Down syndrome (DS) or trisomy 21 (MIM 190685) is the most common genetic disorder with a prevalence of 1 in 660 live births (Jones, 2006). DS is the leading cause of genetically-defined intellectual disability (Contestabile et al., 2010) and its phenotype is complex and variable among individuals, who may present with a combination of dysmorphic features (Ahmed et al., 2005; Pavarino-Bertelli et al., 2009), congenital heart disease (Abbag, 2006), neurological abnormalities such as early manifestations of Alzheimer’s disease (Lott & Head, 2005), immunological impairments (Ram & Chinen, 2011), elevated risk of specific types of leukemia (Hasle et al., 2000), and other clinical complications (Venail et al., 2004).
Trisomy 21 can be caused by three types of chromosomal abnormalities: free trisomy, translocation, or mosaicism. Mosaicism accounts for the minority of DS cases (about 1%) and is characterized by some cells containing 46 chromosomes and others, 47 chromosomes. Translocations are attributed to 3-4% of the cases, with Robertsonian translocation involving chromosomes 14 and 21 being the most common type. Finally, free trisomy occurs in about 95% of cases (Ahmed et al., 2005; J.M. Biselli et al., 2008b) and is characterized by the presence of three complete copies of chromosome 21.
Free trisomy, the main chromosomal abnormality leading to DS, is caused by the failure of normal chromosome 21 segregation during meiosis (meiotic nondisjunction) (Hassold & Hunt, 2000). The parental origin of the extra chromosome 21 is maternal in about 80% of cases (Jyothy et al., 2001), and most (about 77%) occur during the first maternal meiotic division in the maturing oocyte, before conception (Antonarakis et al., 1992).
2. Meiosis and chromosomal segregation
Faithful transmission of a genome from one generation to another depends on the mechanism of cell division in which each pair of replicated chromosomes is separated and equally distributed to mother and daughter cells. Meiosis generates haploid gametes through a specialized cell division process that consists of one round of DNA replication followed by two cell divisions. The first division, meiosis I (MI), involves the segregation of homologous chromosomes from each other, whereas meiosis II (MII) involves the segregation of the sister chromatids (Hassold & Hunt, 2000).
Timing of chromosome attachment and loss of cohesion is essential to faithful chromosome segregation. During MI, the cohesion between sister chromatid arms assures physical attachment by the chiasmata of homologous chromosomes, ensuring their alignment on the meiosis-I spindle, and maintains them at the site of recombination. Chiasmata are resolved at anaphase I by the loss of cohesion between the arms of sister chromatids in the homologous chromosomes; the chromosomes then segregate to opposite poles of the cell. Cohesion, however, must be maintained at centromeres between sister chromatids beyond meiosis I to prevent premature chromatid separation (predivision) and ensure proper attachment of the sister chromatids to opposite spindle poles in meiosis II (Barbero, 2011; Sakuno & Watanabe, 2009; Vogt et al., 2008).
The centromeric cohesion during meiosis I results from the attachment of kinetochores of sister chromatids to only one spindle pole (Sakuno & Watanabe, 2009). Kinetochores are situated on opposite sides of the centromeric heterochromatin at the centromeres of each sister chromatid and they capture and stabilize microtubules for the formation of kinetochore fibers, only then they are capable of chromosome bi-orientation during the metaphase and chromosome segregation during the anaphase of meiosis (Vogt et al., 2008).
During cell division, several chromosomal mal-segregation mechanisms can occur. Classical nondisjunction is due to the failure to resolve chiasmata between homologous chromosomes, whereby both homologues segregate together. In addition, premature resolution of chiasmata or the failure to establish a chiasma between a pair of homologues results in the independent segregation of homologues at MI, which leads to an error if both segregate to the same pole of the MI spindle. A MI error can also involve the segregation of sister chromatids, rather than homologous chromosomes, whereby the premature separation of sister chromatids at MI can result in the segregation of a whole chromosome and a single chromatid to one of the poles. At MII, errors result from the failure of sister chromatid separation (Hassold & Hunt, 2000).
