CIDE gene expression in adipose tissue, liver, and skeletal muscle from obese and lean pigs

Abstract

The expression of the cell death-inducing DNA fragmentation factor α-like effector (CIDE) family including Cidea, Cideb, and Cidec was significantly increased in mouse and human models of obesity. However, there was less information on these genes’ expression in pigs. Here, we hypothesized that different fat accumulation between lean (Duroc×Landrace×Yorkshire gilts, DLY) and obese (Lantang) pigs was attributed to porcine CIDE-modulating lipid metabolism. Our data showed that Cidea and Cidec were expressed at a high level in adipose tissue, and at a relatively high level in skeletal muscle, whereas Cideb was mainly expressed in the liver in both breeds of pig. Lantang pigs had higher white adipose and skeletal muscle Cidea and Cidec mRNA abundance, and hepatic and muscle Cideb mRNA than DLY pigs. Lipid metabolism-related genes including sterol regulatory element binding protein 1c (SREBP-1c), hepatocyte nuclear factor-4α (HNF-4α), peroxisome proliferator-activated receptor γ coactivator-1α (PGC-1α), fatty acid synthase (FASN), diacylglycerol O-acyltransferase 1 (DGAT1), and DGAT2 showed a higher expression level in adipose tissue from obese pigs than in that from lean pigs. Lantang pigs exhibited higher mRNA abundance for liver SREBP-1c, HNF-4α, and PGC-1α, and higher skeletal muscle SREBP-1c, HNF-4α, PGC-1α, and DGAT2 expression, as compared with DLY pigs. However, the perlipin2 mRNA levels in adipose tissues, liver, and skeletal muscle were significantly lower in obese pigs than in their lean counterparts. Furthermore, plasma non-esterified fatty acid (NEFA), glucose, and triacylglycerol (TAG) levels were greater in obese pigs than in lean pigs. Finally, data from correlation analysis further found that CIDE mRNA expression was positively correlated with back fat thickness (BFT), abdominal fat mass (AFM), and the levels of NEFA, TAG, and glucose in the two breeds. Collectively, these data revealed that the porcine CIDEs possibly modulated lipid metabolism and contributed to the development of fat deposition and obesity in Lantang pigs.

1. Introduction

Excessive fat deposition affects animal health and production efficiency, and constitutes a health risk to human consumers. Thus, modulation of fat deposition in adipose tissues of pigs is good for both animals and customers (Jiang et al., 2007). Lantang pig is a native breed (obese-type) of South China, whose carcass contains more fat content than hybrid pigs (Lu et al., 2008; Chen et al., 2010). DLY is the cross breed of three lean-type pigs, Duroc, Landrace, and Yorkshire; the lean percentage of DLY reaches 63%–65% (Lan et al., 2004). Lantang and DLY pigs show an obvious difference in total adipose mass and therefore offer an attractive comparison for studying the mechanism of obesity.

Adipose tissue, liver, and skeletal muscles play important roles in body lipid metabolism in animals (Ahima and Flier, 2000; Hulver et al., 2003; Leonhardt and Langhans, 2004). In pigs, adipose tissue is the central organ for fat synthesis and deposition (O'Hea and Leveille, 1969). Bernlohr et al. (2002) demonstrated that fat accumulation in animals depends on levels of triacylglycerol (TAG) synthesis and storage and levels of lipid mobilization and fatty acid oxidation. Excess fatty acids and glucose are converted into TAG after food intake. TAG, a major energy storage form, is stored in lipid droplets, so that other intracellular organelles can avoid lipotoxicity caused by fatty acid (Girousse and Langin, 2012). The lipid droplet is an important subcellular organelle responsible for lipid storage, and the sizes of unilocular lipid droplets reveal the lipid storage capacity and have a positive association with the development of obesity (Bell et al., 2008).

