Introduction

Lipid deposition in pig is a very complex trait that is likely to be controlled by many genes [1] and is determined by a complex balance between lipogenesis and lipolysis [2–4]. In the balance, fatty acids are intermediate products and must be activated to acyl coenzyme A before participating in most catabolic and anabolic reactions [5]. Through investigation in last decades, it was confirmed that the acyl coenzyme A synthetase (ACS) was the main enzyme to catalyze the activation reactions. ACS family includes multiple isoforms classified by their substrate specificities for fatty acids of varying chain length [6]. In mammals, long-chain acyl-CoA synthetase (ACSL) catalyzes the ATP-dependent acylation of fatty acids into long-chain acyl CoAs, which is the first step in lipid metabolism after fatty acid entry into the cell [7]. The ACSL isoform, an important ACS family member, plays an essential role in both lipid biosynthesis and fatty acid degradation [8], and also plays regulatory roles in numerous reactions, including, for example, protein modification [9], intracellular protein transport [10], protein kinase C activation [11], nuclear thyroid hormone receptor modulation [12], and cell proliferation [13].

There are five cloned isoforms of ACSL: ACSL1, ACSL3, ACSL4, ACSL5, and ACSL6 [9]. These isoforms are expressed not only specially in some certain tissues, but also differently at some certain developing ages. ACSL1, ACSL4, and ACSL5 are all found in liver and adipocytes [14–16], whereas both of ACSL3 and ACSL6 are expressed in brain [17, 18]. ACSL4 expression is most abundant in steroidogenic tissues and ACSL5 in intestine [14, 15]. After birth of rats, just the ACSL1 expression increased by fourfold in heart, whereas the ACSL3 decreased and the other ACSL isoforms remained stable [19]. Furthermore, during 3T3-L1 adipocyte differentiation, the expression of ACSL1 gene increases by about 160-fold, while the expression of other isoforms is unchanged [20]. Transgenic mice with overexpressed ACSL1 in heart increased TAG content [21]. Furthermore, overexpression of ACSL1 mRNA and protein by more than fivefold over controls caused a twofold increase in TAG content in mouse liver [22]. So the ACSL1 overexpression has convincingly been thought to increase TAG accumulation.

The pig ACSL1 gene has been mapped on chromosome 15, near the SW1989 microsatellite [23]. It is split into 20 exons and the resulting cDNA encompass 3,133 bp, of which 2,097 bp correspond to the coding region. However, it is unclear whether the trait of fat accumulation in pigs is relative to the ACSL1 gene expression in some organs. We proposed a hypothesis that the varying phenotype of fat traits in pig might be due to different quantity of the ACSL1 expression or polymorphism of SNPs in ACSL1 gene region. The aim of this study was to investigate the differences of ACSL1 expression and genome polymorphisms in several pig breeds that have different intramuscular fat (IMF) contents, and to provide basic molecular information for the further research on the function of the ACSL1 gene in pig.

Materials and methods

Experimental materials

This experiment covered three breeds of pig: Tibet pig (TP) from Tibet Agricultural and Animal Husbandry College, Diannan small ear pig (DSP) and large white (LW) from Xishuang Banna city, Yunnan province, China. 30 castrated boars (10 each group) were slaughtered when they were 6-month age. Tissue samples were collected from liver, backfat, longissimus dorsi muscle at the last rib. The samples were immediately frozen in liquid nitrogen, and then were stored at −80 °C. Ear samples were collected from populations of TP (n = 67), DSP (n = 54) and LW (n = 56) which was used to detect SNPs in the region of ACSL1 gene.

DNA, RNA extraction and cDNA preparation

Genomic DNA were isolated from ear tissues with the extraction procedure as described [24], dissolved in TE solution and preserved in −20 °C refrigerator.

Total RNA was isolated from the liver, backfat and longissimus dorsi muscle tissues with TRIZOL® Reagent (Invitrogen, San Diego, CA, USA) using the method of the manufacturer’s instructions. The RNA solutions were checked for concentration and purity using an NanoDrop 2000 Biophotometer (Thermo scientific, USA) at 260/280 nm absorbance ratio (range 1.8-2.0 indicates a pure RNA sample) and in a 1 % agarose gel to verify its integrity.

After treatment with DNase I, the total RNA was reverse transcribed to cDNA in a reaction volume of 20 μL containing 2 μg total RNA, 50 μmol oligo-d(T) 15 as a primer and 10 nmol dNTP mix. These mixtures were heated at 70 °C for 5 min and incubated on ice for 2 min. After that, 200 U ImProm-II™ reverse transcriptase (Promega, USA), 40 U RNase inhibitor (Promega Biotech Co., Ltd.) and 4 μL 5× reaction buffer were added to the each mixture and was incubated at 42 °C for 60 min, and then inactivated by heating at 70 °C for 15 min.