3. The origin of maternal chromosome 21 nondisjunction
The molecular mechanisms involved in meiotic nondisjunction leading to trisomy 21 are still poorly understood and the only well-established risk factor for DS is advanced maternal age at conception (35 years or older) (Allen et al., 2009; Jyothy et al., 2001; Lamb et al., 2005). Studies have suggested many explanations for the maternal age-associated increase in aneuploidy. One model attributes the effect of advanced maternal age to the uterine environment, indicating that there might be an age-related decline in the ability to recognize and then abort trisomic fetuses (Aymé & Lippman-Hand, 1982; Stein et al., 1986). However, the observation that the advanced maternal age effect is restricted to chromosome 21 nondisjunction of maternal origin, but not associated with cases resulting from sperm or post-zygotic mitotic errors, suggests that the uterus is the source of the age effect (Allen et al., 2009).
On the other hand, Zheng & Byers (1993) proposed that age-dependent trisomy 21 results primarily from a mechanism that favors maturation and utilization of euploid oocytes over the pre-existing aneuploid products of mitotic (premeiotic) nondisjunction at an early stage of the reproductive lifespan. In addition, decreased expression of checkpoint proteins in aging oocytes (Vogt et al., 2008) and failure to effectively replace cohesion proteins that are lost from chromosomes during aging (Chiang et al., 2010) also are pointed out as risk factors for predisposing oocytes to errors in chromosome segregation.
A link between altered recombination and maternal age-related nondisjunction has been described. It was observed that recombination is reduced among nondisjoined chromosomes 21 at MI, and this reduction seems to be age-related (Sherman et al., 1994). Lamb et al. (1996) proposed that at least two “hits” are required for chromosome 21 nondisjuntion: (1) the establishment in the fetal ovary of a susceptible pattern of meiotic recombination, and (2) the abnormal processing of susceptible chromosomes in the adult ovary. The second “hit” would involve degradation of a meiotic process (e.g., a spindle component, a sister chromatid cohesion protein, a meiotic motor protein, a checkpoint control protein) that increases the risk of improper segregation for these susceptible bivalents (Hassold & Sherman, 2000). Further studies have shown susceptible patterns of chromosome 21 meiotic recombination, including pericentromeric and telomeric exchanges, described as maternal risk factors for DS even in young DS mothers (Gosh et al., 2009; Lamb et al., 2005).
Besides advanced maternal age, the age of the maternal grandmother at the time of birth of the mother has also been pointed out as a risk factor for the occurrence of DS. At an advanced age, the grandmother's reproductive system may fail to make the essential proteins needed for proper meiotic segregation in the germ cells of her daughter, leading to nondisjunction of chromosome 21 during the embryogenesis of DS child’s mother when she was in the grandmother's womb (Malini & Ramachandra, 2006). However, more recent studies failed to support the suggestion that advanced age of the DS grandmother is responsible for meiotic disturbances in her daughter (Allen et al., 2009; Kovaleva et al., 2010).
Although the risk of bearing a child with DS increases substantially with increasing maternal age, many DS children are born to mothers aged less than 35 years-old, suggesting other risk factors influencing DS etiology. In 1999, James et al. produced the first evidence that the occurrence of DS independent of maternal age is associated with DNA hypomethylation due to impairments in folate metabolism.
4. Folate metabolism
Folate represents an essential nutrition component in the human diet, and is involved in many metabolic pathways, mainly the folate metabolism, i.e., a single-carbon transfer from one molecule to another through a series of interconnected biochemical reactions. Folate is a generic term for a family of compounds present in most foods, e.g., legumes, leafy greens, some fruits, vegetables (e.g., spinach, broccoli, asparagus, and lettuce), liver, milk, and dairy products (Lin & Young, 2000). Humans, as all mammals, are unable to synthesize folate, thus its ingestion, either from normal diet or nutritional supplements, is very important. After intestinal absorption, natural folate, known as polyglutamate, requires reduction into monoglutamate by conjugases in the small intestine before it can be absorbed. On the other hand, in its synthetic form, folic acid exists as monoglutamate and does not need to be reduced for release into the blood and cellular uptake (Bailey & Gregory, 1999; Hall & Solehdin, 1998). Another disadvantage of natural food folate is its poor stability especially under typical cooking conditions, which can substantially reduce the vitamin content before it is even ingested, a significant additional factor limiting the ability of natural food folates to enhance folate status (McNulty & Pentieva, 2004; McNulty & Scott, 2008).
Folate metabolism is a complex metabolic pathway that involves multiple enzymes and water-soluble B vitamins such as folate, vitamin B6 and vitamin B12, that play key roles as enzyme cofactors or substrates in this metabolism. It includes two main cycles: purine and pyrimidine synthesis, necessary for synthesis and repair of DNA, and DNA methylation, an epigenetic process that acts on the control associated with gene expression and genomic stability essential for normal cellular methylation reactions (Figure 1).
