Mapping, Tissue Distribution and Polymorphism of Porcine Retinol Binding Protein Genes (RBP5 and RBP7)*

The retinoids (vitamin A and its derivatives) play a critical role in vision, growth, reproduction, cell differentiation and embryonic development. Using the IMpRH panel, porcine cellular retinol binding protein genes 5 and 7 (RBP5 and RBP7) were assigned to porcine chromosomes 5 and 6, respectively. The complete coding sequences (CDS) of the RBP5 and RBP7 genes were amplified using the reverse transcriptase polymerase chain reaction (RT-PCR) method, and the deduced amino acid sequences of both genes were compared to human corresponding proteins. The mRNA distributions of the two genes in adult Wuzhishan pig tissues (lung, skeletal muscle, spleen, heart, stomach, large intestine, lymph node, small intestine, liver, brain, kidney and fat) were examined. A total of nine single nucleotide polymorphisms (SNPs) were identified in two genes. Three of these SNPs were analyzed using the polymerase chain reaction-restriction-fragment length polymorphism (PCR-RFLP) method in Laiwu, Wuzhishan, Guizhou, Bama, Tongcheng, Yorkshire and Landrace pig breeds. Association analysis of genotypes of these SNP loci with economic traits was done in our experimental populations. Significant associations of different genotypes of RBP5-A/G, RBP5-A/G and RPB5-T/C loci with traits including maximum carcass length (LM), minimum carcass length (LN), marbling score (MS), back fat thickness at shoulder (SBF), meat color score (MCS) and hematocrit (HCT) were detected. These SNPs may be useful as genetic markers in genetic improvement for porcine production. (


INTRODUCTION
The vitamin A (retinol) and its active derivatives (retinal and retinoic acid) play an essential role in vision, growth, reproduction, cell differentiation and embryonic development.Retinoic acid (RA) regulates multiple biological processes including cell proliferation and differentiation by modulating the transcription of numerous target genes.This function is carried out by retinoic acid receptor (RAR) and retinoid X receptor (RXR) which as homodimer or heterdimer bind to specific response elements in promoter regions of the target genes (Chambon, 1996;Budhu and Noy, 2002).Because of the chemical instability and quite low solubility in aqueous medium, the retinol and its derivatives (retinal and retinoic acid) must be bound to a specific kind of proteins during the process of absorption, transport and excretion.Those proteins include plasma retinol binding protein (RBP), cellular retinol binding proteins (RBPs) and cellular retinoic acid binding proteins (CRABPs).The RBP5 and RBP7 are members of the RBPs family.
Pork quality and production are essentially affected by the number and size of muscle fibers that are determined by prenatal and postnatal stages, respectively (Tang, 2007).Because the RBPs play a pivotal role in retinol and its derivative's absorption, transport, metabolism and homeostasis, those proteins must influence cell differentiation and development of muscle fiber and adipose.So the RBPs may be candidate genes affecting pork quality, production and fatty deposit.

Chromosome mapping using IMpRH
The INRA-University of Minnesota porcine radiation hybrid (IMpRH) panel (118 clones) was employed to assign porcine RPB5 and RBP7 genes to porcine chromosomes.The contigs assembling of RBP5 and RBP7 genes were performed as described by Wang et al. (2006a).The Primer Premier 5.0 (PremierBiosoft.com)was employed to design gene specific primers.The primer sequences, primer binding regions and expected PCR product sizes for each gene are shown in Table 1.The PCR was performed in 10 μl volume containing 25 ng template DNA, 2.5 mM MgCI 2 , 1×PCR buffer, 0.35 units Taq DNA polymerase (TaKaRa Taq r , TaKaRa Biotechnology, Dalian, China), 50 μΜ of each dNTP, and 0.25 μΜ of each primer.Each PCR also contains two positive control DNA samples of pig and hamster each and a negative control without DNA.All PCRs were carried out in PTC-220 DNA Engine Dyad Cycler (MJ Research).The PCR temperature profile was 5 min at 94°C, followed by 30 cycles of 30 s at 94°C, 30 s annealing (temperatures for each primer set are shown in Table 1) and 30 s at 72°C, and a final extension of 5 min at 72°C.The fragment of each gene was cloned into the pGEM-T Easy vector (Promega Corporation, USA) and sequenced commercially to get an intron sequence of each gene.The PCR products were separated in a 1.5% agarose gel stained with 0.5 μg/ml ethidium bromide.The PCR results were analyzed using the IMpRH mapping tool (http://www.toulouse.inra.fr/lgc/pig/hybrid/htm)and twopoint RH analysis was used for identification of linkage groups with a LOD score threshold of 5.0.