Changes of lipid homeostasis by over-expression or deletion of specific genes often result in obesity. Genetically modified animal models have highlighted that expression of several adipogenic and lipogenic genes, including fatty acid synthase (FASN), diacylglycerol O-acyltransferase (DGAT), sterol regulatory element binding protein 1c (SREBP-1c), and peroxisome proliferator-activated receptor γ coactivator-1α (PGC-1α), played important roles in fatty acid synthesis, TAG synthesis, and lipid storage (Liu et al., 2008; Malaguarnera et al., 2009). Over the past decade, it has been reported that the cell death-inducing DNA fragmentation factor α-like effector (CIDE) family plays an important role in lipid and fat metabolism (Zhou et al., 2003; Gong et al., 2009; Yonezawa et al., 2011). Previous studies have reported that animals with a deficiency in Cidea, Cideb, or Cidec exhibited a typical lean phenotype with high energy expenditure, high levels of plasma TAG and non-esterified fatty acid (NEFA) reduction (Zhou et al., 2003; Li et al., 2007; Nishino et al., 2008), and altered genes’ expression in various metabolic and signaling networks (Li et al., 2010). These studies suggest that the CIDE family plays an important role in TAG synthesis, lipid storage, and the development of obesity (Gong et al., 2009). However, all the studies on CIDEs have been carried out in mice and humans; there was far less information on the expression of these genes in pigs.

In the present study, we evaluated mRNA abundance of the adipose, liver, and skeletal muscle tissues for CIDEs and several lipid metabolism genes in both genetically obese and lean pigs. Our aim was to explain the relationship between porcine CIDE gene expression and lipid accumulation. We think that this basic molecular information might be useful for further investigation of the functions of CIDEs in pig models.

2. Materials and methods

2.1. Animals and sample collection

Eight castrated male Duroc×Landrace×Yorkshire pigs of similar liveweight ((20.21±0.57) kg) as well as eight Lantang pigs, also castrated males of similar liveweight ((16.12±0.63) kg), were used in this study. All pigs were raised in the pig farm of the animal facilities of the Institute of Animal Science in the Guangdong Academy of Agricultural Sciences, China and were provided feed and watered ad libitum. The diets for the two breeds were made up of different proportions of ingredients because the growth rates of the two breeds were different (Tables 1 and 2). The trials were terminated when the pigs of DLY and Lantang reached 100 and 70 kg body weight, respectively. An ultrasonic instrument (Renco, USA) was used to determine back fat thickness (BFT) at the 1st, 10th, and last ribs before slaughter. Blood sample was obtained from the anterior vena cava using vacuum tubes (using ethylenediaminetetra-acetic acid (EDTA) as anticoagulant) and centrifuged for 5 min at 3000g at 4 °C, and the plasma was separated and stored at −20 °C following slaughter. The left-side carcass was dissected into abdominal fat following the procedure of Walstra and Merkus (1995). Subcutaneous adipose tissue (subcutaneous back fat) and liver and skeletal muscle (longissimus muscle) samples were immediately removed and frozen in liquid nitrogen, and stored at −80 °C.

Table 1.

Composition of experimental diets fed to DLY and Lantang pigs

Group	Corn (%)	Soybean meal (%)	Wheat bran (%)	Fish meal (%)	Rice bran (%)	Zeolite powder (%)	Methionine (%)	Calcium hydrophosphate (%)	Salt (%)	Premix* (%)	Total (%)
DLY	67.38	23	4	0	0	0.96	0	0.6	0.06	4	100
Lantang	56.00	18	15	1	5	0.90	0.1	0	0	4	100

^*Premix provided the following nutrients per kilogram of diet: vitamin A 1300 IU, vitamin D₃ 150 IU, vitamin E 11 mg, vitamin K₃ 0.5 mg, thiamin 1 mg, riboflavin 2 mg, pyridoxine 1 mg, vitamin B₁₂ 6 µg, niacin 7.5 mg, pantothenic acid 7 mg, biotin 0.05 mg, folacin 0.3 mg, choline chloride 300 mg, Fe 50 mg, Zn 50 mg, Mn 2 mg, Cu 3.5 mg, and I 0.14 mg

Table 2.