Quantitative analysis of ACSL1 mRNA expression

The expression quantity of ACSL1 gene was measured using real-time PCR. Primers were designed using Primer Premier 5.0 software spanning one intron to avoid genomic DNA contamination. ACSL1 (NM_001167629) primers were: 5′-GCA GGC ATT TCT CAT AGC G-3′ and 5′-TCC CTC CCC AGT CTC AGC AT-3′. We selected the housekeeping gene glyceraldehyde-3-phosphate dehydrogenase (GAPDH) as the internal standard [25]. GAPDH primers were 5′-GGT CAC CAG GGC TGC TTT TA-3′ and 5′-CCT TGA CTG TGC CGT GGA AT-3′. Real-time PCR amplification was conducted using Bio-Rad CFX96 System (Bio-Rad, USA). The gene expression quantity was calculated with the method of 2−ΔΔCt (ΔΔCt = ΔCttarget gene − ΔCt housekeeping gene) [26]. A cDNA pool of all samples was used as a calibration.

SNP screening

Primers for SNP identification were designed using Primer Premier 5.0 software allowing the amplification of a region of 5′ flanking region and coding region (from exon 1 to 20). The targeted regions, primer sequence and the amplicon size are shown in Table 1. The PCR products amplified from 10 pigs each group were pooled and sequenced directly to identify SNPs. Chromas Pro and DNAMAN6.0 were used to analyze the sequencing results.

Table 1 Sequence, targets and amplicon size of the primers used for SNP identification of the pig ACSL1 gene
Full size table

SNP genotyping

For SNPs found in sequence alignment, the software of NEBcutter V2.0 (http://tools.neb.com/NEBcutter2/) was used online to search for special restriction enzymes. The HinfI, HhaI, PvuII, Tasl were selected for genotyping loci of 5′ flanking T-1191C, exon 6 G173A, exon 14 C36T and exon 17 T46C respectively. The reaction conditions were shown in Table 2.

Table 2 Restriction enzyme digestion loci and reaction conditions
Full size table

Statistical analysis

The expression levels were analyzed by one-way ANOVA with repeated measures using SAS9.1 software (SAS Inst. Inc., Cary, NC). The results are presented as mean ± standard error. Significant and extreme differences were set at P < 0.05 and P < 0.01, respectively. χ2 test was used to analyze the distribution of genotypes and the differences in alleles frequencies.

Results

ACSL1 mRNA expression level in the three tissues among the three breeds of pig

The ACSL1 mRNA expression level was different among breeds and tissues as shown in Fig. 1. In both DSP and TP, the ACSL1 gene expression was maximal in liver tissue and mid in backfat and minimal in longissimus dorsi muscle, and the differences among the three tissues were extremely significant differences (P < 0.01). But in LW, the expression difference between liver and backfat tissue was not significant (P > 0.05). The ACSL1 expression level in liver tissue of DSP and TP was significantly higher than that of LW (P < 0.01), and there was no significant differences of the gene expression in liver between DSP and TP (P > 0.05). However, the gene expression in backfat tissue of LW was higher than that of DSP and TP (P < 0.05), and no significant difference was seen in backfat of DSP and TP (P > 0.05). The expression level in longissimus dorsi muscle was the highest in DSP, and was higher than in TP and LW (P < 0.05). And the expression was significantly higher in TP than in LW (P < 0.05).

Fig. 1
figure 1

The ACSL1 mRNA expression level in three tissues of the three pig breeds. Note Error bars represent SE. Letters on bars denote the difference of expression level with significantly difference (P < 0.05). Break range on Y axis is 0.1–0.5 being omitted. DSP diannan small ear pig (n = 10). TP Tibet pig (n = 10), LW large white (n = 10)

Full size image

SNP identification

Four SNPs, 5′ flanking region (T-1191C), exon 6 (G173A), exon 14 (C36T), and exon 17 (T46C), were identified by sequencing. The mutations of exon 6 G173A, exon 14 C36T and exon 17 T46C caused no amino acid changes, so they were all synonymous mutations. No restriction enzymes were found to identify the locus of 5′ flanking region T-1191C, so we designed a forward mismatched primer for identify the mutation using the dCAPS Finder 2.0 software (http://helix.wustl.edu/dcaps/dcaps.html). The mismatched primer sequence was 5′-CAA AAA TAT CAT CTC AAC TTA GAG T-3′, which was used for PCR-RFLP analyzing for the T-1191C with the reverse primer of 5′-FR2 (listed in Table 1). PCR product size of the primers was 218 bp, which could be digested by restriction enzyme Hinf I on the loci of T-1191C. The PCR products amplified using the primers ACSL-P5 and ACSL-P12 listed in Table 1 respectively were used to be digested by HhaI and Pvu II to genotype of the locus of exon 6 G173A and exon 14 C36T. Another forward primer was designed for amplification of exon 17 region with the reverse primer of ACSL-P15 listed in Table 1, and the sequence was 5′-ACA GGC TCA CTT CGC AGG TAG AT-3′. And the PCR product (292 bp) was digested by Tasl to detect the genotype of the exon 17 T46C.

