Identification of ALOX12B Gene Expression, Evolution, and Potential Functional Sites in Horn Development of Sheep

Abstract

The growth and development of horns are primarily controlled by the skin. The ALOX12B gene is crucial for epidermal barrier function and may have a significant impact on horn growth. The purpose of this study was to investigate the expression of ALOX12B across different sheep breeds and tissues by utilizing RNA sequencing. Additionally, potential functional sites were identified in conjunction with whole genome sequencing. Our findings revealed that ALOX12B was highly expressed in the scurred horn group as opposed to the normal horn group (SHE). ALOX12B expression was also notably high in the skin across several species. Eight loci that may influence horn size were indicated in this study. Through the alignment of the ALOX12B protein sequence from 16 species, 15 amino acid sites were identified specifically expressed in horned animals. In conclusion, this study established a connection between ALOX12B and horn size and identified a series of functional sites that may serve as molecular markers for reducing the presence of horns in Chinese sheep breeds.

1. Introduction

Sheep hold significant importance as a livestock species in China, contributing significantly to food security and livestock products [1]. However, as animal husbandry advances, the presence of sheep horns increases feeding costs, poses risks to fencing facilities, and can harm both livestock and feeding personnel [2], leading to hidden dangers in sheep production management. Although members of the breeding industry believe that there is no direct correlation between the production performance of horned and hornless sheep, the trend has been towards the use of hornless breeds because of their ease of management [3]. The current artificial dehorning method creates animal welfare concerns, increases workload, causes physiological damage to individual sheep, and negatively impacts production performance. Therefore, genetic selection for hornless sheep has become a research priority [2].

Horns represent a crucial evolutionary feature in ruminants [4], playing an important role in self-defense, confronting predators, and competing for mating rights [5,6]. As derivatives of the skin, their growth and development are primarily regulated by the skin. However, from an economic value perspective, horns do not directly affect economic value features such as milk production, meat quality, or fur. Consequently, despite their evolutionary significance, horns are not considered an economically relevant trait in livestock production. Several genes connected to horn growth have been identified, such as RXFP2, HOXD, NRDC, DCN, MYL2, INSL3, FOXL2, SFRP2, KRT1, and WNT3 [1,6,7,8,9,10]. However, the formation and regulatory mechanisms of sheep horns are still unclear. Our previous research on sheep horns assessed the gene expression variations between normal and scurred horn groups and discovered, through Gene Ontology (GO) enrichment analysis, that ALOX12B contributes to the skin development process.

The ALOX12B (arachidonate 12 lipoxygenase, 12R type) gene encodes the protein 12R-LOX, which is involved in fatty acid metabolism. 12R-LOX is mainly synthesized in the epithelial tissues (e.g., skin) and is essential to epidermal barrier function [11,12,13,14]. ALOX12B also exhibits immunosuppressive activity by reducing antigen presentation in T cells. Research on this gene has predominantly focused on humans, with very little research conducted on sheep. Genetic evidence has described this gene as the pathogenic gene of autosomal recessive congenital ichthyosis, which is a severe keratinization disease [11]. This indicated that ALOX12B had a significant role in skin keratinization [15]; its missense and nonsense mutations are widely present in various cancers, including skin, colorectal [15,16], and cervical [17]. However, no research has been conducted to document that ALOX12B has a relationship with horn size.

The present research aims to investigate the expression, evolution, and potential functional sites of ALOX12B by multi-omics analysis, to explore the molecular mechanisms of ALOX12B, and to provide valuable molecular markers for reducing the presence of horns in Chinese sheep breeds.

2. Results

2.1. Sample-Specific Expression of ALOX12B in Different Groups

ALOX12B, located on chromosome 11, has a transcript: XM_004012678.5. This transcript was used for subsequent analysis. The experimental procedure is shown in Figure 1A. Using RNA sequencing (RNA-seq) data from self-tested Tibetan sheep (n = 6), this study indicated that the scurred horn group could express ALOX12B at a more elevated level (p-value = 0.026) (Figure 1B). This result showed that the scurred horn group (n = 3, horn lengths ranging from 0 to 12 cm) exhibited higher levels of exon expression than the normal horn group (SHE, n = 3, horn lengths > 12 cm) (Figure 1C). Further analysis of ALOX12B gene expression and GC percent was performed using the University of California Santa Cruz (UCSC) database (Figure 1D). Expression of all exons could be observed in the scurred horn group, whereas only minimal expression of Exon 1, Exon 2, Exon 4, and Exon 15 could be observed in the SHE group. Additionally, skin tissue exhibited a similar trend of exon expression as horn tissues but at much lower levels than those observed in horn tissues. In contrast to skin and horn tissues, these exons were largely not expressed in adipose, muscle, stomach, or kidney tissues.