Nutrient contents of experimental diets fed to DLY and Lantang pigs

Group	Digestive energy (MJ/kg)	Crude protein (%)	Available phosphorus (%)	Lys (%)	Met+cysteine (%)
DLY	13.38	16.0	0.25	0.76	0.58
Lantang	12.04	15.1	0.17	0.72	0.63

2.2. Real-time quantitative PCR analysis

Real-time quantitative polymerase chain reaction (PCR) was done as we described previously (Tian et al., 2016). Briefly total RNA was isolated from subcutaneous adipose and liver tissue samples using TRIzol reagent (Invitrogen Co., Carlsbad, CA, USA) according to the manufacturer’s instructions. All complementary DNAs (cDNAs) were synthesized from 1 μg of total RNA using a reverse transcription kit (TaKaRa, Tokyo, Japan) according to the manufacturer’s recommendations. Then the synthesized cDNA was diluted (1:10, v/v) and real-time quantitative PCR amplification was performed with SYBR green (TaKaRa, Tokyo, Japan) and specific primers for pig messenger RNA (mRNA) sequences (Table 3). Conditions for real-time PCR were an initial denaturation at 95 °C for 180 s, followed by 40 cycles at 95 °C for 15 s and 58 °C for 30 s, with a final elongation at 72 °C for 30 s. Each sample for each gene was amplified three times, in three independent wells, in order to have technical replicates. To normalize expression data, we used multiple internal control genes as described by Vandesompele et al. (2002). The expression stability was evaluated by the M value and pairwise variations of geNorm (Version 3.5; PrimerDesign Ltd., Southampton, Hampshire, UK). We found that β-actin had a lower M value than glyceraldehyde-3-phosphate dehydrogenase (GAPDH), both below 1.5. Thus, β-actin was ranked as the most stably expressed gene. Furthermore, the pairwise variations of β-actin and GAPDH from subcutaneous adipose and liver tissue samples of Lantang and DLY pigs were below the threshold (0.150) that required the inclusion of an additional normalization gene. Therefore, β-actin and GAPDH could be used for normalization. In the present study, the expression of the target genes relative to the β-actin was analyzed by the 2^−ΔΔCt method. ΔCt=Ct (target gene)−Ct (β-actin) and ΔΔCt=ΔCt (Lantang pigs)−ΔCt (DLY pigs), where Ct is cycle threshold.

Table 3.

Primer sequences used in this study

Gene	Sequence (5'→3')	Product size (bp)	GenBank accession
Cidea	Forward: CACCGTGGTAGATACAGAGG	292	NM_001112696.1
	Reverse: GGACAGGAACCGCAACA
Cideb	Forward: TGGGGACTCTGATGCTGAA	284	NM_001112688.1
	Reverse: CCCGTAGAATGTGGCTTTG
Cidec	Forward: CGGTGCCTACTCCCTTTCCT	184	NM_001112689.1
	Reverse: TGGGTCTTTGCCCTTGGT
Perilipin2	Forward: GCTGGCGACATCTACTCA	250	NM_214200.2
	Reverse: AAGTCCACAACAGAACCCTA
DGAT1	Forward: AGGACGGACACGACGAT	287	NM_214051.1
	Reverse: GAACGCAGTCACAGCAAA
DGAT2	Forward: TCCTGTCTTTCCTCGTGC	131	NM_001160080.1
	Reverse: ACCTTTCTTGGGCGTGT
HNF-4α	Forward: ATCGCCACCATCGTCAA	200	NM_001044571.1
	Reverse: CCTCACCCTTTCCACTACCA
SREBP-1c	Forward: AAGCGGACGGCTCACAA	121	NM_214157.1
	Reverse: GCAAGACGGCGGATTTATT
PGC-1α	Forward: TCACCACCCAAATCCTTAT	295	NM_213963.2
	Reverse: ATTCTTCCCTCTTCAGCCT
FASN	Forward: CCTGGGAAGAGTGTAAGCA	108	NM_001099930.1
	Reverse: GGAACTCGGACATAGCG
β-actin	Forward: CATCGTCCACCGCAAAT	210	NC_010445
	Reverse: TGTCACCTTCACCGTTCC

2.3. Analysis of biochemical variables in plasma

Plasma concentrations of high density lipoprotein (HDL), low density lipoprotein (LDL), glucose, cholesterol, NEFA, and TAG were measured using commercial kits purchased from the Nanjing Jiancheng Institute of Bioengineering, China.

