Clinical aspects

Description

	• Knowledge of the disease-related gene and nature of the lesion in it are important for understanding the manifestations of a disease and its management.
	• Next generation sequencing is important for diagnosis of neurogenetic disorders.
	• Even in the absence of knowledge of gene defect involved, molecular diagnostics can provide useful information about a disease with a genetic component.

If the identity of the defective gene and the nature of the gene are known, linkage studies are unnecessary. The family members can be tested directly for the presence of the lesion. If the identity of the affected gene is known but the lesion within it is not, several mutation screening methods can be used to detect the lesion to ensure that the subsequent predictive tests will be fast and reliable. Even if a disease-related gene is unknown, linkage analysis can be attempted to track the disease through the family pedigree. Such studies are probabilistic and their reliability depends on the mode of transmission of the disease, the penetrance of the disease, the pedigree structure, and the availability of family members.

Advances in molecular diagnostics are aimed not only at detection of disease, but also at elucidating pathomechanisms. In addition, these advances lay the groundwork for rational therapies. Discovery of genes for various diseases is important, although not essential, for the development of gene therapy.

Although tests are done to detect and confirm a disease, negative DNA tests are also important. It implies that the search must go on for other causes of the disease. A negative DNA test does not rule out other genetic causes.

Indications

Indications for the use of molecular diagnostics for neurogenetic disorders are:

• Unambiguous determination of the presence or absence of a mutation in an affected person or a carrier. • Determination of genetic risk to relatives of the person tested. • Differential diagnosis of disorders with similar phenotypes. • Desire to take preventative measures when diagnosis prior to onset of clinical manifestations can provide that opportunity.

Neurogenetic disorders that can be diagnosed by commercially available molecular diagnostic tests are listed in Table 1. Most of these can be diagnosed by direct mutation analysis whereas some deletions can be diagnosed by fluorescence in situ hybridization. Others, such as Norrie disease, require a DNA test incorporating single-strand conformation polymorphism and DNA sequencing. Diagnosis of some of the diseases will be discussed in more detail in the following text.

The European Federation of Neurological Societies guidelines on the molecular diagnosis of neurogenetic disorders are designed to provide practical help for the general neurologist to make appropriate use of molecular genetics in diagnosing neurogenetic disorders (14; 06; 07; 15).

Table 1. Neurogenetic Diseases That Have Been Diagnosed by Commercially Available Molecular Diagnostics

• Adrenoleukodystrophy • Alzheimer disease • Amyloid polyneuropathy type 1 • Angelman syndrome • Autism spectrum disorder • Canavan disease • Charcot-Marie-Tooth types 1A and X-linked • Dentatorubral pallidoluysian atrophy • Down syndrome • Duchenne/Becker muscular dystrophy • Emery-Dreifuss muscular dystrophy • Fascioscapulohumeral muscular dystrophy • Fragile X syndrome • Friedrich ataxia • Gaucher disease • Hereditary neuropathy with pressure palsies • Huntington disease • Hyperkalemic periodic paralysis • Kallmann syndrome • Kearns-Sayre syndrome • Kennedy disease (X-linked spinobulbar muscular atrophy) • Leber hereditary optic neuropathy • Lesch-Nyhan syndrome • Limb-girdle muscular dystrophy • Medium chain acyl-CoA dehydrogenase deficiency • Miller-Dieker lissencephaly • Mitochondrial encephalopathy with lactic acidosis, seizures, and stroke-like episodes • Multiple endocrine neoplasia 2b • Myoclonic epilepsy and ragged-red fibers • Myotonic dystrophy • Neurofibromatosis type 2 • Neuropathy, ataxia, retinitis pigmentosa • Norrie disease (monoamine oxidase deficiency) • Ornithine transcarbamylase deficiency • Pelizaeus-Merzbacher disease • Prader-Willi syndrome • Spinocerebellar ataxias (SCA-1,2,3,7,8) • Tay-Sach disease • von Hippel-Lindau disease

Alzheimer disease. Although usually referred to as a neurodegenerative disorder, familial form of Alzheimer disease has a genetic basis. Various molecular diagnostic tests for Alzheimer disease are shown in Table 2.

