Genomic disorders: An overview

INTRODUCTION — Genomic disorders are diseases that result from the loss or gain of chromosomal/DNA material (copy number variations [CNVs]). There are a number of well-delineated genomic disorders that can be divided in two categories: those resulting from copy number losses (deletion syndromes) and copy number gains (duplication syndromes).

An overview of genomic disorders is presented here. Specific syndromic disorders are reviewed separately, as indicated in the sections below.

Separate topics also discuss:

●Chromosomal abnormalities – (See "Chromosomal translocations, deletions, and inversions".)

●Cytogenetics – (See "Tools for genetics and genomics: Cytogenetics and molecular genetics".)

●DNA sequencing – (See "Next-generation DNA sequencing (NGS): Principles and clinical applications".)

●Definitions of terms – (See "Genetics: Glossary of terms".)

COPY NUMBER VARIATIONS

●Structural variation – Structural genetic variation refers to a class of sequence alterations spanning typically more than 1000 base pairs (one kilobase of DNA or kb) [1]. However, studies have indicated that these genetic variations can be smaller in the range of up to 450 base pairs and that most individuals have at least 1000 such variations [2].

These structural genetic variations include:

•Quantitative variants such as copy number variations (CNVs)

•Sequence rearrangements (such as those observed among immunoglobulins)

•Other less-common variations, including chromosomal rearrangements that may or may not alter the genome contents but may disrupt gene function and result in disease

●CNVs – CNVs are the most prevalent type of structural variation. They are DNA segments spanning thousands to millions of bases whose copy number varies among different healthy individuals [3,4].

It is estimated that 50 percent of the genome is constituted by repeated sequences [5]. These CNVs can be recurrent when mediated by regions where low copy repeats share more than 90 percent identity or where short-segmental duplications exceed 1 kilobase in size. Other CNVs can be nonrecurrent and associated with DNA blunt ends, micro-homologies, or even insertions that can lead to more complex rearrangements [6].

These submicroscopic genomic differences in the number of copies of one or more sections of DNA are the result of DNA gains or losses.

•Gains – Copy number gains can be the result of duplications, triplications, or even multiple copy number gains.

•Losses – Most deletions are the result of copy number loss at one locus (heterozygous deletions), but in some instances the loss can affect copies at both loci (homozygous deletions).

•Transmission – CNVs are commonly inherited but can occur de novo (as a new event). These were initially considered rare events resulting from sporadic mutation(s) and correlated with specific Mendelian diseases [7,8]. These misperceptions about their rarity and absolute disease linkage were primarily due to technical limitations precluding genome-wide assessments in large cohorts. Advances in technology have shown that deviation from the diploid state is widespread and contributes substantially to genetic diversity.

Some studies have suggested that CNV differences in the human genome are as extensive as 20 percent, although this may be an overestimation [9,10] and probably closer to 5 percent. It is estimated conservatively that most individuals carry an average of three large-scale CNVs [4]. The number of known CNVs that contribute to disease pathogenesis continues to increase.

•Hotspots – The physical distribution of CNVs appears to be nonrandom across the genome, with both CNV hot and cold spots reported [9,11]. CNV frequency is greatest in regions of segmental duplication (a 4- to 10-fold enrichment for CNVs), consistent with nonallelic homologous recombination as a primary mechanism for the causation of CNVs [3,12-15].

CNVs are more commonly observed in gene-rich regions. CNVs appear to be enriched in specific gene families, including immune and inflammatory response genes, cell signaling and cell adhesion molecules, structural proteins, and olfactory receptors [9]. Most of these differences probably represent benign CNVs that reflect normal variation with no apparent clinical consequence [16].

•Pathogenicity – CNVs can be pathogenic if they involve dose-sensitive genes (haplo-insufficient or triplosensitive) or if they influence or disrupt genomic regions through regulatory elements [17,18]. Gene dosage sensitivity information can be interrogated in curated databases such as ClinGen (ClinGen Dosage sensitivity page) [19]. Some pathogenic CNVs cause syndromic disorders with consistent phenotypic features (eg, deletions of the elastin gene in Williams syndrome, duplications of the PMP22 gene in Charcot-Marie-Tooth disease type 1A [CMT1A]), while others are associated with disease susceptibility or resistance (eg, cancer, HIV infection, autoimmune disorders, autism).

