Revealing liver specific microRNAs linked with carbohydrate metabolism of farmed carp, Labeo rohita (Hamilton, 1822)

Revealing liver specific microRNAs linked with carbohydrate metabolism of farmed carp, Labeo rohita (Hamilton, 1822)

Genomics xxx (xxxx) xxx–xxx Contents lists available at ScienceDirect Genomics journal homepage: www.elsevier.com/locate/ygeno Original Article Re...

3MB Sizes 0 Downloads 8 Views

Genomics xxx (xxxx) xxx–xxx

Contents lists available at ScienceDirect

Genomics journal homepage: www.elsevier.com/locate/ygeno

Original Article

Revealing liver specific microRNAs linked with carbohydrate metabolism of farmed carp, Labeo rohita (Hamilton, 1822) Kiran D. Rasala, Mir Asif Iquebalb, Amrendra Pandeya, Parmeswari Beheraa, Sarika Jaiswalb, Manohar Vasama, Sangita Dixita, Mustafa Razab, Lakshman Sahooa, Samiran Nandia, U.B. Angadib, Anil Raib, Dinesh Kumarb, Naresh Nagpurec, Aparna Chaudharic, ⁎ Jitendra Kumar Sundaraya, a b c

Fish Genetics and Biotechnology Division, ICAR - Central Institute of Freshwater Aquaculture, Bhubaneswar, Odisha 751 002, India Centre for Agricultural Bioinformatics (CABin), ICAR-Indian Agricultural Statistics Research Institute, Library Avenue, PUSA, New Delhi 10012, India Fish Genetics and Biotechnology Division, ICAR-Central Institute of Fisheries Education, Mumbai 400053, Maharashtra, India

ARTICLE INFO

ABSTRACT

Keywords: Micro-RNAs Labeo rohita Next-generation sequencing Carbohydrate metabolism

The role of microRNA in gene regulation during developmental biology has been well depicted in several organisms. The present study was performed to investigate miRNAs role in the liver tissues during carbohydrate metabolism and their targets in the farmed carp rohu, Labeo rohita, which is economically important species in aquaculture. Using Illumina-HiSeq technology, a total of 22,612,316; 44,316,046 and 13,338,434 clean reads were obtained from three small-RNA libraries. We have identified 138 conserved and 161 novel miRNAs and studies revealed that miR-22, miR-122, miR-365, miR-200, and miR-146 are involved in carbohydrate metabolism. Further analysis depicted mature miRNA and their predicted target sites in genes that were involved in developmental biology, cellular activities, transportation, etc. This is the first report of the presence of miRNAs in liver tissue of rohu and their comparative profile linked with metabolism serves as a vital resource as a biomarker.

1. Introduction MicroRNAs (miRNAs) are small RNA (non-coding) with 19–25 nucleotides length and usually found in 3’ UTR (untranslated region) region of the genes [1]. They are mainly responsible for regulating genes at the post-transcriptional level and playing a crucial function in developmental biology. The biogenesis of microRNAs and their mode of action during gene regulation are very well reviewed [2]. The miRNA and their targets have been reported in several organisms such as in plants, animals including fishes [3,4]. The miRNA binds to the target gene of mRNA in the UTR region via sequence complementarity and regulates its expression pattern. Earlier studies discovered miRNAs in various model and non-model fishes using traditional techniques of gene cloning, as well as the use of advanced techniques such as transcriptome sequencing and computational tools [5–10]. Those studies suggested that miRNAs play an important regulatory role in the

biological process, cell proliferation and differentiation, organ development, signal transduction, apoptosis, and immunity. The miRNA expression modulates due to the different types of environmental stressors and it has been well reported in mammals and plants [11]. Most of the miRNAs are highly conserved across the species, which is the important property of miRNA. The recent progress in the sequencing technologies and advancement in computational tools has facilitated to get an in-depth understanding of functions of the genes/ genomes [12,13]. The miRNA has been successfully identified in the various model as well as non-model aquaculture species using this sequencing technology. Rohu carp, Labeo rohita, is an economically important freshwater aquaculture species in southern Asia. This is the most popular and favorite food fish among carp and has high market value due to their delicious taste. In aquaculture, feed cost accounts > 50% and feed mostly comprise of 30–35% protein, which in turn results in the

Abbreviations: RT, Reverse Transcriptase; qPCR, Real-Time PCR; RT-qPCR, Reverse transcriptase qPCR; LR, Labeo rohita; AMPK, AMP-activated protein kinase; TPM, Transcripts Per Million; G6PC, glucose-6-phosphatase; PK, pyruvate kinase; PCK1, phosphoenol pyruvate carboxykinase; JAK/STAT, Janus kinase/signal transducers and activators of transcription; DEMs, Differentially expressed miRNAs; PYGL, glycogen phosphorylase; PGM, phosphoglucomutase ⁎ Corresponding author. E-mail address: [email protected] (J.K. Sundaray). https://doi.org/10.1016/j.ygeno.2019.07.010 Received 27 April 2019; Received in revised form 11 June 2019; Accepted 16 July 2019 0888-7543/ © 2019 Elsevier Inc. All rights reserved.

Please cite this article as: Kiran D. Rasal, et al., Genomics, https://doi.org/10.1016/j.ygeno.2019.07.010

Genomics xxx (xxxx) xxx–xxx

K.D. Rasal, et al.

inflation of feed cost. Indian aquaculture dominated by omnivorous carps (80%), which are supposed to be better carbohydrate utilizer, and show optimum growth at 30–35% dietary protein inclusion level. Therefore, the higher demand for protein by carps can be accomplished by understanding the mechanism of carbohydrate metabolism. The gluconeogenic pathway (gluconeogenesis) is widely reported in the liver, is an important strategy to regulate the blood glucose concentration in fish. In earlier studies, liver-specific miR-33 and miR-122 were identified in rainbow trout, which were associated with metabolism including glucose homeostasis [14,15]. Subsequently, evidence in fishes implicated that, miR-122 mainly involved in the regulation of metabolism in the liver and found to be conserved among the vertebrates [16–18]. Accumulated studies revealed that the liver is a vital organ, which control the carbohydrate metabolism through maintaining glucose homeostasis by tightly regulated system. Recently, glucose metabolism associated 124 miRNAs were identified in the liver of blunt Snout Bream (Megalobrama amblycephala) using NGS and identified miR-192, miR-128, miR-205-5p, and miR-100-5p as a regulator of genes of starch-associated metabolism [19] and reported miR34a as a vital miRNA in carbohydrate metabolism [20,21]. However, the exact mechanism of glucose intolerance responses in teleost at the physiological and molecular level is not fully understood. Thus, the advancement of sequencing technology and computational tools could be helpful to investigate small RNAs (miRNAs), their target in the genes/pathways associated with carbohydrate metabolism. However, there is no report on the characterization of the microRNAs in the farmed carp rohu, Labeo rohita. In this study, we have attempted to investigate the involvement of microRNA in the liver of farmed carp rohu fed with different level of carbohydrate regime using high-throughput sequencing technology. The identified miRNAs through sequencing and computational tools in genes associated with developmental biology, cellular activities such as transportation, and signaling would allow for the understanding role of epigenetics in developmental biology. This study will also help to develop miRNAs as a biomarker and development of miRNAs-inhibitor for functional level studies in rohu.