Table 2. Diagnostic Tests for Alzheimer Disease

Test	Basis and specimen	Comments
ApoE	ApoE genotyping to look for alleles associated with Alzheimer disease (E2, E3, E4) Blood, buccal swab	Adjunct to clinical assessment of dementia. Sensitivity 75% and positive predictive value 100%. Rules out Alzheimer with 95% accuracy in more than 60% of patients over 60 years of age who present with dementia.
Pronto ApoE	ApoE genotyping Blood	For detecting ApoE polymorphism in clinical trials of cholinesterase inhibitors in Alzheimer disease.
ADmark Tau and amyloid beta42	Tau protein CSF	Increased levels of tau and decreased levels of beta amyloid beta 42 are found in Alzheimer disease. CSF tau distinguishes between patients with AD and nondemented control subjects with 63% sensitivity and 89% specificity.
ADmark combined	Includes tau/amyloid beta42 and ADmark® ApoE CSF and blood
Mito-Load assay	mtDNA mutations Blood	Adjunct to increase the accuracy of clinical diagnosis.
ADmark PS-1	PS-1 gene mutations Blood	Detects chromosome mutations that are associated with early onset familial Alzheimer disease and is usually reserved for symptomatic patients with family history of the disease.
Amyloid precursor protein	Amyloid precursor protein mutation analysis Blood

A major portion of the heritability of Alzheimer disease remains unexplained by the currently known disease genes. Most of this "missing heritability" may be accounted for by rare sequence variants, which can now be assessed in unprecedented detail by availability of high-throughput sequencing technologies (03).

Biomarkers of Alzheimer disease. No diagnostic biomarker with 100% sensitivity has yet been developed for Alzheimer disease. The ApoE e4 allele is a risk factor rather than a disease gene and is useful for predicting response to certain drugs for Alzheimer disease.

There are several ongoing studies for biomarkers of Alzheimer disease. Most of these investigate pathomechanism and course of the disease. Thus far, no single biomarker can be used to diagnose Alzheimer disease definitively. Decreased cerebrospinal fluid amyloid beta42 concentration, but not amyloid beta40 concentration is a biomarker for Alzheimer disease. This amyloid beta42 concentration decrease in cerebrospinal fluid likely reflects precipitation of amyloid beta42 in amyloid plaques in brain parenchyma. This pathogenic plaque deposition begins years before the clinical expression of dementia in Alzheimer disease. Normal aging and the presence of the APOE4 allele are the most important known risk factors for Alzheimer disease. Cerebrospinal fluid amyloid beta 42 findings in the aged persons are consistent with acceleration by the APOE4 allele of pathogenic amyloid beta 42 brain deposition starting in later middle age in those with normal cognition. High plasma concentrations of amyloid beta40, especially when combined with low concentrations of amyloid beta42, indicate an increased risk of dementia.

CSF examination is not practical for population screening, and serum-based tests are considered, but they are less specific for neurologic disorders. The CSF biomarker signature of Alzheimer disease, defined by low amyloid beta1-42 and high tau in an autopsy-confirmed Alzheimer disease cohort and confirmed in the cohort followed in Alzheimers Disease Neuroimaging Initiative for 12 months, detected a mild form of the disease in a large, multisite, prospective clinical investigation and predicted conversion from mild cognitive impairment to Alzheimer disease (40). PET scanning using innovative imaging agents targeting amyloid beta and tau are promising to be useful for diagnosis of Alzheimer disease.

Proteomic approaches, microcapillary liquid chromatography mass spectrometry of proteins labeled with isotope-coded affinity tags, have been applied to quantify relative changes in the proteome of human CSF obtained from the lumbar cistern. Quantitative proteomics of CSF from Alzheimer disease patients compared to age-matched controls, as well as from other neurodegenerative diseases, will enable the generation of a roster of proteins that may serve as specific biomarker panels for Alzheimer disease. The plasma levels of 2 proteins, complement factor H and alfa2-M, are elevated in Alzheimer disease, and as they are present in amyloid plaques, the elevations may correlate with disease severity. Further studies will be required to fully validate the usefulness of these proteins as diagnostics. Biomarkers are needed that can detect Alzheimer disease in presymptomatic individuals (05).

