Identification and characterization of circRNA during muscle development in different breeds of cattle CURRENT

Background Black cattle are a new breed of cattle that are developed by applying modern biotechnology, such as somatic cloning, and conventional breeding methods to Luxi cattle. It is very important to study the function and regulatory mechanism of circRNAs in muscle differentiation among different breeds to improve meat quality and meat production performance and to provide new ideas for beef cattle meat quality improvements and new breed development. Therefore, the goal of this study was to sequence and identify circRNAs in muscle tissues of different breeds of cattle. We used RNA-seq to identify circRNAs in the muscles of two breeds of black cattle (Black and Luxi). Results We identified 14640 circRNAs and found 655 differentially expressed circRNAs. We also analysed the classification and characteristics of circRNAs in muscle tissue. GO and KEGG analyses were used on the parental genes of circRNAs. They were mainly involved in a variety of biological processes, such as muscle fibre development, smooth muscle cell proliferation, bone system morphogenesis, tight junctions and the MAPK, AMPK and mTOR signalling pathways. In addition, we used miRanda to predict the interactions between 15 circRNAs and 12 miRNAs. Based on the above assays, we identified circRNAs (circ0001048, circ0001103, circ0001159, circ0003719, circ0003794, circ0003721, circ0003720, circ0001519, circ0001530, circ0005060, circ0006589, circ0000181, circ0000190, circ0010558, circ0010577) that may play an important role in the regulation of muscle growth and development. Conclusion Our results provide more information about circRNAs regulating muscle development in different breeds of cattle and lay a solid foundation for future experiments. of circRNA in muscle differentiation and combined these findings with the characteristics of different breeds to cultivate fast-growing and high meat production rate beef cattle varieties to increase the speed of beef cattle breeding and provide a theoretical basis for the development of China's beef industry. and 38 corresponding circRNAs. Comparing these results with previous research results, significant differences were observed in the expression of circRNAs related to the muscle development of different breeds of cattle, suggesting that circRNAs may play an important role in muscle development. Whether these circRNAs have specific functions and what the functional mechanism is need to be studied further. real-time quantitative PCR; SPONG: the interaction network between miRNA and circRNA.


Abstract
Background Black cattle are a new breed of cattle that are developed by applying modern biotechnology, such as somatic cloning, and conventional breeding methods to Luxi cattle. It is very important to study the function and regulatory mechanism of circRNAs in muscle differentiation among different breeds to improve meat quality and meat production performance and to provide new ideas for beef cattle meat quality improvements and new breed development. Therefore, the goal of this study was to sequence and identify circRNAs in muscle tissues of different breeds of cattle. We used RNA-seq to identify circRNAs in the muscles of two breeds of black cattle (Black and Luxi).

Results
We identified 14640 circRNAs and found 655 differentially expressed circRNAs. We also analysed the classification and characteristics of circRNAs in muscle tissue. GO and KEGG analyses were used on the parental genes of circRNAs. They were mainly involved in a variety of biological processes, such as muscle fibre development, smooth muscle cell proliferation, bone system morphogenesis, tight junctions and the MAPK, AMPK and mTOR signalling pathways. In addition, we used miRanda to predict the interactions between 15 circRNAs and 12 miRNAs. Based on the above assays, we identified circRNAs (circ0001048, circ0001103, circ0001159, circ0003719, circ0003794, circ0003721, circ0003720, circ0001519, circ0001530, circ0005060, circ0006589, circ0000181, circ0000190, circ0010558, circ0010577) that may play an important role in the regulation of muscle growth and development.

Conclusion
Our results provide more information about circRNAs regulating muscle development in different breeds of cattle and lay a solid foundation for future experiments.

