csorba, JI Questa, Q Sun, C Dean. . Antisense COOLAIR mediates the coordinated switching of chromatin states at FLC during vernalization.Proc Natl Acad Sci USA. 2014; 111: , pp.16160–16165. , doi: 10.1073/pnas.1419030111

29 

    BL Raveendra, S Swarnkar, Y Avchalumov, X-A Liu, E Grinman, K Badal, et al. Long noncoding RNA GM12371 acts as a transcriptional regulator of synapse function. Proc Natl Acad Sci USA2018; 115: , pp.E10197–E205. , doi: 10.1073/pnas.1722587115

30 

    PA Spadaro, CR Flavell, J Widagdo, VS Ratnu, M Troup, C Ragan, et al. Long noncoding RNA-directed epigenetic regulation of gene expression is associated with anxiety-like behavior in mice. Biol Psychiat. 2015; 78: , pp.848–859. , doi: 10.1016/j.biopsych.2015.02.004

31 

    SY Ng, L Lin, BS Soh, LW Stanton. . Long noncoding RNAs in development and disease of the central nervous system. Trends Genet. 2013; 29: , pp.461–468. , doi: 10.1016/j.tig.2013.03.002

32 

    S Gepshtein, XY Li, J Snider, M Plank, D Lee, H Poizner. . Dopamine function and the efficiency of human movement. J Cognitive Neurosci. 2014; 26: , pp.645–657. , doi: 10.1162/jocn_a_00503

33 

    M Cerovic, R d'Isa, R Tonini, R Brambilla. . Molecular and cellular mechanisms of dopamine-mediated behavioral plasticity in the striatum.Neurobiol Learn Mem. 2013; 105: , pp.63–80. , doi: 10.1016/j.nlm.2013.06.013

34 

    CM Coleman, WS Neckameyer. . Substrate regulation of serotonin and dopamine synthesis in Drosophila. Invert Neurosci. 2004; 5: , pp.85–96. , doi: 10.1007/s10158-004-0031-y

35 

    SC Daubner, T Le, S Wang. . Tyrosine hydroxylase and regulation of dopamine synthesis. Arch Biochem Biophys. 2011; 508: , pp.1–12. , doi: 10.1016/j.abb.2010.12.017

36 

    M Yang, Y Wei, F Jiang, Y Wang, X Guo, J He, et al. MicroRNA-133 inhibits behavioral aggregation by controlling dopamine synthesis in locusts.PLoS Genet.2014; 10: , pp.e1004206, doi: 10.1371/journal.pgen.1004206

37 

    P Ruiz-Vazquez, FJ Silva. . Aberrant splicing of the Drosophila melanogaster phenylalanine hydroxylase pre-mRNA caused by the insertion of a B104/roo transposable element in the Henna locus. Insect Biochem Mol Biol. 1999; 29: , pp.311–318. , doi: 10.1016/s0965-1748(99)00002-8

38 

39 

    X Guo, Z Ma, L Kang. . Two dopamine receptors play different roles in phase change of the migratory locust.Front Behav Neurosci. 2015; 9: , pp.80, doi: 10.3389/fnbeh.2015.00080

40 

    Z Ma, W Guo, X Guo, X Wang, L Kang. . Modulation of behavioral phase changes of the migratory locust by the catecholamine metabolic pathway. Proc Natl Acad Sci USA. 2011; 108: , pp.3882–3887. , doi: 10.1073/pnas.1015098108

41 

42 

    W Guo, X Wang, Z Ma, L Xue, J Han, D Yu, et al. CSP and takeout genes modulate the switch between attraction and repulsion during behavioral phase change in the migratory locust.PLoS Genet.2011; 7: , pp.e1001291, doi: 10.1371/journal.pgen.1001291

43 

    M Burrows, SM Rogers, SR Ott. . Epigenetic remodelling of brain, body and behaviour during phase change in locusts.Neural Syst Circuits.2011; 1: , pp.11, doi: 10.1186/2042-1001-1-11

44 

    L Wang, HJ Park, S Dasari, S Wang, JP Kocher, W Li. . CPAT: Coding-potential assessment tool using an alignment-free logistic regression model. Nucleic Acids Res. 2013; 41: , pp.e74, doi: 10.1093/nar/gkt006

45 

    CM Coleman, WS Neckameyer. . Serotonin synthesis by two distinct enzymes in Drosophila melanogaster. Arch Insect Biochem Physiol. 2005; 59: , pp.12–31. , doi: 10.1002/arch.20050

46 

    WS Neckameyer, K White. . A single locus encodes both phenylalanine hydroxylase and tryptophan hydroxylase activities in Drosophila. J Biol Chem. 1992; 267: , pp.4199–4206

47 

    B Zhang, L Gunawardane, F Niazi, F Jahanbani, X Chen, S Valadkhan. . A novel RNA motif mediates the strict nuclear localization of a long noncoding RNA. Mol Cell Biol. 2014; 34: , pp.2318–2329. , doi: 10.1128/MCB.01673-13

48 

    SR Lin, R Xiao, PQ Sun, XD Xu, XD Fu. . Dephosphorylation-dependent sorting of SR splicing factors during mRNP maturation. Mol Cell. 2005; 20: , pp.413–425. , doi: 10.1016/j.molcel.2005.09.015