Blood amyloid beta levels in humans do not reflect the amount of amyloid plaques in the brain and are neither sensitive nor specific for the clinical diagnosis of sporadic Alzheimer disease. In contrast to plasma, red blood cells in Alzheimer disease are characterized by low-density with increased volume and enhanced amyloid beta40 content (25).

Down syndrome. Down syndrome is a genetic disorder caused by the inheritance of 3 copies of the 21st chromosome. Individuals with trisomy 21 display complex phenotypes with differing degrees of severity. Numerous reliable methods have been established to diagnose the initial trisomy in these patients, but the identification and characterization of the genetic basis of the phenotypic variation in individuals with trisomy remains challenging. Methods that can accurately determine genotypes in trisomic DNA samples are expensive and require specialized equipment and complicated analyses. Genotyping can determine single nucleotide polymorphisms in trisomic genomic DNA samples in a simple and cost-effective manner. Duplex real-time PCR assays, based on relative quantification of DSCR4 gene on chromosome 21, can be used for rapid prenatal diagnosis of Down syndrome from amniotic fluid samples in clinical settings. The SensiGene T21 assay, which is based on massively parallel shotgun sequencing of maternal blood plasma, can detect trisomy 21 with 100% sensitivity and 99.7% specificity (13). This test needs to be validated in larger clinical studies.

Muscular dystrophies. From a genetic point of view, the best known of these is Duchenne-Becker muscular dystrophy.

Duchenne and Becker muscular dystrophy. The location of the dystrophin gene is on chromosome Xp21. This has been confirmed by DNA polymorphism linking. This is the largest gene known in the human genome, spanning 2.5 megabases. Most Duchenne muscular dystrophy cases are caused by out-of-frame mutations in the dystrophin gene followed by absence of dystrophin. In contrast, most Becker muscular dystrophy cases result from in-frame mutations that allow the expression of truncated partially active protein. An oligonucleotide-based array comparative genomic hybridization platform is useful in detecting submicroscopic copy-number changes involving the Duchenne muscular dystrophy gene as well as in providing more precise breakpoint identification at high-resolution and with improved sensitivity.

Multiplex polymerase chain reaction represents a sensitive and accurate method for deletion detection of most of the cases of Duchenne muscular dystrophy and Becker muscular dystrophy.

Duchenne muscular dystrophy-specific fluorescent in situ hybridization probes are useful for the detection of carriers of Duchenne muscular dystrophy gene deletions in preference to polymerase chain reaction.

Complete sequencing of Duchenne muscular dystrophy gene focuses on small point mutations and other mechanisms underlying complex rearrangements, which are relevant to prognosis, and may be useful in guiding new therapeutic approaches by defining the genetic defects (29). The Centers for Disease Control and Prevention-based Genetic Testing Reference Material Coordination Program has created a comprehensive Duchenne muscular dystrophy/Becker muscular dystrophy reference material panel using polymerase chain reaction and DNA sequence analysis (26). These samples are freely available from Coriell Institute for Medical Research (Camden, NJ) for use in quality assurance, proficiency testing, and research.

Some patients with Duchenne muscular dystrophy have mutations that alter RNA processing and/or expression. Identification of such mutations will likely require analysis of nonexomic sequence sources, including RNA and genomic noncoding DNA. In a case study, RNAseq enabled quick and accurate detection of a noncoding mutation in Duchenne muscular dystrophy (18).

Facioscapulohumeral dystrophy. This is one of the most common muscular dystrophies. DUX4, a gene responsible for facioscapulohumeral muscular dystrophy (FSHD), is normally silenced because it sits next to a telomere, a protective DNA sequence that caps the ends of chromosomes (41). As telomeres shorten with age, DUX4 expression climbs, which may explain the late onset of FSHD. DUX4 is upregulated over 10-fold in FSHD myoblasts and myotubes.

Spinal muscular atrophy. Molecular diagnosis of spinal muscular atrophy is based on use of polymerase chain reaction methods for detection of mutations in survival motor neuron 1 (SMN1) gene, the most frequent of which is absence of exon 7 in approximately 95% of patients. High-resolution melting analysis, a novel polymerase chain reaction without probe, has been applied successfully for detection of exon 7 deletions and other intragenic mutations in SMN1 (33).

