British Poultry Science, 2015 Vol. 56, No. 2, 175–183, http://dx.doi.org/10.1080/00071668.2015.1008994

Variation in the chicken LPIN2 gene and association with performance traits Y. HUANG, C. ZHANG, W. ZHANG, P. ZHANG, X. KANG,

AND

W. CHEN

College of Livestock Husbandry and Veterinary Engineering, Henan Agricultural University, Zhengzhou, Henan, P. R. China

Abstract 1. The objective of the study was to investigate the distribution of LPIN2 variants and haplotypes among breeds and perform an association analysis of the variants and haplotypes with the broiler traits in chickens. 2. Six breeds were used to study the variation and distribution of chicken LPIN2, and an F2 resource population was used to measure growth traits, carcass traits, meat quality traits and serum biochemistry parameters. 3. A c.-599G>A variant was located in the promoter region of LPIN2 and c.444G>A and c.1730A>T (E577D) coding variant mutations were detected. Linkage disequilibrium tests showed that these three variants were under moderate linkage disequilibrium in the 6 breeds and 7 haplotypes were constructed. The distribution of variation/haplotypes presented clear differences among breeds. 4. Association analysis showed that c.-599G>A was associated with leg muscle weight, jejunum length, ileum length, leg muscle fibre density and leg muscle fibre diameter; c.444G>A was associated with spleen weight, ileum length, body weight at hatch and metatarsus length at 8 weeks; c.1730T>A had significant effects on chicken liver weight, heart weight, body weight at 10 weeks, serum albumin and glucose. 5. Diplotypes were significantly associated with body weight at hatch, heart weight, pancreas weight, duodenum length, leg muscle fibre density and lactate dehydrogenase.

INTRODUCTION LPIN2 is a member of the Lipin family, which was first identified as having 60% similarity with Lpin1 in mouse (Peterfy et al., 2001). Lipin proteins exhibit especially high sequence conservation in the N- and C-terminals, known as the N-LIP and C-LIP domains, respectively (Han et al., 2006; Donkor et al., 2007; Harris et al., 2007; Liu et al., 2010; Wang et al., 2012), and possess Mg2+-dependent phosphatidate phosphatase (PAP) activity (Donkor et al., 2007; Gropler et al., 2009; Harris and Finck, 2011). Functional analysis in mouse revealed that Lpin2 plays an important role as a hepatic PAP-1 enzyme. Mouse Lpin2 protein levels can be up-regulated by fasting, obesity and consumption of a high-fat diet, suggesting a potential role in the metabolism of fatty acids (Dwyer et al., 2012; Gropler et al., 2009; Ryu et al., 2011).

Research suggests that LPIN2 could play an important role in fat metabolism (Grimsey et al., 2008; He et al., 2009; Reue, 2009). LPIN2 was expressed in 3T3-L1 preadipocytes along with LPIN1 and diminishes with differentiation (Grimsey et al., 2008). However, Lpin2 could not compensate for Lpin1 function in adipose tissue of fatty liver dystrophy mice, which suggested that Lpin2 provides a different regulatory function during adipogenesis (Reue and Dwyer, 2009). He et al. (2009) also found a significant association of variation in pig LPIN2 with back-fat thickness. Some reports showed that LPIN2 has other important functions that are not reflected from tissue expression patterns (Ferguson et al., 2005; AL-MOSAWI et al., 2007; Milhavet et al., 2008). Three LPIN2 mutations in human could result in Majeed syndrome, a rare recessive Mendelian disease characterised by recurrent episodes of fever and inflammation in bone and skin and

Correspondence to: Wen Chen, College of Animal Science and Veterinary Medicine, Henan Agricultural University, Zhengzhou, Henan 450002, P. R. China. E-mail: [email protected] Accepted for publication 26 November 2014.

© 2015 British Poultry Science Ltd

176

Y. HUANG ET AL.

congenital dyserythropoieticanemia (Ferguson et al., 2005; AL-MOSAWI et al., 2007). LPIN2 mutations were also associated with psoriasis, a skin inflammatory disease in human (Milhavet et al., 2008). These reports show that LPIN2 is an anti-inflammatory enzyme that controls triacylglycerol (TAG) synthesis, JNK/AP-1 pathway activation and ultimately the up-regulation of proinflammatory genes (Valdearcos et al., 2012). To date, there were no reports regarding the effects of variation in chicken LPIN2. Based on the LPIN2 (JN012098.2) sequence, the distribution of three variation sites (one promoter variant and two exonic variants) among six breeds was determined, and the associations of LPIN2 variants with broiler traits in a Gushi (GS) F2 resource population were analysed.

MATERIALS AND METHODS Breeds Six breeds were used to study the variation and distribution of chicken LPIN2. These were White Plymouth Rock (WP), White Leghorn (WL), Henan Game (HG), Gushi (GS), Lushi (LG) and Silkie (SK). WP and WL are foreign breeds, WP is a fast-growing broiler and WL is a high-yield layer breed. HG, GS, LG and SK are Chinese native breeds. HG is a fighting chicken in Henan; GS is a Henan native breed, with yellow skin, yellow plumage and green shank characteristics; LG is a Henan native breed laying green shell eggs; SK is an officially recognised breed with black skin, feather-crest and polydactyl traits. Based on Trask’s suggested sample size for ensuring the discovery of both common and rare polymorphisms in a population (Trask et al., 2011), 10 individuals were randomly selected (as unrelated as possible) from each breed. Anticoagulant (heparin sodium) blood was collected from a wing vein and kept at −20 °C. Construction of GS F2 resource populations The GS F2 resource population was constructed with a cross of Anka (fast-growing broiler breed) and GS chickens (slow-growing Henan native breed) as previously described (Han et al., 2011; Li et al., 2013). Briefly, the F2 resource population was generated with GS chickens (24 hens and two cocks) and Anka broilers (12 hens and 4 cocks). To construct the F2 resource population, 9 F1 hens were selected from each of 7 families (6 unrelated rooster families and one half sib); the 63 F1 hens were mated with 7 F1 cocks from 7 families. Over two hatches at two-week intervals, the resource population was established. It consisted of 42 grandparents, 70 F1 parents and 860 F2 individuals. F2 chickens were cage-raised with