of the small RNAs are 16–23 nt in length. However, in 40%, maximal miRNAs are 21–23 nt in length. 2.2. Identification of conserved and novel miRNA targets (known and novel miRNAs) The filtered reads (sample wise) were aligned and mapped against the closely related reference genome of common carp, Cyprinus carpio inferring 58.79% in LR_20%, 41.86% in LR_40%, and 66.14% LR_60% of reads of miRNA were mapped against the common carp genome. For identification of miRNAs in liver tissues of rohu, the candidate mature sequences from all the three libraries were compared with the deposited database of miRBase using the miRCat tool. Through this, 133, 116, 130 known and 119, 95, 109 novel miRNA sequences were characterized from control and treated -samples, respectively and their distribution depicted using Venn diagram (Fig. 2) [Table S1 & S2]. Comparative analysis data shows that among the known miRNAs, 45 miRNAs are conserved, whereas in novel group 206 miRNAs are present in all treatment groups. The Mfold web server being used for secondary structure prediction. All the mature miRNA have lengths of 16 to 25 nt. In contrast to the length of miRNAs, the length of a novel and known miRNA hairpin structure significantly ranging from 55 nt to 101 nt. The G/C content of the candidate miRNAs was also evaluated where the miRNA precursors had G/C% content of 23% to 73% in known miRNA and 30% to 64% in novel miRNA, with an average of 55.92%. Apart from the above-mentioned characteristics, the negative minimal folding free energy (MFE), adjusted minimal folding free energy (AMFE), and minimal folding free energy index (MFEI) of the candidate miRNA precursors were calculated. MFE is an important criterion to determine the stability of the perfect or near-perfect secondary hairpin structure of pre-miRNAs. The studies reported that the lower the value of MFE, the higher is the thermodynamic stability of the secondary structure of the corresponding precursor sequence [22]. The negative MFE of precursor miRNAs varied significantly with a range of −14 kcal/mol to −65.00 kcal/mol in novel miRNA −16 kcal/mol to −55 kcal/mol in known miRNA. The MFEI of each potential miRNA precursor was calculated for the precise discrimination of the miRNA from other types of small RNAs. Since other RNAs such as mRNA, rRNA, tRNA may also form similar hairpin structures, we used MFEI to distinguish other RNAs or RNA fragments. In the present prediction, the newly identified premiRNAs have MFEI values ranging from 1.86 to 0.49, with an average of about 0.87. These values were significantly higher compared to those reported for tRNAs (0.64), rRNAs (0.59), and mRNAs (0.62–0.66), reflecting that newly identified potential L. rohita miRNAs are probably true miRNAs than any other type of non-coding RNA. The location of the mature miRNA within the 5′-arm or 3′-arm of the stem-loop hairpin structure of potential miRNAs was also identified, where 184 novel mature miRNAs were located on the 3′-arm of the secondary hairpin stem-loop structure of pre-miRNAs and remaining 169 were located at the 5′-arm end. Secondary structure of six selected novel miRNA which is mainly involved in metabolic pathways are shown in Fig. 3; i.e., Labeo-miR-5076, Labeo-miR-5235, Labeo-miR-5219, Labeo-miR-5304, Labeo-miR-5380, and Labeo-miR-5003.

2. Results 2.1. Construction of cDNA library for small RNA-Seq To identify the miRNAs expression profiles in the liver of rohu, the small RNAs cDNA libraries were generated from total RNAs of liver tissues of the three treatment groups fed with different carbohydrate regimes. Using Illumina HiSeq500 sequencing platform, a total of 22,612,316; 44,316,046; and 13,338,434 raw reads were obtained from control and treated (40% and 60%) samples respectively. The data were submitted in NCBI with SRA accession numbers SRR8862885; SRR8862886; SRR8862884. In a step forward to obtaining high-quality reads, the 3′ and 5′ adapters, low-quality tags, reads with sequences shorter than 15 nt and reads containing poly-N were removed. After discarding the adapters and low-quality reads, 18,533,090; 33,525,776; and 8,263,055 high quality cleaned reads were obtained in two treatment groups including control. The small RNA sequences were compared with the RFam database and resulting 7,714,269; 18,893,822; and 3,441,633 reads of non-coding RNAs in treated along with control groups were annotated and removed. The abundance of generated reads mapped to the Rfam database is shown in Fig. 1. Remaining reads were processed for removal of known repeat database mapping reads against Repbase database resulting 9,854,416; 14,505,135; and 4,529,621 numbers of unique reads in control and two treatment groups respectively and details are available in Table 1. Finally, reads 5,253,367; 7,262,926; and 2,853,809 were retained for miRNA analysis after considering reads with length between 15 and 30 nt in 20% (control), 40%, 60% respectively. The size distributions of the small RNAs reads in three libraries (40%, 60% and control 20%) were similar, where most

2.3. Differential expression (DE) of miRNAs The differential expression analysis of known and novel miRNAs was carried out in all three treatment groups. The comparison data of all two treatments with control showed differentially expressed miRNAs in LR_40%, and LR_60% as compared to control LR_20%, respectively (Table 2). A total 42 differentially expressed miRNAs related to carbohydrate metabolism were considered for further analysis (Fig. 4), thereinto, 24 numbers of known and 18 novel miRNAs targeting carbohydrate metabolism related pathways are found to be expressed differentially in a significant rate. The circos plot also depicted up and down regulation miRNAs in different treatment groups as compared to 2

Genomics xxx (xxxx) xxx–xxx

K.D. Rasal, et al.

Fig. 1. (a), (b) and (c) Family distribution abundance of 20, 40, and 60% high CHO through Rfam database.

control (Fig. 5). The known miRNA belongs to miR-365, miR-216, miR130, miR-15, miR-456, miR-20, miR-23, miR-29, miR-200, miR-139, miR-1338, miR-146, miR-2188, miR-199, miR-22, miR-122, and miR100 showed differential expression at the control and treated tissues. Finally, the randomly selected five miRNAs were subjected to RT-qPCR in all liver tissue samples of rohu.

2.4. Co-expression analysis of miRNAs-targeted genes In order to explore the co-expression network of miRNAs with their targeted genes, we have selected 31 upregulated miRNAs in all the treated groups and made a cluster of miRNAs-target genes (Fig. 6). Interaction between the 31 miRNAs and 183 number of target genes shown, notably, one miRNAs regulates multiple genes as revealed 3

Genomics xxx (xxxx) xxx–xxx

K.D. Rasal, et al.

Table 1 Read statistics of the liver tissue small RNA sequencing of L. rohita. Sample ID

LR20 (Control)

LR40

LR60

Total reads Read with Adapters

22, 612, 316 20, 204, 809 (96.0%) 1, 118, 622 (7.9%) 7576 19,081,909 18, 533, 090 7, 714, 269 9, 854, 416

44, 316, 046 43, 485, 470 (97.9%) 8, 947, 765 (14.3%) 10,328 34,529,923 33, 525, 776 18, 893, 822 9, 854, 416

13, 338, 434 10, 737, 939 (87.2%) 2, 252, 398 (12.3%) 4102 8,483,688 8, 263, 055 3, 441, 633 4, 429, 621

Red length < 15 bp Too many N's Adapter Trimmed Reads High quality reads Reads Mapped Reads Unmapped

through network studies. For example, Labeo-miR-5050 target genes like relaxin/insulin-like family peptide receptor, insulin receptor b, phospholipase A2, facilitated glucose transporter, ADP-ribosylation factor interacting protein 2a, patatin-like phospholipase domain, etc. and Labeo-miR-5069 targets genes such as fatty acid amide hydrolase 2a, galactoside-binding, soluble, fibroblast growth factor receptor substrate 2b, carbohydrate kinase domain containing, phospholipase C, annexin A3b, lipase, actin-related protein 2/3 complex, ATPase, Ca++ transporting, ubiquitous, F-box protein 16, fucosidase, alpha-L- 1, galactosidase, insulin receptor substrate 1, procollagen, GATA zinc finger domain containing 2Ab, F-box protein 9, etc.

