Promoter haplotype combinations for the human PDGFRA gene are associated with risk of neural tube defects

https://doi.org/10.1016/j.ymgme.2003.11.003Get rights and content

Abstract

Recent animal studies suggested that deregulated expression of the platelet-derived growth factor receptor alpha (PDGFRα) may contribute to the failure of normal neural tube closure (NTC). There is also suggestive evidence that the promoter haplotype of the PDGFRA is associated with genetic susceptibility in human neural tube defects (NTDs). The purpose of our study was to investigate the association between promoter haplotype combinations of the human PDGFRA gene and risk for NTDs in a Hispanic population from the Texas-Mexico border region. This population has a considerably higher prevalence of NTDs (16/10,000 live births) than that generally reported in the United States (8–10/10,000 live births). In the present study, NTDs were defined as spina bifida or anencephaly. The haplotype of PDGFRA gene promoter was determined by direct DNA sequence analysis. Two novel haplotypes, H2ε and H1β, were found. We observed significant differences among variable haplotype groups from in vitro transient transfection studies in U2-OS osteosarcoma cell and two other cell lines (HeLa cell and MCF7 cell). Result from our case-control study demonstrated that the frequencies of haplotypes with low transcription activity were significantly higher in NTD mothers than that observed in control mothers (odds ratio=2.2, 95% CI=1.0–4.6). Infants with at least one low activity allele showed slightly higher risk (odds ratio=1.5, 95%=0.8–3.1). Our study suggests that the reduced transcriptional activity of PDGFRA gene could increase the risk of having an NTD-affected pregnancy.

Introduction

Neural tube defects (NTDs) represent a group of severe congenital anomalies characterized by the failure of complete neural tube closure during embryonic development. The preventive effect of folic acid on NTDs occurrence has been established by several trials [1], [2], [3]. However, the mechanism by which folic acid acts to prevent NTDs remains unclear. NTDs are considered to be etiologically multifactorial, which require both genetic susceptibility and environmental risk factors. Mutations and polymorphisms in folate pathway genes such as methylenetetrahydrofolate reductase (MTHFR), methionine synthase (MTR), and methionine synthase reductase (MTRR) have been broadly studied and in some studies show an association with NTD risk [4], [5]. However, none of these genes alone has been found to be a highly significant risk factor for NTDs in a wide range of population studies [6], [7], [8]. Folate transport genes are also considered candidate genes for NTDs [9], [10]. These studies, taken collectively, suggest a more complex mechanism of genetic susceptibility underlying the risk for NTDs.

Platelet-derived growth factor (PDGF) is a potent mitogen for mesenchymal and glial cells. A voluminous literature suggests that this protein is involved in embryogenesis, neoplasia, atherosclerosis, and wound-healing [11]. On a more fundamental level, PDGF has also been shown to regulate cell growth and survival, contribute to cell morphology and cell movement, such as scattering, chemotaxis or deposition of extracellular matrix. To date, four subunits of PDGF have been found (PDGF-A, -B, -C, and -D). There are two major types of receptors for PDGF, which are referred to as the PDGF α and β receptors, both belong to the family of tyrosine kinase receptors. PDGFα receptor binds three subunits (PDGF-A, -B, and -C), while PDGFβ receptor binds the PDGF-B and -D subunits [12], [13]. Both PDGF A chain and α receptor mRNA are expressed in the pre-implantation embryo, suggesting that PDGFRα plays a specific role in early embryonic development [14]. PDGF-A is expressed in most embryonic epithelial tissues, while PDGFRα is expressed in adjacent mesenchyme [15].

Studies on the spontaneous deletion mouse model, Patch (Ph), indicated that NTDs could occur as a result of lack of PDGFRα activity secondary to a deletion of this gene. The Ph mutation is an extensive (50–400 kb) deletion on mouse chromosome 5 which encompasses the pdgfra gene [16]. The heterozygous mice exhibited defective melanocyte migration leading to a distinctive white patch on the trunk. The ph/ph embryos presented with spina bifida along the entire anterio–posterior (A/P) axis. Soriano has generated mutant mice carrying a targeted null mutation of pdgfra gene. The homozygotes died during embryonic development and exhibited incomplete cephalic closure similar to that observed in a subset of Patch mutants, while no pigmentation phenotype was observed in heterozygotes. This study also observed increased apoptosis in the pathways followed by migrating neural crest cells, suggesting that PDGFs and their receptors may affect cell survival and patterning during early embryogenesis [17].

Recently, a 380 kb yeast artificial chromosome (YAC) containing the human PDGFRα gene (PDGFRA) was injected into pdgfra knockout mice, and the transgene was observed to be correctly expressed in oligodendrocyte progenitors, as well as in many mesoderm and neural crest-derived tissues. The human YAC completely rescued the profound craniofacial defects and spida bifida, as well as prolonging the survival time of the knockout mice. However, this YAC failed to rescue the congenital anomalies associated with Patch mutant mice [14]. Thus, it remains unclear whether the lack of pdgfra is the unique cause of Patch phenotype.