2.2. Differential Expression of ALOX12B in Tissues, Species, Sex, and Breeds

In order to clarify the differential expression of ALOX12B, RNA-seq information of different humans (n = 9810), pigs (n = 2651), cattle (n = 4359), and sheep (n = 2915) tissues were collected from public databases, and we found expression differences of ALOX12B in 16 independent tissues. Overall, ALOX12B was significantly expressed in the skin of all species, with the highest TPM expression level in humans, followed by sheep. High expression of ALOX12B was also observed in the testis of both sheep and pigs. Sheep and pigs had similar expression levels.

Additional research in the public data (n = 2915) was conducted to determine whether there were sex differences in ALOX12B (Figure 2B). High expression of ALOX12B was observed in the skin, with no significant sex differences found. At the same time, the expression level of ALOX12B was high in the testis. In the blood and lung tissues, although there were sex differences, its expression level was relatively low or almost nonexistent.

Additionally, the expression level of ALOX12B in skin tissue was compared among different sheep breeds (Figure 2C). These ten sheep breeds were divided into three groups (n = 2915): hornless, scurred, and SHE. The findings demonstrated that ALOX12B expression was highest in hornless sheep, then scurred sheep, and lowest in SHE sheep. As a scurred breed, Tan sheep had a slightly higher expression of the ALOX12B gene than the Bashibai sheep. These results indicated that the gene expression of hornless and scurred horn sheep was slightly different from each other.

2.3. Species Evolution and Structure Prediction of the ALOX12B Protein

The phylogenetic relationships of 16 species are shown in Figure 3A. The ALOX12B protein evolution was found to align with the species evolution, with horned species such as cattle, deer, and sheep clustering together and hornless species such as humans, mice, dogs, and camels clustering together. The horned and hornless clusters were well separated, indicating that the structure of the ALOX12B protein is similar and that there has been some protein conservation in horned animals like cattle and deer. The 15 amino acids located at positions 25, 63, 67, 89, 215, 251, 326, 335, 347, 352, 395, 466, 500, 617, and 623 are distinctive to horned animals, which may be vital in determining whether sheep have horns. The amino acids 251, 395, 466, 500, and 623 were located in the α-helix with higher confidence. These positions could make a substantial impact on horns (Figure 3B).

2.4. Allele-Specific Expression Analysis of ALOX12B

In Figure 4, five allele-specific expression (ASE) sites were identified: ASE1 (chr11:27537079), ASE2 (chr11:27537095), ASE3 (chr11:27537929), ASE4 (chr11:27542320), and ASE5 (chr11:27543384). These ASE sites were located on exons, with ASE1 and ASE2 on Exon 1, ASE3 on Exon 3, ASE4 on Exon 8, and ASE5 on Exon 14. In the scurred and SHE groups, all sites’ allele counts had the same trend: the alternative counts were lower than the reference counts. Thus, these ASE sites may be strongly associated with the horn phenotype.

2.5. Identification of Potential Functional Sites in ALOX12B

In analyzing the 2915 public data, the result of three-dimensional principal component analysis (3D-PCA) showed significant separation between horned and hornless sheep breeds (Figure 5A). This result indicated that ALOX12B can, to some extent, distinguish between the two groups, and the sites in this gene region contributed to the regulation of horn development. A total of nine loci with an F-statistic (Fst) > 0.15 were screened by analyzing the public whole genome sequencing (WGS) datasets (n = 3125), and only chr11:27543585 was located on Exon 5 among these loci (Figure 5B, Supplementary Table S3). Through analyzing the self-tested WGS datasets (n = 34), single nucleotide polymorphisms (SNPs, n = 18) were selected. As can be seen in Figure 5C, by intersecting sites with five ASEs and 37 loci (Fst values > 0.05), eight loci were identified (g.27543814G>A, g.27544036T>C, g.27544102G>A, g.27544781A>G, g.27544827T>C, g.27544877C>T, g.27546104C>G, g.27546898T>C), located in the intron region, but their Fst values were not greater than 0.15 (Figure 5C). Among these, chr11:27546104, which had the highest Fst value, showed significant differences in horn length across sheep with different genotypes (Figure 5D). These functional gene loci showed significant differences between horned and hornless sheep and may be crucial in horn size, thereby influencing horn length.