Fig. 2. Venn diagram showing overlapping mature miRNAs after treatment 20 (control), 40 and 60% of high CHO in the liver tissues.

even though a small variation in fold change values of three libraries. 3. Discussion MicroRNAs are playing an important role in diverse physiological functions and biogenesis pathways in several organisms including fishes [7,23]. The role of miRNAs has been revealed in several aquaculture non-model species due to the advent of the NGS technology and bioinformatics algorithms [24]. Similarly, comparative genomics approach is useful to investigate or depict the complexity and heterogeneity of glucose homeostasis via comparing all carbohydrate metabolism-related genes in the metabolic pathway among species such as fish, chickens, mice, frogs, and humans and revealed 66% of these genes are conserved among them [25]. Advancement in the genomics and NGS has enabled to identify small RNAs including a large number of the novel as well as conserved miRNAs. In the present study, we attempted to investigate the involvement of microRNAs in the liver of farmed carp rohu fed with different level of carbohydrate regime using Illumina sequencing technology. The rohu is an important carp species in the Indian sub-continent. Thus, genomic and bioinformatics tools are applied to understand the biology of farmed carp rohu immensely. However, there is no report of miRNAs present in the farmed carp with reference to its physiology functions. In this study, using Illumina pairend sequencing methods, a total of 138 conserved (known) and 161 novel putative miRNAs were investigated in the liver tissues of the farmed carp rohu. Moreover, it was implicated, that few sets of miRNAs were differentially expressed (DE) in liver tissues in different treatment groups, suggesting liver tissue-specific behavior of rohu miRNAs based on carbohydrate level. In earlier studies, NGS technology has been successfully utilized to identify the role of miRNAs in common carp as well as blunt snout bream (Megalobrama amblycephala), revealed expression pattern of miRNAs are species and tissues dependent. The liver is the main organ which possesses endocrine physiology and helps in food digestion and metabolic hormones such as Insulin, glucagon, glucagon-like peptide, pancreastatin, somatostatin, and pancreatic peptide has been responsible for digestion [26,27]. In this study, we could categorize 19 known/conserved miRNAs such as, miR-365, miR-10, miR-216, miR130, miR-15, miR-456, miR-20, miR-99, miR-34, miR-23, miR-200, miR-141, miR-139, miR-146, miR-2188, miR-21, miR-199, miR-22, miR-122, and miR-100, showing significant differential expression in the treatment group including control. In earlier studies, miR-122 found to be associated with liver-specific miRNAs and associated with metabolic re-programming [28]. The miR-122 found to be tissue-specific miRNAs and highly expressed in the liver cells. For example; miR-122 and let-7 highly abundant miRNAs in the liver of M. amblycephala [29]. It has been revealed that miR-122 targets glycolytic genes such as

2.5. Target prediction and gene ontology The 3’ UTR sequences of zebrafish from TargetScanFish (v 6.2) database were used with the help of RNAhybrid software to understand the function of conserved and novel miRNAs in the liver tissues of rohu. A total numbers of 43,764 transcripts targeted by 331 miRNAs were identified through computational analysis. About 65% of the target genes could be annotated due to the absence of annotated database for rohu. The gene ontology (GO) was performed using the identification of target genes through functional enrichment, which classifies biological processes based on miRNA-gene regulatory networks. This has resulted in 43,764 predicted targets genes by miRNAs and was classified in a total of 22 biological processes, 11 molecular functions and 11 in cellular components (Fig. 7). The target genes were identified in various carbohydrate related metabolic processes such as carbohydrate metabolic process, CDP- diacyl glycerol-serine-O-phosphatidyl transferase activity, fatty-acyl-CoA binding, glucoronyl-galactosyl-proteoglycan 4 alpha N-acetyl glucosaminyl transferase activity, GTP binding, GTPase activity, ATP binding, NADPH-hemoprotein reductase activity1-phosphatidyl inositol 4-kinase activity, GDP dissociation inhibitor activity, Dolichyl-phosphate-beta-D-mannosyl transferase activity, mannosyl transferase activity, NAD + ADP-ribosyl transferase activity, NAD + binding, UDP-glucose 4-epimerase activity, NADPH hemoprotein phosphatidyl inositol phospholipase C activity, phosphopyruvate GTPase activator activity, phosphatidyl inositol phospholipase C activity. The functional analysis of those clustered miRNAs found to be linked with metabolic, biological and cellular processes as shown by using WEGO map. 2.6. RT-qPCR validation of differentially expressed miRNAs (DEMs) The expression profile of randomly selected 5 DEM from all DEMs which includes mir-22, mir-122, mir-365, mir-200 and mir-146 a known miRNA and U6 was used as housekeeping miRNA. These DEMs were significant in glucose or protein metabolism during high carbohydrate treatment. Interestingly, for all five of the differential expressed novel DEMs identified by small RNA-Seq analysis, higher and lower than log2 values (Fold change) were selected (Fig. 8). The five miRNAs expression profiles were consistent with the results obtained by NGS, 4

Genomics xxx (xxxx) xxx–xxx

K.D. Rasal, et al.

Fig. 3. Stem-loop structure of newly identified miRNAs involved in carbohydrate metabolic pathway. The actual size of the precursors may be slightly shorter or longer than the one shown in the figures. (a) Labeo-miR-3204 (b) Labeo-miR-5076 (c) Labeo-miR-5201 (d) Labeo-miR-5219 (e) Labeo-miR-5235 (f) Labeo-miR-5339.

pyruvate kinase (PK) in the liver, which showed a strong association with miRNAs expression [30]. In our studies, miR-122 and miR-21, the most abundant miRNAs in all treatment group, but it was down-regulated in control (20%) and 60% starch diets. The let-7 miRNA found normally expressed in all treatment group, as previously reported that highly conserved and it regulates cell proliferation and differentiation [31]. It was reported that let-7 controlled the insulin and glucose homeostasis in the liver [32]. But, surprisingly, in our studies, let-7 expression was normal at different treatment. In earlier studies, miR-375 was exclusively found in the pancreatic cells of zebrafish embryos [33]. In this studies, miR-375 was down-

regulated in control and 40% CHO diet, and normally expressed in a higher level of carbohydrate diet. The miR-125b also reported being involved in modulating metabolic activities through its over-expression [34]. But in our studies, miR-125 was normally expressed in all three treatments. Similarly, it has been reported that the secretory function of insulin-producing cells was controlled by miR-9 [35]. Similarly, miR-34a was important miRNAs in carbohydrate metabolism and functional level studies reported in M. amblycephala fed with the inclusion of a high 45% starch diet [21]. They inhibited the expression of miR-34a using IP injection of miR-34a inhibitor (antagomiR-34a). Their functional enrichment analysis demonstrated that 5

miRNA ID

Labeo-miR-5104 Labeo-miR-5111 Labeo-miR-5076 Labeo-miR-5125 Labeo-miR-5156 Labeo-miR-5309 Labeo-miR-5066 Labeo-miR-5249 Labeo-miR-5070 Labeo-miR-5105 Labeo-miR-5117 Labeo-miR-5048 Labeo-miR-3204 Labeo-miR-3215 Labeo-miR-5123 Labeo-miR-5037 Labeo-miR-3349 Labeo-miR-3217 Labeo-miR-5072 Labeo-miR-5116 Labeo-miR-3210 Labeo-miR-3011 Labeo-miR-3332 Labeo-miR-3285 Labeo-miR-3214 Labeo-miR-3202 Labeo-miR-5175 Labeo-miR-3376 Labeo-miR-3364 Labeo-miR-3379 Labeo-miR-3199 Labeo-miR-5050 Labeo-miR-5069 Labeo-miR-5224 Labeo-miR-3236 Labeo-miR-3376 Labeo-miR-3364 Labeo-miR-3379 Labeo-miR-3310 Labeo-miR-3310