CDS sequences and tissue distribution
Based on the contig and its intron/exon boundary of each gene, specific primers to amplify the complete coding sequences (CDS) of each gene were designed.Twelve different tissues: lung, skeletal muscle, spleen, heart, stomach, lymph code, small intestine, liver, brain, kidney and fat were harvested from one adult Wuzhishan Miniature pig, then frozen in liquid nitrogen and stored at -80 o C. Total RNAs were extracted using the Trizol reagent kit (Life Technologies, Grand Island, NE, USA) and precipitated with ethanol according to the manufacture's protocol.The syntheses of the first-stand cDNA and the reverse transcriptase polymerase chain reaction (RT-PCR) were performed as described by Yu et al. (2004).As amplifying the CDS of RBP5 and RBP7 genes, RT-PCR was carried out using the cDNA template derived from reverse transcription of pig liver and skeletal muscle total RNAs, respectively.The PCR products were purified with TIANGEN Midi Purification Kit [TIANGEN BIOTECH (Beijing) Co., LTD] and cloned into the pGEM-T Easy vector (Promega Corporation, USA).Single clone was randomly selected and sequenced commercially.ORFs of the two genes were identified and the amino acid sequences were deduced with the program Seqman (DNA star, Madison, WI, USA).The PCR was performed as described above except for with 27 cycles.The β-actin was used as an internal control.Fivemicrolitre PCR products were size-separated in a 1.5% agarose gel stained with ethidium bromide to detect the expression profile.

SNPs identification and association analysis
The SNPs analysis was performed in five indigenous Chinese breeds (Laiwu, Wuzhishan, Bama, Guizhou and Tongcheng) and two commercial breeds (Yorkshire and Landrace).The genomic DNA extraction was according the standard phenol-chloroform method.As assembling the contig of each gene, the potential SNPs loci emerged.Using the Primer Premier 5.0 (PremierBiosoft.com),those potential SNPs were detected if there were restriction enzyme sites.The potential SNPs having restriction enzyme sites were assayed using PCR-RFLP method with 24 individuals randomly selected from Tongcheng (8), Yorkshire (8) and Landrace (8) breeds.For the possible SNPs without restriction enzyme site, the PCR products were directly sequenced in 12 individuals that were randomly selected from Tongcheng (3), Wuzhishan (3), Laiwu (3) and Bama (3) breeds.For the confirmed SNPs showing the RFLPs, the method was further employed to detect genotypes in Laiwu, Wuzhishan, Bama, Guizhou, Tongcheng, Yorkshire and Landrace breeds.Ten-microlitre digested products were electrophoresed in a 3% agarose gel for genotyping.
A two-step analysis was performed to assess the association between genotype and trait.Firstly, a preliminary generalized linear model was applied to eliminate system effects including sex, combination (different genotypes) and batch (different slaughter groups) using the SAS software package (SAS Inst.Inc., Cary, NC): Where Y ijkl is the phenotypic value of specific trait, μ the population mean, C i the combination effect, B j the batch effect, S k the sex effect, (CB) ij the interaction effect between combination and batch, (BS) jk the interaction effect between batch and sex and e ijkl the residual error effect for each observation.Secondly, the standardized residual value obtained from the model above was then used to partition the effect of the genotype.Significant difference between genotypes was tested after a Bonferroni correction (Wang et al., 2006b).