Triple repeat disorders. Triple repeats are dynamic mutations because they show instability of alleles through generations, and the mutation rate is related to the copy number of repeats. The mutability of 1 of these sequences in the offspring is different from that in the parents. Thirteen neurologic diseases due to expansion of trinucleotide repeats are shown in Table 3. A distinctive feature of all these diseases is a phenomenon called anticipation, whereby the symptoms appear at earlier ages and with greater severity in successive generations. This phenomenon is due to the instability of expanded DNA sequence as it passes through the germ line and the number of repeats increases. Once these repeats reach a critical threshold, a hairpin structure is formed that disrupts normal DNA replication. This leads to a vicious circle of replication errors and further expansion. Expanded trinucleotide repeats are found in the general population and may be associated with age-related deterioration in health, suggesting that these repeats may confer susceptibility to common diseases that do not display a single-gene major inheritance.

Molecular diagnosis of triple repeat disorders is based on polymerase chain reaction amplification of the triple repeat in genomic DNA obtained from peripheral blood nucleated cells. This is a reliable as well as cost-effective method for the identification of the cytosine, adenine, and guanine expansion mutation.

Huntington disease. The gene for Huntington disease, which has been cloned, was mapped to the short arm of chromosome 4 using linkage analysis by polymorphic DNA markers. The mutation contains an unstable trinucleotide repeat (cytosine, adenine, and guanine) within a gene in the 4p16.3 chromosome. Because the tip of chromosome 4 contains 50 to 100 genes, researchers have not yet been able to precisely localize the Huntington disease gene. Nevertheless, it is known that the disease-causing mutation expands the length of a repeated stretch of amino acid glutamine in the gene's product, the Huntingtin protein. Although this development may trigger the onset of Huntington disease, other genetic, neurobiological, and environmental factors may also contribute to the progression of the illness and underlying neuronal degeneration. A Huntingtin-associated protein has been identified that binds to huntingtin; this binding is enhanced by an expanded polyglutamine repeat, the length of which correlates to the age of disease onset. The Huntingtin-associated protein is enriched in the brain, suggesting a possible role for selective brain pathology in Huntington disease development.

Predictive testing for Huntington disease had been available for some time before the Huntington disease gene was cloned. In these procedures, polymorphic markers, flanking the Huntington disease gene and located some distance from it, were used to track the disease allele through affected pedigrees. This indirect method yielded probabilistic results. Direct mutation analysis of the Huntington disease gene is now possible and gives more accurate results. Measurement of the number of cytosine, adenine, and guanine repeats in the Huntington disease gene represents an effective, direct test with which to confirm the clinical diagnosis in difficult cases.

Although polymerase chain reaction-based techniques enable presymptomatic diagnosis, the ethical aspects of this screening remain controversial. Diagnosis of individuals who are destined to develop this disease, for which no cure exists, is considered to place intense psychological stress on the patient. Surprisingly, adverse effects are rare among individuals who receive the news that Huntington disease will develop in the future. The explanation for this situation probably lies in the psychological robustness of those taking the tests and the careful protocols that are followed. Genetic testing for Huntington disease is a success story so far and should serve as a model for presymptomatic testing of other adult-onset presymptomatic disorders. The region around and within the cytosine, adenine, and guanine repeat sequence in the Huntington disease gene is a hot spot for DNA polymorphisms, which can occur in up to 1% of subjects tested for Huntington disease. These polymorphisms may interfere with amplification by polymerase chain reaction, and so have the potential to produce a diagnostic error. PCR protocols are available for rapid molecular diagnostics of triplet expansion diseases.

Retrospective and clinical data from individuals with manifest Huntington disease show psychiatric and cognitive symptoms that are common in Huntington disease gene carriers, with earlier onsets associated with longer CAG repeats (32). Of patients with Huntington disease, 42.4% reported at least 1 psychiatric or cognitive symptom before motor symptoms, with depression most common. Each nonmotor symptom was associated with significantly reduced total functional capacity.