Linkage disequilibrium (LD) revealed 21 SNPs that had high LD scores (Figure 5E). There were two LD blocks containing five SNP loci. LD block1 contained chr11:27538478, chr11:27538651, chr11:27538664, chr11:27538669, and chr11:27540125. LD block3 contained chr11:27543814, chr11:27543949, chr11:27544036, chr11:27544102, and chr11:27544205. LD block2 contained seven SNP sites, which included chr11:27540587, chr11:27540723, chr11:27540798, chr11:27541102, chr11:27541963, chr11:27542091, and chr11:27542269. These findings indicated that these loci were highly likely to affect the growth of horns.

3. Discussion

In previous studies, many genes have been reported to be correlated with the growth of horns. However, research using CRISPR/Cas9 technology revealed that although the RXFP2 gene was partially disrupted, the horn type of sheep remained unaffected [18]. These analyses partially demonstrate that horn development is not merely dictated by a single gene or locus [19].

The transcriptional regulation of ALOX12B is still not fully understood. A study showed that during skin barrier formation, 12R-LOX could oxidize O-linoleoyl-ω-hydroxyceramide to 9R-hydroperoxy-linoleoyl-ω-hydroxyceramide; then eLOX3 (epidermal lipoxygenase-3) further converted 9R-hydroperoxy-linoleoyl-ω-hydroxyceramide to epoxyalcohol, promoting the partial hydrolysis of linoleate in O-linoleoyl-ω-hydroxyceramide [20]. This biochemical process allows for the covalent attachment of free β-hydroxyl groups to the proteins of the cornified cell envelope (CE), ultimately contributing to the formation of the corneocyte lipid envelope (CLE) [21]. Together, the CLE and CE constituted the necessary structure of the epidermal water barrier to reduce water loss. The absence of CLE was found in 12R-LOX (null) mice, demonstrating that 12R-LOX could promote the formation of CLE [22]. These studies provide a preliminary basis for understanding ALOX12B’s role in skin keratinization. However, to date, research on the role of ALOX12B in the growth and development of horns is limited. Although this gene’s mutations are related to skin keratinization, the specific genetic mechanism is unknown.

Our study showed the expression differences of ALOX12B between SHE and scurred groups in sheep for the first time and its expression in different sheep breeds and tissues. Our results indicated that compared with the SHE group, the ALOX12B was significantly upregulated in the scurred group and expressed at lower levels in the skin. Overall, the most elevated expression levels were recorded in human and sheep skin. It was speculated that the ALOX12B gene was strongly associated with the formation of horns, with horn size influenced by its expression level. Overexpression of ALOX12B may even inhibit horn growth. However, there is a limitation in our study to analyze sex differences for ALOX12B due to the lack of data on male sheep horns. In addition, there were differences between the SHE and scurred groups in the screened SNP sites. However, these results were not verified experimentally. This could potentially compromise the reliability of our findings. Therefore, experimental verification will be performed in a subsequent analysis of this gene. Meanwhile, this study was currently using horn length phenotypic data from Small-tailed Han sheep (n = 34) to consider the heritability of ALOX12B. In the future, we will further study ALOX12B by combining Tibetan sheep phenotypic studies with genetic analyses, aiming to achieve a more thorough comprehension of the genetic factors that contribute to horn development. This will contribute to a clearer understanding of the relationship between genotype and phenotype in horn development.

Allele-specific expression (ASE) analysis, based on RNA-seq, is widely utilized in pigs [23], cattle [24], and mice [25]. However, there is a relative paucity of ASE research on sheep. This study identified five ASEs, yet no common intersection was observed with the loci (F-statistic values > 0.05) and SNPs. This discrepancy may be attributed to the fact that ASE primarily recognizes cis-trans regulation, whereas SNP primarily identifies mutation sites. Furthermore, the ongoing refinement of genomic information and the analysis of multiple samples will likely provide additional research opportunities in this field, contributing to a deeper comprehension of the mechanisms regulating gene expression in general or specific to horn development.