the same maize–soya bean diet, which contained 11.90 MJ/kg of ME and 190 g/kg CP from 0 to 8 weeks old and 12.13 MJ/kg of ME and 170 g/kg of CP after 8 weeks old. The following growth traits were measured: body weight (BW, at 2-week intervals), metatarsus length (ML, at 4-week interval) and metatarsus girth (MG, at 4-week intervals). All animals received humane care as outlined in the Guide for the Care and Use of Agricultural Animals in Research and Teaching. F2 chickens were slaughtered at 84-d old. Anticoagulation blood (3 ml) was collected from chicken heart for DNA extraction by a phenolchloroform method. Non-anticoagulant blood (3 ml) was collected from chicken heart for serum separation. Carcass traits were measured and leg muscle and breast muscle were collected for the analysis of the meat quality traits. The serum samples were separated by centrifugation at 3000 g for 15 min at 4 °C and frozen at −80 °C. The serum biochemical parameters were measured by an automatic biochemistry analyser (Hitachi 747, Tokyo, Japan). The measurement of these traits is detailed in Han et al. (2011, 2012) and Li et al. (2013). Analysed growth traits included BW, ML and MG at different ages. Analysed carcass traits include carcass weight, semi-eviscerated weight (the weight of evisceration plus heart, liver, kidney, gizzard, proventriculus, abdominal fat), eviscerated weight, subcutaneous fat thickness, abdominal fat weight, breast muscle weight, leg muscle weight (LMG), heart weight, liver weight (LVW), gizzard weight, spleen weight (SPW), pancreas weight (PW), metatarsus and claw weight, duodenum length (DL), jejunum length (JL), ileum length (IL) and caecal length (CL); Analysed meat quality traits included breast muscle water loss rate, leg muscle water loss rate, breast muscle fibre density, leg muscle fibre density (LFD), breast muscle fibre diameter and leg muscle fibre diameter (LFA); serum biochemistry parameters include total cholesterol, high-density lipoprotein cholesterol, low-density lipoprotein cholesterol, TAG, total protein, albumin (ALB), lactate dehydrogenase (LDH), amylase (AMY), globulin and glucose (GLU). The determination and genotyping of three variation sites The variation sites locating in the promoter region of chicken LPIN2 (NW_003763686.1) were identified by directly PCR-sequencing of LP1 amplicons in GS, WL, WP and LS chickens (one individual per breed). The Created Restriction Fragment Length Polymorphism PCR (Created-PCR-RFLP) method was used to genotype the c.-599G>A (rs16082507) variant (with LP2 primer set and

LPIN2 VARIANTS AND BROILER TRAIT ASSOCIATIONS

XagI enzyme) in different breeds and in the F2 resource population. In the 3ʹ end of the forward primer, a point mismatch was introduced to create a XagI restriction site (CCTNN^NNN A GG, the framed “A” is the variation site and the underlined “C” is the mismatch base) for carriers of the c.599A allele, while the non-carriers of the c.-599A allele lacks this restriction site. The c.444G>A (rs316031739) and c.1730A>T (rs318147579) mutations are exonic variations, which were identified by cDNA cloning of chicken LPIN2 (Zhang et al., 2014). The c.444A→G change forms a natural MspI restriction site, while the c.1730A→T change forms a natural MboI restriction site. So, the c.444G>A (with LP3 primer set and MspI enzyme) and c.1730A>T (with LP4 primer set and MboI enzyme) variation sites were genotyped by the PCR-RFLP method in different breeds and in the GS F2 resource population. The related primers are presented in Table 1. PCR conditions for LP1-4 primers were as follows: the initial denaturation step was for 5 min at 94°C, followed by 30 cycles of 30 sec at 94°C, 30 sec at 53–61°C (Table 1), 40 sec at 72°C; and a final extension step of 7 min at 72°C. LP2-3 PCR products were enzyme-digested and separated by 2.0– 3.0% agarose gel. Representative individuals with different enzyme-digested genotypes in each locus were sequenced to confirm the base variation. Analysis software TFSEARCH (http://www.cbrc.jp/research/db/ TFSEARCH.html) was used to predict the effect of promoter variation on transcription factor binding site (Heinemeyer et al., 1998). ESE finder program (Version 3.0, http://rulai.cshl.edu/cgi-bin/tools/ ESE3/esefinder.cgi?process=home) was used to predict Exonic Splicing Enhancers (ESE) (Smith et al., 2006). Shesis (http://analysis.bio-x.cn/ myAnalysis.php) was used to conduct the Linkage disequilibrium tests (Yong and He, 2005). Phase program 2.1 was used to reconstruct haplotypes with three variation sites in breeds and the F2 resource population, respectively (Stephens and Donnelly, 2003). Statistical analysis Data were analysed using SAS 6.0 software as described (Han et al., 2012; Li et al., 2013). Table 1. Name LP1 LP2 LP3 LP4 1