Chromosome assignment
Using the IMpRH panel, we mapped the porcine RBP5 being closely linked (two-point-analysis) to the microsatellite marker SW963 (29cR; LOD score 16.78) that is located to q 21 -q 24 of SSC5, thus the most likely chromosomal location for RBP5 is also within this region.The porcine RBP7 is tightly linked (two-point-analysis) to marker SW1355 (25cR; LOD score 13.01) on SSC6, and its deduced cytogenetic position is assigned to 6q 12 -q 21 .The human RBP5 and RBP7 genes were assigned to 12p 13.31 and 1q 36.22 , respectively.Our results are completely in accordance with comparative mapping data available for the porcine and human genomes (Goureau et al., 1996).

CDS of porcine RBP5 and RBP7 genes
Using the primer pairs of CDS PL and CDS PR of each gene (Table 1), 728 bp and 558 bp cDNA fragments of porcine RBP5 and RBP7 genes were amplified by RT-PCR.ORFs were identified and nucleotide sequences of cDNA fragments of porcine RBP5 and RBP7 genes were 85% and 92% identical to human RBP5 (GenBank accession no.NM_031491.1)and RBP7 (GenBank accession no.NM_052960.1)mRNAs, respectively.Analysis of the cDNA sequences of porcine RBP5 and RBP7 genes revealed that: (1) The cDNA fragment of porcine RBP5 contains a 408 bp long ORF that encodes a protein of 135 residues with an estimated molecular mass of 15.8 kDa and an isoelectric point (pI) of 5.76.It contains a 82 bp long 5 'untranslated region (5'-UTR) and a 238 bp long 3'untranslated region (3'-UTR); (2) The porcine RBP7 cDNA fragment consists of a 405 bp long ORF, a 56 bp long 5'-UTR and a 97 bp long 3'-UTR.It is predicted to encode a protein of 134 amino acids with an estimated molecular mass of 15.4 kDa and a pI of 6.59.The deduced amino acid sequence of porcine RBP5 is 88% identical to human RBP5 (GenBank accession no.NP_113679.1).The deduced amino acid sequence of RBP7 is 94% and 92% identical to human RBP7 (GenBank accession no.NP_443192.1)and house mouse (Mus musculus) RBP7 (GenBank accession no.NP_071303.1),respectively.The CDS sequences of porcine RBP5 and RBP7 genes were deposited in the GenBank (GenBank accession nos.EF-208120 and EF-208119, respectively).
Within human RBPs, several amino acid residues (K40, T53, R58, W106 and Q108) in close contact with retinol in holo-RBPs are identical or chemically conserved, with the exception of Gln 108 which is replaced by a His residue in both human RBP5 and RBP7.The side chain of Q108 hydrogen bonds the hydroxyl group of retinol in RPB1and RBP2, and the Q-H replacement at the position might have a functional significance (Folli et al., 2001(Folli et al., , 2002)).The amino acid residues of porcine RBP5 and RPB7 at positions 40, 53, 58 106 and 108 are identical with the human corresponding RBPs (K40, T53, R58, W 106 and H108).

The tissue distribution
The RT-PCR was performed to detect the porcine RBP5 and RBP7 genes' expression patterns in 12 tissues of adult Wzhishan pig.The results are given in Figure 1.The mRNA expression level of porcine RBP5 was the highest in liver followed by spleen, large intestine and lymph node, whereas expressions in lung, heart, little intestine, kidney and fat were weak.This is not similar with human RBP5 (alias CRABIII) gene expression that its mRNA is most abundant in adult kidney and liver tissues, low in spleen lymph node and appendix (Folli et al., 2001).The porcine RBP7 mRNA expression level was slightly high in fat, weak in spleen, lymph node, liver, brain and kidney, and very weak in lung and skeletal muscle.This is similar with house mouse Rbp7 expression pattern that is highly expressed in white adipose tissue and mammary gland, but low in heart, brain and kidney (Conforti et al., 2000).Human RBP7 (alias CRBP IV) mRNA is expressed widely in 61 different tissues with the highest abundance in adult kidney, and the next richest in adult heart, adult transverse colon, fetal heart and fetal spleen (Folli et al., 2002).
In general, porcine RBP5 and RBP7 have different expression patterns.In lung, spleen, large intestine, lymph code and liver tissue, the RBP5 mRNA level is much higher than that of RBP7.The RBP5 expressed in heart, large intestine and small intestine, but no RBP7 mRNA was detected in these tissues.This reflects the functional differences of the two proteins in transport retinol.

