Elsevier

Brain, Behavior, and Immunity

Volume 67, January 2018, Pages 308-313
Brain, Behavior, and Immunity

Full-length Article
HLA-G coding region polymorphism is skewed in autistic spectrum disorders

https://doi.org/10.1016/j.bbi.2017.09.007Get rights and content

Highlights

  • HLA-G alleles are skewed in ASD children.

  • HLA-G alleles seen in ASD result in immune activation.

  • The skewing of HLA-G alleles seen in ASD might play a pathogenic role.

Abstract

Different isoforms of HLA-G protein are endowed with a differential ability to induce allogenic tolerance during pregnancy. As prenatal immune activation is suggested to play a role in the onset of autistic spectrum disorders (ASD), we evaluated HLA G*01:01-*01:06 allelic polymorphism in a cohort of Italian children affected by ASD (N = 111) their mothers (N = 81), and their healthy siblings (N = 39). DNA sequencing analysis of HLA-G exon 2, 3 and 4 was used to obtain HLA-G allelic frequencies; alleles distribution was compared with that of two control groups of Caucasoid couples of multiparous women and their partners from Brazil and Denmark. HLA-G distribution was significantly different in ASD children compared to both control groups (Brazilian pc = 1 × 10−4; Danish pc = 1 × 10−3). Since HLA-G distribution was similar in the two control groups, their data were pooled. Results indicated that HLA-G*01:01 was significantly less frequent (pc = 1 × 10−4; OR:0.5, 95%CI: 0.3–0.7) whereas HLA-G*01:05N was significantly more frequent (pc = 2 × 10−3; OR:7.3, 95%CI: 2.4–26.6) in ASD children compared to combined controls. Finally, no clear pattern emerged when HLA-G allelic distribution was analyzed in healthy sibs. Notably, HLA-G allelic distribution found in ASD mothers was similar to that observed in the control subgroup of women with recurrent miscarriages, whilst it was significantly different compared to women without miscarriages (pc = 6 × 10−4 df = 12). Since HLA-G*01:01 is associated with the elicitation of KIR-mediated tolerogenic responses and HLA-G*01:05N correlates with NK cells activation, results herein indicate that an immune activating milieu during pregnancy is more likely observed in association with the development of ASD, similarly to what occurs in women with recurrent miscarriages.

Introduction

An uncontrolled immune response is suspected to impair the central nervous system (CNS) development as observed in children with a diagnosis of Autism Spectrum Disorders (ASD) (Saresella et al., 2009, Westover et al., 2011, Onore et al., 2012). In line with this, a maternal immune attack on foetal tissues is suggested to be involved in the pathogenesis of ASD (Warren et al., 1996). This hypothesis is reinforced by the observation that autoantibodies (indicating an adaptive immune process) targeting neuronal components of the foetal brain, are observed in a subset of mothers of ASD patients (Braunschweig et al., 2013, Croen et al., 2005, Patterson, 2011, Singer et al., 2008).

Natural killer (NK) cells are effectors of innate immunity that are present in the peripheral blood and, during pregnancy, can also be identified within the uterine mucosa, at the foetal/maternal interface. NK cells are characterized by the expression of a receptor family (KIR) that interact with classical HLA-C and non-classical HLA-G molecules on foetal trophoblast. The net effect of these interactions determines whether NK will or will not be activated. A skewing of the KIR-HLA complexes, in which activating molecules prevail, was recently shown in ASD children (Torres et al., 2012) and their mothers (Guerini et al., 2014). HLA-G, in particular, is a non-classical HLA Class Ib molecule expressed by the placental and decidual trophoblast cells (Yelavarthi et al., 1991) as well as by foetal cells (Van Wijk et al., 2001), which is responsible for the induction of allogenic tolerance during pregnancy. The immunomodulatory functions of HLA-G are mediated by the interaction with KIR receptors on NK cells, including KIR2DL and ILT-2 (Favier et al., 2010); these interactions lead to suppression of T cell lymphocyte proliferation and stimulation of interferon (IFN)-c production by uterine NK cells (UNK), with the net effect of favouring foetal tolerance (Akhter et al., 2012, Matter and Sharif, 2013, Peter von Dadelszen, 2008). Besides being expressed on placental cell surface, HLA-G molecules can also be secreted. Soluble HLA-G (sHLA-G) induces the apoptosis of both activated NK and CD8 T-cells and a decreased amount of sHLA-G is seen in serum of women undergoing recurrent spontaneous abortions (Hviid et al., 2002). These observations suggest that the HLA-G/NK interaction at the maternal–foetal interface is critical in determining the pregnancy outcome.

Different HLA-G protein isoforms were recently identified; in fact, four membrane-bound HLA-G isoforms (G1, G2, G3, and G4) and three different soluble forms (G5, G6 and G7) -resulting by alternative splicing and very few amino acid polymorphisms (Castelli et al., 2014)- can be distinguished. To date, 44 HLA-G alleles encoding 14 distinct functional proteins have been described. Nevertheless, only four variation sites in the coding region of the HLA-G locus are related to the amino acid exchange frequently found in worldwide populations: (i) the +292 A/T SNP at exon 2 (codon 31), exchanging a threonine for serine in the G∗01:03 allele; (ii) the +755 C/A SNP at exon 3 (codon 110), exchanging leucine/isoleucine in the G∗01:04 allele group; (iii) the DeltaC deletion at exon 3 in the G∗01:05N allele; and (iv) the +1799 C/T SNP (codon 258), exchanging threonine/methionine in the G∗01:06 allele (Donadi et al., 2011). These isoforms are functionally different, as they are endowed with a differential ability to induce tolerance. Finally, a nucleotide insertion of 14 bp in the regulatory 3′UTR region of HLA-G gene (HLA-G∗14bp+) is shown to reduce tolerogenic activity (Christiansen et al., 2012, Hylenius et al., 2004), and suggested to be involved in ASD development (Guerini et al., 2015).

