Identification of sex-biased MiRNA markers informative of heat-past events
Tosca A. van Gelderen, Jerome Montfort, José Antonio Álvarez-Dios, Francesc Piferrer, Julien Bobe, Laia Ribas
TL;DR
This study identifies miRNAs in fish gonads that can indicate past exposure to high temperatures, which can influence sex ratios.
Contribution
The study discovers conserved sex-biased miRNAs across multiple fish species that are informative of heat-induced developmental changes.
Findings
Three and twelve miRNAs in ovaries and testes, respectively, reflect past thermal events in European sea bass.
Thirty-three conserved sex-biased miRNAs were identified across nine fish species.
Specific miRNAs like miR-223-3p and miR-143-3p show thermosensitive expression patterns linked to heat exposure.
Abstract
Elevated temperatures during early developmental stages play a pivotal role in the fate of the adult sexual phenotype of fish populations, usually leading to male-skewed sex ratios. This is the case with European sea bass (Dicentrarchus labrax), one of the most important species in the European aquaculture industry. To unveil informative markers of the past thermal events, we investigated changes in the miRNome within the gonads of this species. Consequently, we exposed European sea bass to elevated temperatures (21ºC) during early development (from 7 to 68 days post fertilization). After one-year post-heat treatment growing at natural temperature, a miRNA-sequencing analysis was conducted in the ovaries and testes of juvenile fish. The examination of miRNA expression levels identified three and twelve miRNAs in ovaries and testes, respectively, reflecting past thermal events. To assess…
Genes, proteins, chemicals, diseases, species, mutations and cell lines named across the full text — each resolved to its canonical identifier and authoritative record.
Click any figure to enlarge with its caption.
Figure 1
Figure 2
Figure 3
Figure 4
Figure 5
Figure 6- —Consejo Superior de Investigaciones Cientificas (CSIC)
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsGenetic and Clinical Aspects of Sex Determination and Chromosomal Abnormalities · MicroRNA in disease regulation · Reproductive biology and impacts on aquatic species
Introduction
The European sea bass (Dicentrarchus labrax) is important for aquaculture. Since 2016, aquaculture has accounted for 96% of the total European sea bass production in comparison with that obtained from fisheries [1, 2]. To ensure successful growth and development in cultured fish, it is essential to maintain optimal rearing. For economic production efficiency, European sea bass females are the desired sexual phenotype as they are favored due to their larger size, which can be up to 30% larger than their male counterparts [3]. However, farmed European sea bass is known to produce populations consisting of 70–100% males, due to increased temperatures during early stages of development in the hatcheries [4, 5]. Various studies have attempted to identify the ideal rearing temperature for European sea bass larvae but it is known that temperatures exceeding 16 °C during the thermal sensitive window (∼ 0 to 150 days post fertilization, dpf) lead to male-skewed populations [6, 7].
Unlike species with genotypic sex determination, some teleost fish, including the European sea bass, exhibit polygenic sex determination (PSD) [8]. In PSD, environmental factors like temperature [9], stress [10], hypoxia [11], density [12], chemical compounds [13] and even the activation of the immune system [14] play a substantial role in shaping the final sexual phenotype of individuals. These external influences impact gene expression through various epigenetic mechanisms (i.e., histone modification, DNA methylation and, post-transcriptional regulators, like miRNAs) [15]. The first description of epigenetics in sex regulation due to temperature in fish was given in European sea bass more than a decade ago. DNA methylation of the gonadal aromatase cytochrome P450, family 19, subfamily A (cyp19a1a) promoter —a key gene for ovarian development in non-mammalian vertebrates— was altered in adults previously reared at a high temperature during early development, favoring population masculinization [16]. Since then, research has been deployed to better understand the connection between the epigenome and sexual phenotypes. This is the case, for example, in pufferfish (Takifugu rubripes), or in half-smooth tongue sole (Cynoglossus semilaevis) [17, 18].
The function of miRNAs —small non-coding RNAs that play a vital role in post-transcriptional gene regulation— is to act as an epigenetic mechanism and serve as a bridge between the environment and the phenotype. In the last years, the importance of miRNAs in fish has flourished and thus research related to miRNA in cultured fish has grown exponentially. Many studies regarding temperature alterations and miRNA expression have been described in various species and tissues in fish. For example, after short-term and acute high or low temperature exposures (18 days), 29 miRNAs were differentially expressed (DE) in the common carp (Cyprinus carpio) liver [19]. In the Antarctic notothenioid (Trematomus bernacchii), adapted to extreme climates of temperatures around − 1 °C, an acute heat stress was able to alter twelve miRNAs in the gills, mainly pertaining to cellular stress response [20]. Long-term effects on the miRNome were shown in zebrafish (Danio rerio) [21] and in Atlantic cod (Gadus morhua) [22]. Further, adult zebrafish exposed to high temperature for 21 days changed the miRNome with a consecutive recovery in which miRNA expressions returned to their original state [23]. In particular, in European sea bass exposed at high temperature during different stages of early development, miRNA expression was assessed, showing an alteration of heat-sensitive miRNAs in which the immune and the reproductive systems were involved [24].
One of the particularities that makes miRNAs good markers for conserved developmental processes is their high level of conservation across evolution —sequence might be conserved but function could change across evolution— [25]. miRNAs emerged during early evolution, i.e., in metazoans, and some miRNAs such as miR-100, miR-125 and let-7 have been described as ancient miRNAs in bilaterians [26]. Thus, ever since their initial discovery, miRNAs have been used as tools in extensive applications in human health diagnostics, primarily for tumor detection [27]. In the context of aquaculture, and since the first miRNA discovery in zebrafish [28], numerous studies have been undertaken to explore and characterize the miRNA repertoire in other teleost species. In fish, miRNA expression patterns were proposed as indicators to improve productivity by, for example, optimizing selection for breeding programs. This is the case of rainbow trout (Onchorhynchus mykiss), in which circulating miRNAs in the blood were recently identified as a non-invasive approach to detail the metabolic and reproductive states [29]. In Nile tilapia (Oreochromis niloticus) miR-1, miR-206, and miR-133a abundance was interpreted as markers to link genetic polymorphism data with miRNA targets [30]. By using public databases, a set of miRNAs (i.e., miR-9-3p, miR-135c, miR-9-5p, miR-30b, miR-122 and miR-92a-3p) was suggested as a collection of potential markers for cold tolerance in fish [31].