The neighbor-joining (NJ) method offers a notable advantage in computational efficiency, enabling the rapid construction of evolutionary trees while maintaining high accuracy for extensive datasets. This capability facilitates the effective identification of evolutionary relationships among species. Although NJ may encounter limitations in complex scenarios, such as evolutionary rate heterogeneity and model simplification issues, it was selected for this study’s purpose of constructing an evolutionary tree. The method successfully differentiated between hornless and horned animals. The 15 amino acid sites identified may have significant effects on the evolution of horns.

The elevated expression of ALOX12B in the skin and ram testis tissues observed in this study suggested its potential involvement in both physiological and reproductive processes in rams. Given ALOX12B’s role in lipid metabolism and inflammatory responses, its expression in the testis may be associated with sperm production, sexual maturation, or seasonal reproductive cycles. Considering the typically seasonal nature of ram fertility, ALOX12B could potentially play a role in regulating seasonal variations in reproductive performance. Future research could explore the seasonal fluctuations in ALOX12B expression to ascertain whether environmental factors such as light exposure or temperature, known to influence reproductive cycles in various livestock species, affect its activity. Additional experimental validation is necessary to elucidate the specific function of ALOX12B in these processes.

ALOX12B was also expressed in small amounts in the brain, whereas it was expressed relatively low or almost nonexistent in other tissues (e.g., lipids, heart, large intestine, lungs). This is consistent with our previous findings in pigs, cattle, humans, and sheep. Horn growth and development are tightly linked to skin growth, while ALOX12B protein’s expression in keratinized animals shows a similar trend; this indicates a strong relationship between ALOX12B and horn growth.

4. Materials and Methods

4.1. Ethics Statement

The animal research protocol received approval from the Science Research Department of the Institute of Animal Sciences, Chinese Academy of Agricultural Sciences, Beijing, China (IAS-CAAS), under acceptance number IASCAAS-AE-03.

4.2. Animal and Sample Collection

RNA-seq datasets were previously acquired by the research team (PRJNA1003277) [26]. The research team collected the soft horn tissue of Tibetan sheep (female, 3.5 to 5 years old, collected in Dangxiong, Tibet, China). The samples (n = 6) were divided into two groups: the scurred horn group (n = 3), with horn lengths ranging from 0 to 12 cm, and the normal horn group (SHE, n = 3), with horn lengths greater than 12 cm (Supplementary Table S1). All tissues were stored in deep cryopreservation tubes and then immediately frozen using liquid nitrogen.

4.3. RNA-Seq Data Analysis

We additionally obtained 2915 publicly available sheep RNA-seq datasets from two databases: the National Center for Biotechnology Information (NCBI) database (https://www.ncbi.nlm.nih.gov/, accessed on 23 July 2023) and the European Bioinformatics Institute (EBI) database (https://www.ebi.ac.uk/, accessed on 23 July 2023). To ensure a robust analysis, a standard workflow was employed to process the data: initially, data were quality controlled and trimmed utilizing Trim Galore (v.0.6.7). Then, sequences were aligned to the reference genome (ARS-UI_Ramb_v2.0) by STAR (v.2.7.7a) [27]. Finally, the clean datasets were generated for in-depth investigation, achieving over 85% unique mapping reads and more than 20,000,000 clean reads. Normalized gene expression levels and exon modeling were performed by calculating the number of transcripts per million exons (TPM) using the Stringtie’s prepDE.py script (v.2.1.5) [28]. The generated data were carried out by featureCounts (v.2.0.1), and the original count matrix was extracted [29]. The GATK (v.4.2.5.0) pipeline, with the default parameters, was applied to Allele-specific expression (ASE) [30,31].

4.4. Expression Pattern Analysis of the ALOX12B Gene

We used ggplot2 (v.3.5.1) to generate a boxplot for the self-tested RNA-seq datasets (n = 6). The dexseq_prepare_annotation2.py script (from https://github.com/vivekbhr/Subread_to_DEXSeq, accessed on 5 March 2024) was used to initialize the genome annotation (GTF) file, enabling the analysis of disparities in ALOX12B exon expression between the two groups. After that, the formatted GTF file and the counts matrix were processed using the load_SubreadOutput.R. Subsequently, an analysis of differential gene expression in exons was performed.