Background
CircRNA is a unique kind of noncoding RNA that has no 5' terminal cap or 3' terminal poly (a) tail and presents a closed ring structure [1]. It was first discovered in 1976 by Kolakofsky and Sanger in plant viroid and parainfluenza virus particles by electron microscopy that showed closed loop, circular RNAs [2,3]. It was found that most of the circRNAs originate from exons and a few from introns. In terms of function, circRNAs mainly adsorb miRNAs through the "molecular sponge" mechanism, thus inhibiting the regulatory role of miRNAs and enhancing the expression levels of target genes [4 − 6]. In recent years, circRNA has become a new research hotspot in the field of scientific research. It has been found that circRNAs are involved in many biological processes, including growth and development, and diseases, among others. At present, research on circRNAs is mainly focused on cancer. There are almost no reports about the regulation by circRNAs of the development of bovine skeletal muscle. Li found that the circRNAs circFUT1O and circFGFR4 could regulate the proliferation and differentiation of bovine skeletal muscle cells by absorbing miR-133a and miR-107, respectively [7,8]. Wei found that circLMO7 can regulate the differentiation and apoptosis of bovine skeletal muscle cells by absorbing miR-378a-3p [9]. In conclusion, circRNAs play an important role in the development of bovine muscle.
Black cattle are the first embryo transfer calves in China and are obtained by vitrified frozen somatic cell cloning embryos. After careful cultivation by researchers, Black cattle bulls were used in breeding. Researchers overcame the shortcomings of female Luxi cattle by hybrid breeding combined with molecular marker-assisted breeding, adopted black hair and obtained bull semen, and improved Luxi cattle by hybridization with another breed so that the offspring combined the excellent characteristics of both and improved production performance. In 2015, Black cattle were identified as a new species group by authoritative experts and established a Chinese type base for new breed cultivation. Luxi cattle are one of the five local breeds of beef cattle in China, with high meat production capacity and tender meat quality, and are well known as "five flowers and three layers of meat". We selected Black cattle and Luxi cattle for this research study, identified circRNAs in muscle tissue, analysed their genomic characteristics, expression differences, etc. Through high-throughput sequencing technology combined with functional verification tests, we revealed the function and regulatory mechanism of circRNA in muscle differentiation and combined these findings with the characteristics of different breeds to cultivate fast-growing and high meat production rate beef cattle varieties to increase the speed of beef cattle breeding and provide a theoretical basis for the development of China's beef industry.

Results
Apparent differences in muscle fibres in different breeds of beef cattle The muscle fibres of the longest muscle in the back of Black cattle and Luxi cattle were significantly different in the apparent observation of paraffin sections stained by HE (Fig. 1). The length of single muscle fibres of Black cattle was significantly longer than that of Luxi cattle, and the number of nuclei in each muscle fibre was also greater. The boundary between the muscle fibres of Black cattle was clearer and rounder than that of Luxi cattle. IPP software analysis showed that there were significant differences in the muscle fibre diameter, length and weight (P < 0.05) but no significant differences in other muscle fibre properties (P > 0.05) ( Table 1).

Differential Expression Of Circrnas
In this study, 655 differential circRNAs and 467 parental genes (attachment 2: Table S2) (Table 4). One parental gene was involved in the regulation of multiple pathways.
According to the statistics of the pathway enrichment analysis of the 49 parental genes (Fig. 7 involved in muscle processes. The pathways related to meat growth and development were also regulated, but the degree of enrichment was different. According to the interaction data and differential expression of miRNAs and circRNAs, the interaction network data files were generated and imported into Cytoscape software. The attributes of the target circRNAs were visualized in the network, and the topological attributes of some networks are marked (Fig. 8) Table   S6). The circRNAs with the most target miRNAs were circ0013807 (37), circ0006152 (27), and circ0008394 (26). According to the above differential expression and GO/KEGG/PPI enrichment analyses, we identified 7 genes of different sources related to muscle growth, corresponding to 15 circRNAs and 12 target miRNAs (Table 5). It was found that there are multiple binding sites of miRNAs in some circRNAs (such as MYBPC1 and miR-11986b, RYR1 and miR-10171-3p) sequences. After a miRNA is adsorbed, it cannot regulate its corresponding target gene, thus a circRNA acts as a miRNA molecular sponge.

Confirmation Of Circrna Expression By Qrt-pcr
To verify the expression level of differentially expressed circRNAs, we randomly selected six highly expressed circRNAs and detected their expression level by qRT-PCR (attachment 7: Table S7). These results were consistent with the trends observed in RNA-seq data (Fig. 9a), with a correlation coefficient R2 = 0.9982 indicating that the RNA-seq results were reliable (Fig. 9b).