SNP genotype frequency

The genotype frequency and allele frequency of each site of ACSL1 gene in TP (n = 67), DSP (n = 54) and LW (n = 56) breeds were shown in Table 3. The results showed that polymorphisms of T-1191C, G173A, and C36T were only identified in TP and DSP populations, whereas polymorphism of T46C was found in the three breed populations. The χ2 test showed that four locus were staying in Hardy–Weinberg equilibrium (P > 0.05) in TP population. But in DSP and LW populations, only the T46C site was staying in Hardy–Weinberg equilibrium (P > 0.05). The C allele frequency of T-1191C site and the T allele frequency of C36T site in TP population were higher than those in DSP (P > 0.05), and the G allele frequency of G173A site in TP population was significantly higher than that in DSP (P < 0.01). In T46C site, the C allele frequencies of TP and DSP populations were extremely significant higher than that of LW (P < 0.01).

Table 3 Genotype frequency and gene frequency of the four SNPs in three pig breeds
Full size table

Discussions

The native breeds such as TP in the Qinghai-Tibet Plateau and DSP in south of Yunnan province of China have good adaptation to the harsh conditions and have superior meat quality [27–29]. The two indigenous pigs have high IMF content, whereas LW, a introduced pig breed, possesses good performances in growth rate and lean ratio, but has a low IMF [29, 30]. The ACSL1 expression level was 1.8- and 2.2-fold higher in livers of TP and DSP respectively, and was 4- and 6-fold higher in longissimus dorsi muscles of TP and TSP respectively, comparing with LW. The data in vitro and in vivo had indicated that the ACSL1 was linked to the storage pathway of lipid metabolism in liver and that it might act to channel fatty acids into triglyceride synthesis rather than into β-oxidation and energy production [22]. Our data suggest that the DSP and TP breeds with higher capacity of adipose deposition need more ACSL1 to synthesize long-chain acyl-coenzyme A for the synthesis of triglyceride, which may be the reason of higher IMF content in TP and DSP breeds than in LW. Otherwise, the expression level in backfat of LW was higher than that in TP and DSP. Those results suggested that the expression of ACSL1 in different tissues may be regulated by different transcription factors. For instance, ACSL1 in liver is regulated by PPARα [31, 32] that mainly expressed in liver [33, 34], whereas it is regulated by PPARγ [35] in adipose tissue [36, 37]. Otherwise, it is also possible that more ACSL1 needed for β-oxidation in backfat of LW than the other breeds. The partitioning of intracellular fatty acids between storage pathways and β-oxidation is also controlled by regulation of the mitochondrial acyl-CoA transporter carnitine palmitoyltransferase-1 [38].

In this study, we identified four SNPs in the region of pig ACSL1 gene. The 5′ flanking region T-1191C, exon 6 G173A and exon 14 C36T sites had not been reported in previous works and the polymorphisms of the three locis were only found in Chinese indigenous pig breeds. For the 5′ flanking T-1191C, the binding sites of transcription factors were predicted by the software CONSITE online (http://asp.ii.uib.no:8090/cgi-bin/CONSITE/consite). The results indicated that the mutation of T → C led sites for combining with SOX17 and broad-complex-1 transcription factors. The SOX17 transcription factor mainly acts as a regulator for sustaining self-renewal of hematopoietic stem cells [39, 40]. The broad-complex-1 is a zinc-finger protein found in insects such as Bombyx mori, Drosophila melanogaster, and it has not been found the homologisation sequences in mammals. No polymorphisms of the exon 6 G173A and exon 14 C36T sites were found in LW and also in Duroc, Landrace, Piétain, LW and Iberian populations [41]. So, we suggested that the mutations of the two sites may be particular to the TP and DSP populations. The polymorphism of exon 17 T46C was detected in the three pig breeds and in other breeds such as Duroc, Landrace [41], but the mutation in exon 17 region was not found in Piétain and Iberian populations [41].

In conclusion, the pig ACSL1 gene expresses differently in different tissues and pig breeds. The acyl coenzyme A activated by ACSL1 can enter any metabolic both lipid biosynthesis and fatty acid degradation [5], the mechanism of higher IMF content in fatty-type pigs may be due to more ACSL1 needed for lipogenesis in liver and longissimus dorsi muscle tissues, and the thinner of backfat thickness in lean-type pigs may be due to more ACSL1 needed for β-oxidization in backfat. We found four mutant loci in the two indigenous pigs, but only one of them was multiple in LW populations. The allele frequencies of the four sites were significantly different in indigenous and introduced pig breeds. Those four mutations may be relative to fat deposition in backfat and IMF. However, further studies will be needed such as correlation analysis between ACSL1 expression level and the traits of backfat thickness and IMF content, genotypes of SNPs and lipid deposition traits.