Notwithstanding the existence of many studies regarding the miRNA roles in the reproductive system, little is still known about the presence of many miRNAs in fish, particularly in the context of sex development. The elucidation of conserved miRNAs within the gonads of teleost fish species may provide insights for the identification of markers relevant to sex development. In light of the relative lack of research on the European sea bass species in this field, delving into the European sea bass miRNome promises to offer valuable insights into the intricate relationship between the environment, miRNA expression patterns and the sexual phenotype. Thus, the aims of the present study were: (1) to enrich the knowledge of miRNA present in the European sea bass gonads; (2) to identify sex specific markers in fish; (3) to discover heat-recorders of environmental past events like temperature. To achieve our goals, miRNA-sequencing strategies of European sea bass gonads subjected to high temperatures were used. Furthermore, we undertook a comparative analysis of our European sea bass miRNA data with obtained profiles of public gonadal miRNAs databases of nine different species [32]: zebrafish, three-spined stickleback (Gasterosteus aculeatus), striped catfish (Pangasianodon hypophthalmus), Japanese medaka (Oryzias latipes), black bullhead (Ameiurus melas), European perch (Perca fluviatilis) and eastern mudminnow (Umbra pygmaea) and two Holostean species: bowfin (Amia calva) and spotted gar (Lepisosteus oculatus) [33]. This study highlights the identification of potential miRNAs as candidates for markers of heat-induced masculinization in fish gonads, as well as miRNAs exhibiting sexually dimorphic responses in the gonads across various fish species. These findings offer promising potential as tools for predicting historical environmental events linked to masculinization caused by exposure to high-temperature.
Materials and methods
Experimental design
The fish used in this study were siblings of those used in previous studies in which a significant male-skewed sex ratio (70.45% males to 29.55% females, *p-*value < 0.001) was observed after high temperature treatments during early development [34, 35].
One-day-post hatch (dph) European sea bass larvae were purchased from a commercial Spanish hatchery. Eggs and newly hatched larvae were kept in 19-liter cylindrical PVC containers, which were placed inside 650-liter fiberglass tanks filled with seawater. The containers were fitted with nylon mesh at the bottom, with pore sizes adjusted to match the growth stage of the larvae. Larvae were divided in two 650-l tanks and maintained at 17 °C, a temperature known to avoid temperature effects on sex ratio, for the first 20 dph. Subsequently, the temperature in one tank was gradually raised to 21 °C (high temperature, HT group), while it was lowered to 15 °C in the other (low temperature, LT group). In both cases, the temperature was adjusted at a rate of 0.5 °C per day. At 68 days post-hatching (dph), the temperature in the LT group was progressively increased to match that of the HT group, and thus all tanks returned to natural temperatures until fish achieved the sex differentiation (∼ 150–280 dpf). Around 220 dph, the temperatures of both groups were allowed to follow natural fluctuations. Therefore, the only difference in rearing conditions between the LT and HT groups was the temperature experienced during the 20–68 dph period. At 400 dpf, fish were sacrificed by an overdose of benzocaine, and biometric data were obtained. Gonads were dissected and quickly flash frozen in liquid nitrogen and kept at -80 °C.
European sea bass larvae were divided into four groups: female control temperature (FCT), female high temperature (FHT), male control temperature (MCT), and male high temperature (MHT), in duplicated tanks.
Fish were reared under daily supervision of the following environmental conditions: pH (∼ 7.9), salinity (∼ 37.8 ppt), oxygen saturation (85–100%) and with a water renewal rate of 30% vol·h − 1. The rearing density was maintained at a low level to prevent any potential impact on sex ratios, therefore it was kept at safe ranges during all developmental stages: eggs (6000–10,000 eggs·l − 1), larvae (200–350 larvae·l − 1), juveniles (10–20 fish·l − 1) and adults (11–12 kg·m − 3) (details can be found in [36]). Larvae were manually fed three times a day, starting with Artemia AF, followed by amino acid-enriched Artemia EG (INVE Aquaculture, Belgium), and then transitioning to dry food (ProAqua, S.A., Spain) of appropriate pellet sizes as the fish grew. Juveniles were fed ad libitum twice a day.
A total of 14 fish were selected: N = 8 males (four fish per treatment, two for each duplicated tank), N = 6 females (three fish per treatment, two from one duplicated tank and one from the other tank). In order to be able to study the heat effects uniquely, we checked the selected fish not to present biometric differences between study groups within the same sex [34, 35]; FCT: 151 g ± 25 g, 21 cm ± 1.29 cm; FHT: 169 g ± 28 g, 22 cm ± 0.95 cm; MCT 77 g ± 17.2 g, 17 cm ± 1.19 cm; MHT: 79 g ± 9 g, 17 cm ± 1 cm, *p-*values > 0.05). Animals were sacrificed via an overdose of phenoxyethanol, followed by spinal cord dislocation.
miRNA extraction
miRNAs of 14 gonads (eight testes and six ovaries) were obtained by total mRNA by using miRNAs isolation commercial kit (Qiagen miRNA, 217004) and RNeasy MinElute Cleanup Kit to purify high quality miRNAs (Qiagen miRNA, 74204). Quality of the samples was assessed by BioAnalyzer (2100 Bioanalyzer, Agilent Technologies) with ratio 260/280 = 2.07 ± 0.04 and 260/230 = 1,73 ± 0,24) and by the RNA Integrative Number (RIN) measured with a Bioanalyzer (Agilent Technologies, USA) with values > 9.2 in testis. In ovary, RIN numbers were not considered due to naturally high levels of 5s rRNA and tRNA [37, 38].
miRNA sequencing
In total, 14 libraries were constructed from European sea bass gonad samples. Library preparation was performed by NEBNext Small RNA Library Prep Set for Illumina (Multiplex Compatible) kit following manufacture instructions, using sequencing Lane (1 × 50, v4, HiSeq) single-end mode with a read length of 50 bp at the Genomics Unit of the Center for Genomic Regulation (CRG) service in Barcelona.