49 

    X Ji, Y Zhou, S Pandit, J Huang, H Li, CY Lin, et al. SR proteins collaborate with 7SK and promoter-associated nascent RNA to release paused polymerase. Cell. 2013; 153: , pp.855–868. , doi: 10.1016/j.cell.2013.04.028

50 

    U Homberg. . Neuroarchitecture of the central complex in the brain of the locust Schistocerca gregaria and S. americana as revealed by serotonin immunocytochemistry. J Comp Neurol. 1991; 303: , pp.245–254. , doi: 10.1002/cne.903030207

51 

    B Wendt, U Homberg. . Immunocytochemistry of dopamine in the brain of the locust Schistocerca gregaria. J Comp Neurol. 1992; 321: , pp.387–403. , doi: 10.1002/cne.903210307

52 

    U Homberg. . Neurotransmitters and neuropeptides in the brain of the locust. J Comp Neurol. 2002; 56: , pp.189–209. , doi: 10.1002/jemt.10024

53 

    N Gil, I Ulitsky. . Regulation of gene expression by cis-acting long non-coding RNAs. Nat Rev Genet. 2020; 21: , pp.102–117. , doi: 10.1038/s41576-019-0184-5

54 

    P Yan, S Luo, JY Lu, X Shen. . Cis- and trans-acting lncRNAs in pluripotency and reprogramming. Curr Opin Genet Dev. 2017; 46: , pp.170–178. , doi: 10.1016/j.gde.2017.07.009

55 

    JB Heo, S Sung. . Vernalization-mediated epigenetic silencing by a long intronic noncoding RNA. Science. 2011; 331: , pp.76–79. , doi: 10.1126/science.1197349

56 

    TT Sun, J He, Q Liang, LL Ren, TT Yan, TC Yu, et al. Lncrna GClnc1 promotes gastric carcinogenesis and may act as a modular scaffold of WDR5 and KAT2A complexes to specify the histone modification pattern. Cancer Discov. 2016; 6: , pp.784–801. , doi: 10.1158/2159-8290.CD-15-0921

57 

    B Chen, Y Zhang, X Zhang, S Jia, S Chen, L Kang. . Genome-wide identification and developmental expression profiling of long noncoding RNAs during Drosophila metamorphosis.Sci Rep. 2016; 6: , pp.23330, doi: 10.1038/srep23330

58 

    I Sarropoulos, R Marin, M Cardoso-Moreira, H Kaessmann. . Developmental dynamics of lncRNAs across mammalian organs and species. Nature. 2019, 571: , pp.510–514. , doi: 10.1038/s41586-019-1341-x

59 

    Q Sun, Q Hao, KV Prasanth. . Nuclear long noncoding RNAs: key regulators of gene expression. Trends Genet. 2018, 34: , pp.142–157. , doi: 10.1016/j.tig.2017.11.005

60 

    KC Wang, YW Yang, B Liu, A Sanyal, R Corces-Zimmerman, Y Chen, et al. A long noncoding RNA maintains active chromatin to coordinate homeotic gene expression. Nature. 2011; 472: , pp.120–124. , doi: 10.1038/nature09819

61 

    S Lin, XD Fu. . SR proteins and related factors in alternative splicing. Adv Exp Med Biol. 2007; 623: , pp.107–122. , doi: 10.1007/978-0-387-77374-2_7

62 

    SR Lin, G Coutinho-Mansfield, D Wang, S Pandit, XD Fu. . The splicing factor SC35 has an active role in transcriptional elongation. Nat Struct Mol Biol. 2008; 15: , pp.819–826. , doi: 10.1038/nsmb.1461

63 

    JC Long, JF Caceres. . The SR protein family of splicing factors: master regulators of gene expression. Biochem J. 2009; 417: , pp.15–27. , doi: 10.1042/BJ20081501

64 

    V Tripathi, JD Ellis, Z Shen, DY Song, Q Pan, AT Watt, et al. The nuclear-retained noncoding RNA MALAT1 regulates alternative splicing by modulating SR splicing factor phosphorylation. Mol Cell. 2010; 39: , pp.925–938. , doi: 10.1016/j.molcel.2010.08.011

65 

    AK Sapra, ML Anko, I Grishina, M Lorenz, M Pabis, I Poser, et al. SR protein family members display diverse activities in the formation of nascent and mature mRNPs in vivo. Mol Cell. 2009; 34: , pp.179–190. , doi: 10.1016/j.molcel.2009.02.031

66 

    D Bernard, KV Prasanth, V Tripathi, S Colasse, T Nakamura, Z Xuan, et al. A long nuclear-retained non-coding RNA regulates synaptogenesis by modulating gene expression. EMBO J. 2010; 29: , pp.3082–3093. PMC2944070 , doi: 10.1038/emboj.2010.199

67 

    B Chen, S Li, Q Ren, X Tong, X Zhang, L Kang. . Paternal epigenetic effects of population density on locust phase-related characteristics associated with heat-shock protein expression. Mol Ecol. 2015; 24: , pp.851–862. , doi: 10.1111/mec.13072

68 

    S Chen, P Yang, F Jiang, Y Wei, Z Ma, L Kang. . De novo analysis of transcriptome dynamics in the migratory locust during the development of phase traits. PLoS One. 2010; 5: , pp.e15633, doi: 10.1371/journal.pone.0015633

69 

    MJ Burns, GJ Nixon, CA Foy, N Harris. . Standardisation of data from real-time quantitative PCR methods–evaluation of outliers and comparison of calibration curves.BMC Biotechnol.2005; 5: , pp.31, doi: 10.1186/1472-6750-5-31