Folate requires several transport systems to enter the cells and the one best characterized is the reduced folate carrier (RFC1), an enzyme located on intestinal cell membranes that carries out the transport of 5-methyltetrahydrofolate (5-methyl-THF) to the interior of a variety of cells, representing an important determinant of folate concentration in the interior of cells (Nguyen et al., 1997). In addition to the folate transport system, several genes and their respective enzymes play important roles in folate metabolism. The D
Methylenetetrahydrofolate reductase (MTHFR) is responsible for the conversion of 5,10-methylene-THF to 5-methyl-THF, the main circulating form of folate that donates methyl groups for homocysteine (Hcy) remethylation into methionine. This latter reaction is catalyzed by the enzyme Methionine synthase (MTR), which requires vitamin B12 or cobalamin (Cbl) as a cofactor, and results in the formation of S-adenosylmethionine (SAM), the primary methyl (CH3) donor for DNA methylation reactions (Finkelstein & Martin, 2000). SAM is demethylated to form S-adenosylhomocysteine (SAH) and then hydrolyzed to form adenine and Hcy. The DNA methyltransferase (DNMTs) enzymes catalyze the transfer of the methyl group, obtained from conversion of SAM into SAH, to position 5’ of cytosine residues located mainly in dinucleotide cytosine-guanine (CpG) (Bestor, 2000; De Angelis et al., 2008).
Methionine synthase reductase (MTRR), an enzyme codified by the
Betaine-homocysteine methyltransferase (BHMT) catalyses the conversion of Hcy to methionine by an alternative pathway of remethylation using the amino acid bethaine as methyl donor. When the Hcy folate-dependent remethylation catalyzed by the MTR enzyme is impaired by genetics or environmental factors, the BHMT enzyme plays an important role maintaining the homeostasis of Hcy (Pajares & Pérez-Salab, 2006).
In the transsulfuration cycle, Hcy is converted into cystathionine by Cystathionine β-synthase (CβS), a vitamin B6-dependent enzyme, and then into cysteine (Kraus et al., 1998). Under normal physical conditions, all Hcy is remethylated into methionine or catalyzed into cystathionine. The increase of Hcy concentration represents impairment in folate metabolism and thus in methylation reactions (Fenech, 2002).
Besides the enzymes that act directly on folate metabolism, cobalamin-transporting proteins also play an important role in this metabolic pathway, since the MTR enzyme is cobalamin-dependent. The enzyme Transcobalamin 2 (TC2) is synthesized in the intestinal villi and binds itself to Cbl in the interstitial fluid. This formed complex goes into the intestinal villi microcirculation and then reaches the systemic circulation. This circulation distributes the vitamin to all tissues where specific receptors on cell membranes bind and internalize the TC2-Cbl complex by endocytosis (Quadros et al., 1999; Seetharam & Li, 2000).
5. Folate metabolism, genomic stability, and maternal risk for chromosome 21 nondisjunction
Based on evidence that stable centromeric DNA chromatin may depend on the epigenetic inheritance of specific centromeric methylation patterns and on the binding of specific methyl-sensitive proteins to maintain the higher order DNA architecture necessary for kinetochore assembly (Karpen & Allshire, 1997), James et al. (1999) hypothesized that pericentromeric hypomethylation, resulting from impaired folate metabolism secondary to polymorphism of the
As described above, the MTHFR enzyme plays an important role in regulating DNA methylation through the reduction of 5,10-methylene-THF to 5-methyl-THF (Figure 1). The 677 C→T polymorphism is known to decrease the affinity of the enzyme for the flavin-adenine-dinucleotide (FAD) cofactor, decreasing enzyme activity (Guenther et al., 1999; Yamada et al., 2001). The
Another common polymorphism in the
In addition to the
As to the
A common polymorphism in the
Johnson et al. (2004) described a 19-base pair (bp) deletion polymorphism in intron-1 of the
The conflicting results shown by literature have raised the suggestion that the presence of individual polymorphisms in genes involved in folate metabolism might not increase the risk of having a child with DS, although the effect of combined risk genotypes might modify their individual effect and increase DS risk (J.M., Biselli et al., 2008a; Brandalize et al., 2010; Coppedè et al., 2006; Coppedè et al., 2009 ; da Silva et al., 2005; Martínez-Frías, et al., 2006; Scala et al., 2006; Wang et al., 2008). Moreover, there is evidence that the significance of genetic polymorphisms seems to depend on interactions with nutritional factors (Papoutsakis et al., 2010; Stover & Caudill, 2008).
