| Edited by Kiichi Fukui. Shu-Hong Zhao: Corresponding author. E-mail: shzhao@mail.hzau.edu.cn |
Imprinted genes, which are expressed monoallelically depending on their parental origin, play important roles in the mammalian development, growth, and behavior (Kelsey et al., 1998; Isles et al., 2005). Since the close relationship between genomic imprinting and function of placenta during mammalian embryo growth has been found, placenta, as a key tissue, was used to illuminate the evolution of genomic imprinting (Mochizuki et al., 1996; Monk et al., 2006; Wood et al., 2006). The organization of placental connections to the maternal blood supply varies remarkably across species. Human placentation involves significant invasion of endometrial tissue layers. The maternal endometrium is breached by the placenta in the mouse, with less deeply invaded compared to humans. The pig has diffuse epitheliochorial placentation, with no invasion of the endometrium (Zhao et al., 2004). Therefore, investigation of the imprinting status of imprinted genes in different type of placentas is of great interest to understand the possible roles of those genes in placenta function. So far, most genes, which imprinted in placenta, were found in human and mouse (Reik et al., 2003). There are few reports on identification of imprinted genes in the porcine placenta until now. Thus it is necessary to detect genes imprinted in the porcine placenta to offer new data and insights to the comparative genome research as well as the porcine placenta functional genomics.
The gene GATM (L-arginine: glycine amidinotransferase), encoding the first and rate-limiting product in creatine biosynthesis, is also a target of the estrogen receptor (Zhu et al., 2001). Creatine and its phosphorylated form play essential roles in the energy metabolism of muscle (Markus et al., 2000). In 2003, Sandell et al. reported that Gatm is imprinted in the placenta of mouse, and expresses exclusively from the maternal allele (Sandell et al., 2003). However, GATM escapes genomic imprinting in human placenta (Miyamoto et al., 2005). Paternally expressed gene 10 (PEG10), as an endogenous gene, is expressed extensively in some embryonic tissues as well as in the placenta (Ono et al., 2001). It is highly conserved across mammalian species, and is essential for placenta formation. Furthermore, Peg10-deficient mice showed early embryonic lethality (Ono et al., 2006).
In this study, we characterized the porcine GATM gene, identified several cSNPs of GATM and PEG10 genes, investigated allele frequencies in different pig breeds, and analyzed the imprinting status in placentas on days 75 and 90 of gestation.
Placenta samples of 60 piglets from 6 healthy Yorkshire pregnant gilts were collected on days 75 and 90 of gestation (3 sows at each stage). After the sows were necropsied, the uteruses were removed immediately and the placentas were collected, washed briefly with PBS, rapidly frozen in liquid nitrogen, and then stored at –80°C. Genomic DNA and total RNA were extracted from the samples according to standard phenol/chloroform procedure and TRIzol reagent kit, respectively. Total RNAs from heterozygous individuals were treated with DNaseI by TURBO DNA-free kit (Ambion, Austin, TX) and then was used for RT-PCR.
Blast (http://www.ncbi.nlm.nih.gov/blast/) searches were carried out with the cDNA of human GATM for the porcine expressed sequence tags (ESTs). ESTs shared at least 85% similarity to the corresponding human cDNA sequence, were assembled into contig and then used to design the gene specific primers. The RT-PCR method was used to clone the entire GATM coding sequence (CDS). The primer pairs G-1, G-2, G-3, G-4, G-5 and G-6 for RT-PCR were listed in Table 1. Total RNA was extracted from the Meishan placenta on days 90 of gestation using the TRIzol reagent kit. The amino acid sequences were deduced with the program Seqman (DNA Star, Madison, WI, USA).
 View Details | Table 1. Primers designed for amplification of the porcine GATM and PEG10 genes |
Alignment of ESTs revealed several possible cSNPs in the two genes. PCR-RFLP analysis and sequencing of PCR products were used to confirm two SNPs. Primer pairs for SNP amplification were G-4 and P-2 (Table 1). After sequencing confirmation of the PCR products, genetic variation C/T and T/C substitutions were detected by PCR-RFLP using the BseLI and TaqI restriction enzymes. DNA samples of unrelated animals from 7 (for GATM) and 6 (for PEG10) breeds (Table 2, Table 3) were genotyped. A Chi-square test on the allele frequencies was performed using SAS version6.12.