Fragile X syndrome. Molecular diagnosis of fragile X syndrome has been developed using Southern hybridization methods or polymerase chain reaction approaches. Cytogenetics, which has been the major diagnostic tool used for fragile X syndrome until the discovery of the FMR1 gene, is no longer required to establish the diagnosis but is still considered to be a supplementary test to rule out any other chromosomal abnormalities in a child with mental retardation. DNA testing was found to be more sensitive than cytogenetic analysis (100% vs. 50%) and more cost effective. One drawback of DNA-based diagnosis is that IQ scores cannot be predicted, and for this the emerging technique of FMR1 protein quantification in target tissues should be considered. Simple polymerase chain reaction combined with blood spot analysis could be a reliable, inexpensive test for fragile X syndrome that is feasible for a large-scale screening of male subjects with mental retardation, but Southern blot assay with mixed deoxyribonucleic acid is appropriate for screening female subjects. A cheap as well as rapid PCR-based screening method has been developed for identification of all expanded alleles of the fragile X gene in newborns and is suitable for screening of high-risk populations (42). A dual-mode single-molecule fluorescence assay has been developed that enables acquisition of 2 parallel, independent measures of CGG repeat number (09). This strategy may be useful for identifying heterozygosity or for screening collections of individuals.

Spinocerebellar ataxias. Disorders of over 31 genes underlie cerebellar and brain stem dysfunction in spinocerebellar ataxias. Molecular diagnostics, by identification of the causative molecular deficits, enables determination of the different spinocerebellar ataxia subtypes and facilitates genetic counseling (31).

Next-generation sequencing has been shown to be useful in 50 heterogeneous patients with ataxia who had been extensively investigated and were refractory to diagnosis (36). Functional analysis of data predicted 13 different mutations in 8 different genes: PRKCG, TTBK2, SETX, SPTBN2, SACS, MRE11, KCNC3, and DARS2 of which 9 were novel, including one causing a newly described recessive ataxia syndrome.

Table 3. Neurologic Diseases Due to Expansion of Trinucleotide Repeats

Disease	Site of repeat	Effect on gene expression
Fragile X syndrome (FRAXE) Also FRA 16A and FRAXF	CGC repeats 5' to translation initiation site	Failure of expression of FMR-1 or FMR-2 genes
Huntington disease	CAG repeats within gene	Normal gene expression
Myotonic dystrophy	Cytosine, thymine and guanine repeats in 3' untranslated region of DM-1 gene	Normal or decreased
Spinal and bulbar muscular dystrophy (Kennedy syndrome)	CAG repeat within androgen-receptor gene	Normal gene expression
Spinocerebellar ataxia type 1	CAG repeats within gene	Normal gene expression
Spinocerebellar ataxia type 2	CAG repeats within gene	Normal gene expression
Spinocerebellar ataxia type 3 (SCA-3)	CAG repeats within gene	Normal gene expression
Spinocerebellar ataxia type 7 (SCA-7)	CAG repeats within gene	Normal gene expression
Spinocerebellar ataxia type 8 (SCA-8)	CAG/CTG repeats	Affects an antisense transcript that could in turn regulate expression of an overlapping protein-coding gene
Dentatorubral-pallidoluysian atrophy	CAG repeats within gene	Normal gene expression
Haw River syndrome *	CAG repeats within gene	Normal gene expression
Machado-Joseph disease	Expansion of a CAG repeat in the open reading frame of a novel gene on chromosome 14q32	Normal gene expression
Friedreich ataxia	Guanine, adenine, and adenine triple repeat	Reduced frataxin expression neuro-degeneration

*Haw River syndrome resembles Dentatorubral-pallidoluysian atrophy clinically except for the absence of myoclonic seizures but differs histopathologically by the presence of extensive demyelination of the subcortical white matter, basal ganglia calcification, and neuroaxonal dystrophy.

Hereditary spastic paraplegias. These disorders, caused by mutations in multiple genes, are characterized by progressive spasticity and weakness of the lower limbs, and may be difficult to differentiate from other neurogenetic disorders such as hereditary ataxias. Dealing with these complex situations requires high-quality, curated mutation databases for developing adequate diagnostic guidelines and correct interpretation of genetic testing (04).