6. Folate metabolism, genomic stability, and genetic polymorphisms
There is increasing evidence of association between polymorphisms in folate and Hcy metabolizing genes and levels of chromosome damage. The
Moreover, Piskac-Collier et al. (2011) recently demonstrated that lymphocytes from lung cancer patients showed a considerably increased frequency of cytogenetic damage in the presence of
7. Predisposition to chromosome malsegregation in young DS mothers and its association with folate-metabolizing gene polymorphisms
Studies with women who have a DS child at a young age have suggested that they present genetic predispositions to chromosome malsegregation in both somatic and germ line cells. Migliore et al. (2006) observed increased frequency of binucleated-micronucleated lymphocytes in women who had a DS child before 35 years of age, and fluorescence in situ hybridization analysis revealed that micronuclei were mainly originating from chromosomal malsegregation events, including chromosome 21 malsegregation. Further studies from their group confirmed increased chromosome damage in blood cells of young DS mothers and showed a significant correlation between micronucleated cells and both
8. Folate supplementation and DS prevention
Two important emerging areas of nutrition science are nutrigenomics, which refers to the effect of diet on DNA stability, and nutrigenetics, which refers to the impact of genetic differences between individuals on their response to a specific dietary pattern, functional food, or supplement for a specific health outcome. On these terms, two premises are important: (a) inappropriate nutrient supply can cause considerable levels of genome mutation and alter the expression of genes required for genome maintenance, and (b) common genetic polymorphisms may alter the activity of genes that affect the bioavailability of micronutrients and/or the affinity for micronutrient cofactors in key enzymes involved in DNA metabolism or repair, resulting in a lower or higher reaction rate (Bull & Fenech, 2008; Fenech, 2005).
As mentioned before, the folate-dependent biosynthesis of nucleotide precursors for DNA synthesis and genome methylation is dependent on the availability of many vitamins, including B12, B6, niacin, riboflavin, and minerals (zinc, cobalt), and is subject to regulation by other nutrients, such as iron and vitamin A, not directly involved in DNA or SAM biosynthesis (Stover, & Caudill 2008). Therefore, impairments in one-carbon metabolism, and the SAM cycle in particular, induced by nutritional deficiencies and/or genetic polymorphisms that encode folate-dependent enzymes, alter genome methylation patterns and gene expression levels (Stover, 2004; Stover, & Caudill 2008).
Since 1992, supplementation with 0.4 mg/daily of folic acid is recommended for women of childbearing age for the prevention of neural tube defects (Centers for Disease Control, 1992). Barkai et al. (2003) observed that families at risk for neural tube defects present with a higher frequency of DS cases and vice-versa, suggesting that both disorders are influenced by the same folate-related risk factors. However, two issues ought to be considered in the prevention of DS by folic acid: the dose and the timing of folic acid intake (Scala et al., 2006). It has been proposed that genomic instability is reduced at plasma folate concentrations above 34 nmol/L and Hcy concentrations below 7.5 μmol/L; these concentrations can only be reached with the ingestion of more than 0.4 mg/day of folic acid (Fenech, 2002). A report of a decreased occurrence of DS offspring in mothers supplemented with high doses of folic acid (6 mg/day) (Czeizel & Puho, 2005) supports the hypothesis of an involvement of folate in the etiology of DS. Concerning the timing of folate intake, it should be remembered that maternal MI errors in the primary oocyte may occur in a process that begins during fetal life and ends at the time of ovulation, whereas MII errors occur at the time of fertilization (Yoon et al., 1996). Therefore, it is likely that only MII errors would be immediately affected by folic acid intake in adult women (Ray et al., 2003).
Currently available literature suggests that abnormal folate metabolism is associated with increased maternal risk for DS, with a complex interaction between genetic polymorphisms, environmental factors (i.e., nutritional factors), and epigenetic processes. However, given the complexity of the folate pathway, these complex interactions cannot be easily understood and none of the polymorphisms studied so far can be used in genetic counseling to predict the maternal risk for having a DS child ( Coppedè et al., 2009 ). However, nutrigenetics and nutrigenomics are promising areas for evaluating the possibility of DS prevention with folic acid supplementation associated with susceptible genotypes. Thus, further large-scale studies are necessary to better understand the complex association between chromosomal 21 nondisjunction and folate metabolism.