Discussion
In this study, we first examined the apparent differences in muscle fibres of different breeds of beef cattle. The results showed that there were obvious differences in the apparent observation of the muscle fibres of HE stained paraffin sections of the longissimus dorsi of Black cattle and Luxi cattle.
The length of a single muscle fibre of Black cattle was significantly longer than that of Luxi cattle, and the number of nuclei in each muscle fibre was also greater. The border between the muscle fibres of Black cattle was clearer and rounder than that of Luxi cattle. IPP software analysis showed that there were significant differences in the diameter and length of muscle fibres (P < 0.05) but no significant differences in the number, density or area of other muscle fibres (P > 0.05). The occurrence of these differences may be the key factors leading to the differences in meat production performance and meat quality of the two breeds of cattle after birth, which was also the research basis of this study to explore the underlying molecular regulatory mechanism.
We used RNA-seq technology to study the expression of circRNA in the longissimus dorsi muscle of different breeds of beef cattle. A total of 14640 circRNAs and 4201 parental genes were detected. showed that they are involved in protein ubiquitination, which is part of the protein modifications that regulate cell metabolism within eukaryotes [10]. MYL1 is a crucial protein for adequate skeletal muscle function and belongs to the myosin family [11]. The ubiquitin proteasome system (UPS) is mainly responsible for the increased protein breakdown observed in muscle wasting. The Ube family of E3 ligases is a class of enzymes (i.e., troponin I, myosin heavy chains and actin) that can guide the degradation of major contractile proteins. Their catalytic activity depends on the covalent binding of polyubiquitin chains catalysed by a specific E2 on the substrate [12]. Studies have shown that UPS can control almost any muscle mass and recovery process in catabolism. The muscle-specific E3 ligase UBE family participates in the targeting of actin, myosin, troponin and other major contractile proteins [13], indicating that the high expression of circRNAs plays a certain role in muscle development and redifferentiation.
According to the fold change > 1. circRNAs can play important roles by regulating the transcription and expression of their parental genes [20]. At present, there is a relatively limited understanding of the details of the formation of circRNA and its functional mechanism. circRNA can be obtained by transcription of protein-coding genes or intergenic regions [21]. The formation of circRNA from a protein-coding gene is caused by the variable splicing of the parental gene [22]. Therefore, there should be a certain correlation between a circRNA and its parental gene expression. We found that one source gene may produce multiple circRNA subtypes. For example, the MYBPC1 gene can produce nine different circRNA subtypes. We obtained the FPKM value of the two varieties and found that both were differentially expressed. Although one source gene may produce multiple circRNA subtypes at the same time, only 3 or 4 of them have high expression levels, and the rest have low expression levels, which indicates that the cyclization of RNA in muscle is strictly regulated. To further understand the biological function and molecular function of the parental genes of significantly differentially expressed circRNAs, we predicted the interaction between circRNAs and miRNAs and constructed a network from the interaction data. The interaction network showed that a single miRNA may be correlated with multiple differentially expressed circRNAs, and there have been reports that circRNAs can competitively adsorb miRNAs [23]. Based on the high-throughput sequencing results, we selected 15 circRNAs related to muscle development as candidate circRNAs (circ0001048, circ0001103, circ0001159, circ0003719, circ0003794, circ0003721, circ0003720, circ0001519, circ0001530, circ0005060, circ0006589, circ0000181, circ0000190, circ0010558, circ0010577). In addition, the target miRNAs were predicted, the corresponding mRNA targets of the miRNAs were predicted, and the circRNA-miRNA-mRNA network was constructed to further study the regulation of muscle development. We will further verify this network in future experiments, which provides a new basis for the study of muscle development in cattle.
In addition to the above findings, there were pathways that were not found in our study that are known to be important and enriched for many parental genes and some that have been reported in previous studies to regulate muscle growth and development. The miRNAs that have been shown to be expressed specifically or preferentially in muscles are called muscle-specific miRNAs (muscle-specific microRNAs, myomiRs) [24] and include miR-1, miR-206, miR-128, miR-483, miR-2425-5p,

Identification of differentially expressed circRNAs
We used SRPBM as a standardized method to quantify the expression of circRNA, and DEseq2 was used to analyse the differential expression of circRNA [30]. In pairwise comparisons, circRNAs with P<0.05 and absolute multiple change value greater than 1.5 were considered to be significantly differentially expressed, and finally, the number of upregulated and downregulated circRNAs was obtained.
Enrichment analysis of GO, KEGG and PPI pathways GO analysis and KEGG pathway analysis of the parental genes of differentially expressed circRNAs were used for annotation. The Blast2GO method was used for GO function analysis. KOBAS software was used to test the statistical enrichment of differential gene expression in the KEGG pathway. When P<0.05, GO terms and KEGG pathways were considered to be significantly enriched.

Prediction of miRNA targets of circRNAs
To explore the function of circRNAs and predict which circRNA acts as a miRNA sponge, we used miRanda (3.3a) (http://www.microrna.org/microrna/home.do) to predict the target relationship [31]. In

Ethics approval and consent to participate
All experimental designs and procedures in this study were performed in strict accordance with the 2(11):e363.

Additional Files
Additional file 1: Table S1. Total circRNA detected in Black cattle and Luxi cattle.
Additional file 2: Table S2. Differentially expressed circRNAs in the two comparison groups.
Additional file 3: Table S3. The results of the GO functional enrichment analysis for source genes of circRNAs.
Additional file 4: Table S4.The screened results of GO enrichment analysis.
Additional file 5: Table S5. Detailed results of KEGG pathway analysis for source genes of circRNAs.
Additional file 6: Table S6. Detailed results of predicted the interactions between circRNAs and miRNAs.
Additional file 7: Table S7. Primer sequences for qRT-PCR of randomly selected circRNAs. Figure 1 HE staining of muscle tissue paraffin sections.

Figure 2
Expression level of sample circRNAs Enrichment degree of the identified parental genes in different GO terms Figure 6 Enrichment results of the KEGG analysis Interaction between circRNAs and miRNAs.

Figure 9
Linear fitting of RNA-seq and qRT-PCR circRNA expression data