Mitochondrial disorders affecting the nervous system. Mitochondria generate energy for cellular processes by producing adenosine triphosphate through oxidative phosphorylation. These organelles contain their own extrachromosomal DNA, which is distinct from DNA in the nucleus. Diseases, particularly those affecting organs with high-energy requirements such as the brain and the muscles, have been linked to defects in the mtDNA. Features that suggest a mitochondrial origin for disease are maternal inheritance and defect in mitochondrial oxidative phosphorylation. There is increasing acceptance of the role of defective mitochondrial energy production and the resulting increased level of free radical production in the pathogenesis of various neurodegenerative disorders such as Huntington disease, Parkinson disease, and amyotrophic lateral sclerosis. These defects may contribute to both excitotoxic and oxidative damage. The evidence comes from a similarity to known mitochondrial disorders including delayed and variable age of onset, slow progression, and symmetric degeneration of localized groups of neurons. More than 50 mtDNA mutations are known and some mtDNA-associated diseases with neurologic implications are shown in Table 4.

A complete human mtDNA sequence is available and a set of sensitive and specific molecular genetic tests has been developed for several mitochondrial diseases. In suspected mitochondrial disease, a small portion of the muscle biopsy is frozen in liquid nitrogen for DNA isolation. Because most mtDNA mutations occur in relatively few families, it is important to do a comprehensive analysis of mtDNA by single-strand conformation polymorphism and sequencing to rule out a mtDNA mutation as a cause of symptoms in an individual. The diagnosis is based on integrated clinical genetic and biochemical information. The unequivocal establishment of the diagnosis of mitochondrial disease by mtDNA examination is a prerequisite for proper genetic counseling and, eventually, for treatment. The data generated by the DNA 500 and DNA 1000 assays using the Agilent 2100 bioanalyzer (Agilent Technologies, Palo Alto, California) show a lower percentage error and a better reproducibility as compared to those obtained by the conventional methods such as restriction fragment length polymorphism analysis.

Comprehensive molecular analysis of mitochondrial respiratory chain disorders should include qualitative identification of the mutation and quantitative measurement of both the degree of mutant heteroplasmy and the total amount of mtDNA. Following identification of clinical syndromes or characteristic respiratory chain complex deficiencies, direct sequencing of the specific causative nuclear gene(s) can be performed on white blood cell DNA.

Although identification of pathogenic mutations is usually accepted as definitive, the large number of candidate nuclear genes and the involvement of 2 genomes frequently provide challenges for successful molecular diagnostic confirmation, but next generation sequencing technologies should improve diagnostic accuracy (19).

A rapidly expanding catalog of neurogenetic disorders has encouraged a diagnostic shift towards early clinical whole exome sequencing (WES). Adult primary mitochondrial diseases frequently exhibit neurologic manifestations that overlap with other nervous system disorders. However, mitochondrial DNA (mtDNA) is not routinely analyzed in standard clinical WES bioinformatic pipelines. A study reanalyzed 11,424 exomes enriched with neurologic diseases, and it detected pathogenic mtDNA variants and different mtDNA mutations in 64 exomes, 11 of which were considered disease causing based on the associated clinical phenotypes (38). These findings highlight the diagnostic uplifts gained by analyzing mtDNA from WES data in neurologic diseases.

Table 4. Neurologic Disorders Due to Mitochondrial DNA (mtDNA) Mutations

I. Mitochondrial myopathies
	Defect:
	• Large single deletions in mtDNA that arise sporadically by an unknown mechanism. A frequent point mutation occurs at base pair 3243
	Clinical features:
	• Chronic progressive external ophthalmoplegia • Kearns-Sayre syndrome, plus retinopathy cardiac conduction defects and ataxia
II. Mitochondrial encephalomyopathies
	Defect:
	• tRNALys gene mutation • tRNALeu(UUR) gene mutation (base pair 3243 in 80% of cases) • Mutations at base pair 8993 in the adenosine triphosphate 6 gene
	Clinical features:
	• Myoclonic epilepsy with ragged-red fibers • Mitochondrial encephalopathy with lactic acidosis, seizures, and stroke-like episodes (MELAS syndrome) • Neuropathy, ataxia, and retinitis pigmentosum. Similar mutations are present in Leigh disease.