2.4. Statistical analysis

Data were presented as mean±standard error of the mean (SEM). Analysis was performed using GraphPad Prism Version 5 (GraphPad Software Inc., San Diego, CA, USA). Significance was predominantly established using a two-tailed Student’s t-test. P<0.05 was considered a statistically significant difference. Correlation analysis between CIDE mRNA expression levels and apparent index (BFT, abdominal fat mass (AFM), NEFA, TAG, and glucose) was calculated using the Pearson’s correlation coefficient of the IBM SPSS Statistics 22 software (IBM Corp., Armonk, NY, USA). P<0.05 and P<0.01 were considered as significant and highly significant correlations, respectively.

3. Results

3.1. Animal performance

As shown in Table 4, Lantang pigs exhibited significantly greater BFT and AFM than DLY pigs.

Table 4.

Growth performance of DLY and Lantang pigs

Group	Initial body weight (kg)	Final body weight (kg)	Average back fat thickness (cm)	Abdominal fat mass (kg)
DLY	20.21±0.57	101.60±2.40	2.10±0.24	0.61±0.06
Lantang	16.12±0.63	68.90±1.83	4.05±0.03*	1.57±0.06*

Data are expressed as mean±SEM, with n=8.

^* P<0.05 vs. DLY pigs

3.2. Expression of CIDEs in adipose, liver, and skeletal muscle tissues

As shown in Fig. 1, porcine Cidea and Cidec mRNAs were highly expressed in white adipose tissue. Both were expressed at a relatively high level in skeletal muscle. However, Cidea was not expressed in liver, and Cidec mRNA was expressed at a much lower level in liver. Cideb mRNA was mainly expressed in porcine liver tissue, at a lower level in adipose tissue, and was not detected in skeletal muscle. Moreover, obese pigs (Lantang) had a significantly higher Cidea and Cidec mRNA levels in adipose and skeletal muscle tissues, and a higher Cideb in adipose and liver than lean breeds (DLY).

Fig. 1.

CIDE gene expression patterns Relative mRNA levels of Cidea, Cideb, and Cidec in adipose tissue, liver, and skeletal muscle of Langtang and DLY pigs. Data were expressed as mean±SEM, with n=8. ^* P<0.05, ^** P<0.01, ^*** P<0.001 vs. DLY pigs

3.3. Expression levels of genes responsible for lipid metabolism

The mRNA levels of PGC-1ɑ and SREBP-1c in adipose tissue from Lantang pigs increased 3.67-fold (P<0.01) and 10-fold (P<0.001), respectively, in comparison with DLY pigs. mRNA abundances for HNF-4α, FASN, DGAT1, and DGAT2 mRNAs were significantly higher in adipose tissue from the Lantang breed than in that from DLY pigs. We observed that Lantang pigs exhibited higher PGC-1ɑ, SREBP-1c, and HNF-4α mRNA expression in liver tissue than the DLY breed (P<0.05, P<0.01, and P<0.05, respectively). FASN, DGAT1, and DGAT2 mRNA expression tended to be greater in liver tissue from obese pigs than in that from lean pigs, although at a non-significant level by the Student’s t-test. In skeletal muscle, the amounts of PGC-1ɑ, SREBP-1c, and DGAT2 mRNAs in Lantang pigs were higher than those in DLY pigs. However, the mRNA levels of perlipin2 in adipose, liver, and skeletal muscle from obese pigs were significantly decreased, as compared with their lean counterparts (Fig. 2).

Fig. 2.

Relative mRNA levels of SREBP-1c, HNF-4α, PGC-1α, FASN, DGAT1, DGAT2, and perlipin2 in adipose tissue, liver, and skeletal muscle from Lantang and DLY pigs Data were expressed as mean±SEM, with n=8. ^* P<0.05, ^** P<0.01, ^*** P<0.001 vs. DLY pigs

3.4. Assessment of biochemical variables in different swine breeds

No difference in the level of HDL, LDL, or cholesterol was observed between Lantang and DLY breeds (Figs. 3d‒3f). However, plasma NEFA and TAG as well as glucose levels in Lantang pigs were significantly higher than those in DLY breeds (Figs. 3a‒3c).

Fig. 3.