Mendelian disease associations — CNVs can be responsible for Mendelian diseases associated with gains and losses of genetic material, even at the exonic level. Examples include the following:

●Contiguous gene deletions or duplications as seen in Williams-Beuren syndrome, 22q11 deletion-syndrome, Smith-Magenis syndrome, and Potocki-Lupski syndrome. These are examples of recurrent CNVs and mediated by nonallelic homologous recombination (NAHR) occurring at sites of low-copy repeats. (See "Williams syndrome" and "Microdeletion syndromes (chromosomes 12 to 22)", section on '17p11.2 deletion syndrome (Smith-Magenis syndrome)' and "Microduplication syndromes", section on '17p11.2 duplication syndrome (Potocki-Lupski syndrome)'.)

●Deletions of genes or portions of genes (exons) leading to Mendelian-inherited genetic disorders, including disorders that are autosomal dominant (eg, deletions of the CREBBP gene in Rubinstein-Taybi syndrome) and X-linked recessive (eg, deletions of the dystrophin gene in Duchenne muscular dystrophy) [20]. (See "Microdeletion syndromes (chromosomes 12 to 22)", section on '16p13.3 deletion syndrome (Rubinstein-Taybi syndrome)' and "Duchenne and Becker muscular dystrophy: Genetics and pathogenesis", section on 'Genetics and pathogenesis'.)

Associations with complex traits — CNVs can also be associated with more complex traits or syndromes in which there are combinations of genetic and environmental factors [2]. Examples include a number of disorders associated with auto-inflammatory disease. Some of these CNVs and their association with disease have been replicated by other studies but not always found to be significant and perhaps may be dependent of the population studied [21].

●A deletion at the complement component 4 (C4) locus that confers a 1.6- to 5.3-fold risk for systemic lupus erythematosus [22]. (See "Epidemiology and pathogenesis of systemic lupus erythematosus".)

●Deletions in FCGR3B associated with granulomatosis with polyangiitis [23] and other autoimmune/inflammatory conditions such as rheumatoid arthritis [24,25]. (See "Pathogenesis of antineutrophil cytoplasmic autoantibody-associated vasculitis".)

●A deletion of defensin-beta 4 (DEFB4) associated with increased risk of colonic Crohn disease [26]. A deletion of 20 kb upstream of the IRGM gene has also been associated with Crohn disease and psoriasis [21,27].

●A 32 kb deletion involving two genes that encode envelope proteins (LCE3B and LCE3C) has been associated as a susceptibility factor for psoriasis [28]. DEFB4 deletions have also been associated with psoriatic arthritis [29].

●Another example of complex disorders includes the increased frequency of de novo germ-line CNVs in patients with autism spectrum disorder (ASD) and schizophrenia [30,31]. CNVs in multiple areas of the human genome potentially involved in autism pathogenesis have been described. (See "Autism spectrum disorder (ASD) in children and adolescents: Terminology, epidemiology, and pathogenesis", section on 'Genetic factors'.)

This line of research has also identified specific genes and pathways involved in ASD and related syndromes (eg, duplications of the 15q11-q13 region, deletions and duplications of 16p11.2, 1q21 duplications) [32-34]. (See "Microdeletion syndromes (chromosomes 1 to 11)" and "Microdeletion syndromes (chromosomes 12 to 22)".)

In addition, some conditions are associated with multiple CNVs, which may explain their variable phenotypes [35]. As an example, in a retrospective study, array comparative genomic hybridization (CGH) was used to evaluate CNVs in 2312 children with developmental disabilities who already had one predefined CNV and in 8329 children without developmental disabilities [36]. This study found that, in comparison to controls without developmental disabilities, individuals with developmental disabilities had an increased number of additional CNVs. This increase in CNVs may have played a causative role in the disabilities (eg, by causing disruption of a new gene or altering gene dosage), or may be an indirect marker of susceptibility to genomic damage. (See 'Array comparative genomic hybridization' below.)

Causes of CNVs — Low copy repeats are stretches of repetitive DNA sequences (segmental duplications) approximately 10 to 300 kilobases in size that share ≥95 percent homology [21]. Erroneous pairing of these highly homologous regions can cause misalignment and unequal recombination during meiosis. This can lead to duplication and deletion of chromosomal material resulting in CNVs. This process is known as nonallelic homologous recombination (NAHR) (figure 1), the most common mechanism for the formation of genomic rearrangements [37,38].

NAHR can result in either deletions or duplication via the same mechanisms and due to low copy repeats mediated nonallelic homologous recombination. A classic example is the case of Charcot-Marie-Tooth Type I, a peripheral neuropathy caused by duplications of the PMP22 gene on chromosome 17p11.2. The same region when deleted leads to a different neuropathy known as Tomaculous neuropathy also known as hereditary neuropathy with liability to pressure palsies (HNPP). (See "Microduplication syndromes".)