Epilepsy. Several genetic loci of epilepsy are known, but commercial DNA diagnostics for asymptomatic mutations are not available to test for the known epilepsy genes for the following reasons:

	• Because of the large number of mutations associated with each gene, a considerable amount of DNA sequencing may be needed for accurate identification.
	• Diagnostic procedures for this situation would not be cost effective.
	• Most of the disorders for which genes have been identified are relatively benign conditions that can be treated with existing antiepileptic drugs, and identification of a specific mutation would not change the therapy.

However, an example of the usefulness of the knowledge of the genetic basis of epilepsy is seen in progressive myoclonic epilepsy. The gene for this encodes cystatin B, a protease inhibitor. A related protein, cystatin C, has been found in deposits in affected brain arteries in cerebral amyloid angiopathy. Patients with this type of epilepsy respond to the antiepileptic drug sodium valproate but show toxic effect to phenytoin. The identification of mutant genes encoding cystatin B may help us to understand the differential response to these 2 drugs. This knowledge also provides a biochemical pathway and a molecular target for devising treatment of epilepsy. Pharmaceutical companies are stepping up programs that utilize genomic strategies for epilepsy drug discovery. Supposing that a specific mutation, such as a calcium channel defect, is linked to human epilepsy, it would be simple to make a molecular diagnosis. Such a patient would be expected to respond to a drug aimed at the calcium channel. Molecular diagnosis of epilepsy is expected to be available within the next 5 years.

Whole exome sequencing has shown that SCN2A mutations are an important genetic cause of Ohtahara syndrome, an early-onset epileptic encephalopathy (34). In view of a broad spectrum of disorders associated with SCN2A mutations, genetic testing these mutations should be considered for children with different epileptic conditions. Exome sequencing was shown to be a useful diagnostic tool in severe early-onset epilepsy, with positive novel genetic etiology shown in 7% of patients in 1 series, and the frequency was highest (17%) in patients with epileptic encephalopathies (22).

Hereditary metabolic disorders with neurologic manifestations. Several hereditary metabolic disorders can be diagnosed by laboratory methods of detecting biochemical disturbances. A gene panel for lysosomal storage disorders results in an increased diagnostic yield in comparison to classical biochemical testing because the panel is able to cover a wider range of diseases. This method has been proposed as the initial test in cases of an nonspecific presentation of these diseases, whereas enzymatic testing remains the first choice in patients with a more distinctive clinical presentation (17). Genes for some of these disorders have been localized, and molecular diagnostic methods have been used for the detection of these diseases as shown in the following examples.

Lesch-Nyhan syndrome. This is an X-linked recessive disease due to deficiency of hypoxanthine-guanine phosphoribosyltransferase resulting in an increase of purine biosynthesis. The native HGPRT gene has been localized to chromosome Yq26-q27. It is possible to detect carrier females as well as affected males using enzyme assay, or DNA diagnostic methods, or both.

Gaucher disease. Molecular diagnosis of Gaucher disease is possible by duplex-polymerase chain reaction test. This involves amplification of the genomic DNA and hybridization to oligonucleotide probes. Signals are detected by chemiluminescence. The disease is worth diagnosing because it can be treated.

Molecular diagnosis of Gaucher disease is complicated by the presence of a pseudogene near the true gene on chromosome 1q21. Selective polymerase chain reaction amplification of the true gene can be accomplished by specially designed oligonucleotide primers.

Pompe disease (infantile acid maltase deficiency). DNA analysis is now the standard diagnostic procedure for confirmation, for carrier detection, and for genetic counseling. For enzymatic analysis, use of glycogen as a natural substrate in the reaction mixture adds to the selectivity of using mixed leukocyte preparations as diagnostic material.

Hereditary motor and sensory neuropathies. The molecular basis of several of these neuropathies is now known. The most important of these are Charcot-Marie Tooth disease and hereditary neuropathy with liability to pressure palsies. Peripheral myelin protein is a component of peripheral nervous system myelin and these disorders are associated with mutations of peripheral myelin protein 22 gene and have been called myelin disorders or "myelinopathies."