Plasma levels of NEFA (a), TAG (b), glucose (c), total cholesterol (d), LDL-cholesterol (e), and HDL-cholesterol (f) in Lantang and DLY pigs

The results show higher plasma levels of NEFA, TAG, and glucose in Lantang pigs than in DLY pigs. Data were expressed as mean±SEM, with n=8. ^* P<0.05, ^** P<0.01, ^*** P<0.001. NS: not statistical significant; NEFA: non-esterified fatty acid; TAG: triacylglycerol; LDL: low density lipoproteins; HDL: high density lipoproteins

3.5. Relationship between CIDE mRNA expression and BFT, AFM, NEFA, TAG, and glucose

The results of correlation analysis between CIDE mRNA expression and BFT, AFM, NEFA, TAG, and glucose are presented in Table 5. Cidea mRNA expression in adipose is positively correlated with BFT, AFM, TAG, and glucose (P<0.01) in the two breeds of pig. Cidea mRNA expression in skeletal muscle had a positively significant correlation with BFT, AFM, and NEFA (P<0.05). The correlations of Cidec mRNA expression level in adipose but not in liver or skeletal muscle, and BFT, AFM, NEFA, TAG, and glucose were positive and highly significant (P<0.01) in the two breeds. Expression of Cideb mRNA in both adipose and liver tissues correlated positively with BFT, AFM, NEFA, TAG, and glucose in two porcine breeds (P<0.01).

Table 5.

Correlation analysis between CIDE mRNA expression and BFT, AFM, NEFA, TAG, and glucose in adipose, liver, and skeletal muscle

Parameter	Cidea mRNA expression		Cideb mRNA expression		Cidec mRNA expression
	Adipose	Muscle	Adipose	Liver	Adipose	Liver	Muscle
BFT	0.788**	0.580*	0.707**	0.751**	0.743**	0.120	0.214
AFM	0.818**	0.570*	0.668**	0.614**	0.668**	0.246	0.170
NEFA	0.460	0.514*	0.636**	0.549**	0.618**	−0.265	0.383
TAG	0.685**	0.488	0.665**	0.659**	0.675**	0.123	0.237
Glucose	0.652**	0.385	0.810**	0.583**	0.582**	0.103	0.067

BFT: back fat thickness; AFM: abdominal fat mass; NEFA: non-esterified fatty acid; TAG: triacylglycerol.

^* P<0.05,

^** P<0.01

4. Discussion

In the present study, we have observed several interesting findings regarding obese Lantang pigs exhibiting higher CIDE gene expression than lean DLY pigs. Some expressions of lipid metabolism genes-related CIDEs also were up-regulated or down-regulated in Lantang pigs, as compared with DLY pigs, and the correlation analysis data found that porcine CIDE mRNA abundances were positively associated with BFT, AFM, NEFA, TAG, and glucose in the two breeds. These observations were considered novel as there were no reports, to our knowledge, comprehensively evaluating the difference of porcine adipose, liver, or skeletal muscle CIDE mRNA expression between obese and lean pigs, or elucidating the relationship between the roles of CIDEs and fat accumulation.

Earlier studies have reported that Cidea was mainly expressed in the heart, and at a lower level in the brain, skeletal muscle, thymus, appendix, lymph nodes, and bone marrow, but neither in normal adult human nor mouse liver tissue (Inohara et al., 1998; Zhou et al., 2003). Our study found that porcine Cidea was highly expressed in white adipose and at a relatively high level in skeletal muscle tissue, but was not detected in liver, although Li et al. (2009) showed that porcine Cidea can be detected in the liver of Tongcheng and Large White pigs. Danesch et al. (1992) demonstrated that Cidec was an adipocyte-specific marker gene. Consistently, we found that porcine Cidec gene mRNA was expressed at a high level in white adipose tissue, at a relatively high level in skeletal muscle, and at a lower level in liver. It has been reported that Cideb mRNA was expressed in many tissues with the highest levels in the liver (Inohara et al., 1998). Our result agreed with the report that porcine Cideb was mainly expressed in the liver, but was not detected in skeletal muscle. The differential tissue distribution patterns of the three CIDE genes can imply that they may have different functions.