Bioinformatics and statistical analysis
Alignment was done with Prost [39]! using the Dicentrarchus labrax genome assembly (version: dlabrax2021), and Oryzias latipes was used as reference genome for annotating the miRNAs. For the annotation, three mismatches were allowed. miRNAs without annotation were searched in the FishmiRNA database and complemented [32]. Raw reads were normalized using the DESeq2 package in R and hierarchical clustering of the sample groups was determined by a Principal Component Analysis (PCA) using the plotPCA library in the DESeq2 package [40]. Differential expression (DE) was determined by the DESeq2 package in R software and DE was considered when a minimum of 10 samples presented count reads. miRNAs with less than 10 reads in a sample were discarded. Significant differentially expressed miRNAs (DE-miRNAs) were identified using an adjusted p-value (adj p-value) cutoff of < 0.05, with logFC = 0. The adjusted p-value was calculated using the “results” function on the dataset obtained by the DESeqDataSetFromMatrix function, which was included in the differential expression analysis. Visualization of differential expression was accomplished by a heatmap using the pheatmap package (version 1.0.12) in R. Target genes of DEM were determined using TargetScan (v. 6.0).
Quantitative PCR validation for miRNA-seq validation
Validation of the miRNA sequencing data was done by quantitative PCR (qPCR) of twelve selected sequenced miRNAs in all the samples used by RNA-seq. qPCR product lengths were 18–22 nucleotides and the efficiencies of the primers of the twelve selected miRNA were between 98 and 110%. A gradient qPCR was used to determine the correct annealing temperature for each primer. cDNA was generated using the miRNA 1st-Strand cDNA Synthesis Kit (Agilent Technologies) following manufacturer’s instructions. Firstly, cDNA synthesis was performed by the polyadenylation reaction. qPCR was performed using the qPCRBio SyGreen Blue Mix low ROX (PCR Biosystems). A mix of 5 µL 2x qPCRBIO SyGreen Blue mix, 0.5 µL forward primer, 0.5 µL universal reverse primer (Agilent Technologies), 2 µL of cDNA and up to 10 µL of water was prepared for each sample. The amplification program was as follows: 95 °C for 2 min, 95 °C for 10 min, followed by 40 cycles of 95 °C for 5 s and 60 to 65 °C for 20–30 s. qPCRs were run in triplicate in optically clear 384-well plates (CFX-386, Touch BioRad). Melting curve analysis was done at the end of each PCR program to ensure the uniqueness of the product.
For real-time PCR analysis, the comparative threshold cycle approach (2–ΔΔCT) was used to examine the expression level of miRNAs, followed by a Student’s t-test [41, 42]. To test validation, qPCR data and miRNA-seq data were analyzed by Pearson correlation analysis.
The sequences of the forward primers for the selected miRNAs were as follows: ola-miR-146a-3p: ATCTATGGGCTCAGTTCTTTTG, ola-miR-7132b-3p: TGAGGCGTTTAGAACAAGTTCA, ola-miR-143-3p: TGAGATGAAGCACTGTAGCTC, ola-miR-21-5p: TAGCTTATCAGACTGGTGTTGG, ola-miR-26a-2/3-b-5p: TTCAAGTAATCCAGGATAGGCT, ola-miR-199a-5p: CCCAGTGTTCAGACTACCTGTTC, ola-miR-199a-3p: ACAGTAGTCTGCACATTGGTT, ola-223-3p: TGTCAGTTTGTCAAATACCCCA, ola-192-5p: ATGACCTATGAATTGACAGCC, ola-miR-194a-5p: TGTAACAGCAACTCCATGTGGA, ola-miR-726-5p: GGAATTCCGCTAGTTCTGAACT, ola-miR-726-3p: TTCACTACTAGCAGAACTCGG. The small nuclear RNA dre-U6: ACTAAAATTGGAACGATACAGAGA was used as the reference gene.
Conservation analysis
Expression data from miRNAs in ovary and testis were downloaded from the nine fish species in which gonadal miRNAs data were available from FishmiRNA [32], which represent a wide diversity of reproduction systems: domesticated zebrafish —PSD system— [43], three-spined stickleback—XX/XY chromosomal system— [39, 44], bowfin — XX/XY or ZZ/ZW chromosomal system — [45], striped catfish —XX/XY chromosomal system— [46] (NCBI SRA: PRJNA256963), Japanese medaka —XX/XY chromosomal system, dmrt1Y being the sex determining gene— [47, 48], spotted gar —gonochoristic— [49, 50], black bullhead —gonochoristic— [51] (NCBI SRA: PRJNA730692), European perch —XX/XY chromosomal system— [45, 52] and eastern mudminnow —gonochoristic— [45, 53].