The human PDGFRA gene, consisting of 23 exons and encoding a 123 kDa protein of 1089 amino acids, is located at chromosome 4q12. Herrmann et al. [18] identified seven polymorphisms within the PDGFRA promoter and non-coding regions that exist as specific haplotypes [−1630 I/D (+/−AACTT), A-1506G, C-1390G, G-956A, C-908A, G-793T, +69 I/D (+/−GA)]. C-908A is located at a putative GCF and SP1 site, C-1390G is located at a putative CUP2 site. Joosten et al. [19] grouped the haplotypes into H1 and H2 (H2α, H2β, H2γ, and H2δ) according to seven single nucleotide polymorphisms (SNPs) in the distal promoter region, as well as a 2 bp insertion/deletion in the non-coding region of exon 1. H1, H2γ, and H2δ are characterized by low transcriptional activity, while the H2α and H2β haplotypes have high transcriptional activity. Among patients with spina bifida, the low transcriptional activity haplotype (H1) was under-represented, while heterozygotes of the high transcriptional activity haplotypes (H2α/H2β) were over-represented. Heterozygotes with the two most common haplotypes, H1 and H2α, were significantly under-represented in the control group but over-represented in the spina bifida patients. These findings indicated that promoter haplotype combinations that affect the transcriptional activity of human PDGFRA might be associated with the risk of spina bifida.

The purpose of the present study is to investigate whether specific promoter haplotypes of PDGFRA gene are involved in regulating the risk of NTDs in a large Hispanic population in the Texas-Mexico border. Furthermore, we were interested in determining whether high transcriptional activity or low transcriptional activity leads to a higher NTD risk.

Section snippets

Study population

A detailed description of the study population has been previously reported [5], [7]. Subjects were recruited from a Hispanic population living among the 14 counties along the Texas-Mexico border region that delivered or terminated their pregnancies between June 1, 1995 and September 30, 1998. Case infants and fetuses were identified through hospitals, birthing centers, ultrasound centers, abortion centers, pre-natal clinics, genetics clinics, and birth attendants (midwives and non-hospital

Result

We cloned six different haplotypes from control control mothers and one NTD mother, designated as H1α, H1β, H2α, H2β, H2γ, and H2ε (Fig. 1). We found two novel haplotype combinations. H1α was originally designated as H1 [19]. The haplotype H1β was found in four control mothers and three control infants, and in one case and his/her mother. All were heterozygotes. H2ε haplotype was observed in one case mother as a homozygote. Two case mothers were H1α/H2ε heterozygotes. One control infant was a

Discussion

Our study is the first to report that mothers with two copies of low promoter activity alleles of PDGFRA gene showed a 2.2-fold higher risk of having an NTD-affected pregnancy. However, this increase was not observed in NTD-affected infants. Infants with at least one low activity allele had a slightly higher NTD risk than those who had two high activity alleles. The observed differences could be real or may have occurred by random variation. We cannot discriminate between the alternatives

Acknowledgements

This work was supported in part by grants from the National Institutes of Health (DE12898 and ES09106) and March of Dimes Grant FY-01-542. Additional funding for this project was provided in part through Cooperative Agreements No. U50/CCU613232 and U59/CCU913241 from the Centers for Disease Control and Prevention. Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the Centers for Disease Control and Prevention or that of the NIH, or

References (30)

  • L.D Botto et al.

    5,10-Methylenetetrahydrofolate reductase gene variants and congenital anomalies: a HuGE review

    Am. J. Epidemiol.

    (2000)
  • G.M Shaw et al.

    Maternal preconceptual vitamin use, genetic variation of infant reduced folate carrier (A80G), and risk of spina bifida

    Am. J. Med. Genet.

    (2002)
  • C.H Heldin et al.

    Possible in vivo effect and clinical utility of platelet-derived growth factor and PDGF antagonists

    Transplant. Proc.

    (1993)
  • G.B Afink et al.

    Molecular cloning and functional characterization of the human platelet-derived growth factor α receptor gene promoter

    Oncogene

    (1995)
  • C Betsholtz et al.

    Developmental roles of platelet-derived growth factors

    Bioessays

    (2001)
  • Cited by (28)

    • Neural Tube Defects

      2015, Principles of Developmental Genetics: Second Edition
    • Genetic Basis of Neural Tube Defects

      2009, Seminars in Pediatric Neurology
      Citation Excerpt :

      A long list of candidate genes derived from animal studies has been investigated in human NTDs and only a few yielded positive results with no major causative gene identified.1,44 Table 1 lists candidate genes that were derived from animal studies and that were found to be associated with human NTDs.64-81 These genes are involved in a wide range of functions at all steps of neural tube formation and patterning.

    • Chapter 1 Modeling Neural Tube Defects in the Mouse

      2008, Current Topics in Developmental Biology
      Citation Excerpt :

      From the analysis of un and Ph mutant mouse lines, one genetic risk factor may be PDGFRα mutation. Interestingly, polymorphisms in the PDGFRα gene have also been associated with NTDs in humans as specific polymorphisms that alter transcription from the PDGFRα promoter correlate with a predisposition to spina bifida (Joosten et al., 2001; Zhu et al., 2004). Other potential candidate genes that interact with mutant Pax1 to cause spina bifida in mouse models include Transcription factor E2a (Tcf2a) and Forkhead box C2 (FoxC2; Furumoto et al., 1999; Joosten et al., 2005).

    View all citing articles on Scopus
    View full text