Sr·No

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40

NA NA NA NA NA NA NA NA NA NA NA NA mir-365 mir-29 NA NA mir-202 mir-29 NA NA mir-10 mir-193 mir-155 mir-146 mir-29 mir-216 NA mir-133 mir-133 mir-375 mir-200 NA NA NA mir-454 mir-133 mir-133 mir-375 mir-122 mir-22

miRNA family Novel Novel Novel Novel Novel Novel Novel Novel Novel Novel Novel Novel Known Known Novel Novel Known Known Novel Novel Known Known Known Known Known Known Novel Known Known Known Known Novel Novel Novel Known Known Known Known Known Known

miRNA type CCCCAGCGGTAGTCAGCTG CGAGACTCAGACTTTA CAAAGTGCTCACAGTGCA CTCTCGTTCCCGCTCTG GATTGACAGTCAGACCT TGGAGTGTGACAATGGT ATGGCAGAATCTGAGA TCAATTTTGTGGGTTTC ATTCTCTTGGCTGTCTGATT CCCCTCAGTTCCTCCA CGGCGCCCGCGGCGGG AGGTGCTGTAAGCTTA TAATGCCCCTAAAAATCCTTAT TAGCACCATTTGAAATCAGTGT CTCCGGGCCGCGCGCT AGATTATGAGATCTGAGGGTC TTCCTATGCATATACCTCTTT TAGCACCATTTGAAATCGGTT ATTGATATCTGAGACC CGCTCTGGGGGTCGGT TACCCTGTAGATCCGAATTTGT AACTGGCCTACAAAGTCCCAGT TTAATGCTAATCGTGATAGGGG TGAGAACTGAATTCCATAGAT TAGCACCATTTGAAATCAGTG TAATCTCAGCTGGCAACTGTGA GTACAGTACTGTGATA TTTGGTCCCCTTCAACCAGCT TTGGTCCCCTTCAACCAGCTGT TTTGTTCGTTCGGCTCGCGTT TAATACTGCCTGGTAATGATG AGTGCACCCGACCGTTGGCTCG ATGTTTGCCTGCAATCTGCCCT TAGCAGCGCGTCATGGTTTTC TAGTGCAATATTGCTTATAGGG TTTGGTCCCCTTCAACCAGCT TTGGTCCCCTTCAACCAGCTGT TTTGTTCGTTCGGCTCGCGTT TGGAGTGTGACAATGGTGTTTG AAGCTGCCAGCTGAAGAACTG

Mature miRNA Sequence

Table 2 Differentially expressed miRNAs associated with carbohydrate metabolism in L.rohita.

9.71 9.52 7.83 7.62 6.15 5.91 5.57 5.31 5.25 3.83 3.83 3.78 3.64 3.11 3.11 3.11 3.04 2.69 2.25 2.04 2.00 1.98 1.96 1.65 1.63 1.54 1.51 0.11 −2.75 0.30 0.62 −0.63 −2.24 −0.27 1.25 0.11 −2.75 0.30 1.51 −0.38

Log fold change value (20 vs 40 CHO) Up Regulated Up Regulated Up Regulated Up Regulated Up Regulated Up Regulated Up Regulated Up Regulated Up Regulated Up Regulated Up Regulated Up Regulated Up Regulated Up Regulated Up Regulated Up Regulated Up Regulated Up Regulated Up Regulated Up Regulated Up Regulated Up Regulated Up Regulated Up Regulated Up Regulated Up Regulated Up Regulated Normally Expressed LR20A_Specific Normally Expressed Normally Expressed Normally Expressed LR20A_Specific Normally Expressed Normally Expressed Normally Expressed LR20A_Specific Normally Expressed Up Regulated Normally Expressed

Remark −9.09 −2.63 5.07 −1.43 −5.55 −0.53 −3.22 −1.19 4.22 0.06 0.36 −1.38 −0.22 −0.61 2.45 −1.46 −3.74 −0.16 0.97 −1.17 −1.96 0.03 −0.42 −0.19 0.64 −0.98 −1.29 3.94 2.76 2.28 0.31 2.20 1.98 1.91 1.86 3.94 2.76 2.28 0.61 0.12

Log fold change value (20 vs 60 CHO)

Down regulated Down regulated Up-regulated Normally Expressed Down regulated Normally Expressed Down regulated Normally Expressed Up Regulated Normally Expressed Normally Expressed Normally Expressed Normally Expressed Normally Expressed Up Regulated Normally Expressed LR_20 specific Normally Expressed Normally Expressed Normally Expressed Down Regulated Normally Expressed Normally Expressed Normally Expressed Normally Expressed Normally Expressed Normally Expressed Up Regulated Up Regulated Up Regulated Normally Expressed Up Regulated Up Regulated Up Regulated Up Regulated Up Regulated Up Regulated Up Regulated Normally Expressed Normally Expressed

Rmark

K.D. Rasal, et al.

Genomics xxx (xxxx) xxx–xxx

6

Genomics xxx (xxxx) xxx–xxx

K.D. Rasal, et al.

Fig. 4. The heatmap visualization for the expression of the miRNAs involved in carbohydrate metabolic process. Different colors are indicates different P-value (−1 to 1).

most of the differentially expressed genes were enriched in the insulin, peroxisome proliferator-activated receptor (PPAR), AMP-activated protein kinase (AMPK), and Janus kinase/signal transducers and activators of transcription (JAK/STAT) signaling pathways, which mainly linked with the metabolism. Other studies also implicated that miR-34a as a key miRNAs in metabolism programming, which associated with genes in rodents and humans associated with non-alcoholic fatty liver disease and type 2 diabetes (T2D) [36]. It has been also seen that, miR34a involved in the process of insulin signaling and hyposecretion in a mammalian type 2 diabetes (T2D) model [37]. In our studies, miR-34 was down-regulated in all the treatment groups of fishes. This indicates that miR-34a may be involved in glucose metabolism and its activity is inhibited in high percentages of carbohydrate diets. Earlier studies suggested that the expression profile of miR-122 and miR-200 was dominant in the liver and revealed its role in fatty acid and cholesterol metabolism [38]. We could identify that, most of the miRNAs associated with glucose utilization (60%) and showed high expression profile. Only a few miRNAs showed high expression level in other treatment groups including control. Due to the absence of rohu genome sequence data in the database, the small RNAs that were mapped against model organism genome (zebrafish), which further subjected to novel miRNAs prediction using the computational tool based on phylogeny. Interestingly, we could identify that novel miRNAs are weakly expressed, which was consistent with the earlier report that conserved miRNAs showed high expression as compared to novel miRNAs. The differential analysis of miRNAs was shown in the heatmap plot. Among, a total of 24 known miRNAs was differentially expressed

Fig. 5. Circos plot showing overlap among filtered genes for up and down regulation (log2FC > 2, average expression ≥5, and p ≤ .05). Purple curves link identical miRNAs genes. Light orange areas within the inner arc indicate unique miRNA genes. Dark orange areas within the inner arc represent miRNA genes that are differentially expressed in another groups. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.) 7

Genomics xxx (xxxx) xxx–xxx

K.D. Rasal, et al.

Fig. 6. Network diagram of upregulated genes involved in carbohydrate metabolic pathway. Solid lines denote association between the miRNAs and the target genes. The nodes represent target genes; the circle nodes represent miRNAs. 8

Genomics xxx (xxxx) xxx–xxx

K.D. Rasal, et al.