Previous studies reported that with CIDE deficiency mice showed a typical lean phenotype and a markedly lower adiposity index. Plasma TAG and NEFA levels, and the size of white adipocytes in CIDE-deficient mice were significantly reduced, compared with those in wild-type mice (Li et al., 2007; Nishino et al., 2008; Toh et al., 2008; Gong et al., 2009). In addition, it has been shown that over-expressed Cidec can increase lipid droplet size and enhance the accumulation of lipids (Keller et al., 2008), whereas deletion of Cidea by RNA interference in human adipocyte stimulated lipolysis (Nordström et al., 2005). In our results, we found that mRNA expressions of Cidea and Cidec in adipose tissue and skeletal muscle, and Cideb in adipose and liver from obese pigs were significantly higher than that from the lean breed. These data indicated that higher mRNA abundance for Cidea and Cidec in adipose tissue and skeletal muscle, and Cideb in liver from Lantang breeds, might play an important role in the development of fat deposition.

The molecular basis for CIDE-modulated lipid metabolism has been elucidated. CIDE deletion in mice resulted in significant reductions of the expression of SREBP-1c and its downstream target genes (ACC1, FASN, Elovl6, and SCD1) (Abu-Elheiga et al., 2003; Li et al., 2007), which were required for fatty acid synthesis (Shimomura et al., 1999) and TAG synthesis (Horton et al., 2002). Similarly, we found that CIDE-enriched Lantang pigs exhibited greater mRNA abundance for SREBP-1c and FASN. Therefore, it has been indicated that CIDEs could positively modulate the expression of SREBP-1c and its downstream target genes, and lead to increased TAG secretion and lipogenesis (Gong et al., 2009). PGC-1α and HNF-4α mRNA expression, which were critical to de novo lipogenesis and gluconeogenesis (Herzig et al., 2001) and controlling the expression of Cidea and Cideb (Hallberg et al., 2008; Chen et al., 2010; Yu et al., 2013), were higher in adipose and liver tissue from obese pigs than in those from their lean counterparts. At the same time, skeletal muscle PGC-1α mRNA abundance from Lantang pigs was increased compared to the DLY pigs. In addition, adipose DGAT1 and DGAT2 mRNA expression, which play a central modulation role in animal fat deposition (Nishizuka, 1992), were higher in Lantang pigs that in DLY pigs. Skeletal muscle DGAT2 mRNA abundance was also increased in DLY pigs. The present data uncovered that the mRNA levels of perilipin2 in adipose, liver, and skeletal muscle tissue from CIDE-enriched Lantang pigs were lower than those from DLY pigs. Our result was consistent with the report by Singaravelu et al. (2013) who demonstrated that Cideb over-expression resulted in a significant down-regulation of perilipin2 protein levels. In addition, a study by Li et al. (2012) indicated that Cideb and perilipin2 played opposite roles in modulating TAG accumulation, with Cideb as a positive regulator and perilipin2 as a negative regulator of lipid droplet size in hepatocytes. Collectively, these results revealed that the CIDE gene family modulated fat deposition through controlling the expression of lipid metabolism-related genes.

CIDE-sufficient Lantang pigs also exhibited higher plasma NEFA, TAG, and glucose levels than DLY pigs. Our results agreed with the reports that CIDE deficiency resulted in a reduction of plasma TAG and fatty acid levels (Zhou et al., 2003; Li et al., 2007; Toh et al., 2008). We also found that Lantang pigs had greater BFT and AFM than DLY pigs (Lu et al., 2008). Therefore, there may be a close association between CIDE mRNA expression and fat deposition. The further correlation analysis results uncovered an interesting finding that Cidea mRNA expression in adipose was positively correlated with BFT, AFM, TAG, and glucose, and in skeletal muscle was positively correlated with BFT, AFM, and NEFA in the two breeds of pigs. Adipose and liver Cideb mRNA abundance and BFT, AFM, TAG, NEFA, and glucose had a positively significant correlation in the two breeds. Cidec mRNA in adipose but not in liver or skeletal muscle was positively correlated with BFT, AFM, TAG, NEFA, and glucose in Lantang and DLY pigs. We made a correlation analysis between porcine CIDE mRNA and fat deposition-related factors. These results further suggested that CIDE genes contributed to fat deposition of the fatty Lantang pigs.

5. Conclusions

In conclusion, our data suggest that the porcine CIDEs possibly modulated lipid metabolism and contributed to the development of fat deposition and obesity.