We evaluated HLA G∗01:01-∗01:06 allelic polymorphisms in a well-characterized cohort of Italian children affected by ASD and their relatives to investigate whether specific HLA-G alleles, which might play a role in ASD-associated immune activation, can be identified.

Section snippets

Patients and controls

A total of 111 ASD children (92 males/19 females; mean age: 8.7 ± 4.4 years) born in peninsular Italy and of Italian descent, 81 of their mothers, and 39 of their healthy siblings (13 males/26 females; mean age: 11 ± 5.9 years) were enrolled at the Don Gnocchi Foundation, at the C. Mondino National Neurological Institute and at the ANFFAS (National Association of the Families of People with Intellectual and/or Relational Disability).

In the 3 months preceding study enrollment, all ASD children

HLA-G distribution in ASD children and control group

HLA-G allelic distribution was analyzed both in the group of 111 ASD children and in their healthy siblings and was compared to that reported in two Caucasoid population of multiparous women and their partners from Brazil (N = 167) (Nardi et al., 2012) (Pop 1) and Denmark (N = 93) (Hviid et al. 2002) (Pop 2) (Fig. 1). A statistically different HLA-G distribution was observed when ASD children were compared to either Pop 1 (pc = 1 × 10−4 df = 12) or Pop 2 (pc = 1 × 10−3 df = 12). In particular, the HLA-G∗0105N

Discussion

HLA-G coding region polymorphisms were analyzed in ASD children, in their healthy siblings and in 81 of their mothers; results were also compared to those obtained in two different datasets of multiparous women with different ethnic backgrounds. Results showed the presence of a significant skewing of HLA-G polymorphisms in ASD children, being the HLA-G∗0105N allele significantly more and the HLA-G∗0101 allele significantly less frequent, respectively. Notably, the same skewing was detected in

Conflict of interest statement

No author declares any conflict of interest.

Acknowledgments

Supported by Ricerca Corrente 2015 and Ricerca Finalizzata 2009 [Italian Ministry of Health].

References (37)

  • R.P. Warren et al.

    Strong association of the third hypervariable region of HLA-DRB1 with autism

    J. Neuroimmunol.

    (1996)
  • A. Akhter et al.

    In vitro up-regulation of HLA-G using dexamethasone and hydrocortisone in first-trimester trophoblast cells of women experiencing recurrent miscarriage

    Tissue Antigens

    (2012)
  • C.L. Aldrich et al.

    HLA-G genotypes and pregnancy outcome in couples with unexplained recurrent miscarriage

    Mol. Hum. Reprod.

    (2001)
  • B.M. Altevogt et al.

    Autism and the environment: challenges and opportunities for research

    Pediatrics

    (2008)
  • American Psychiatric Association

    Diagnostic and Statistical Manual of Mental Disorders

    (2013)
  • F. Arjmand et al.

    The balance of the immune system between HLA-G and NK cells in unexplained recurrent spontaneous abortion and polymorphisms analysis

    Immunol. Res.

    (2016)
  • H.O. Atladóttir et al.

    Association of family history of autoimmune diseases and autism spectrum disorders

    Pediatrics

    (2009)
  • D. Braunschweig et al.

    Autism-specific maternal autoantibodies recognize critical proteins in developing brain

    Transl. Psychiatry

    (2013)
  • Cited by (19)

    • Understanding the genetic contribution of the human leukocyte antigen system to common major psychiatric disorders in a world pandemic context

      2021, Brain, Behavior, and Immunity
      Citation Excerpt :

      We and others have demonstrated the association of both HLA-G polymorphism and/or expression in psychiatric disorders including Autism Spectrum Disorders, schizophrenia and bipolar disorder, within an early developmental context. In consecutive articles, Guerini et al., demonstrated that genetically determined low expression of the tolerogenic HLA-G molecules at the fetal-mother interface, possibly led to prenatal immune activation, is associated with ASD risk (Guerini et al., 2015, 2018a, 2018b, 2019). Schizophrenia is also widely believed to be a neuro-developmental psychiatric disorder, although the role of HLA-G polymorphism and expression has been less investigated.

    • HLA-G allelic distribution in Sardinian children with Autism spectrum disorders: A replication study

      2019, Brain, Behavior, and Immunity
      Citation Excerpt :

      Finally, possibly owing to the limited sample size, no allelic skewing was observed between sASD children and their healthy siblings. HLA-G distribution was finally analyzed in 91 sASD mothers, 98 women with recurrent miscarriages (WRM) and 130 multiparous women without miscarriages (WWM) (see Guerini et al., 2018). Results confirmed that, whereas no differences in HLA-G alleles were observed in cASD, sASD and WRM (pc = 0.51 and pc = 0.24, df = 12 respectively), the distribution of these alleles was significantly different in WWM mothers (sASD mothers vs. WWM, pc = 1.5 × 10−2, df = 12).

    View all citing articles on Scopus
    View full text