Fig. 7. Functional enrichment of differentially expressed miRNA in different treatments: Conserved miRNAs, (b) Novel miRNAs.

in the liver tissues of control as well as treated tissues. The novel miRNAs coded as 16 were found to be differentially expressed. The miRNA belongs to miR-200 and miR-20 family showed significant expression at all levels. The heighten expression profile of miR-206, miR-1, miR-199 and miR-23a were reported significantly upon a short period of fasting in fast muscle, while decreased the expression of miR-206 and miR-1 were detected in the slow muscle of juvenile pacu (Piaractus mesopotamicus) [39]. They also investigated the gene targets by those miRNAs i.e. negative expression profile of mRNA-target genes such as mTOR for miR199; IGF-1 for miR-1, miR-206, and miR-199 and MFbx and PGC1a for miR-23a, respectively. The miR-33 and miR-122 were identified as liver-specific and which associated with lipid as well as glucose metabolism [14]. In our studies expression for miR-199, in starch fed fishes, liver tissues showed normal expression. The lipid and glucose metabolism regulated post-prandially by few miRNAs, which were reported by Mennigen and their co-workers. Their studies imply that miR-103, miR-143, and miR-107 have conserved targets among fish and humans

[17]. Earlier, it was demonstrated that miR-122b and miR-33 were found to be up-regulated, while miR-122a was down-regulated. miR-33 and miR-122b were involved in biological processes in hepatic insulin pathway [40]. Further, Mennigen and their co-workers analyzed the expression profile of miRNA-33 and miRNA-122a/b using RT-qPCR in rainbow trout fingerlings upon endogenous to exogenous feeding and found down-regulation of those miRNAs [18,41]. In addition, modulation of the metabolic gene regulation was observed upon microRNA122 inhibition (LNA-122 inhibitor). The present study indicate that inhibition of function of the miRNAs results in an alteration in lipid as well as glucose metabolic activities. In the present study, we could identify that miR-33 was normally expressed in three treatment where starch diet (40%, 60% Starch level) were utilized. Similarly, in other organisms, several miRNAs have been reported to be involved in metabolic activities in the liver such as miR-103 and miR-107 [42], miR181a [43], and miR-143 [44]. In humans, miR-802 has been depicted to associate with obesity and its down-regulation improves insulin activity and glucose tolerance [45]. 9

Genomics xxx (xxxx) xxx–xxx

K.D. Rasal, et al.

Fig. 8. qPCR analysis of selected miRNAs based on Ct value.

[47]. It was suggested that miRNAs play a vital role in the insulin signaling pathway through the PPAR pathway by targeting metabolically related genes [21,48]. The miR-33 associated with sterol regulatory element-binding protein (SREBP) genes for glucose metabolism regulation via targeting important regulatory enzymes of gluconeogenesis such as glucose-6-phosphatase (G6PC) and phosphoenolpyruvate carboxykinase (PCK1). Also, overexpression of miR-33 reduced the level of enzymes such as glycogen phosphorylase (PYGL) and phosphoglucomutase (PGM) involved in glycogenolysis [48]. Our results also depicted the involvement of miRNAs in molecular and cellular processes.

In order to quantify the differential expression of miRNAs in the liver tissues of different treatment groups, we have randomly selected five miRNAs which include mir-22, mir-122, mir-365, mir-200 and mir146 for qPCR. The qPCR analysis confirmed that the identified miRNAs (five miRNAs) are differentially expressed in all treatment group including control samples. The results were found to be consistent with transcriptome data generated by sequencing. As the identification of miRNAs target is essential to understand its biological functions, the bioinformatics approach was utilized as reported in several studies [1]. The miRNAs regulates or modulates various target genes via binding at multiple sites of the 3’ UTR region of mRNAs and also it has been reported that it can binds to 5’ UTR region and coding part of mRNAs, which enhances gene regulation significantly. The specific functions of miRNAs and their targets were predicted using annotation with the zebrafish genome. Our analysis showed that most of the miRNAs in the present study are orthologous with the zebrafish. We could identify 43,764 genes being targeted (multiple targets in single genes) by the novel as well as conserved miRNAs. The single miRNA targets found in multiple sites of the genes. Through miRNA-mRNA target network studies also indicated that single miRNAs regulate many genes associated with biological processes. In earlier studies, miRNAs are identified in several mammalian species and revealed that those miRNAs playing an important role in biological processes [23]. The targets in metabolically related genes such as PGM1, GPD2, and PCK1, which modulating metabolic functions such as insulin resistance and glyceroneogenesis [46]. In this study, we have specifically identified that miRNAs targeted genes which possess metabolic functions. Those identified miRNAs possess diverse annotated regulatory functions in carbohydrate metabolism. In the present studies, the predicted targets were classified as per their functions and most of the miRNAs targets belonged to the metabolic pathways such as carbohydrate/glucose metabolism, lipid metabolism and amino acid metabolism. We could identify that DEGs enriched in peroxisome proliferator-activated receptor (PPAR), AMP-activated protein kinase (AMPK), and Janus kinase (JAK) or Signal transducers and activators of transcription (STAT) signaling pathways. These targets specifically involved in anabolic as well as catabolic activities such as ATP/GTP binding, and others. In line with earlier studies, we could identify that miRNAs involved in MAPK pathways and insulin signaling pathways [21]. Evidence suggested that, PPAR involved in insulin activities thereby control obesity via controlling insulin signaling in humans

4. Conclusion Overall, the present study revealed first of its kind of miRNAs in rohu liver tissues associated with carbohydrate metabolism and identified 138 known (conserved) and 161 novel putative miRNAs. The differential expression profile of the liver-specific miRNAs revealed that distinct sets of miRNAs were expressed in the liver tissue of rohu. This study enriches the rohu miRNAs database and their linkage with the metabolism provides an important resource as a biomarker. This work will also help to conduct specific miRNAs studies via developing its antagomir for inhibition. This study could be a basis to identify microRNAs related SNPs and decipher other events such as RNA editing. 5. Materials and methods 5.1. Experimental design and sampling The farmed carp rohu (Average weight 40 ± 10 g) were cultured in the farm ponds of ICAR-Central Institute of Freshwater Aquaculture, Bhubaneswar, Odisha, India. All the experimental procedures were performed with the approval of the Institute Biosafety Committee (IBSC) on ethics. The fish were captured and transferred in aerated cement tanks (500 L) for acclimatization and were fed twice with a normal diet before the experiment. Three diets with three different levels and types of carbohydrate (20% starch (control), 40% starch, and 60% starch (Table 3) were formulated using low-cost fine ingredients. At the time of formulation, all dietary ingredients were mixed for about 30 min and further, it was blended for 10 min and pelleted through hand pelletizer having 0.2 mm diameter. The pellets were further 10

Genomics xxx (xxxx) xxx–xxx

K.D. Rasal, et al.

removed using Trimmomatic (v 0.36) in single end mode with minimum read length 15 nt and quality cut-off 20 [49]. Meanwhile, to other small non coding RNAs such as rRNA (ribosomal RNA), snRNA (small nuclear RNA), tRNA (transfer RNA), snoRNA (small nucleolar RNA) etc., the sample-wise reads were analyzed by NCBI-BLAST (v 2.6.0) against the customized Rfam database (v12, 2) [50], the known mature miRNA and mature-star sequence database, in which miRNA and their precursor sequences have been removed. Again, the remaining quality reads were mapped against repeat sequences (known) in Repbase (Version 21) using NCBI-BLAST. Then after, the matched tags have been excluded from further analysis resulting in the removal of known repeat database mapping reads. Further, read tags were processed via removal of < 15 nt and > 30 nt and resulting read population is considered as the clean putative miRNA population.