Seed region reads from all the gathered datasets were filtered using a threshold of ≥ 1 count per million (CPM). Common miRNA seed regions between species were visualized in an upset plot using the UpSetR package (version 1.4.0). To identify highly expressed miRNA seeds in the ten studied fish species, the relative abundance was shown by calculating the percentage of miRNA abundance in the whole miRNome of the tissue, either ovary or testis. The top conserved miRNA seeds were visualized in a dotplot using the ggplot2 package (version 3.4.2). Sex-specific overexpression of mature miRNAs was determined by calculating the testis normalized read count divided by the ovary normalized read count per species. When a miRNA was not expressed in ovary or testis, a value of 1 was assigned to avoid division by 0. Afterwards, the log2 ratio>|1| was calculated for all values. Expression of conserved main sequences or mature miRNA names was considered conserved when complying to a threshold of log2ratio≥|1| in at least five species, i.e., 50% of the studied cases, among which presence in D. labrax was mandatory. To show the evolutionary relation between the teleost species used in the present study, a phylogenetic tree was generated using PhyloT (v2) website tool.
Results
miRNA sequencing and annotation
On average, 10.3 and 6.7 million sequences per library were obtained in testes and ovaries, respectively. The total number of sequences exceeded 146 million; ∼93 and ∼ 54 million for testes and ovaries, respectively. A total of 298 mature miRNAs were identified after the alignment to the Dicentrarchus labrax genome (version dlabrax2021) and after performing the annotation by using the Oryzias latipes genome (version ASM223467v1) (Dataset 1). IsomiRs, identified by Prost!, resulted in 31 out of 298 miRNAs (Dataset 1). Ten mature miRNAs were not aligned to the genome but were identified using FishmiRNA [32] and manually annotated. Thus 100% of the obtained sequences were annotated.
miRNA-seq validation
To validate the miRNA-seq data, qPCR analysis was performed on the ovaries and testes samples of twelve selected sequenced miRNAs in all the samples used by RNA-seq. When comparing the log2FC as obtained by qPCR and miRNA-seq for twelve miRNAs, we obtained an R^2^ = 0.9426 and p-value = 1.6e-07 (Suppl Fig. 4), thus validating our data.
Sexual dimorphisms in the MiRNA expression
PCA analysis separated the two sexes, independently of the treatment, achieving a percentage of variance explained of 58% (Suppl Fig. 1A). When comparing testis vs. ovary, 69 miRNAs were found to be DE in European sea bass (adj p-value < 0.05), of which 33 were upregulated in ovary and 36 were upregulated in testis (Fig. 1, Dataset 2). The five most upregulated miRNAs in ovary were miR-734-3p, miR-130c-1-5p, miR-187-3p, miR-155-5p and miR-9-4-3p while in testis were miR-726-3p, miR-724-5p, miR-135b-5p, miR-184a-3p, and miR-726-5p.
Fig. 1. Heatmap of 69 differentially expressed miRNAs (DEM) in European sea bass (Dicentrarchus labrax) testes in comparison to ovaries (MCT vs. FCT). The color scale ranges from red to blue, which red shows relative upregulation in testis and blue shows relative upregulation in ovary. Both DEM and samples (N = 7; N = 4 males and N = 3 females) were grouped by hierarchical clustering based on sex
Heat sensitive MiRNAs
FHT vs. FCT and MHT vs. MCT miRNA expression clustering was shown in a PCA (Suppl Fig. 1B and C). The two components together explained 57% and 69% of the variance in ovary and testis, respectively.
Almost one year after the end of the heat exposure period development, three miRNAs remained DE (adj p-value < 0.05) in ovary, one upregulated, miR-192-5p, and two downregulated, miR-143-3p and miR-146a-3p (Dataset 2). In testis, twelve DE miRNAs (DEM) were identified (adj p-value < 0.05, Fig. 2, Dataset 2), of which four (e.g., miR-194a and miR-223-3p) and eight (e.g., miR-92a-5p, miR-726-3p, miR-724-5p, miR-143-3p, miR-129-5p) miRNAs were upregulated and downregulated, respectively. Based on the expression patterns of the DEMs shown in the heatmap (Fig. 2), the MHT fish clustered into two distinct groups: MHT1 and MHT2 grouped separately from MHT3 and MHT4, which clustered with the control males. This clustering suggests that MHT1 and MHT2 were likely genetic females that might have been masculinized due to heat treatments.
Fig. 2. Heatmap of 12 differentially expressed miRNAs (DEM) in European sea bass (Dicentrarchus labrax) differentiated testes one year after heat exposure. The color scale ranges from red to blue, where red shows relative upregulation in male high temperature (MHT) males and blue is relative upregulation in male control temperature (MCT). The cutoff values were, for adjusted p-vales < 0.05 and log2 FC = 0
Reproduction-related genes were predicted as target genes of male DEM (Dataset 3), such as tdrd6 (miR-1388-5p, miR-194a-5p, miR-2187a-3p, miR-9-4-3p and miR-92a-1-5p), sox3 (miR-129-5p, miR-223-3p, miR-7133-3p), pgrmc1 (miR-365-2b-3p, miR-7133-3p, miR-724-5p) and gper1 (miR-129-5p, miR-1388-5p, miR-2187a-3p, miR-7133-1-3p, miR-724-5p, miR-9-4-3p, miR-92a-1-5p). Supplementary Figure S2 shows the number of predicted target genes for each of the identified miRNAs in the Reproduction process GO term.
miRNA conservation
In order to determine conserved expression across evolution in many fish species, miRNA gonadal data from nine fish species were used together with that of European sea bass of the present study. By using an upset plot, data showed three intersecting sets (Fig. 3, Dataset 4): (1) the number of common seed region sequences between the ten species; (2) those seeds that were conserved in less than ten species and (3) the miRNA seeds per each species. In total, 210 seed regions belonging to 105 miRNAs in ovaries (Fig. 3A) and in testes 184 seed region sequences belonging to 98 miRNAs were expressed in all ten species (Fig. 3B). The evolutionary distances between species are shown in a phylogenetic tree in Suppl Fig. 3 made up of two clades of the studied species Holostei and Teleostei, emerged ∼ 250 and ∼ 310 million years, respectively.