The public RNA-seq datasets were downloaded from the GTEx project, including pigs (http://piggtex.farmgtex.org/, accessed on 15 August 2023, n = 2651), cattle (https://cgtex.roslin.ed.ac.uk/, accessed on 6 October 2023, n = 4359), and humans (https://gtexportal.org/home/datasets, accessed on 17 January 2024, n = 9810) [32,33,34]. Together with the previous public sheep datasets (n = 2915), all RNA-seq datasets were merged and categorized into 16 tissues. ALOX12B’s mean TPM values are shown in Table S2.

The sheep breed information from public RNA-seq datasets (n = 2915) was collected, excluding those with smaller sample sizes, to further explore the specific expression variations of ALOX12B. These datasets from ten breeds of sheep were classified into three subgroups: hornless (Spanish churra sheep, Hu sheep, Bashibai sheep, Texel sheep), scurred (Chinese merino sheep, Gansu alpine fine wool sheep, Minxian black fur sheep, Tan sheep), and SHE sheep (Rambouillet sheep, Tibetan sheep). All significance statistics were calculated using the t-test.

For the public sheep RNA-seq datasets (n = 2915), principal component analysis (PCA) was conducted by prcomp function in stats (v 4.3.3) with the default parameters, and the three-dimensional (3D) PCA scatterplot was visualized by scatterplot3d (v0.3.44) with the parameter “pch = 19”. The percentage of variance explained for each principal component was PC1 = 65.93%, PC2 = 18.07%, and PC3 = 9.93%.

4.5. Cross-Species Comparison of ALOX12B Protein

A total of 16 species’ ALOX12B protein sequence files were retrieved from NCBI, including sheep, goats, camels, dogs, cattle, deer, mice, and humans. Using the MEGA (v 11.0.11), we constructed a protein evolutionary tree for ALOX12B [35]. The ClustalW method (with default parameters) [36,37], the Jones–Taylor–Thornton model, and the neighbor-joining method were applied to align raw data and build an evolutionary tree for the ALOX12B protein. To identify the presence of amino acids in critical positions within ALOX12B, AlphaFold2 (with default parameters) was applied to predict the 3D structure of ALOX12B from UniProt (https://www.uniprot.org/, accessed on 20 June 2024) [38].

4.6. Identifying SNP Loci from WGS

The public WGS data were obtained from NCBI for 3125 sheep, including PRJEB30931, PRJNA304478, PRJNA325682, PRJNA479525, PRJNA480684, PRJNA509694, PRJNA624020, PRJNA675420, PRJNA779188, PRJNA783661, and PRJNA822017 [39,40,41,42,43,44,45,46,47,48]. These datasets were classified into two groups: horned and hornless. In order to distinguish SNPs that differed significantly between the two groups, Fst values were calculated by VCFtools (v.0.1.16).

For the experiment, we measured the horn length of 34 Small-tail Han sheep, specifically the average length of the left and right horns, and performed WGS on them. Sequencing reads, and adapters were trimmed by Trimmomatic (v.0.39). Raw sequence quality was determined by using FastQC (v.0.12.1) [49]. After that, BWA (v.0.7.17) and Picard (v. 3.1.1) were used for mapping, sorting, and deduplication [50,51]. Here, the default parameters in the GATK (v.4.2.5.0) pipeline were applied to SNP prediction, while mutation annotation was conducted by SnpEff (v.4.3) [52]. All statistical analyses of significance were performed using a t-test. For a more in-depth analysis, the filters “--maf 0.45” and “--min-meanDP 5” in VCFtools (v.0.1.16) [53] were employed, followed by the use of LDBlockShow (v.1.40) [54] to the determination of SNP chains.

5. Conclusions

This study showed for the first time that ALOX12B expression is related to sheep horn size and type and identified 15 conserved sites in keratinous animals and eight SNPs (Fst > 0.05) associated with horn size in the SHE and scurred horn groups, which may serve as potential molecular markers for genetic selection to reduce horns. Specific amino acid sites (n = 15) that may be significant in the evolutionary development of the ruminant horn were identified in the ALOX12B protein. Additionally, ALOX12B was found to be highly expressed in the testis, suggesting a potential role in reproductive traits. These findings provide valuable insights for improving sheep, particularly for controlling horn size and improving reproductive traits.