Table 3 Depicts the ingredients of different level of carbohydrate treatments. Ingredients

Treatment_1 (20%)

Treatment_2 (40%)

Treatment_3 (60%)

Fish meal Casein Gelatinised starch Glucose Cellulose Fish oil Veg oil Binder Mineral mix Vitamin mix Vitamin C BHT

20 10.2 20

20 10.2 40

20 10.2 60

0 40 4 3 0.5 1 1 0.2 0.1 100

0 20 4 3 0.5 1 1 0.2 0.1 100

0 0 4 3 0.5 1 1 0.2 0.1 100

5.4. Discovery of the conserved and novel miRNAs Due to the absence of whole genome sequence information of Labeo rohita in the database, sample-wise repeat reads (filtered) were mapped against to the reference genome of common carp, Cyprinus carpio, and subsequently, precursor sequences were identified using miRCat from the package sRNA-workbench (v3.2.) (http://srna-workbench.cmp.uea. ac.uk/mircat/). Relevant parameters used for this approach are minimum abundance (6), minimum free energy (−25.0 kCal/mol), maximum gaps (3 nt), maximum genome hits (50), minimum GC (30%) and p-Value (0.05). Furthermore, to identify the conserved (known) and novel putative miRNAs in the three treatment groups, the resulted data from three libraries were compared to conserved miRNAs of miRBase and also assigned the miR family is based on which the conserved and novel putative miRNAs were segregated. The structures of miRNAs and precursor RNAs were drawn using the Vienna RNA package [51]. The identified novel putative miRNAs were analyzed for understanding its secondary structures, which plays a crucial role for target mRNA recognition. The binding affinity between microRNAs and mRNAs based on the secondary structure of pre-miRNA and their precursor's free energy values. We have used Mfold algorithm for checking and generating of the fold-back secondary structure via submitting miRNA precursor sequences (http://unafold.rna.albany.edu/?q=mfold). The mfold software predicts the miRNA secondary structures along with the minimum free energy of the RNA sequences.

subjected to air dry and fed to the fish at 3% of body weight. The experiment was set-up in triplicate along with control containing 30 individuals in each tank. The tank water was partially cleaned and refilled with fresh water and the fish were fed twice per day (10 am and 5 pm). After 45 days of the experiment, liver tissue samples were collected from three experimental fishes each treatment. The fishes were anesthetized in a water tank containing the 100 mg/L concentration of MS-222 (Tricainemethanesulfonate, Sigma Aldrich). Three fishes were collected from each treatment in triplicates representing as biological replicates. The tissue samples were immediately collected and frozen in liquid nitrogen (LN2) and subsequently stored at −80 °C for further processing for RNA isolation. 5.2. RNA isolation, cDNA library construction and sequencing The total RNA was isolated from the collected liver tissue samples for the generation of small RNA library using Quick-RNA Miniprep plus kit (ZYMO Research) as per the manufacturer's instruction. The quality and quantity of the isolated RNA were checked to ensure the RNA quality met the criteria for sequencing using 1% denaturing RNA agarose gel and NanoDrop, respectively. The small RNA-Seq libraries were constructed from the isolated total RNA using IlluminaTruSeq small RNA Library Preparation Kit as per manufacturer's instruction. The library generation procedure includes adapter ligation, RT (reverse transcription), amplification using PCR and gel purification. Summarily, 3′ adapter of RNA is particularly modified to target miRNAs isolation including other small non-coding RNAs, which resulted due to enzymatic cleavage by Dicer or other enzymes. Here, adaptors mainly ligated to RNA molecule end having 3’ Hydroxyl group, subsequently, RT-qPCR used to generate cDNA (Single-stranded). Further, cDNA is amplified using universal primer having one of the index sequences. After that, the amplified PCR product with index sequences was size selected and purified on 6% TBE gel. The size selected and purified cDNA libraries processed to analyze using high sensitivity D1000 Screen tap in 4200 Tape station system (Agilent Technologies) as per manufacturer's directives. Further, SE Illumina quality libraries having mean peak size were loaded onto the NextSeq 500 machine for cluster creation and high-throughput sequencing. The sequencing was carried out using 1 × 75 bases chemistry on IlluminaNextSeq 500.

5.5. Differential analysis of miRNA and target identification After aligning onto the respective genome, the read counts were used for expression profiling to identify differentially expressed miRNA involved in carbohydrate metabolism. miRNA with at least five read support in any of the samples were considered for the differential analysis and reads were normalized using TPM (Transcripts Per Million). Finally, fold change values were obtained from TPM values and the fold change higher than 1.5 and below −1.5 were considered as differentially expressed miRNAs. In addition, the miRNA targets were predicted using RNAhybrid [52], a well-known miRNA target identification tool. Well annotated zebrafish genome was used for target identification in which the 3’ UTR sequences of zebrafish were fetched from the TargetScanFish (v6.2) Database [53] and subsequently targets were identified. 5.6. Co-expression network analysis of miRNAs-targeted genes The differential expressed miRNAs were selected based on their target genes associated with metabolic activity for the co-expression network analysis. The miRNAs-targeted mRNAs network constructed using Cytoscape v3.7.1 software [54]. In this analysis, we have selected up-regulated miRNA targeted genes which showed a high node degree and proximity centrality. Here, closeness centrality measures the number of interactions among miRNAs and genes. The Cytoscape

5.3. Sequence analysis The raw reads were processed for escalating sequence quality by removing adapters and low quality reads. Initially, the reads containing adapters were screened using Cutadapt (v 1.12) (https://cutadapt. readthedocs.io/en/stable/) and subsequently, low-quality reads were 11

Genomics xxx (xxxx) xxx–xxx

K.D. Rasal, et al.

software was used for visualization of the miRNA-gene networks via enriching with them expression pattern. [12]

5.7. qPCR for selected miRNAs

[13]

To validate the differential expression of miRNAs in the liver of L. rohita, randomly five miRNAs were selected and their relative expressions were profiled using quantitative stem-loop RT-qPCR. The total RNA was isolated from the liver tissue of same experiment using Trizol reagent, subsequently, cDNA libraries were generated using RT (reverse transcription) kit (Thermo Scientific, USA) as per the manufacturer's instruction. The qPCR was performed in a 20 ul reaction volume, which containing 2 ul of cDNA and 10 ul of SYBR Green Supermix (GeneCopoeia, USA), 2 ul of forward primer and 2 ul of Universal reverse adapter primer with the condition of 35 cycles followed by Initial denaturation 95 °C for 10 mins, Denaturation 95 °C for 10 s, Annealing 57 °C for 20 s and extension 72 °C for 15 s using ABI 7500 PCR machine (Applied Biosystems). The primers were designed based on mature miRNA sequence obtained from NGS sequencing data and list of primers are available in details (Table S3). The U6 gene was used as an internal control (housekeeping gene). For each sample, the samples were tested in triplicate. The specific amplifications were confirmed using PAGE (polyacrylamide gel electrophoresis). The miRNA expression levels were normalized to that of the housekeeping gene and the results were analyzed. Supplementary data to this article can be found online at https:// doi.org/10.1016/j.ygeno.2019.07.010.

[14] [15]

[16]

[17] [18] [19]

[20]

[21]

Declaration of Competing Interest

[22]

All authors declare that they don't have any competing interests.

[23]

Acknowledgments

[24]

Authors are thankful to Director, ICAR-Indian Agricultural Statistics Research Institute, New Delhi, India; Director, ICAR-Central Institute of Fisheries Education, Mumbai, India and Director, ICAR - Central Institute of Freshwater Aquaculture, Bhubaneswar, Odisha, India for providing research facility.