70 

    A Fire, S Xu, MK Montgomery, SA Kostas, SE Driver, CC Mello. . Potent and specific genetic interference by double-stranded RNA in Caenorhabditis elegans. Nature. 1998; 391: , pp.806–811. , doi: 10.1038/35888

71 

72 

    Y Luo, X Wang, D Yu, L Kang. . The SID-1 double-stranded RNA transporter is not required for systemic RNAi in the migratory locust. RNA Biol. 2012; 9: , pp.663–671. , doi: 10.4161/rna.19986

73 

    J He, Q Chen, Y Wei, F Jiang, M Yang, S Hao, et al. MicroRNA-276 promotes egg-hatching synchrony by up-regulating brm in locusts. Proc Natl Acad Sci USA. 2016; 113: , pp.584–589. , doi: 10.1073/pnas.1521098113

74 

    Y Wang, P Yang, F Cui, L Kang. . Altered immunity in crowded locust reduced fungal (Metarhizium anisopliae) pathogenesis.PLoS Pathog.2013; 9: , pp.e1003102, doi: 10.1371/journal.ppat.1003102

75 

    M Hirosawa, M Hoshida, M Ishikawa, T Toya. . MASCOT: Multiple alignment system for protein sequences based on three-way dynamic programming. Bioinformatics. 1993; 9: , pp.161–167. , doi: 10.1093/bioinformatics/9.2.161

76 

    X Wang, X Fang, P Yang, X Jiang, F Jiang, D Zhao, et al. The locust genome provides insight into swarm formation and long-distance flight. Nat Commun. 2014; 5: , pp.2957, doi: 10.1038/ncomms3957

77 

    GM Daubner, A Cléry, S Jayne, J Stevenin, FHT Allain. . A syn–anti conformational difference allows SRSF2 to recognize guanines and cytosines equally well. EMBO J. 2012; 31: , pp.162–174. , doi: 10.1038/emboj.2011.367

21 Feb 2020

* Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out. *

Dear Dr Kang,

Thank you very much for submitting your Research Article entitled 'Long noncoding RNA PAHAL modulates locust behavioural plasticity through the feedback regulation of dopamine biosynthesis' to PLOS Genetics. Your manuscript was fully evaluated at the editorial level and by independent peer reviewers. The reviewers appreciated the attention to an important topic but identified some aspects of the manuscript that should be improved, and which may require a re-review of the revised manuscript.

We therefore ask you to modify the manuscript according to the review recommendations before we can again consider your manuscript for acceptance. Your revisions should address the specific points made by each reviewer.

In addition we ask that you:

1) Provide a detailed list of your responses to the review comments and a description of the changes you have made in the manuscript.

2) Upload a Striking Image with a corresponding caption to accompany your manuscript if one is available (either a new image or an existing one from within your manuscript). If this image is judged to be suitable, it may be featured on our website. Images should ideally be high resolution, eye-catching, single panel square images. For examples, please browse our archive. If your image is from someone other than yourself, please ensure that the artist has read and agreed to the terms and conditions of the Creative Commons Attribution License. Note: we cannot publish copyrighted images.

We hope to receive your revised manuscript within the next 30 days. If you anticipate any delay in its return, we would ask you to let us know the expected resubmission date by email to plosgenetics@plos.org.

If present, accompanying reviewer attachments should be included with this email; please notify the journal office if any appear to be missing. They will also be available for download from the link below. You can use this link to log into the system when you are ready to submit a revised version, having first consulted our Submission Checklist.

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org.

Please be aware that our data availability policy requires that all numerical data underlying graphs or summary statistics are included with the submission, and you will need to provide this upon resubmission if not already present. In addition, we do not permit the inclusion of phrases such as "data not shown" or "unpublished results" in manuscripts. All points should be backed up by data provided with the submission.

PLOS has incorporated Similarity Check, powered by iThenticate, into its journal-wide submission system in order to screen submitted content for originality before publication. Each PLOS journal undertakes screening on a proportion of submitted articles. You will be contacted if needed following the screening process.

To resubmit, you will need to go to the link below and 'Revise Submission' in the 'Submissions Needing Revision' folder.

[LINK]

Please let us know if you have any questions while making these revisions.

Yours sincerely,

John Ewer

Associate Editor

PLOS Genetics

Gregory P. Copenhaver

Editor-in-Chief

PLOS Genetics

Reviewer's Responses to Questions

Comments to the Authors:

Please note here if the review is uploaded as an attachment.

Reviewer #1: Summary:

Zhang et al. have presented an interesting and thorough mechanistic investigation into the role of phenylalanine hydroxylase (PAH) in brain biogenic amine metabolism and transcriptional regulatory networks. The aggregation switching behavior in locusts has been studied from a variety of perspectives and the authors build upon a significant number of studies that specifically indicate the role of dopamine and biogenic amines in this phenotypic transition. I found this paper in general to be methodologically strong and integrative, and the experiments across domains (targeted / high-throughput gene expression analysis, metabolic measurements, PCR, behavioral analysis) were linked clearly. The transcriptional activation experiments were very nice, and demonstrated that the authors were committed to exploring the exact functional of the PAHAL lncRNA – they passed a high level of evidence to show that the PAHAL transcriptional activation was relevant in multiple contexts (rather than e.g. simply identifying a putative lncRNA transcript from RNA-seq data and publishing a hypothesis paper). My comments are all in service of clarifying the experiments, results, and interpretations of the authors.