Other mechanisms leading to CNVs include nonhomologous end joining and microhomology-mediated break-induced replication, although discussion of these is beyond the scope of this chapter [39].

Despite a great deal of knowledge about the structural details of how CNVs occur, we do not know what predisposes certain individuals to develop these changes more than other individuals.

Of interest, in a large study of patients with developmental disabilities, parental data provided information about whether CNVs were inherited or arose de novo [36]. This study suggested that CNVs were more likely to arise de novo in the syndromic disorders (eg, Williams-Beuren syndrome), whereas CNVs were more likely to be inherited in the disorders with variable phenotype (eg, intellectual disability). A potential explanation may be that reproductive fitness is reduced in individuals with the more severe syndromic disorders. (See "Microdeletion syndromes (chromosomes 1 to 11)", section on '7q11.23 deletion syndrome (Williams syndrome)'.)

Interpreting CNVs — The interpretation of CNVs has steadily improved due to use of large control databases that allow a direct comparison with apparently normal controls. Examples of these databases include the Database of Genomic Variants (DGV), large sequencing projects such as the 1000 Genomes Project, or ClinVar [40-42]. Other databases that include phenotypic information, such as DECIPHER (DatabasE of Chromosomal Imbalance and Phenotype in Humans using Ensembl Resources), are becoming valuable resources for findings detected by array-comparative genomic hybridization [43]. DECIPHER is a web-based resource that contains array and clinical data deposited by over 247 member centers around the world. The Deciphering Developmental Disorders study performed in the United Kingdom has evaluated 14,000 children with severe developmental delay and their parents with microarray and exome sequencing [44]. These large-scale studies are extremely valuable tools for the interpretation of CNV data. Another popular tool available for curation of CNVs and their gen e content is ClinGen. This is an open clinical genome resource that establishes proper annotation of CNVs based on information assessed from multiple tiered resources and curated by experts [45].

Interpreting the pathogenicity of CNVs can be quite challenging in the presence of multiple CNVs in one individual. Some studies have shown that multiple rare CNVs, either inherited or de novo, can compound the clinical severity [36,46]. It is also important to highlight the presence of common CNVs in the general population. It is unclear whether many of these common polymorphisms may play a role in common disorders [47].

DISEASE MECHANISMS — There are different potential mechanisms that can lead to disease in genomic disorders secondary to deletions and duplications. The main mechanism is related to changes in dosage sensitive genes. "Haploinsufficiency" ("haplo" means half) defines the concept where loss or gain of one allele of a gene may lead to abnormal protein production or function, thereby causing disease [48].

Deletions can interfere with the required amount of a protein dose, resulting in disease. One example is Williams-Beuren Syndrome, which is caused by a microdeletion in chromosome 7q11.23 involving multiple genes, including the elastin gene [49,50]. Having only one-half of the normal dose of elastin is enough to cause disruption of the arterial elastic fibers, leading to aortic narrowing and multiple other arterial abnormalities. The deletions in Williams-Beuren syndrome, similar to what happens in many other genomic disorders, can be of different sizes. The loss typically encompasses 1.55 Mb, but in some cases the deletion can be more extensive or even smaller. The size difference is due to the different LCRs around the critical region that can be involved in mediating the rearrangements. (See "Chromosomal translocations, deletions, and inversions", section on 'Deletions' and "Microdeletion syndromes (chromosomes 1 to 11)", section on '7q11.23 deletion syndrome (Williams syndrome)' and "Valvar aortic stenosis in children".)

The importance of gene dosage in determining the effect of CNVs was illustrated in the sex differences from a large study of CNVs in patients with developmental disabilities [36]. When compared with females, males had more of the "variable phenotype genomic disorders" (eg, intellectual disability) but not syndromic disorders (eg, autism spectrum disorders). Females may be protected from these more genetically multifactorial disorders due to sex chromosome bias (ie, the protection of females from weakly deleterious mutations on one X chromosome by the normal corresponding genes on the other X chromosome).

Duplications may cause disease if the genes involved in the duplications are triplosensitive or if the copy number gain disrupts the reading frame of a gene, leading to abnormal protein synthesis. Another mechanism includes unmasking of recessive alleles [51-53]. When a recessive allele is deleted, it could potentially uncover a pathogenic variant in the remaining copy (gene) on the other chromosome. This would lead to disease because there is no normally functioning copy left for the affected gene.