 View Details | Table 2. Genotype and allele frequencies of GATM in seven pig breeds |
 View Details | Table 3. Genotype and allele frequencies of PEG10 in six pig breeds |
The pig × rodent somatic cell hybrid panel (SCHP) and the INRA-University of Minnesota porcine radiation hybrid (IMpRH) panel were employed for localization of GATM and PEG10 genes to the pig genome. Primers were designed from the 3’-untranslated region (G-5 for GATM; P-1 for PEG10) (Table 1). The PCR fragment was sequenced to verify the correct amplification. PCR products were separated in a 2.0% agarose gel stained with 0.5ug/ml ethidium bromide. The results were statistically analyzed by the tools provided at http://www.toulouse.inra.fr/lgc/pig/hybrid.htm (Yerle et al., 1996) and http://www.toulouse.inra.fr/lgc/pig/RH/IMpRH.htm (Milan et al., 2000), respectively.
All 60 samples were used to detect heterozygous animals of GATM and PEG10 cSNPs. For GATM, the primer pair G-4 was used to amplify the genomic DNA and a new primer pair that spans one intron (G-6) was used to amplify the cDNA. The cDNA was separately subjected to 34, 28 and 25 cycles of PCR amplification. For PEG10, the primer pair P-2 was used to amplify the genomic DNA and cDNA. The PCR products were electrophoresed in a 2.0% agarose gel stained with 0.5ug/ml ethidium bromide after digestion by BseLI and TaqI respectively.
A 2210-bp cDNA contig (GenBank: EF612462) was assembled after sequencing PCR products and colonies from Meishan placenta cDNA on days 90 of gestation. The porcine GATM coding region shared 91% and 89% sequence similarity with human (NM_001482), and mouse (NM_025961) GATM. A 1272-bp open reading frame (ORF) was predicted by online software in NCBI. This ORF encoded a polypeptide of 424 amino acids, with a molecular mass of 48359.54 Da and an isoelectric point of 7.839. The porcine GATM amino acid sequence shared 94% similarity with human (NP_001473) and mouse (NP_080237) corresponding sequences. This result was in consistent with previous reports on direct sequencing of human and pig amino acid sequences (Humm et al., 1994).
Statistical evaluation of SCHP showed significant correlation between the GATM PCR amplicon and loci on chromosome 1 (SSC1). The region with the highest probability was 1q12-21 with probability 0.818 and correlation 0.9282. The radiation hybrid panel was then used to increase the mapping resolution. The GATM gene was significantly linked (LOD > 3.99) to three markers (SW2432, SWR702, SKMC1) on chromosome 1. The retention fraction was 20%. The closest marker was SW2432 (55cR, LOD = 7.33), which has been mapped at 69.3cM of SSC1 (Yerle et al., 1996). The PEG10 gene was significantly linked (LOD > 3.89) to three markers (SWR915, SWR68, SSC8B04) on chromosome 9. The retention fraction was 21%. The closest marker was SWR915 (77 cR, LOD = 5.51), which has been mapped to SSC9. Human GATM and PEG10 were assigned to HSA 15q21.1and HSA 7q21 respectively (http://www.ncbi.nlm.nih.gov/). This information was in consistent with the established conservation of synteny between pig and human through chromosome painting (Goureau et al., 1996).
For GATM, the 704 bp 3’-UTR PCR product amplified using the primer pair G-4 was digested by BseLI. Allele frequencies for the BseLI polymorphism were significantly different in native Chinese Meishan and Yushan breeds compared with Yorkshire and Duroc breeds (P < 0.01). Interestingly, most of the native Tibetan pigs (18/20) were heterozygous animals at this locus. For PEG10, the primer pair P-2 were used to amplify the samples, and the PCR products were digested with TaqI restriction enzyme to distinguish different alleles. Allele frequencies for TaqI polymorphism were significantly different in native Chinese Erhualian breed compared with Yorkshire pig.
We firstly examined the allelic expression of porcine GATM using a SNP identified within 3’-untranslated region (G-4). Six 90 days and three 75 days heterozygotes were found and used for imprinting analysis (704 bp for allele A, 609 and 95 bp for allele G, Fig. 1). Allelic expression of GATM was examined by RT-PCR in the heterozygotes with primer pair G-6 which span an intron (Table 1). This eliminated the possibility of amplification of contaminated genomic DNA. RT-PCR followed by BseLI digestion revealed biallelic expression of GATM gene both in placentas on days 75 and 90 of gestation (396 bp for allele A, 278 and 118 bp for allele G, Fig. 1).