Fig. 3. Upset plot of conserved seed regions in ovary (A) and testis (B) in ten teleost species analyzed: Danio rerio, Dicentrarchus labrax, Gasterosteus aculeatus, Amia calva, Pangasianodon hypophthalmus, Oryzias latipes, Lepisosteus oculatus, Ameiurus melas, Perca fluviatilis and Umbra pygmaea. In ovary (green), 210 seed region sequences were conserved in all ten species. In testis (purple), 184 seed region sequences were conserved in all ten species
Abundancy of gonadal MiRNAs expression across evolution
To identify highly expressed miRNA patterns throughout evolution in fish, the relative abundance of the top 10 most expressed miRNAs was calculated (Fig. 4). In ovary, five miRNAs were highly conserved: miR-143-3p, miR-26a-5p, let-7a-5p miR-100-5p and miR-30a-5p in all ten species (Fig. 4A) whereas in testis, two miRNAs, namely miR-143-3p and miR-26a-5p, were identified in all ten species (Fig. 4B). miRNAs expressed in all ten species are represented by dots connected in a continuous line, while the other miRNAs, which were among the top 10 most abundant in the ovary or testis, are plotted as separated dots in various colors. The ten highest expressed miRNAs per species in ovary and testis can be found in Dataset 5.
Fig. 4. Dot plot of the top 10 most abundant miRNAs in ovary (A) or testis (B) of ten different teleost species. miRNAs expressed in all ten species are represented by a continuous line. The other miRNAs that were in the top 10 most abundant miRNAs in the ovary or testis were plotted as separated dots in different colors. The 10 most abundant miRNAs per species in ovary and testis can be found in Dataset 5
miRNA gonadal markers
To identify sex-skewed miRNA expression in different fish species and miRNAs markers as heat recorders, the following analyses were performed. First, miRNAs were identified with a log2 ratio>|1| of testis vs. ovary in at least five or more species resulting a total of 33 miRNAs (Table 1). From those 21 identified as female-biased miRNA markers and 12 as male-biased markers because they were more expressed in either ovaries or testes in at least five species across evolution.
Table 1. Sex-skewed miRNA expression in different fish species and the effects of heat treatment. MiRNAs were identified with log2 ratio>|1| of seed region expression in testis vs. ovary in at least five or more fish species, among which European sea bass. Furthermore, each miRNA was significantly differential expressed (DEM, adjusted-p value < 0.05) in our data. European sea bass DEM comparisons were: (1) male high temperature (MHT) vs. male control temperature (MCT); (2) female high temperature (FHT) vs. female control temperature (FCT); (3) MCT vs. FCT; (4) MHT vs. FHT. Negative values mean the upregulation in ovary whereas positive values mean upregulation in testes for each species. Data in the table was organized based on the MiRNA conservation in the ovaries or testes along fish evolution. The DEM column shows the comparisons (1, 2, 3, and/or 4) in which each miRNA was differentially expressed for the European sea bass data. ‘count O’ and ‘count T’ represent the number of times the miRNA was differentially expressed in the ovaries or testes, respectively, in the studied species. The miRNA name corresponds to the seed region annotation aligned with European sea bass (Dincentrarcus labrax). Abbreviations: Dla: Dicentrarchus labrax; Ola: Oryzias latipes; Aca: Amia calva; Gac: Gasterosteus aculeatus; Dre: Danio rerio; Loc: Lepisosteus oculatus; Phy: Pangasianodon hypophthalmus; Pfl: Perca fluviatilis; Ame: Ameiurus melas; Upy: Umbra pygmaea; O: ovaries; T: testesmiRNASexdlaolaacagacdrelocphypflameupyDEMCount OCount TmiR-140-3p♀-1,72-1,14-1,011,37-2,43-1,161,19-2,83 + 452miR-429a-3p♀-3,05-2,81-3,05-2,5-2,861,153 + 451miR-140-5p♀-2-1,55-1,63-1,23-2,27-2,363 + 450miR-191-5p♀-2,05-2,07-1,18-1,29-1,943 + 450miR-425-5p♀-2,88-2-1,12-1,31-2,823 + 450miR-155-5p♀-4,15-3,82-1,18-1,59-4,263 + 450miR-223-3p♀-1,11-4,16-1,1-1,8-1,561 + 450miR-142-3p♀-1,17-1,22-1,97-2,991,86-4,3451miR-34a-5p♀-2,52-2,2-1,71,522,12-2,28-1,59442miR-148a-3p♀-1,28-1,45-2,45-2,151,19-1,03-1,45351miR-200b-3p♀-1,88-2,513,6-2,86-2,74-2,47341miR-146a-5p♀-1,08-1,15-1,651,29-1,72-2,01-2,79261miR-192-5p♀-2,06-1,45-1,324,91-2,992,383,71-2,53243miR-200a-3p♀-1,48-2,59-3,973,8-2,7-2,27-3,28-3,6561miR-27a-3p♀-1,11-2,721,34-1,231,26-1,37-1,561,6853miR-7132b-5p♀-1,37-1,23-1,6-2,26-4,432,31,2952miR-200b-5p♀-3,01-3,383,16-3,7-4,04-3,13-5,3651miR-1388-3p♀-1,4-1,93-2,06-1,66-1,51-5,8150miR-33-5p♀-1,47-1,44-21,062,743,81-1,19-2,4543miR-223-5p♀-1,91-3,59-2,41,54-7,073,75-3,7742miR-193a-2-5p♀-1,99-2,83-2,16-4,45-5,9740miR-143-3p♂1,251,251,531,912,381 + 2 + 3 + 405miR-129-5p♂2,651,932,992,211,364,277,821 + 3 + 407miR-724-5p♂3,85-1,985,41,831,1-3,124,741 + 3 + 425miR-499a-5p♂2,23-3,31,293,961,95-1,982,482,283,513 + 427miR-135b-5p♂3,761,793,341,69-3,7635,134,253 + 417miR-182a-5p♂1,84-1,192,263,7-3,063,8123 + 425miR-7-5p♂2,854,36-3,051,141,883,823 + 415miR-135b-3p♂2,762,022,14-2,991,065,193 + 415miR-132b-5p♂2,112,291,891,09-2,39-1,96,19325miR-2187a-5p♂3,441,592,321,943,61405miR-212a-5p♂2,123,04-1,243,18-3,362,813,39-6,7225miR-15c-3p♂1,93-2,381,572,62,862,29-2,7715