SNPs identification and association analysis
The genomic DNA sequences that contain the possible SNP loci were amplified using the primer pairs of SNP-N/N n PL and SNP-N/N n PR shown in Table 1.The SNP loci A/G 63 , T/C intro1-90 and A/G 517 of RBP5 were confirmed by digestion with enzymes Cac8I, Mwo I and Fnu4H I,  4. The SBF and MCS of pigs with genotype RBP5-GA 517 are significant thicker and better than the pigs with genotype AA.The LN and LM of pigs with genotypes RBP5-CC intron1-90 and -TC intron1-90 are significantly longer than that of pigs with genotype TT.At RBP5-A/G 63 , the means of LN and LM of pigs with genotypes GG and GA are significantly higher than the individual's with genotype AA (p<0.01).For the meat quality traits of MCS and MS, pigs with genotypes GG and GA are much better than pigs with genotype AA (p<0.01) at -A/G 517 and -A/G 63 loci.Furthermore, except for the traits of HCT in genotype RBP5-TC intron1-90 and of MS in genotype RBP5-AG 63 all remaining traits of pigs with heterozygous genotypes are always the best among the three genotypes at the three SNP loci.This may be great useful in porcine breeding and production.The RBP5 gene is assigned to SSC 5q 21-q 24 with the closest marker as of SW963 (position 95.5 cM).On this chromosome (position 107-113 cM), three significant QTLs (average back fat, last rib back fat and lumbar back fat, cm) at the 5% chromosome-wise level associated with back fat thickness were identified (Malek et al., 2001), and a significant QTL (position 135.8 cM) affecting the carcass length (cm) was also detected (Lee et al., 2003).These support our findings although further confirmation in a large population is needed.
As porcine RBPs play a profound role in cell differentiation and embryonic development, they probably affect muscle fiber and adipose cell differentiation and development both at prenatal and postnatal stages.This may be the cause of genotypes RBP5-A/G 63 , A/G 517 and -   T/C intron1-90 have significant association with economic traits such as SBP, LM, LN, MS, HCT and MCS.Whether these SNPs have really affected some traits associated with biological process or just linked to other major genes need further investigation.

Figure 1 .
Figure 1.The RT-PCR results of porcine RBP5 and RBP7 genes.The number 1 to 12 represent lung, skeletal muscle, spleen, heart, stomach, large intestine, lymph code, small intestine, liver, brain, kidney and fat respectively.The β-actin was used as an internal control.1

Figure 2 .
Figure 2. Three genotypes of RBP5-A/G 63 in Chinese indigenous Laiwu breed.The genotypes (AA, GA and GG) are shown in the bottom.M: 100 bp ladder.
The values are presented in mean±standard error.SBF = Backfat thickness at the shoulder, MCS = Meat color score, LM = Maximum carcass length.LN = Minimum carcass length, MS = Marbling score, HCT = Hematocrit.Means within a column with different superscripted capital letters differed at p<0.01 and means within a column with different superscripted small letters differed p<0.05.

Table 1 .
Primers used in PCR and RT-PCR

Table 2 .
Results of SNP S assaying of porcine RBP5 and RBP7 genes 1 The enzymes were from EEW ENGLAND Biolabs (Beijing), Ltd. 2 Number of individuals sequenced.

Table 3 .
Allele AA AA AA GA GG GA GA GA GA GA M