Other disease mechanisms include the following:

●Interference with regulatory elements outside of genes, as in brachydactyly type A2 and duplications outside the BMP2 gene [54].

●Interference with imprinted genes, as in the case of paternal duplications of 11p15 that lead to Beckwith-Wiedemann syndrome [55]. (See "Beckwith-Wiedemann syndrome", section on 'Genetics and pathogenesis'.)

Arrays that contain single nucleotide polymorphisms (SNPs) can further aid in the identification of imprinting disorders caused by uniparental disomy. One such example is Angelman syndrome and uniparental disomy (UPD) caused by isodisomy. Isodisomies result from either nondisjunction in meiosis II or by postzygotic duplication (monosomy rescue). (See "Basic genetics concepts: Chromosomes and cell division", section on 'Meiosis'.)

A small number of Angelman syndrome cases are the result of UPD. In those cases there is absence of the maternal contribution for a region of chromosome 15 (15q11-q13). These cases are typically associated with monosomy rescue (duplication of a chromosome from a monosomy 15 zygote), where there are two identical copies of the paternal chromosome 15 and no maternal contribution. Other examples of UPD can be seen in Prader-Willi syndrome. (See "Prader-Willi syndrome: Clinical features and diagnosis".)

Contiguous gene syndromes — Contiguous gene syndromes can occur when large CNVs affect several contiguous genes [56,57].

As examples:

●Williams-Beuren Syndrome, also called Williams syndrome, is caused by a 1.5-1.8 Mb deletion on chromosome 7q11 that typically encompasses 26 to 28 genes. (See "Williams syndrome", section on 'Genetics'.)

●In WAGR syndrome (Wilms tumor, Aniridia, Genitourinary anomalies, and mental Retardation), clinical features are attributable to the loss of individual genes by a large deletion: deletions of WT1 are responsible for Wilms tumor, while PAX6 deletions are responsible for the aniridia findings. Both genes are contiguously located within the short arm of chromosome 11. (See "Microdeletion syndromes (chromosomes 1 to 11)", section on '11p13 deletion syndrome (WAGR syndrome)'.)

DETECTION OF GENOMIC DISORDERS — Genomic disorders are typically investigated and detected by array comparative genomic hybridization (array CGH). Gains or losses detected on an array can be confirmed by an independent method such as fluorescence in situ hybridization (FISH), multiple ligation dependent probe amplification (MLPA), or quantitative PCR (Q-PCR). (See "Tools for genetics and genomics: Cytogenetics and molecular genetics" and "Polymerase chain reaction (PCR)".)

Parental FISH or array CGH testing may be considered in cases of selected genomic abnormalities that can be inherited or to interpret CNVs that are variants of uncertain significance and aid with the clinical interpretation. This is relevant for future pregnancies and family planning [58]. Whereas the genomic position of a loss is clear by array testing, gains may be either tandem duplications or insertions. If the latter arise as a consequence of a parental insertional translocation (IT), this may have important implications for future pregnancies [59]. In this case, FISH can be used to determine the nature and localization of the copy number gain.

Next generation sequencing (NGS) technology used for exome sequencing is also suitable to identify CNVs but not widely used for this purpose. Whole exome sequencing (WES) can detect CNVs and sequence variations simultaneously [60,61]. (See "Next-generation DNA sequencing (NGS): Principles and clinical applications".)

Array comparative genomic hybridization — Array comparative genomic hybridization (array CGH), also known as chromosome microarray or microarray-based comparative genomic hybridization, is the gold standard laboratory test for the detection of CNVs that cause genomic disorders (figure 2). Array CGH allows detection of small losses or gains of genomic material down to several kilobases (kb) down to the exonic level. It is widely used in the evaluation of patients with intellectual disabilities and/or congenital malformations [62-67]. (See "Tools for genetics and genomics: Cytogenetics and molecular genetics", section on 'Array comparative genomic hybridization'.)

The two main platforms currently used for CNV detection are oligonucleotide arrays (oligonucleotides are stretches of DNA ranging from 25 to 60 base pairs) (figure 2), and single nucleotide polymorphism arrays (SNP arrays) [68]. There are approximately 10 million polymorphic SNPs throughout the human genome. Both SNPs and oligonucleotide arrays can detect copy number variations, but SNP arrays can be used in addition to determine lack of heterozygosity or presence of homozygosity, as seen in cases of consanguinity [69] and in cases of uniparental disomy [70] when there is inheritance of regions or entire chromosomes from one single parent instead of the normal biparental contribution. SNPs can also detect loss of heterozygosity (LOH) typically seen in somatic cancer cell changes. SNPs can also be very useful for the detection of somatic mosaicism, a situation where two or more cell lines can be present in a single individual, and triploidies, a rare instance when there could be a total of 69 chromosomes (3 haploid sets) that can be detected in the prenatal setting. Some current platforms are combining the use of array CGH and SNPs integrated in one single platform.