[25] [26]

References

[27]

[1] T. Kaeuferle, S. Bartel, S. Dehmel, S. Krauss-Etschmann, MicroRNA methodology: advances in miRNA technologies, Methods Mol. Biol. 1169 (2014) 121–130. [2] M.E. Herkenhoff, A.C. Oliveira, P.G. Nachtigall, J.M. Costa, V.F. Campos, A.W.S. Hilsdorf, D. Pinhal, Fishing into the MicroRNA transcriptome, Front. Genet. 9 (2018). [3] L. Yang, D. Jue, W. Li, R. Zhang, M. Chen, Q. Yang, Identification of MiRNA from eggplant (Solanum melongena L.) by small RNA deep sequencing and their response to Verticillium dahliae infection, PLoS One 8 (2013) e72840. [4] Y. Fu, J. Lan, X. Wu, D. Yang, Z. Zhang, H. Nie, R. Hou, R. Zhang, W. Zheng, Y. Xie, N. Yan, Z. Yang, C. Wang, L. Luo, L. Liu, X. Gu, S. Wang, X. Peng, G. Yang, Identification of Dirofilaria immitis miRNA using illumina deep sequencing, Vet. Res. 44 (2013) 3. [5] K.D. Rasal, P.C. Nandanpawar, P. Swain, M.R. Badhe, J.K. Sundaray, P. Jayasankar, MicroRNA in aquaculture fishes: a way forward with high-throughput sequencing and a computational approach, Rev. Fish Biol. Fish. 26 (2016) 199–212. [6] S.D. Hsu, Y.T. Tseng, S. Shrestha, Y.L. Lin, A. Khaleel, C.H. Chou, C.F. Chu, H.Y. Huang, C.M. Lin, S.Y. Ho, T.Y. Jian, F.M. Lin, T.H. Chang, S.L. Weng, K.W. Liao, I.E. Liao, C.C. Liu, H.D. Huang, miRTarBase update 2014: an information resource for experimentally validated miRNA-target interactions, Nucleic Acids Res. 42 (2014) D78–D85. [7] A.F. Schier, A.J. Giraldez, MicroRNA function and mechanism: insights from zebra fish, Cold Spring Harb. Symp. Quant. Biol. 71 (2006) 195–203. [8] L. Zhao, H. Lu, Q. Meng, J. Wang, W. Wang, L. Yang, L. Lin, Profilings of MicroRNAs in the liver of common carp (Cyprinus carpio) infected with Flavobacterium columnare, Int. J. Mol. Sci. 17 (2016) 566. [9] G. Li, Y. Zhao, L. Wen, Z. Liu, F. Yan, C. Gao, Identification and characterization of microRNAs in the spleen of common carp immune organ, J. Cell. Biochem. 115 (2014) 1768–1778. [10] W. Chi, C. Tong, X. Gan, S. He, Characterization and comparative profiling of MiRNA transcriptomes in bighead carp and silver carp, PLoS One 6 (2011) e23549. [11] K. Lau, K.P. Lai, J.Y. Bao, N. Zhang, A. Tse, A. Tong, J.W. Li, S. Lok, R.Y. Kong,

[28] [29] [30] [31] [32] [33] [34] [35]

[36]

[37]

[38]

12

W.Y. Lui, A. Wong, R.S. Wu, Identification and expression profiling of microRNAs in the brain, liver and gonads of marine medaka (Oryzias melastigma) and in response to hypoxia, PLoS One 9 (2014) e110698. S. Kumari, K.D. Rasal, J.K. Sundaray, S. Nandi, P. Jayasankar, Comparative study of cellular tumor antigen p53 protein of fishes and analysis of its protein interaction network using computational approach, Comput. Mol. Biol. 5 (2015) 1–9. Y.P. Zhu, W. Xue, J.T. Wang, Y.M. Wan, S.L. Wang, P. Xu, Y. Zhang, J.T. Li, X.W. Sun, Identification of common carp (Cyprinus carpio) microRNAs and microRNA-related SNPs, BMC Genomics 13 (2012) 413. J.A. Mennigen, C.J. Martyniuk, I. Seiliez, S. Panserat, S. Skiba-Cassy, Metabolic consequences of microRNA-122 inhibition in rainbow trout, Oncorhynchus mykiss, BMC Genomics 15 (2014) 70. J.A. Mennigen, E. Plagnes-Juan, C.A. Figueredo-Silva, I. Seiliez, S. Panserat, S. Skiba-Cassy, Acute endocrine and nutritional co-regulation of the hepatic omymiRNA-122b and the lipogenic gene fas in rainbow trout, Oncorhynchus mykiss, comparative biochemistry and physiology, Part B, Biochem. Mol. Biol. 169 (2014) 16–24. F. Gomes, L. Watanabe, S. Nozawa, L. Oliveira, J. Cardoso, J. Vianez, M. Nunes, H. Schneider, I. Sampaio, Identification and characterization of the expression profile of the microRNAs in the Amazon species Colossoma macropomum by next generation sequencing, Genomics 109 (2017) 67–74. J.A. Mennigen, S. Panserat, M. Larquier, E. Plagnes-Juan, F. Medale, I. Seiliez, S. Skiba-Cassy, Postprandial regulation of hepatic microRNAs predicted to target the insulin pathway in rainbow trout, PLoS One 7 (2012) e38604. J.A. Mennigen, S. Skiba-Cassy, S. Panserat, Ontogenetic expression of metabolic genes and microRNAs in rainbow trout alevins during the transition from the endogenous to the exogenous feeding period, J. Exp. Biol. 216 (2013) 1597–1608. L.H. Miao, Y. Lin, W.J. Pan, X. Huang, X.P. Ge, M.C. Ren, Q.L. Zhou, B. Liu, Identification of differentially expressed Micrornas associate with glucose metabolism in different organs of blunt snout bream (Megalobrama amblycephala), Int. J. Mol. Sci. 18 (2017). P. Liang, S. Liu, F. Xu, S. Jiang, J. Yan, Q. He, W. Liu, C. Lin, F. Zheng, X. Wang, W. Miao, Powdery mildews are characterized by contracted carbohydrate metabolism and diverse effectors to adapt to obligate biotrophic lifestyle, Front. Microbiol. 9 (2018) 3160. L.H. Miao, Y. Lin, X. Huang, W.J. Pan, Q.L. Zhou, B. Liu, M.C. Ren, X.P. Ge, L.K. Pan, In vivo analysis of miR-34a regulated glucose metabolism related genes in Megalobrama amblycephala, Int. J. Mol. Sci. 19 (2018). G.R. Prabu, A.K. Mandal, Computational identification of miRNAs and their target genes from expressed sequence tags of tea (Camellia sinensis), Genomics, Proteomics Bioinforma. 8 (2010) 113–121. T.T. Bizuayehu, I. Babiak, MicroRNA in teleost fish, Genome Biol. Evol. 6 (2014) 1911–1937. B. Yan, B. Liu, C.D. Zhu, K.L. Li, L.J. Yue, J.L. Zhao, X.L. Gong, C.H. Wang, microRNA regulation of skin pigmentation in fish, J. Cell Sci. 126 (2013) 3401–3408. G. Zhang, E. Dervishi, B.N. Ametaj, Milk fever in dairy cows is preceded by activation of innate immunity and alterations in carbohydrate metabolism prior to disease occurrence, Res. Vet. Sci. 117 (2018) 167–177. E. Wienholds, W.P. Kloosterman, E. Miska, E. Alvarez-Saavedra, E. Berezikov, E. de Bruijn, H.R. Horvitz, S. Kauppinen, R.H. Plasterk, MicroRNA expression in zebrafish embryonic development, Science 309 (2005) 310–311. W.P. Kloosterman, A.K. Lagendijk, R.F. Ketting, J.D. Moulton, R.H. Plasterk, Targeted inhibition of miRNA maturation with morpholinos reveals a role for miR375 in pancreatic islet development, PLoS Biol. 5 (2007) e203. M. Girard, E. Jacquemin, A. Munnich, S. Lyonnet, A. Henrion-Caude, miR-122, a paradigm for the role of microRNAs in the liver, J. Hepatol. 48 (2008) 648–656. L. Cui, H. Hu, W. Wei, W. Wang, H. Liu, Identification and characterization of MicroRNAs in the liver of blunt snout bream (Megalobrama amblycephala) infected by Aeromonas hydrophila, Int. J. Mol. Sci. 17 (2016) 1972. A.M. Liu, Z. Xu, F.H. Shek, K.F. Wong, N.P. Lee, R.T. Poon, J. Chen, J.M. Luk, miR122 targets pyruvate kinase M2 and affects metabolism of hepatocellular carcinoma, PLoS One 9 (2014) e86872. B. Zhang, X. Pan, G.P. Cobb, T.A. Anderson, microRNAs as oncogenes and tumor suppressors, Dev. Biol. 302 (2007) 1–12. R.J.A. Frost, E.N. Olson, Control of glucose homeostasis and insulin sensitivity by the Let-7 family of microRNAs, Proc. Natl. Acad. Sci. 108 (2011) 21075–21080. N.N. Baroukh, E. Van Obberghen, Function of microRNA-375 and microRNA-124a in pancreas and brain, FEBS J. 276 (2009) 6509–6521. F.J. Jin, P. Han, M. Zhuang, Z.M. Zhang, L. Jin, Y. Koyama, Comparative proteomic analysis: SclR is importantly involved in carbohydrate metabolism in Aspergillus oryzae, Appl. Microbiol. Biotechnol. 102 (2018) 319–332. V. Plaisance, A. Abderrahmani, V. Perret-Menoud, P. Jacquemin, F. Lemaigre, R. Regazzi, MicroRNA-9 controls the expression of Granuphilin/Slp4 and the secretory response of insulin-producing cells, J. Biol. Chem. 281 (2006) 26932–26942. X. Xie, X. Meng, X. Zhou, X. Shu, H. Kong, Research on therapeutic effect and hemorrheology change of berberine in new diagnosed patients with type 2 diabetes combining nonalcoholic fatty liver disease, Zhongguo Zhong yao za zhi = Zhongguo zhongyao zazhi = China Journal of Chinese Materia Medica 36 (2011) 3032–3035. L. Zhang, S. He, S. Guo, W. Xie, R. Xin, H. Yu, F. Yang, J. Qiu, D. Zhang, S. Zhou, K. Zhang, Down-regulation of miR-34a alleviates mesangial proliferation in vitro and glomerular hypertrophy in early diabetic nephropathy mice by targeting GAS1, J. Diabetes Complicat. 28 (2014) 259–264. X. Yan, L. Ding, Y. Li, X. Zhang, Y. Liang, X. Sun, C.B. Teng, Identification and