miRNA markers for temperature-induced sex-reversal
miRNAs were further filtered based on the heat-treated data obtained in European sea bass in the different comparisons based on sex and heat treatment: (1) MHT vs. MCT, (2) FHT vs. FCT, (3) MCT vs. FCT, and (4) MHT vs. FHT). This resulted in a total of 23 miRNAs (Fig. 5), thirteen of which were female-like miRNA markers, for example, miR-140-3p, miR-429a-3p, miR-191-5p and miR-223-3p, and ten as male-like miRNA markers, for example, miR- miR-143-3p, miR-129-5p, miR-724-5p and 2187a-3p, which were found in at least in one of the studied comparisons. Data in Table 1 were organized based on the miRNA conservation in the ovaries or testes along fish evolution, the DEM column showing the comparisons (1, 2, 3, and/or 4) in which each miRNA was differentially expressed for the European sea bass data. Most of the species showed similar sex-skewed patterns of these miRNA markers, except G. aculeatus and A. melas species, whose expression patterns differed, in particular, from those identified as female-like miRNA markers. ‘Count O’ and ‘Count T’ represent the number of times the miRNA was differentially expressed in the ovaries or testes, respectively, in the studied species.
Fig. 5. Gonadal miRNAs markers identified in this study having had across evolution. miRNAs were filtered based on the heat-treated differential expressed (DEM, adjusted-p value < 0.05) data obtained in European sea bass, in the different comparisons based on sex and heat treatment. miRNAs were identified with log2 ratio>|1| of seed region expression in testis vs. ovary in at least five or more fish species, among which European sea bass. Furthermore, each miRNA was significantly differential expressed (DEM, adjusted-p value < 0.05) in our data. European sea bass DEM comparisons were: (1) Male high temperature (MHT) vs. Male control temperature (MCT); (2) Female high temperature (FHT) vs. Female control temperature (FCT); (3) MCT vs. FCT; (4) MHT vs. FHT. In purple boxes, the female-like miRNAs were represented and in green, the male-like miRNAs with log2 ratio>|1| of seed region expression in testis vs. ovary in at least five or more fish species
For the comparison (2; FHT vs. FCT), two miRNAs were altered in the heat-treated ovaries, miR-192-5p and miR-146a-5p, which were upregulated and identified as female-like markers. Further, miR-143-3p was downregulated in ovaries after the treatment, but also previously identified as male-like marker. From the comparison (1; MHT vs. MCT), a total of four miRNAs altered the expression after one year of the heat treatment in males. Three of them, i.e., miR-143-3p, miR-129-5p and, miR-724-5p, were downregulated in testes subjected to HT in European sea bass but also identified as male-like miRNAs across evolution (Fig. 6).
Fig. 6miRNA markers for temperature-induced sex-reversal proposed as heat-recorders. miRNAs were identified with log2 ratio>|1| of seed region expression in testis vs. ovary in at least five or more fish species, European sea bass among them. Furthermore, each miRNA was significantly differential expressed (DEM, adjusted-p value < 0.05) in our data in the DEM comparisons: (1) Male high temperature (MHT) vs. Male control temperature (MCT); (2) Female high temperature (FHT) vs. Female control temperature (FCT). miRNAs are represented by triangles. Boxes show expression of the miRNA in the ovaries or testes in control groups
Specifically, miR-143-3p was identified as DEM in all the four comparisons studied (1), (2), (3) and (4), indicating that this miRNA was heat sensitive in European sea bass, because it showed a downregulation in both heat-treated testes and ovaries and further, miR-143-3p was highly conserved in testes across evolution. In contrast to the other three male-skewed miRNAs identified, miR-223-3p was upregulated in European sea bass heated males but also expressed in ovaries across evolution, therefore previously identified as a female-like marker.
Discussion
High temperatures can induce masculinization of fish populations. This phenomenon has significant implications for many farmed species, including the European sea bass, in which females grow to a larger size than males. In addition, increasing temperatures in our seas and oceans due to climate change have drastic implications in fish physiology and evolution (Alfonso et al. 2020). Here, we study the effect on the gonadal miRNome of this heat-induced masculinization and further explore the gonadal miRNome of other species in an effort to decipher conserved miRNA sex markers.