Main comments.

I am somewhat confused by Figure 2B. Usually, lncRNA regulate loci in trans, and are transcribed from loci that are distal to lncRNA targets (see comment 2 though). However Figure 2B looks like PAHAL has a transcriptional start site within the 4th intron of PAH, is transcribed in the same direction as PAH (e.g. at Line 102: “ The transcription direction of this gene [PAHAL] is the same as that of the canonical PAH (Fig 1B)”), and includes the 4th intron + 5th exon + part of the 5th intron (the red section). Is this the case? If so, then PAHAL’s sequence would be anti-sense to PAH (e.g. the PAHAL lncRNA would anneal to the sense strand of PAH so it would have sequence homology to the anti-sense PAH). Then I am confused by the primer design in Figure 1C -- the common reverse primer #11 is located within PAH exon 6, which is not within PAHAL as visualized, but apparently this means that Exon 6 is within PAH (as discussed in text, e.g. At Line 301). At Line 101 it is written: “The PAHAL sequence covers the sixth exon and part of the two introns flanking the exon in the PAH gene (Fig 1B and S1 Fig).”. First off, this not what I see in Figure 1B (Line 923: “The red shading indicates the location of PAHAL in PAH locus)”, where the red PAHAL transcript appears to be from the 4th intron to the middle of the 5th intron (and thus doesn’t include the 6th exon as suggested at Line 101). Then the Supplemental Figure S1 appears to be a screenshot of a fasta file of the sequence of the 3' end of PAHAL transcript, highlighting the NLS sequence, but not providing any information about which introns/exons of PAH are included within PAHAL (this Figure S1 should be more clearly informative, as it is used to support the idea at Line 101 “The PAHAL sequence covers the sixth exon and part of the two introns flanking the exon in the 102 PAH gene (Fig 1B and S1 Fig)”, which I am not seeing). So honestly I cannot tell exactly where the PAHAL locus/transcript begins and ends. For example in Figure 1C, the PCR with 10+9 gave no product, when this is what spans Exon 5 to Intron 5, I’d expect this to form a product with the unspliced PAH or PAHAL transcript. The caption mentions that the locus diagrams are not to scale – I think for a complex focal loci in this paper, a scale figure would be more appropriate.

Related to my comment 1: I appreciated the citations in Lines 295-299 about cis-regulatory lncRNA, but are there any citations to which % of lncRNA act in this manner? Otherwise these two cited examples could be the two rare exceptions. Or it could be discussed why we do not / cannot know which fraction of lncRNA may act in this way.

This is a comment about the RNAi experiments and interpretation. Line 116: “We knocked down PAHAL through dsRNA and siRNA interferences to test the regulatory relationship between PAHAL and PAH in the locust brain.”. I think that the unique relationship of PAH-PAHAL makes the dsRNA/siRNA experiments inconclusive, or at least not completely explained. Here is my contention: since dsRNA targeting the PAHAL locus also (by definition) targets the PAH locus, the dsRNA is directly knocking down PAH mRNA (as well as PAHAL) – in fact both transcripts are reduced an almost-identical amount, and follow the same regulatory time course in later experiments. The authors interpret their knockdown experiment solely in terms of “the dsRNA knocks down PAHAL mRNA, and the reduction of PAHAL mRNA reduces PAH”. However it is more parsimonious to suggest that the dsRNA corresponding to the PAH/PAHAL locus knocks down transcription of that entire locus, thus not necessarily acting via the transcriptional loop hypothesized. The piece I might be missing here is if the siRNA causes a “PAHAL-specific knockdown”, as suggested at Line 120. But again, any siRNA corresponding to PAHAL would correspond to an intron/exon of PAH and thus be likely to knock down PAH transcription directly – in other words, the entire logic of this section/experiment is that the dsRNA/siRNA experiments only knock down PAHAL (hence interpreting PAH decreases as mechanistically downstream of PAHAL), but because PAHAL is inside PAH, any RNAi-type experiments would be more simply interpreted as a transcriptional silencing of the PAH locus directly. I believe the correct control (or at least an experimental group to include in these experiments) for these dsRNA experiments would be a dsRNA/siRNA targeting a region in PAH but not PAHAL (e.g. in 1st exon of PAH). This would hypothetically reduce PAH without reducing PAHAL, and if this phenocopies the “PAHAL knockdown” presented here, would be strong support for the author’s interpretation. Also as a related note, just because a RNAi knockdown reduces a metabolite X, this does not mean that that the knocked down gene is a rate-limiting step in the synthesis of X.

I think that the sentence at Line 293-294, that the positive transcriptional influence of PAHAL was verified in “locust, fruit fly, and mice in vivo and in virto” is extremely strong. I think that this framing could be introduced earlier – that the effect of any candidate genes identified through high-throughput approaches would be verified through in vivo and in vitro experiments in multiple species.

The SRSF2 experiments were very nice, and good support for the article’s main hypotheses.