 View Details | Fig. 1. Detection of allelic expression of GATM by PCR-RFLP. (A) and (B) Six and three heterozygous placentas were detected on days 90 and 75 of gestation respectively. (C) and (D) RT-PCR-RFLP analysis of cDNAs in heterozygous placentas. MK is marker. |
Although most imprinted genes identified so far showed clear all-or-none monoallelic expression, some imprinted genes showing weak parental preferences were reported. For example, ORCTL2 gene exhibited “leaky” imprinting in both human fetal kidney and liver (Cooper et al., 1998). Furthermore, Tnfrsf23 gene was also proved to be weakly imprinted in several organs in mouse (Clark et al., 2002). Considering the unequal expression of the parental allele, we used different number of PCR cycles (25, 28, 34), but did not detect clear unequal expression. Then we sequenced the genomic DNA and cDNA from 3 heterozygotes. The peaks of both alleles from the parents were at close range (Fig. 2), indicating there is no preferential expression of any of the alleles of the GATM gene.
 View Details | Fig. 2. Direct sequencing assay of the GATM genomic DNA and cDNA of the heterozygous placenta sample. |
Imprinted genes were frequently conserved across different species of mammals (Jinno et al., 1996). Genomic imprinting has been also confirmed in plants; however, there were not enough data showing the conservation of the imprinted genes across different plant species. Thus, the conservation of imprinted genes across different mammalian species is a special phenomena comparing to other organisms such as plants. Our data showed biallelic expression of GATM in the placenta of pig in two different developmental stages, this was in agreement with the result from human placenta, which had been confirmed to escape genomic imprinting (Miyamoto et al., 2005). This demonstrated that GATM in extant mammals has divergent imprint status.
This is the first report on the imprinting status of PEG10 in the placenta of pig. Through the alignment and sequencing, we found a polymorphism (T/C) in the 3’-UTR of the PEG10, which has been reported by Zhang et al. (Zhang et al., 2006). The primer pair P-2 was used to amplify the DNA samples, three heterozygous individuals were found, two of them were 75 days and one was 90 days placentas. PCR-RFLP analysis of cDNA by TaqI restriction enzymes demonstrated that all three samples were monoallelically expressed (656 bp for allele T, 498 and 158 bp for allele C, Fig. 3). Due to the limitation of the materials, we could not identify the paternal origin of the alleles in this study. However, the monoallelic expression gave a strong clue that this gene is imprinted in the placenta.
 View Details | Fig. 3. Allelic expression detection of PEG10 gene by PCR-RFLP. D1, D2, and D3 were from the genomic DNAs, C1, C2, and C3 were from corresponding cDNAs. D1, D2, C1, and C2 were placentas on days 75 of gestation, D3 and C3 were placentas on days 90 of gestation. MK is the marker. |
Paternally expressed gene 10, which was predominantly expressed in placenta and testis (Clark et al., 2002) was essential for placenta formation and trophoblast differentiation (Ono et al., 2006; Smallwood et al., 2003). Previous study on human placenta revealed that PEG10 gene was down-regulated at early hypoxic stage of placental growth, induced at 11-12 wk, and then the level of expression increased and maintained at a high level (Smallwood et al., 2003). Notably, fetal weight of pig increased dramatically after 33 days (Vonnahme et al., 2004). Thus, we presume that PEG10 play a vital role in the development of fetus. Here, we detected the allele expression status of PEG10 in two stages of placenta, the result could provide useful information for further investigation of the function of this gene in the pig.
In summary, our study showed that GATM gene was expressed biallelically in porcine placenta on days 75 and 90 of gestation, whose imprinting status is not conserved across mouse and pig. PEG10 gene was monoallelically expressed in the porcine placentas on days 75 and 90 of gestation, which is conserved between pig, mouse and human. Our data also contribute to the comparative expression analysis of genes may related to epigenetics.
The RH panel and SCHP were kindly provided by Dr Martine Yerle of INRA, France. We also thank Dr Zhenfang Wu for helping with sample collection. This research was supported by the National Natural Science Foundation of China (30571328), the Program for New Century Excellent Talents in University of Chinese Education Ministry (NCET-05-0669), the Hubei province natural science creative team project (2006ABC008), and Key Project of National Basic Research and Developmental Plan (2006CB102105) of China.
|
Index