Long-term effects, ranging from days to months, of heat treatment on the miRNA expression profile in the gonads have been previously described in fish, long terms. In particular, in Atlantic cod, miR-27c, miR-30c, and miR-200a were found to be DE in juveniles after embryonic or larval heat-treatment. However, no distinction between males or females was addressed [22]. Other studies in zebrafish describing the heat stress by temperature in the gonadal miRNome revealed that miR-22b and miR-301a were altered to be fully recovered 21 days after the exposure [23], thus not being appropriate as a long-term markers. In zebrafish adult females, 23 miRNAs were DE in ovaries after larval heat-treatment, among which, for example, dre-miR-726-3p was downregulated while dre-miR-724-5p was upregulated three months after the heat treatment [21]. Interestingly, these two miRNAs were both found to be downregulated in European sea bass testis one year after heat exposure in the present study, indicating that a conservation in the heat response among species may exist, although the expression pattern might not always be conserved. Similarly, miR-143-3p and miR-92a-3p, which were downregulated in high temperature treated male fish in the present study, were also altered in mice testes exposed for only 25 min to elevated temperatures, highlighting them as potential markers in heat-induced spermatogenesis dysregulation [54]. Higher levels of circulating miR-143 and miR-223 were detected in plasma levels together with cortisol after acute stress response in European sea bass [55]. In the present study, both miRNAs (miR-143 and miR-223) were altered after heat treatment, indicating that they could be potential makers for stress response. A heat-treatment experiment with exact the same conditions in E. sea bass, no differences were observed in the presence and abundance of different cell types between high-temperature and low-temperature fish —females had immature ovaries containing oocytes at the cortical alveolar stage while males had testes containing all germ cell types, including spermatozoa, although no sperm— [35]. Thus, the differences observed in the present study were due to the heat treatment during early development, one year before we measured the alterations in the miRNome, and not due to gonadal development, sperm quality, or puberty. Overall, our data showed that in fish, miRNAs are sensitive to drastic temperature changes in the environment and could play a role in regulating thermal plasticity.
In European sea bass, we observed sexual dimorphism in miRNA expression within fish gonads, with 33 upregulated and 36 downregulated miRNAs in the ovary compared to the testis. To enhance reliability, we cross-referenced our data with existing datasets from the nine species, identifying common gonadal miRNAs across evolution. In testes and ovaries, we have identified 184 and 210 conserved seed region sequences in the ten studied fish species, respectively. Although we have detected the expression of these miRNAs in the gonads in many fish species, these miRNA seed sequences were not sex-specific or tissue-specific, as many miRNAs are highly conserved throughout evolution and body plan development. For example, miR-100 and let-7, were considered to be “ancestral miRNAs”, due to its conservation already described within Metazoan species [26, 56, 57]. Nevertheless, a conservation analysis revealed 74 conserved miRNAs divided into three groups: 23 miRNA families were present in both protostomes and deuterostomes, 46 families conserved exclusively in vertebrates, and 5 families (mir-430, mir-722, mir-724, mir-734, and mir-738) exclusively for fish species [58]. Notably, in our data, miR-143-3p, miR-26a-5p, let-7a-5p, miR-100-5p, and miR-30a-5p in the ovaries, and miR-143-3p and miR-26a-5p in the testes, were the most expressed miRNAs in all ten studied fish species. Similarly, in olive flounder (Paralichthys olivaceus), miR-143, miR-26a and let-7a were also identified as the most expressed miRNAs in both ovaries and testes [59]. In mouse and human, miR-30a-5p was crucial for spermatogonial stem cell differentiation [60] and was used as a marker for men suffering from non-obstructive azoospermia [61]. In our data, however, we observed a conserved overexpression of miR-30a in adult fish ovaries compared to adult testes in the ten studied species. Nevertheless, in Nile tilapia, miR-30a presented higher expression in males at 5 days after hatching (dah) and it was able to downregulate cyp19a1a in gonads [62]. We identified miR-26a both in ovaries and testes, while in Chinese tongue sole (Cynoglossus semilaevis), was identified as transgenerational male marker in sperm [63]. Overall, data showed the complexity of the presence of miRNA across evolution and their multiple roles in regulating many biological processes, even those involved in the reproductive system.
The exploration of miRNA phylogenetic conservation and diversity indicates that miRNAs play crucial roles in animal evolution by influencing phenotypic variation during development [25]. This might explain the different sexual patterns observed in the gonads between the ten fish species of two different clades (i.e., Teleostei and Holostei) separated from one species to others by millions of years of evolution. Further, the fish species used in the present study represent distinct reproduction systems, from XX/XY chromosomal system such as medaka, stickleback, perch and, panga [44, 46, 52, 64], to polygenic system such as European sea bass and domesticated zebrafish [8, 65]. Thus, to detect sexual dimorphic miRNA markers in the fish gonads, we considered a sex-biased miRNA when the expression was found in at least five studied species. In total, we identified thirteen and ten female-like and male-like markers, respectively. For example, miR-429a-3p and miR-140-3p were overexpressed in ovary in at least five studied species. miR-429b-3p was reported to have lower expression in yellow catfish in YY super males compared to XY males [66] and in chicken, miR-140-3p was shown to promote germ cell proliferation while targeting anti-Müllerian hormone amh [67]. Scarce data exist about miR-499, which has been related to cardiac disorders, lung cancer and regulating circadian clock [68–70] in mammals. In fish, miR-499 plays a role in muscle tissue [71–73]. Further, its evolution targeting the intronic region of the myosin heavy chain (MYH14) gene was studied throughout evolution in teleosts [74]. Recently, in the European sea bass, higher levels of circulating miR-499 in plasma differentiating males were identified, when compared to differentiating females [55]. Another example of a male-like miRNA in fish gonads is miR-135b-3p, which is a biomedical marker in humans due to its involvement in many cancer types and disorders like Alzheimer’s disease [75, 76]. Similar to our data, in tilapia, miR-135 presented sexual dimorphism towards males as it was upregulated in testes together with miR-33a, miR-132 and miR-212 [77, 78].