A large body of research suggests that the rate limiting step of Dopamine biosynthesis is Tyrosine Hydroxylase (TH), and TH is used commonly as the defining marker of dopaminergic neurons (e.g. to provide just one citation: ncbi.nlm.nih.gov/pmc/articles/PMC3065393/ ). In the Figure 1a of this manuscript, it is clearly laid out that TH and DDC are both required after PAH for DA synthesis. At Line 72 the authors write that PAH “is a rate-limiting enzyme in the synthesis of DA and other bioamines [34,35]”, but citations 34 & 35 do not specifically support the idea that PAH is indeed rate-limiting for DA or any amine other than Tyrosine. In the Discussion from Lines 259-266 there are good citations relating to the specific role of PAH in locust, these citations could be relevant in the intro as well. Otherwise, I believe this statement that links PAH directly as rate-limiting to DA synthesis is inaccurate (it is TH, not PAH). In general the manuscript is oriented towards the specific role of PAH in DA synthesis, and this ends up neglecting non-PAH players in DA metabolism & PAH’s role in other processes.

In general, whenever a P value is presented, I would appreciate full statistical information (e.g. Line 162 and 168 mention the statistical test used, add information on the N, and test statistic). In many other cases, I was unclear which statistical test was being performed, and how large the sample sizes were, without combining back through Methods/Materials/etc.

Minor comments:

Line 20: As is, this sentence sounds like PAHAL is what "is known for extreme behavioral plasticity", not the locusts. To clarify the sentence, something like "...behavioural adjustment in migratory locusts, a species with extreme behavioral plasticity". Also the latin species name could be included in the abstract, for search reasons and clarity.

Line 22, 42, elsewhere: Repeatedly this lncRNA is described as being “is sense to PAH gene”. This use of the word “sense” may be obscure for those who are unfamiliar with the sense/anti-sense terminology. Perhaps in an early mention, something like “PAHAL has sequence homology to the sense (coding) strand of the PAH locus” or “PAHAL ” (or whatever accurate statement would clarify my uncertainty in the next sentences).

Line 77: Would be helpful to add an estimate of how long "rapidly shift their behaviours" is referring to. E.g. does "rapidly" mean 1 hour, 1 day, 1 week.

Line 128-131: This sentence makes it sound like the phenylethylamine (tyrosine/dopa/dopamine/octopamine) and tryptamine (5HTP/serotonin) biogenic amines are all downstream of PAH. Rather, as pointed out a few lines later, there are two separate branches (one downstream of phenyalanine/PAH, and one downstream of tryptophan). The authors then claim at Line 135 that “These results indicate that 136 PAHAL only affects the DA synthesis in the catecholamine metabolic pathway”, which is incorrect – the knockdown of PAH (whether directly or via PAHAL transcriptional loop) resulted in changes to tyrosine plus some phenylethylamines (l-dopa/dopamine but not octopamine/tyramine), and also changes to 5HTP levels (why do the authors think that this tryptamine would be decreased by PAH knockdown when it is not linked to phenylethylamine metabolism?).So while it is true that “PAHAL only affects the DA synthesis in the catecholamine metabolic pathway”, this section would benefit from an explanation or interpretation of the unexpected changes to 5HTP on the non-catecholamine branch.

Line 145-146: “The result showed that PAH and PAHAL mRNA were mainly expressed in cell bodies” – as opposed to what? The cell body is where all transcription and gene expression occurs (even if RNA is transported or localized elsewhere, it is still “expressed” from the nucleus/body). So I don’t know what the point of this sentence is, as all genes are expressed in the cell body (maybe the authors mean to say that the transcripts are localized only in the cell bodies).

In Figure 2E I would prefer if the Y-axes in Figure 2 were not truncated – this makes the differences between the groups appear to be much larger than they actually are.

Figure 3B -- I do not understand why the "Rlative mRNA" (typo, should be "Relative") starts at 0.015, instead of 1. What is this mRNA being relativized to?

If the raw RNA-seq reads have been uploaded to a public database (e.g. Short Read Archive), this could be included (or perhaps it already here and I missed it).

Grammatical suggestions:

Line 42: “mediated the transcription activation” –> “transcriptional”

Line 57: “exhibits the spatiotemporal specific expression patterns” can delete “the”.

Line 60: “regulate the gene expression” can delete “the”

Line 66: "is" --> "are"

Reviewer #2: Zhang et al., describe the neural expression a locust long non-coding RNA (lncRNA), PAHAL, and provide good evidence that it regulates expression of the gene encoding phenylalanine hydroxylase (PAH) by binding to the SRSF2 splicing factor and recruiting it to the PAH promoter. Evidence is presented that this regulatory mechanism controls dopamine (DA) amounts in the locust brain. The authors show that PAHAL is expressed in the brain (and in the nucleus) and that both it and PAH are elevated when locusts are in crowded conditions and decreased when the insects are isolated. This suggests a role for the lncRNA in the regulation of social behavior. Further, they show that RNAi-mediated reductions in PAHAL result in a transition from group to solitary behavior, supporting the idea that it can regulate such behavior. Additional studies indicate that PAHAL binds to a short segment of the proximal promoter region of PAH, providing a mechanism for regulation of the gene. Using immunoprecipitation assays and brain tissues, the authors show that PAHAL is bound to SRSF2, in vivo, and that SRSF2 amounts increase with locust crowding and decrease in solitary locusts, similar to what is observed for PAHAL and PAH. They use cell-based assays to show that SRSF2 enhances expression of PAH and that it interacts with PAHAL to regulate gene expression. Finally, the authors identify the sequences within PAHAL that bind SRSF2 for its recruitment to PAH as well as exonic-splicing enhancer (ESE) sites within PAH that mediate regulation by SRSF2. All of these results lead to a satisfying model in which PAHAL recruits SRSF2 to the PAH gene to regulate its expression, dopamine signaling and behavior.