General linear model I with fixed effects was used to analyse the relationship between the variants and the growth, carcass, meat quality traits and biochemical parameters of the GS F2 resource population. Linear Model II was applied to carcass traits, where body weight before slaughter (BW12) was taken as a covariate to analyse the effects of sequence variation on carcass traits. yijklm ¼ μ þ Gi þ Sj þ Hk þ Fl þ eijklm

(Model I)

yijklm ¼ μ þ Gi þ Sj þ Hk þ Fl þ bðWijklm  W Þ þ eijklm

(Model II)

Where yijklm is the observed value of the traits measured on the mth animal; µ is the overall population mean; Gi is the genotype effect (i = 1, 3), Sj is the sex effect (j = 1, 2); Hk is hatch effect (k = 1, 2); Fl is the family effect (l = 1, 7); b is the regression coefficient for BW12; Wijklm is the individual BW12; W is the average BW12; eijklm is the random error. Data were analysed with the general linear model in two steps: first, a full animal model was analysed and then a reduced animal model was used to exclude the non-significant effects (P > 0.05). The Bonferroni adjustment method was used to compute P values for pairwise comparisons; P ≤ 0.05 was considered significant. According to the method of Liu (Liu, 1998), both additive and dominance effects were estimated using a regression procedure, where the additive effect was estimated as −1, 0 and 1 for GG (AA, AA), GA (AT, GA) and AA (GG, TT), respectively; and the dominance effect represented as 1, −1 and 1 for GG (AA, AA), GA (GA, TA) and AA (GG, TT) for c.-599G>A, c.444G>A and c.1730A>T, respectively.

RESULTS Detection of polymorphic sites The c.-599G>A (reference to JN012098) mutation was located in the predicted promoter region of chicken LPIN2, which was detected with the LP1 primer set in 4 breeds. TFSEARCH predicted that a c.-599G→A change could lead to the formation of a deltaE transcription factor binding site. DeltaE is also called YY1, which is involved in the repression and activation of many promoters. The c.444G>A and c.1730T>A mutations (reference to

Primers and endonucleases and assay conditions1

Primer sequence (5′–3′) F: F: F: F:

177

AGAAGAGATGGGCAGGGCAG R: TTACAGTTGGTTCCTCGTGAC GACACAATGAGAGACGC C TA R: GGTTCCTCGTGACGGAGAAA ATAGAATCATGTGCTCCAAAG R: TTACCTAAGGGGCTGAACACT GCTTAGAAGTTCACTGCTCC R: CCCAACAGCAATAAAACCAC

Region

Length (bp)

Tm (°C)

Endonuclease

Promoter Promoter Exon4 Exon12

342 195 386 319

61 57 53 54

– XagI MspI MboI

The primers were designed based on the chicken LPIN2 sequence (NW_003763686.1 and JN012098.2); the mismatch base in the LP2 primer set is boxed.

178

Y. HUANG ET AL.

had low frequency in HG and WP chickens, with high frequency in SK chickens; while the c.1730T allele was the minor allele (frequency = 0.1) in GS chickens and the major allele (frequency = 0.9) in HG chickens (Table 2). Linkage disequilibrium tests showed that c.444G>A with c.1730T>A (D’ = 0.505, R2 = 0.189), c.-599G>A with c.444G>A (D’ = 0.334, R2 = 0.023), and c.-599G>A with c.1730T>A (D’ = 0.319, R2 = 0.006) were in moderate linkage disequilibrium. The three variants constitute 7 haplotypes among breeds (Table 3). H1, H2 and H3 were the major haplotypes, shared by at least 5 breeds and accounting for 87.5% of the observations. While H4, H5, H6 and H7 haplotypes were the rare haplotypes shared by 1–2 breeds (with a frequency < 0.05), H4 is a specific haplotype for SK chickens. HG chickens only contain two haplotypes, and H3 is the predominant haplotype (Table 3).

JN012098.2) were also confirmed in the breeds and the GS F2 resource population. The c.1730T>A mutation was predicted to cause an E577D (reference to AEJ89532.1) amino acid change. When the effect of this change was analysed using the SIFT algorithm, this mutation was predicted to be tolerated in terms of protein structure/function. The c.444G>A was a synonymous mutation in exon 4, ESE finder predicted that a c.444G>A variation could lead to the formation of a SRSF2 protein binding site (which may result in a change of exon splice site). The c.444G→A and c.1730T→A variations lead to the loss of MspI and MboI restriction enzyme sites, respectively. The naming of genotypes for three variants The Created PCR-RFLP genotypes for the c.-599G>A variation were denoted as AA (195 bp), GG (174 bp and 21 bp) and AG (195 bp, 174 bp and 21 bp, with the 21-bp band being too short to be detected), respectively. For the c.444G>A variation, PCR-RFLP genotypes were named as AA (386 bp), GG (237 bp and 149 bp) and AG (386 bp, 237 bp and 149 bp), respectively; while PCR-RFLP genotypes for the c.1730T>A variation were denoted as AA (319 bp), TT (204 bp and 115 bp) and TA (319 bp, 204 bp and 115 bp), respectively.