To identify miRNA markers that convey information about both sexual dimorphism and serve as recorders of environmental cues, we considered the sex-biased miRNAs across evolution together with the data from the heated European sea bass. These analyses aided in pinpointing miRNAs that could be accountable for the masculinizing effects induced by high temperatures, detecting a total of five informative miRNAs. miR-223-3p was upregulated in HT males but overexpressed in ovaries in at least five fish species, thus indicating that they might be involved in the masculinization of the gonads. Therefore, it might be hypothesized that the differentiated testes heated with high temperatures during early development might belong to genetic females, similarly to what has been described in heated-temperature female zebrafish presenting a male-like transcriptome [9]. Nevertheless, reliable sexual-genetic makers — which unfortunately are not available in European sea bass —would be required to further confirm the original sex of the heated animals. In Chinese sole tongue, miR-223-3p (referred to as miR-223-y) was upregulated during oocyte maturation. The expression of miR-223-3p was lower in YY super males of yellow catfish (Pelteobagrus fulvidraco) than XY males [66]. Further, male-like miRNAs that showed heat sensitivity were miR-129-5p, miR-724-5p and miR-143-3p, which were downregulated in HT males. In literature, miR-129 was associated with sexual maturity in rainbow trout testis [79], while miR-724-5p (referred to as miR-724-x) showed upregulation during oocyte maturation although it suppressed cyp19a1a [80, 81]. In mice [54], miR-143-3p was shown to be highly upregulated after heat-induced stress, reducing spermatogenesis and thus indicating the thermal sensitivity of this miRNA in the gonads. In the present study females treated with heat, miR-143-3p together with miR-146a-5p were inhibited, while miR-192-5p was upregulated due to temperature one year after the heat. In all, these identified miRNAs may serve as potential epimarkers for predicting past environmental events based on the sexual fish phenotype. The present results demonstrate that, in European sea bass, the miRNome may influence an essential biological process, with consequences for resulting population sex ratios. Although further studies are needed to confirm this, it is tempting to suggest that the miRNAs described herein may be components of the molecular mechanisms connecting environmental temperature, gonadal development, and, consequently, sex ratios.
Conclusions
To gain a better understanding of the intricacies of masculinization induced by heat treatments during early developmental stages in European sea bass, we detected a total of three and twelve miRNAs in the ovaries and testes, respectively. To explore the sexual dimorphic miRNA patterns in the gonads across different fish species, we integrated our findings with available miRNAs from public databases. This analysis identified five evolutionarily conserved miRNAs (miR-146a-5p, miR-143-3p, miR-129-5p, miR-724-5p, and miR-223-3p) that could serve as sex markers. Additionally, all these identified miRNAs were altered following heat treatment in European sea bass, suggesting their potential use as candidates for conserved markers of heat-induced masculinization in fish gonads and across species. These findings hold promise as potential tools for predicting historical environmental events associated with masculinization due to high-temperature treatments in cultured species, but also perhaps in natural populations exposed to a climate change scenario.
Electronic supplementary material
Below is the link to the electronic supplementary material.
Supplementary Material 1: Dataset 1
Supplementary Material 2: Dataset 2
Supplementary Material 3: Dataset 3
Supplementary Material 4: Dataset 4
Supplementary Material 5: Dataset 5
Supplementary Material 6: Supplementary Figures
The reference list from the paper itself. Each links out to its DOI / PubMed record.
- 1Ribas L, Vanezis K, Imués MA, Piferrer F. Treatment with a DNA methyltransferase inhibitor feminizes zebrafish and induces long-term expression changes in the gonads. Epigenetics Chromatin. 2017;10.10.1186/s 13072-017-0168-7PMC 572147729216900 · doi ↗ · pubmed ↗
- 2Zhou H, Zhuang ZX, Sun YQ, Chen Q, Zheng XY, Liang YT et al. Changes in DNA methylation during epigenetic-associated sex reversal under low temperature in Takifugu rubripes. P Lo S ONE. 2019;14.10.1371/journal.pone.0221641 PMC 671151931454376 · doi ↗ · pubmed ↗
- 3Papadaki M, Kaitetzidou E, Papadakis IE, Sfakianakis DG, Papandroulakis N, Mylonas CC et al. Temperature-Biased mi RNA Expression Patterns during European Sea Bass (Dicentrarchus labrax) Development. Int J Mol Sci. 2022, Vol 23, Page 11164. 2022;23:11164.10.3390/ijms 231911164 PMC 957021536232462 · doi ↗ · pubmed ↗
- 4Desvignes T, Bardou P, Montfort J, Sydes J, Guyomar C, George S et al. Fishmi RNA: an evolutionarily supported Micro RNA annotation and expression database for Ray-Finned fishes. Mol Biol Evol. 2022;39.10.1093/molbev/msac 004PMC 882651935020925 · doi ↗ · pubmed ↗
- 5Bir J, Rojo-Bartolomé I, Lekube X, Diaz de Cerio O, Ortiz-Zarragoitia M, Cancio I. High production of transfer RN As identifies the presence of developing oocytes in ovaries and intersex testes of teleost fish. Mar Environ Res. 2023;186 February.10.1016/j.marenvres.2023.10590736774708 · doi ↗ · pubmed ↗
- 6Pasquier J, Cabau C, Nguyen T, Jouanno E, Severac D, Braasch I et al. Gene evolution and gene expression after whole genome duplication in fish: the phylofish database. BMC Genomics. 2016;17.10.1186/s 12864-016-2709-z PMC 487073227189481 · doi ↗ · pubmed ↗
- 7Hertel J, Bartschat S, Wintsche A, Otto C, of the Bioinformatics Computer Lab TS, Stadler PF. Evolution of the let-7 micro RNA Family. RNA Biol. 2012;9:231–41.10.4161/rna.18974 PMC 338458022617875 · doi ↗ · pubmed ↗
- 8Jing J, Wu J, Liu W, Xiong S, Ma W, Zhang J et al. Sex-biased Mi RN As in gonad and their potential roles for testis development in yellow catfish. P Lo S ONE. 2014;9.10.1371/journal.pone.0107946 PMC 416813325229553 · doi ↗ · pubmed ↗