I have no major issues with the manuscript, but the following should be addressed prior to publication:

-The Discussion section should be shortened considerably; it is far too long.

-What was the reason for thinking PAHAL regulated PAH. Was it just the proximity of PAHAL to PAH or was there another reason for believing that PAHAL might be regulatory? The authors may wish to state this explicitly.

-I was initially confused by the PAHAL knockdown experiment because it seemed like the dsRNA would target both PAH and PAHAL. I then realized that the dsRNA was targeted to an intron of PAH, so it ought to be specific for PAHAL. The authors should add a sentence in the appropriate section making this clear so that other readers will not be confused.

-p. 6, Please indicate somewhere in this first paragraph that primer sequences for making dsRNA and siRNAs can be found in supplemental materials.

-p 8, it might be good here to remind the reader about G and S populations at the beginning of this section, even though they were described in the Intro. The average reader won't be as familiar with locust behavior as the authors.

-p 8, the images in Fig. 3C are difficult to interpret unless one happens to be a locust neurobiologist. Which cell bodies are shown? Are the RNAs expressed everywhere in the brain or is there a more specific expression pattern? Do the images represent the whole brain or part of the brain? Also, it looks like PAHAL is predominantly nuclear whereas PAH is cytoplasmic so I'm not sure if colocalization is the right word to use. I would say they are in the same cells with one being nuclear and the other cytoplasmic. Similarly, in Fig. 5B, what brain region is shown?

-p. 10, top, does siRNA or dsRNA for PAH and PAHAL have the same effects on behavior. Perhaps this is known from a previous publication. If so, please mention it.

-p. 11, bottom, please state clearly that SRSF2 was tagged with V5. The wording here is a little unclear.

-p 13, there is reference to Fig 6C which I think should be 7C.

-Fig. 6 is missing the legend for panel F.

-It's almost impossible to read the names of the genes shown in Fig 2C. Please create a larger figure for publication.

-In some places PAHAL is referred to as mRNA. It’s not really a messenger RNA as it doesn’t encode a protein. It seems better to refer to it as RNA, not mRNA.

Reviewer #3: The authors investigate the regulation of the gene phenylalanine hydroxylase (PAH) in the locust, which plays a role in dopamine production in the brain that, in turn, regulates behavioral phase changes (solitary vs. gregarious). A model is proposed wherein a portion of the unspliced PAH transcript is separately transcribed as a sense long noncoding RNA (PAHAL) that positively regulates PAH transcription by recruiting a splicing factor protein that interacts with the PAH promoter. The regulatory potential of long noncoding RNAs is an exciting area of research. The authors present an impressive array of experimental data, including gene expression assays, RNA interference, RNA immunoprecipitation, mouse cell knockouts, and luciferase assays of PAHAL (and interacting factors) effects on transcription. The results of these experiments provide insights into PAH regulation important to dopamine-mediated locust phase transitions, but I have some concerns about the robustness of the described role of the long noncoding RNA, PAHAL, that is proposed to mediate neurotransmitter metabolism – and behavioral plasticity by orchestrating a local transcriptional loop.

Could the PAHAL dsRNA used to target PAHAL also target PAH pre-mRNA (since both are in the nucleus) to directly decrease PAH expression?

Did the nuclear fractionation experiment differentiate PAHAL from PAH pre-mRNA?

For the luciferase assays, is it possible that the PAHAL enhancer effect is simply due to a genomic enhancer element residing in an intron? In other words, could it be that the effect on transcription is not typically RNA-mediated but is rather through DNA looping interactions with the promoter? Can the authors comment on whether/how their experiments preclude this possibility?

Could the RNA pulldown with SRSF2 be capturing pre-mRNA in the nucleus that is being targeted by the spliceosome (including SRSF2, which is associated with nascent RNA and the spliceosome)?

Line 81: I did not see PAH mentioned in the abstract or in a word search of ref 41. Can the link between microRNA-133 and PAH in this sentence be described more clearly? Was PAH knocked down in this prior study?

The senior author has argued that other single factors like a microRNA are responsible for modulating dopamine and behavioral plasticity in locusts in other publications. Can the authors take an integrative view in their discussion to describe how the findings in the present manuscript fit in with their prior findings about dopamine regulation in the locusts?

Overall, I think it is an overstatement to conclude that “PAHAL modulates behavioral aggregation” as stated in the abstract and that “PAHAL mediates PAH transcriptional activation.” Some of my hesitance to accept these statements arises because it seems possible that PAHAL is produced only in response to PAH activation to further enhance transcription (what activates PAHAL?). I would like to see the authors respond to potential sources of uncertainty outlined above in my comments and more explicitly lay out the logic that counteracts such concerns or otherwise introduce more uncertainty to the interpretation of their findings where appropriate.

**********

Have all data underlying the figures and results presented in the manuscript been provided?

Large-scale datasets should be made available via a public repository as described in the PLOS Geneticsdata availability policy, and numerical data that underlies graphs or summary statistics should be provided in spreadsheet form as supporting information.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: None

**********

PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

Reviewer #3: No


23 Mar 2020

Submitted filename: Response_to_Reviewers-zx321.docx

2 Apr 2020

* Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out. *

Dear Dr Kang,

Thank you very much for submitting your Research Article entitled 'Long noncoding RNA PAHAL modulates locust behavioural plasticity through the feedback regulation of dopamine biosynthesis' to PLOS Genetics. Your revised manuscript was fully evaluated at the editorial level and again by independent peer reviewers. The reviewers appreciated your thorough responses to their comments. There is now only one issue pending, which was raised by Reviewer #2; we ask that you address this issue before we can consider your manuscript for acceptance.