Genomics xxx (xxxx) xxx–xxx

K.D. Rasal, et al.

[39]

[40] [41] [42] [43] [44]

[45]

profiling of microRNAs from skeletal muscle of the common carp, PLoS One 7 (2012) e30925. T.G. Paula, B.T.T. Zanella, B.E.A. Fantinatti, L.N. Moraes, B. Duran, C.B. Oliveira, R.A.S. Salomao, R.N.D. Silva, C.R. Padovani, V.B.D. Santos, E.A. Mareco, R.F. Carvalho, M. Dal-Pai-Silva, Food restriction increase the expression of mTORC1 complex genes in the skeletal muscle of juvenile pacu (Piaractus mesopotamicus), PLoS One 12 (2017) e0177679. C. Fernandez-Hernando, Y. Suarez, K.J. Rayner, K.J. Moore, MicroRNAs in lipid metabolism, Curr. Opin. Lipidol. 22 (2011) 86–92. R. Mennigen, M. Colombo-Benkmann, N. Senninger, M. Laukoetter, Endoscopic closure of postoperative gastrointestinal leakages and fistulas with the over-thescope clip (OTSC), J. Gastrointest. Surg. 17 (2013) 1058–1065. M. Trajkovski, J. Hausser, J. Soutschek, B. Bhat, A. Akin, M. Zavolan, M.H. Heim, M. Stoffel, MicroRNAs 103 and 107 regulate insulin sensitivity, Nature 474 (2011) 649–653. H. Bai, Z. Cao, C. Deng, L. Zhou, C. Wang, miR-181a sensitizes resistant leukaemia HL-60/Ara-C cells to Ara-C by inducing apoptosis, J. Cancer Res. Clin. Oncol. 138 (2012) 595–602. S.D. Jordan, M. Kruger, D.M. Willmes, N. Redemann, F.T. Wunderlich, H.S. Bronneke, C. Merkwirth, H. Kashkar, V.M. Olkkonen, T. Bottger, T. Braun, J. Seibler, J.C. Bruning, Obesity-induced overexpression of miRNA-143 inhibits insulin-stimulated AKT activation and impairs glucose metabolism, Nat. Cell Biol. 13 (2011) 434–446. J.W. Kornfeld, C. Baitzel, A.C. Konner, H.T. Nicholls, M.C. Vogt, K. Herrmanns, L. Scheja, C. Haumaitre, A.M. Wolf, U. Knippschild, J. Seibler, S. Cereghini,

[46] [47] [48] [49] [50] [51] [52] [53] [54]

13

J. Heeren, M. Stoffel, J.C. Bruning, Obesity-induced overexpression of miR-802 impairs glucose metabolism through silencing of Hnf1b, Nature 494 (2013) 111–115. E.G. Beale, R.E. Hammer, B. Antoine, C. Forest, Disregulated glyceroneogenesis: PCK1 as a candidate diabetes and obesity gene, Trends Endocrinol Metab 15 (2004) 129–135. W. Liao, X.T. Cui, Reagentless aptamer based impedance biosensor for monitoring a neuro-inflammatory cytokine PDGF, Biosens. Bioelectron. 23 (2007) 218–224. C.M. Ramirez, L. Goedeke, N. Rotllan, J.H. Yoon, D. Cirera-Salinas, J.A. Mattison, Y. Suarez, R. de Cabo, M. Gorospe, C. Fernandez-Hernando, MicroRNA 33 regulates glucose metabolism, Mol. Cell. Biol. 33 (2013) 2891–2902. A.M. Bolger, M. Lohse, B. Usadel, Trimmomatic: a flexible trimmer for Illumina sequence data, Bioinformatics 30 (2014) 2114–2120. S. Griffiths-Jones, A. Bateman, M. Marshall, A. Khanna, S.R. Eddy, Rfam: an RNA family database, Nucleic Acids Res. 31 (2003) 439–441. R. Lorenz, S.H. Bernhart, C. Höner zu Siederdissen, H. Tafer, C. Flamm, P.F. Stadler, I.L. Hofacker, ViennaRNA Package 2.0, Algorithm. Mol. Biol. 6 (2011) 26. J. Krüger, M. Rehmsmeier, RNAhybrid: microRNA target prediction easy, fast and flexible, Nucleic Acids Res. 34 (2006) W451–W454. V. Agarwal, G.W. Bell, J.-W. Nam, D.P. Bartel, Predicting effective microRNA target sites in mammalian mRNAs, eLife 4 (2015) e05005. P. Shannon, A. Markiel, O. Ozier, N.S. Baliga, J.T. Wang, D. Ramage, N. Amin, B. Schwikowski, T. Ideker, Cytoscape: a software environment for integrated models of biomolecular interaction networks, Genome Res. 13 (2003) 2498–2504.