The distribution of variants and haplotypes in the GS F2 resource population The distribution of variants and haplotypes in the GS F2 resource population is presented in Tables 4–7. The frequency of c.-599G, c.444G and c.1730A alleles was 0.31, 0.34 and 0.57 in the GS F2 resource population. Linkage disequilibrium tests in F2 individuals showed that c.444G>A with c.1730T>A (D’ = 0.833, R2 = 0.267) was in strong linkage disequilibrium, c.-599G>A with c.444G>A (D’ = 0.342, R2 = 0.096) was in moderate linkage disequilibrium, while c.-599G>A with c.1730T>A (D’ = 0.026, R2 = 0.000) were in weak linkage disequilibrium. Seven haplotypes were reconstituted with three variants in the F2 resource population (H4 was not detected). H1, H2, H3 and H6 were the major haplotypes, which accounted for 88% of the observations in the F2 population. A total of 19 diplotypes were obtained based on 7

The distribution of variation/haplotypes among breeds The distribution of the three variation sites among breeds are presented in Table 2. For the c.-599G>A variation, the c.-599A allele is the major allele (frequency = 0.89) in the breeds, while the c.-599G allele was only detected from heterozygous individuals, and was not found in HG chickens (Table 2). The distribution of the c.444G>A and c.1730A>T variants presented a clear difference among breeds. The c.444G allele Table 2. Variation sites

The genotypic and allelic distributions of chicken LPIN2 variants among breeds

Location

c.-599G>A(rs16082507)

Promoter

c.444G>A(rs316031739)

Exon4

c.1730A>T(rs318147579)

Exon13

1

Genotype/allele GG GA AA G GG GA AA G AA TA TT T

GS1 2

0(0) 0.40(4) 0.60(6) 0.203 0.2(2) 0.5(5) 0.3(3) 0.45 0.90(9) 0(0) 0.10(1) 0.10

HG

LG

WL

WP

SK

Total

0(0) 0(0) 1.00(10) 0 0(0) 0.10(1) 0.90(9) 0.05 0(0) 0.20(2) 0.80(8) 0.90

0(0) 0.20(2) 0.80(8) 0.10 0.20(2) 0.40(4) 0.40(4) 0.40 0.30(3) 0.40(4) 0.30(3) 0.50

0(0) 0.30(3) 0.70(7) 0.15 0.40(4) 0.10(1) 0.50(5) 0.45 0.50(5) 0.20(2) 0.30(3) 0.40

0(0) 0.20(2) 0.80(8) 0. 10 0(0) 020(2) 0.80(8) 0.10 0.20(2) 0.60(6) 0.20(2) 0.50

0(0) 0.10 (1) 0.90 (9) 0.05 0.60(6) 0.40(4) 0(0) 0.80 0.60 (6) 0.30 (3) 0.10 (1) 0.25

0(0) 0.20(12) 0.80(48) 0.11 0.23(14) 0.28(17) 0.43(29) 0.37 0.42(25) 0.27(17) 0.30(18) 0.44

Abbreviations: GS, Gushi; HG, Henan game; LG, Lushi green-shell; WL, White Leghorn; WP, White Plymouth Rock;SK, Silkie. Genotypic frequency (number). 3 Three variants were bi-allelic, only the frequency of one allele was presented. 2

LPIN2 VARIANTS AND BROILER TRAIT ASSOCIATIONS

179

Table 3. The distributions of haplotypes constructed by three chicken LPIN2 variants among breeds Haplotype1

Name H1 H2 H3 H4 H5 H6 H7 H8 Haplotypes

GS2 3

AAA AGA AAT AGT GAA GGA GAT GGT

0.25(5) 0.45(9) 0.10(2) 0 0.20(4) 0 0 0 4

HG

LG

WL

WP

SK

Total

0 0.05(1) 0.95(19) 0 0 0 0 0 2

0.10(2) 0.40(8) 0.40(8) 0 0 0 0.10(2) 0 4

0.15(3) 0.35(7) 0.35(7) 0 0 0.10(2) 0.05(1) 0 5

0.30(6) 0.10(2) 0.50(10) 0 0.10(2) 0 0 0 4

0.05(1) 0.60(12) 0.15(3) 0.15(3) 0 0.05(1) 0 0 5

0.14(17) 0.33(39) 0.41(49) 0.03(3) 0.05(6) 0.03(3) 0.03(3) 0 7

The order for the haplotype: c.−599G>A – c.444G>A – c.1730A>T. Abbreviations: GS, Gushi; HG, Henan Game; LG, Lushi green shell; WL, White Leghorn; WP, White Plymouth Rock; SK, Silkie. 3 Haplotype frequency (number). 1 2

Table 4. The frequency of chicken LPIN2 haplotypes and diplotypes in the GS F2 resource population Haplotype 1

Diplotype

Name

Type

Number (%)

Type

Number (%)

Type

Number (%)

H1 H2 H3 H5 H6 H7 H8

AAA AGA AAT GAA GGA GAT GGT

365(24) 219(14) 491(32) 15(1) 271(18) 139(9) 26(2)

H1H1 H1H3 H1H6 H2H2 H2H6 H3H3 H3H7 H6H6 H6H8 H7H5

45(5.9) 89(11.7) 70(9.2) 18(2.4) 36(4.7) 51(6.7) 62(8.1) 10(1.3) 4(0.5) 2(0.3)

H1H2 H1H5 H1H7 H2H3 H2H8 H3H6 H3H8 H6H7 H7H7

45(5.9) 13(1.7) 58(7.6) 95(12.5) 7(0.9) 128(16.8) 15(2.0) 13(1.7) 2(0.3)

Total

1526(100)

763(100)

The base order: c.−599G>A – c.444G>A – c.1730A>T.

1

Table 5.