This next version will not be sent out again for review; instead the newly revised version will be evaluated at the editorial level. To assist us in this job, please indicate in your cover letter how you have responded to this comment, and in your manuscript please highlight the text you have added/modified. 

In addition we ask that you upload a Striking Image with a corresponding caption to accompany your manuscript if one is available (either a new image or an existing one from within your manuscript). If this image is judged to be suitable, it may be featured on our website. Images should ideally be high resolution, eye-catching, single panel square images. For examples, please browse our archive. If your image is from someone other than yourself, please ensure that the artist has read and agreed to the terms and conditions of the Creative Commons Attribution License. Note: we cannot publish copyrighted images.

We hope to receive your revised manuscript within the next 30 days. If you anticipate any delay in its return, we would ask you to let us know the expected resubmission date by email to plosgenetics@plos.org.

If present, accompanying reviewer attachments should be included with this email; please notify the journal office if any appear to be missing. They will also be available for download from the link below. You can use this link to log into the system when you are ready to submit a revised version, having first consulted our Submission Checklist.

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org.

Please be aware that our data availability policy requires that all numerical data underlying graphs or summary statistics are included with the submission, and you will need to provide this upon resubmission if not already present. In addition, we do not permit the inclusion of phrases such as "data not shown" or "unpublished results" in manuscripts. All points should be backed up by data provided with the submission.

PLOS has incorporated Similarity Check, powered by iThenticate, into its journal-wide submission system in order to screen submitted content for originality before publication. Each PLOS journal undertakes screening on a proportion of submitted articles. You will be contacted if needed following the screening process.

To resubmit, you will need to go to the link below and 'Revise Submission' in the 'Submissions Needing Revision' folder.

[LINK]

Please let us know if you have any questions while making these revisions.

Yours sincerely,

John Ewer

Associate Editor

PLOS Genetics

Gregory P. Copenhaver

Editor-in-Chief

PLOS Genetics

Reviewer's Responses to Questions

Comments to the Authors:

Please note here if the review is uploaded as an attachment.

Reviewer #1: Summary:

Zhang et al. have revised their paper which investigated phenylalanine hydroxylase (PAH) in the migratory locust (Locusta migratoria).

In general the responses in blue text to my comments were excellent. In several of the responses to my comments, and the responses to Reviewer #3, the authors demonstrate a very reasoned understanding of RNA and neurotransmitter biology.

I appreciated the addition of the RNAi control experiments, the edits to the Figures, and the addition of accessible datasets (which I have not downloaded and investigated in detail, but trust that the authors have uploaded all info into an authorized database).

Other than that, I am satisfied with the revisions and have nothing more specific to add. With the key changes to the cited literature about lncRNA, PAH functionality, and PAHAL experiments, the manuscript is significantly improved.

Reviewer #2: Regarding my comments about Fig. 3, I understand that locust cell bodies are in a cortex surrounding the neuropil; this is similar to Drosophila. Rather, I was referring to the localization of PAH and PAHAL mRNAs within neuronal cell bodies. PAH is cytoplasmic whereas PAHAL is nuclear so strictly speaking they do not co-localize (i.e., they are in different cell compartments). A better way to describe their localization is to say something like: "They are both localized to neurons, but as expected PAH is cytoplasmic whereas PAHAL is nuclear".

And, the authors still need to indicate whether PAH and PAHAL are expressed in all neurons or just a subset (e.g., DA neurons). Fig. 3 makes it look like PAH and PAHAL have broad expression patterns rather than being localized only in a subset of locust neurons. Please clarify.

And, further, where are the aminergic neurons localized in the locust brain? This must be known. A comparison of the observed PAH and PAHAL expression patterns with the known location of aminergic neurons would be informative. I don't expect the authors to verify the location of aminergic neurons, just to point out their general location in the locust brain in comparison to the PAH and PAHAL expression patterns.

Reviewer #3: The authors’ responses to my questions and concerns were thoughtful and thorough. Well done.

**********

Have all data underlying the figures and results presented in the manuscript been provided?

Large-scale datasets should be made available via a public repository as described in the PLOS Geneticsdata availability policy, and numerical data that underlies graphs or summary statistics should be provided in spreadsheet form as supporting information.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

Reviewer #3: No


6 Apr 2020

Submitted filename: Response_to_Reviewers_R2.docx

9 Apr 2020

Dear Dr Kang,

We appreciate the care you have taken in addressing the most recent comments from reviewers and are pleased to inform you that your manuscript entitled "Long noncoding RNA PAHAL modulates locust behavioural plasticity through the feedback regulation of dopamine biosynthesis" has been editorially accepted for publication in PLOS Genetics. Congratulations!

Before your submission can be formally accepted and sent to production you will need to complete our formatting changes, which you will receive in a follow up email. Please be aware that it may take several days for you to receive this email; during this time no action is required by you. Please note: the accept date on your published article will reflect the date of this provisional accept, but your manuscript will not be scheduled for publication until the required changes have been made.

Once your paper is formally accepted, an uncorrected proof of your manuscript will be published online ahead of the final version, unless you’ve already opted out via the online submission form. If, for any reason, you do not want an earlier version of your manuscript published online or are unsure if you have already indicated as such, please let the journal staff know immediately at plosgenetics@plos.org.