Other molecular diagnostic techniques — Other molecular techniques used for the detection of genomic disorders include fluorescence in situ hybridization (FISH), PCR-based studies like Q-PCR (quantitative PCR), and MLPA (Multiple Ligation Dependent Probe Amplification). Protocols using sequencing of the whole genomes (WGS) or exome (WES, sequencing all regions of the genome that encode proteins) will likely become viable options for detecting CNV variations [71,72]. (See "Tools for genetics and genomics: Cytogenetics and molecular genetics".)

●FISH – FISH uses larger stretches of DNA (approximately 50 to 200 kilobases of DNA) labeled with fluorescence reagents to target specific genome regions. The use of FISH requires knowledge of what specific area is being targeted and is dependent upon a clinical diagnosis [73,74]. It is generally reserved for specific situations such as translocations, fusions, or amplifications. (See "General aspects of cytogenetic analysis in hematologic malignancies", section on 'Methods of detection'.)

●MLPA – MLPA uses a cocktail of multiple probes available in kits and targets specific chromosomes or disease regions. A single reaction allows simultaneous hybridization of multiple probes to multiple regions or even multiple exons within a gene.

As the cost of microarrays and other molecular techniques is decreasing, FISH and MLPA are less frequently used.

A copy number gain can be the result of a chromosome duplication immediately adjacent to the area of interest, a marker chromosome (an extra structurally abnormal chromosome), or the result of an insertion or translocation.

●WGS – Whole genome sequencing (WGS) is a more comprehensive tool than whole exome sequencing (WES) for detecting single nucleotide variants and for studying patients with genetic disorders. WGS is also a more efficient tool to assess the presence of CNVs than WES.

As WGS cost lowers and analysis becomes more efficient, this technique may replace the use of chromosome microarray technology [75,76]. The usefulness of this approach was demonstrated in a study that used sequencing from 849 individuals to identify areas of CNVs and their role in gene dosage [71]. Genome sequencing has also been used to map breakpoints in the genome responsible for duplications of chromosomal regions [77].

Sequencing has the potential to provide improved resolution, but the sensitivity and specificity of sequencing for detecting CNVs is variable. WGS can allow simultaneous detection of single nucleotide and copy number variations, although the study cost is too high [71,78]. While WGS can accurately achieve CNV detection, the use of microarrays continues to be the gold standard.

Additional information about WGS and WES methods, collectively referred to as next-generation sequencing (NGS), is presented separately. (See "Next-generation DNA sequencing (NGS): Principles and clinical applications".)

SUMMARY

●Genomic disorders – Genomic disorders are diseases that result from copy number variations (CNVs). The most common and better-delineated genomic disorders are divided in two main categories: those resulting from copy number losses (deletion syndromes) and copy number gains (duplication syndromes). (See 'Introduction' above and "Congenital cytogenetic abnormalities".)

●Copy number variations – CNVs are generally submicroscopic genomic differences in the number of copies of one or more sections of DNA (figure 1). Some pathogenic CNVs cause syndromic disorders with consistent phenotypic features (also known as genomic disorders). Other CNVs are associated with disease susceptibility or resistance. (See 'Copy number variations' above.)

●How they cause disease – The main mechanisms that leads to disease in genomic disorders are deletions (copy number losses) and duplications/triplications (copy number gains) in dosage-sensitive genes. Other disease mechanisms include interference with imprinted genes, unmasking of recessive disorders, or disruption of neighbor regulatory elements outside of genes. (See 'Disease mechanisms' above.)

●Laboratory testing – Genomic disorders are typically detected by array comparative genomic hybridization (array CGH) also known as chromosome microarray. Next generation sequencing (NGS) is likely to become a popular way to detect CNVs, but chromosome microarray testing remains the standard of care. NGS may be done using whole exome sequencing (WES) or whole genome sequencing (WGS). Fluorescence in situ hybridization (FISH) is less commonly used, with a few exceptions such as in copy number gains and to rule out more complex chromosome abnormalities. (See 'Detection of genomic disorders' above.)