The association of chicken LPIN2 c.-599G>A variation with broiler traits in the GS F2 resource population Genotype2

Traits1

LMG (g) JL (cm) IL (cm) LFD (fibres/mm2) LFA (μm) SPW (g)

Genetic effects3

GG (n = 31)

GA (n = 446)

AA (n = 353)

RMSE

P

a

d

RMSE

193 48.3a 47.1a 1223b 34.7ab 3.33

199 51.8b 50.4b 953a 35.9a 2.94

197 52.0b 50.7b 910a 37.9b 2.92

15.0 7.1 6.7 286 6.5 0.96

0.013 0.028 0.021 A variation

The association of the c.-599G>A variation with the related traits in the GS F2 resource population are presented in Table 5. It shows that

c.-599G>A had a significant effect on LMG (P < 0.05), JL (P < 0.05), IL (P < 0.05), LFD (P < 0.01) and LFA (P < 0.01) traits, with a suggested significant effect on the SPW (P < 0.1) trait. GG genotypic values were lower than that of GA and AA genotypes for JL and IL traits, and higher than that of GA and AA genotypes for the LFD trait, while AA genotypic value was higher than that of GG and GA genotypes for the LFA trait (after Bonferroni adjustment). The additive effects for LMG (P < 0.05), JL (P < 0.05), IL (P < 0.05) and LFD (P < 0.01) were significant,

180

Y. HUANG ET AL.

Table 6.

The association of chicken LPIN2 c.444G>A variation with broiler traits in the GS F2 resource population Genotype2

Traits1

Genetic effects3

GG (n = 76)

GA (n = 370)

AA (n = 327)

RMSE

P

a

d

RMSE

3.16b 48.3a 31.6b 3.42ab 24.0 50.0 324

2.98b 50.1ab 30.6a 3.40a 24.8 51.7 317

2.81a 51.1b 30.5a 3.44b 25.0 52.1 324

0.93 6.6 2.5 0.21 3.3 7.2 41

0.007 0.009 0.003 0.036 0.085 0.094 0.061

0.06 −1.88** 0.96** 0.00

0.01 −0.43 0.23 0.02

1.02 8.90 2.81 0.27

SPW (g) IL (cm) BW0 (g) ML8 (cm) DL (cm) JL (cm) BW4 (g) 1

Abbreviations: SPW, spleen weight; IL, ileum length; DL, duodenum length; JL, jejunum length; BW0, body weight at 0 w; BW4, body weight at 4 w; ML8, metatarsus length at 8 w; RMSE, root mean square error. 2 Within a row among three genotypes, means without a common superscript differ (P < 0.05) after Bonferroni adjustment. 3 a, the additive effect; d, the dominant effect; * P < 0.05; ** P < 0.01.

Table 7.

The association of chicken LPIN2 c.1730T>A variation with broiler traits in the GS F2 resource population Genotype2

Traits1

LVW (g) HW (g) IL (cm) BW10 (g) ALB (mmol/L) GLU (mmol/L) DL (cm) CL (cm) BW12 (g) MG8 (cm)

Genetic effects3

AA (n = 256)

TA (n = 447)

TT (n = 134)

RMSE

P

a

d

RMSE

28.1a 6.4a 49.4a 1088a 16.44a 8.94b 24.4 15.7 1331 3.39

28.9b 6.7c 50.4a 1119b 16.57a 8.85b 25.0 16.1 1356 3.42

29.1ab 6.9c 51.9c 1126ab 17.38c 8.00a 25.1 16.4 1371 3.45

3.6 0.9 6.7 147 2.11 3.33 3.3 2.8 180 0.21

0.017 0.001 0.004 0.019 0.000 0.037 0.087 0.081 0.099 0.073

0.57* 0.23** 1.12* 11.59 0.49** −0.32

−0.22 −0.02 −0.14 −1.03 0.18* −0.13

4.92 1.28 8.7 184 2.15 3.46

1

Abbreviations: LVW, liver weight; HW, heart weight; DL,duodenum length; IL, ileum length;CL, cecal length; BW10, body weight at 10 w; BW12, body weight at 12 w; MG8, metatarsus girth at 8 w; ALB, Albumin; GLU, Glucose; RMSE, root mean square error for genotype. 2 Within a row among three genotypes, means without a common superscript differ (P < 0.05) after Bonferroni adjustment. 3 a, the additive effect; d, the dominant effect; * P < 0.05; ** P < 0.01.

and the dominant effect for the LFD trait (P < 0.05) was significant (Table 5). The c.444G>A variation

The information for the association of the c.599G>A variation with the related traits in the GS F2 resource population are presented in Table 6. It shows that the c.444G>A variation had a significant effect on SPW (P < 0.01), IL (P < 0.01), BW0 (P < 0.01) and ML8 (P < 0.05), with a suggested significant effect on DL (P < 0.1), JL (P < 0.1) and BW4 (P < 0.1) traits. AA genotypic values were lower than that of GG and GA genotypes for the SPW trait, and higher than that of the GG genotype for the IL trait and that of the GA genotype for the ML8 trait. GG genotypic value was higher than that of AA and GA genotypes for the BW0 trait (after Bonferroni adjustment). The additive effects at this locus were significant for IL (P < 0.01) and BW0 (P < 0.01) traits (Table 6).

The c.1730T>A variation

The information for the association of the c.-599G>A variation with the related traits in the GS F2 resource population are presented in Table 7. It shows that the c.1730T>A variation had a significant effect on LVW (P < 0.05), HW (P < 0.01), IL (P < 0.01), BW10 (P < 0.05), ALB (P < 0.01) and GLU (P < 0.05) traits, with a suggested significant effect on DL, CL and MG8 (P < 0.01) traits. AA genotypic values were lower than that of the TA genotype for LVW and BW10 traits, and lower than that of the TT and TA genotypes for the HG trait, while TT genotypic values were higher than that of AA and TA genotypes for IL and ALB traits, and lower than that of AA and TA genotypes for the GLU trait (after Bonferroni adjustment). The additive effects at this locus were significant for LVW (P < 0.05), HW (P < 0.01), IL (P < 0.05) and ALB (P < 0.01) traits. In addition, the dominant effect was significant for the ALB (P < 0.05) trait (Table 7).