In the meantime, please log into Editorial Manager at https://www.editorialmanager.com/pgenetics/, click the "Update My Information" link at the top of the page, and update your user information to ensure an efficient production and billing process. Note that PLOS requires an ORCID iD for all corresponding authors. Therefore, please ensure that you have an ORCID iD and that it is validated in Editorial Manager. To do this, go to ‘Update my Information’ (in the upper left-hand corner of the main menu), and click on the Fetch/Validate link next to the ORCID field.  This will take you to the ORCID site and allow you to create a new iD or authenticate a pre-existing iD in Editorial Manager.

If you have a press-related query, or would like to know about one way to make your underlying data available (as you will be aware, this is required for publication), please see the end of this email. If your institution or institutions have a press office, please notify them about your upcoming article at this point, to enable them to help maximise its impact. Inform journal staff as soon as possible if you are preparing a press release for your article and need a publication date.

Thank you again for supporting open-access publishing; we are looking forward to publishing your work in PLOS Genetics!

Yours sincerely,

John Ewer

Associate Editor

PLOS Genetics

Gregory P. Copenhaver

Editor-in-Chief

PLOS Genetics

www.plosgenetics.org

Twitter: @PLOSGenetics

----------------------------------------------------

Comments from the reviewers (if applicable):

----------------------------------------------------

Data Deposition

If you have submitted a Research Article or Front Matter that has associated data that are not suitable for deposition in a subject-specific public repository (such as GenBank or ArrayExpress), one way to make that data available is to deposit it in the Dryad Digital Repository. As you may recall, we ask all authors to agree to make data available; this is one way to achieve that. A full list of recommended repositories can be found on our website.

The following link will take you to the Dryad record for your article, so you won't have to re‐enter its bibliographic information, and can upload your files directly: 

http://datadryad.org/submit?journalID=pgenetics&manu=PGENETICS-D-20-00072R2

More information about depositing data in Dryad is available at http://www.datadryad.org/depositing. If you experience any difficulties in submitting your data, please contact help@datadryad.org for support.

Additionally, please be aware that our data availability policy requires that all numerical data underlying display items are included with the submission, and you will need to provide this before we can formally accept your manuscript, if not already present.

----------------------------------------------------

Press Queries

If you or your institution will be preparing press materials for this manuscript, or if you need to know your paper's publication date for media purposes, please inform the journal staff as soon as possible so that your submission can be scheduled accordingly. Your manuscript will remain under a strict press embargo until the publication date and time. This means an early version of your manuscript will not be published ahead of your final version. PLOS Genetics may also choose to issue a press release for your article. If there's anything the journal should know or you'd like more information, please get in touch via plosgenetics@plos.org.


20 Apr 2020

PGENETICS-D-20-00072R2

Long noncoding RNA PAHAL modulates locust behavioural plasticity through the feedback regulation of dopamine biosynthesis

Dear Dr Kang,

We are pleased to inform you that your manuscript entitled "Long noncoding RNA PAHAL modulates locust behavioural plasticity through the feedback regulation of dopamine biosynthesis" has been formally accepted for publication in PLOS Genetics! Your manuscript is now with our production department and you will be notified of the publication date in due course.

The corresponding author will soon be receiving a typeset proof for review, to ensure errors have not been introduced during production. Please review the PDF proof of your manuscript carefully, as this is the last chance to correct any errors. Please note that major changes, or those which affect the scientific understanding of the work, will likely cause delays to the publication date of your manuscript.

Soon after your final files are uploaded, unless you have opted out or your manuscript is a front-matter piece, the early version of your manuscript will be published online. The date of the early version will be your article's publication date. The final article will be published to the same URL, and all versions of the paper will be accessible to readers.

Thank you again for supporting PLOS Genetics and open-access publishing. We are looking forward to publishing your work!

With kind regards,

Laura Mallard

PLOS Genetics

On behalf of:

The PLOS Genetics Team

Carlyle House, Carlyle Road, Cambridge CB4 3DN | United Kingdom

plosgenetics@plos.org | +44 (0) 1223-442823

plosgenetics.org | Twitter: @PLOSGenetics

This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
https://www.researchpad.co/tools/openurl?pubtype=article&doi=10.1371/journal.pgen.1008771&title=Long noncoding RNA <i>PAHAL</i> modulates locust behavioural plasticity through the feedback regulation of dopamine biosynthesis&author=&keyword=&subject=Research Article,Biology and Life Sciences,Agriculture,Pests,Insect Pests,Locusts,Biology and Life Sciences,Organisms,Eukaryota,Animals,Invertebrates,Arthropoda,Insects,Locusts,Biology and life sciences,Biochemistry,Nucleic acids,RNA,Non-coding RNA,Long non-coding RNAs,Biology and life sciences,Genetics,Gene expression,Gene regulation,Small interfering RNAs,Biology and life sciences,Biochemistry,Nucleic acids,RNA,Non-coding RNA,Small interfering RNAs,Biology and life sciences,Genetics,Gene expression,DNA transcription,Biology and Life Sciences,Genetics,Gene Expression,Biology and Life Sciences,Genetics,Gene Expression,Gene Regulation,Transcriptional Control,Biology and Life Sciences,Biochemistry,Metabolism,Metabolic Pathways,Biology and Life Sciences,Developmental Biology,Life Cycles,Nymphs,