LPIN2 VARIANTS AND BROILER TRAIT ASSOCIATIONS

The haplotypic combination

The association of haplotype combinations with the related traits in the GS F2 resource population (Table 8), showed that diplotypes were significantly associated with BW0 (P < 0.01), HW (P < 0.05), PW (P < 0.05), DL (P < 0.01), LFD (P < 0.05), and LDH (P < 0.05) traits. There were suggested associations with SPW (P < 0.1), ALB (P < 0.1) and AMY (P < 0.1) traits (Table 8). The H6H7 diplotype had the biggest value for HW, SPW and LFD traits; H6H6 had the smallest value for BW0 and LDH traits, and the biggest value for PW and ALB traits; H2H2 had the smallest value for DL and ALB traits; H2H6 had the biggest value for BW0 and the smallest value for AMY. In addition, H1H5 had the smallest value for the PW trait, and H1H6 had the smallest value for the HW trait, while H1H2 had the biggest value for the AMY trait and H1H3 had the biggest value for the LDH trait.

DISCUSSION LPIN2 might have a function similar to LPIN1 in fat metabolism (Peterfy et al., 2001) and glucose metabolism (Hariharan et al., 1991; Ryu et al., 2011). It was reported that the human SNP, rs3745012, in LPIN2 was associated with type 2 diabetes, glucose metabolism and fat distribution, and interacted with BMI (body mass index) in determination of type 2 diabetes (Aulchenko et al., 2007). In the present study, chicken LPIN2 variation sites were associated with growth traits including BW4, BW10, BW12, MG8 and ML8 traits, which suggested that LPIN2 variation affects chicken growth and development at multiple growth stages. In addition, the Table 8.

181

association of c.1730T>A with the serum GLU trait suggested that chicken LPIN2 may have a role in glucose metabolism too. Approximately, 90% of nutrient absorption occurs in the small intestine. A longer intestine will increase the intestinal absorption area allowing for the improvement of the absorption ability of nutritional ingredients. Interestingly, three variants in chicken LPIN2 showed a significant association with the IL trait and presented significant (or suggested significant) associations with the length of other intestinal segments (including JL, DL and CL), which provides strong evidence that chicken LPIN2 has an important effect on the development of the small intestine and the absorption of nutritional components. The association of LPIN2 variation with growth traits (including BW4, BW10, BW12, MG8 and ML8) also supports this suggestion. The ileum is the “final section” of the small intestine, which functions primarily to absorb vitamin B12, bile salts, cholesterol and products of digestion that are not processed by the jejunum. LPIN2 was reported to be expressed in the small intestine of chickens (Zhang et al., 2014), and the intestine of mouse (Donkor et al., 2007) and human (Donkor et al., 2007), which suggested that intestinal LPIN2 has a conserved function among species. Donkor et al. suggested that LPIN2 could have a role in the synthesis of membrane phospholipids required for the rapid turnover of the intestinal epithelium (Donkor et al., 2007). It was reported that chicken LPIN1 has a potentially important effect on chicken muscle fibre development (Li et al., 2013). Chicken LPIN1-β is predominantly expressed in muscle tissues, and variants in the 5′ flanking region of chicken LPIN1 were found to be significantly

The association of chicken LPIN2 diplotype with broiler traits in the GS F2 resource population

1

Diplotype HW (g) SPW (g) PW (g) DL (cm) BW0 (g) LFD (fibres/mm2) ALB (mmol/L) LDH (mmol/L) AMY (mmol/L) H1H1 H1H2 H1H3 H1H5 H1H6 H1H7 H2H2 H2H3 H2H6 H3H3 H3H6 H3H7 H3H8 H6H6 H6H7 RMSE P

6.4 6.6 6.4 6.4 6.4 b 6.8 6.5 6.6 6.4 6.8 6.8 6.7 6.7 6.3 7.0 0.9 0.02

2.8 3.3 2.8 2.8 2.8 2.8 3.1 3.0 3.1 2.9 2.9 2.8 3.0 3.3 3.5 0.9 0.10

3.6 3.2 3.4 2.9 3.4 3.3 3.3 3.4 3.4 3.4 3.4 3.3 3.6 3.7 3.2 0.6 0.02

25.5 24.9 25.4 22.8 24.4 24.7 22.5 25.5 24.5 25.1 24.3 24.6 26.6 23.7 25.1 3.3 0.01

30.4 29.9 31.0 30.0 30.5 29.7 31.1 30.4 32.1 30.5 31.0 30.9 32.0 28.5 31.1 2.4 A variation and the haplotype combination showed significant association with chicken LFD and LFA traits, which suggests that LPIN2 has an important effect on muscle fibre development just as does LPIN1. Chicken LPIN1 and LPIN2 may function complementarily. It was hypothesised that LPIN2 plays a role in the regulation of the innate immune system, and a defect in the LPIN2 protein would lead to increased production of proinflammatory signals (Ferguson et al., 2005). The spleen is the largest peripheral immune organ of the chicken and functions in both humoral immunity and cellular immunity. Chicken LPIN2 mutations (c.444G>A and c.-599G>A) were associated with the SPW trait in the current study, suggesting that chicken LPIN2 has an important role in immune function. In conclusion, the distribution of three variants and corresponding haplotypes (c.-599G>A, c.444G>A and c.1730T>A) were examined in different chicken breeds and in a GS F2 resource population. The distribution of these mutations and haplotypes showed clear differences among breeds. LPIN2 showed significant associations with body weight, intestine length, muscle fibre traits and serum GLU traits, which indicates that chicken LPIN2 affects chicken growth traits, carcass traits and serum biochemistry parameters.

ACKNOWLEDGEMENTS Blood samples for breeds were provided by Henan Agricultural University Platform of Domestic Animal Germplasm Resources.

FUNDING Work was supported by the National Natural Science Foundation of China (No. 30771533) and Zhengzhou key laboratory of animal biotechnology.

REFERENCES AL-MOSAWI, Z.S., AL-SAAD, K.K., IJADI-MAGHSOODI, R., EL-SHANTI, H.I. & FERGUSON, P.J. (2007) A splice site mutation confirms the role of LPIN2 in Majeed syndrome. Arthritis and Rheumatism, 56: 960–964. doi:10.1002/art.22431 AULCHENKO, Y.S., PULLEN, J., KLOOSTERMAN, W.P., YAZDANPANAH, M., HOFMAN, A., VAESSEN, N., SNIJDERS, P.J., ZUBAKOV, D., MACKAY, I., OLAVESEN, M., SIDHU, B., SMITH, V.E., CAREY, A., BEREZIKOV, E., UITTERLINDEN, A.G., PLASTERK, R.H., OOSTRA, B. A. & VAN DUIJN, C.M. (2007) LPIN2 is associated with type 2 diabetes, glucose metabolism, and body composition. Diabetes, 56: 3020–3026. doi:10.2337/db07-0338 DONKOR, J., SARIAHMETOGLU, M., DEWALD, J., BRINDLEY, D.N. & REUE, K. (2007) Three mammalian lipins act as phosphatidate phosphatases with distinct tissue expression patterns.

Journal of Biological Chemistry, 282: 3450–3457. doi:10.1074/ jbc.M610745200 DWYER, J.R., DONKOR, J., ZHANG, P., CSAKI, L.S., VERGNES, L., LEE, J. M., DEWALD, J., BRINDLEY, D.N., ATTI, E., TETRADIS, S., YOSHINAGA, Y., DE JONG, P.J., FONG, L.G., YOUNG, S.G. & REUE, K. (2012) Mouse lipin-1 and lipin-2 cooperate to maintain glycerolipid homeostasis in liver and aging cerebellum. Proceedings of the National Academy of Sciences of the United States of America, 109: E2486–95. doi:10.1073/pnas.1205221109 FERGUSON, P.J., CHEN, S., TAYEH, M.K., OCHOA, L., LEAL, S.M., PELET, A., MUNNICH, A., LYONNET, S., MAJEED, H.A. & ELSHANTI, H. (2005) Homozygous mutations in LPIN2 are responsible for the syndrome of chronic recurrent multifocal osteomyelitis and congenital dyserythropoietic anaemia (Majeed syndrome). Journal of Medical Genetics, 42: 551– 557. doi:10.1136/jmg.2005.030759 GRIMSEY, N., HAN, G.-S., O’HARA, L., ROCHFORD, J.J., CARMAN, G. M. & SINIOSSOGLOU, S. (2008) Temporal and spatial regulation of the phosphatidate phosphatases lipin 1 and 2. Journal of Biological Chemistry, 283: 29166–29174. doi:10.1074/jbc.M804278200 GROPLER, M.C., HARRIS, T.E., HALL, A.M., WOLINS, N.E., GROSS, R.W., HAN, X., CHEN, Z. & FINCK, B.N. (2009) Lipin 2 is a liver-enriched phosphatidate phosphohydrolase enzyme that is dynamically regulated by fasting and obesity in mice. Journal of Biological Chemistry, 284: 6763–6772. doi:10.1074/jbc.M807882200 HAN, G.-S., WU, W.-I. & CARMAN, G.M. (2006) The Saccharomyces cerevisiae Lipin homolog is a Mg2+-dependent phosphatidate phosphatase enzyme. Journal of Biological Chemistry, 281: 9210–9218. doi:10.1074/jbc.M600425200 HAN, R.L., LI, Z.J., LI, M.J., LI, J.Q., LAN, X.Y., SUN, G.R., KANG, X.T. & CHEN, H. (2011) Novel 9-bp indel in visfatin gene and its associations with chicken growth. British Poultry Science, 52: 52–57. doi:10.1080/00071668.2010.537310 HAN, R.L., WEI, Y., KANG, X.T., CHEN, H., SUN, G.R., LI, G.X., BAI, Y.C., TIAN, Y.D. & HUANG, Y.Q. (2012) Novel SNPs in the PRDM16 gene and their associations with performance traits in chickens. Molecular Biology Reports, 39: 3153–3160. doi:10.1007/s11033-011-1081-y HARIHARAN, N., KELLEY, D.E. & PERRY, R.P. (1991) Delta, a transcription factor that binds to downstream elements in several polymerase II promoters, is a functionally versatile zinc finger protein. Proceedings of the National Academy of Sciences of the United States of America, 88: 9799–9803. doi:10.1073/pnas.88.21.9799 HARRIS, T.E. & FINCK, B.N. (2011) Dual function lipin proteins and glycerolipid metabolism. Trends in Endocrinology and Metabolism, 22: 226–233. doi:10.1016/j.tem.2011.02.006 HARRIS, T.E., HUFFMAN, T.A., CHI, A., SHABANOWITZ, J., HUNT, D. F., KUMAR, A. & LAWRENCE JR., J.C. (2007) Insulin controls subcellular localization and multisite phosphorylation of the phosphatidic acid phosphatase, lipin 1. Journal of Biological Chemistry, 282: 277–286. doi:10.1074/jbc. M609537200 HE, X.P., XU, X.W. & LIU, B. (2009) Molecular characterization, chromosomal localization and association analysis with back-fat thickness of porcine LPIN2 and LPIN3. Molecular Biology Reports, 36: 1819–1824. doi:10.1007/s11033008-9385-2 HEINEMEYER, T., WINGENDER, E., REUTER, I., HERMJAKOB, H., KEL, A.E., KEL, O., IGNATIEVA, E.V., ANANKO, E.A., PODKOLODNAYA, O.A. & KOLPAKOV, F. (1998) Databases on transcriptional regulation: TRANSFAC, TRRD and COMPEL. Nucleic Acids Research, 26: 362–367. doi:10.1093/nar/26.1.362 LI, S.Y., CHEN, W., KANG, X.T., HAN, R.L., SUN, G.R. & HUANG, Y. Q. (2013) Distinct tissue expression profiles of chicken Lpin1-α/β isoforms and the effect of the variation on muscle fiber traits. Gene, 515: 281–290. doi:10.1016/j. gene.2012.11.075

LPIN2 VARIANTS AND BROILER TRAIT ASSOCIATIONS LIU, B.H. (1998) Statistical Genomics: Linkage, Mapping, and QTL Analysis (New York, USA, CRC Press) LIU, G.H., QU, J., CARMACK, A.E., KIM, H.B., CHEN, C., REN, H., MORRIS, A.J., FINCK, B.N. & HARRIS, T.E. (2010) Lipin proteins form homo- and hetero-oligomers. Biochemical Journal, 432: 65–76. doi:10.1042/BJ20100584 MILHAVET, F., CUISSET, L., HOFFMAN, H.M., SLIM, R., EL-SHANTI, H., AKSENTIJEVICH, I., LESAGE, S., WATERHAM, H., WISE, C., SARRAUSTE DE MENTHIERE, C. & TOUITOU, I. (2008) The infevers autoinflammatory mutation online registry: update with new genes and functions. Human Mutation, 29: 803–808. doi:10.1002/humu.20720 PETERFY, M., PHAN, J., XU, P. & REUE, K. (2001) Lipodystrophy in the fld mouse results from mutation of a new gene encoding a nuclear protein, lipin. Nature Genetics, 27: 121–124. doi:10.1038/83685 REUE, K. (2009) The lipin family: mutations and metabolism. Current Opinion in Lipidology, 20: 165–170. doi:10.1097/ MOL.0b013e32832adee5 REUE, K. & DWYER, J.R. (2009) Lipin proteins and metabolic homeostasis. Journal of Lipid Research, 50 (Suppl): S109–114. doi:10.1194/jlr.R800052-JLR200 RYU, D., SEO, W.-Y., YOON, Y.-S., KIM, Y.-N., KIM, S.S., KIM, H.-J., PARK, T.-S., CHOI, C.S. & KOO, S.-H. (2011) Endoplasmic reticulum stress promotes LIPIN2-dependent hepatic insulin resistance. Diabetes, 60: 1072–1081. doi:10.2337/db10-1046 SMITH, P.J., ZHANG, C., WANG, J., CHEW, S.L., ZHANG, M.Q. & KRAINER, A.R. (2006) An increased specificity score matrix for the prediction of SF2/ASF-specific exonic splicing

183

enhancers. Human Molecular Genetics, 15: 2490–2508. doi:10.1093/hmg/ddl171 STEPHENS, M. & DONNELLY, P. (2003) A comparison of Bayesian methods for haplotype reconstruction from population genotype data. The American Journal of Human Genetics, 73: 1162–1169. doi:10.1086/379378 TRASK, J.A., MALHI, R.S., KANTHASWAMY, S., JOHNSON, J., GARNICA, W. T., MALLADI, V.S. & SMITH, D.G. (2011) The effect of SNP discovery method and sample size on estimation of population genetic data for Chinese and Indian rhesus macaques (Macaca mulatta). Primates, 52: 129–138. doi:10.1007/ s10329-010-0232-4 VALDEARCOS, M., ESQUINAS, E., MEANA, C., PENA, L., GIL-DEGOMEZ, L., BALSINDE, J. & BALBOA, M.A. (2012) Lipin-2 reduces proinflammatory signaling induced by saturated fatty acids in macrophages. Journal of Biological Chemistry, 287: 10894–10904. doi:10.1074/jbc.M112.342915 WANG, X.K., CHEN, W., HUANG, Y.Q., KANG, X.T., WANG, J.P., LI, G. X. & JIANG, R.R. (2012) Identification of the transcript isoforms and expression characteristics for chicken Lpin1. Animal, 6: 1897–1903. doi:10.1017/S1751731112001358 YONG, Y.Y. & HE, L. (2005) SHEsis, a powerful software platform for analyses of linkage disequilibrium, haplotype construction, and genetic association at polymorphism loci. Cell Research, 15: 97–98. doi:10.1038/sj.cr.7290272 ZHANG, C.X., WANG, R.Z., CHEN, W., KANG, X.T., HUANG, Y.Q., WALKER, R. & MO, J. (2014) Gene structure and spatiotemporal expression of chicken LPIN2. Molecular Biology Reports, 41: 4081–4091. doi:10.1007/s11033-014-3278-3

Copyright of British Poultry Science is the property of Taylor & Francis Ltd and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.