Charcot-Marie-Tooth disease. The genes for this disease have been mapped to chromosome 17 (Charcot-Marie-Tooth 1A), chromosome 7 (Charcot-Marie-Tooth X), and another unknown chromosome (Charcot-Marie-Tooth 1C). It has been found that 70% to 80% of patients with a clinical diagnosis of Charcot-Marie-Tooth 1 carry the 17p11.2-12 duplication, implying that an assay for duplication provides a powerful marker for screening suspected patients and family members at risk. Other forms of Charcot-Marie-Tooth are associated with mutations in the myelin protein Z (Charcot-Marie-Tooth 1B) and Cx32 (Charcot-Marie-Tooth X) genes. Thus, mutations in different genes can cause similar Charcot-Marie-Tooth phenotypes. Mutations in the peripheral myelin protein 22 and myelin protein Z genes can also cause the related but more severe neuropathy, Dejerine-Sottas syndrome. All genes so far identified by Charcot-Marie-Tooth investigators appear to play an important role in myelin formation or maintenance of peripheral nerves.

Conventional disease detection depends on electrodiagnostic tests including EMG and nerve conduction velocity measurements. The isolation of genes underlying these conditions has facilitated both the differential and molecular diagnosis of these disorders by single-stranded conformation polymorphism, direct DNA sequencing, or both. Other methods used to detect CMT1A gene duplication are pulsed field gel electrophoresis, restriction fragment length polymorphism, and fluorescence in situ hybridization. Another approach is the use of Southern blot and amplification by polymerase chain reaction of polymorphic poly repeats (microsatellites) located within the duplicated region, or the detection of junction fragments specific for the duplication. A pulsed field gel electrophoresis-based CMT1A DNA test is available. It detects CMT1A duplication in 70% to 90% of the patients. These methods require radioactive markers or other complicated procedures and are time-consuming and labor-intensive. A polymerase chain reaction-based test provides results within 24 hours for detection of a recombinant hotspot associated with CMT1A duplication. Another diagnostic strategy uses highly polymorphic short tandem repeats located inside the CMT1A duplicated region. Combined use of the 3 short tandem repeats allows robust diagnosis, which is almost completely informative. As DNA testing for Charcot-Marie-Tooth 1A becomes more widely available, it may become an accepted part of the evaluation of any patient with a suspected hereditary neuropathy.

Charcot-Marie-Tooth type 1 disease has been associated with 280 mutations in the GJB1 (gap junction protein, beta 1) gene. High-resolution melting analysis is a simple, sensitive, and cost-efficient alternative method to scan for gene mutations in the GJB1 gene.

Hereditary neuropathy with liability to pressure palsies. Hereditary neuropathy with liability to pressure palsies is a dominantly inherited disorder that presents as recurrent mononeuropathies precipitated by trivial trauma. Hereditary neuropathy with liability to pressure palsies has been attributed to a 1.5 mb deletion in 17p11.2 spanning the peripheral myelin protein 22 gene. Underexpression of the peripheral myelin protein 22 gene causes hereditary neuropathy with liability to pressure palsies just as overexpression of peripheral myelin protein 22 causes Charcot-Marie-Tooth 1A. Thus, 2 different phenotypes can be caused by dosage variations of the same gene.

Analysis of DNA can be used to detect the clinically unaffected members of families. Genetic marker screening is an efficient diagnostic strategy; in about 90% of cases, genetic marker screening reveals the presence or absence of 17p deletion in hereditary neuropathy with liability to pressure palsies. Families with hereditary neuropathy with liability to pressure palsies have been identified that lack the 17p deletion and possess 2 normal copies of peripheral myelin protein 22 gene, and it is possible that mutations are present in other parts of the gene that have not yet been examined.

The understanding of the molecular basis of Charcot-Marie-Tooth 1 and related disorders has allowed accurate DNA diagnosis and genetic counseling of inherited peripheral neuropathies and will make it possible to develop rational strategies for therapy. As several loci for Charcot-Marie-Tooth 2 have been identified, the genes responsible for Charcot-Marie-Tooth 2 will most likely be disclosed using positional cloning and candidate gene approaches in the future. Whole genome sequencing has identified all potential functional variants in genes likely to be related to the disease in a family with a recessive form of Charcot-Marie-Tooth disease for which the genetic basis had not been identified, and these variants were genotyped in the affected family members (28).

Real-time quantitative polymerase chain reaction is sensitive for identifying peripheral myelin protein 22 (PMP22) gene copy number in CMT1A duplication and hereditary neuropathy with liability to pressure palsies gene deletion.

Inherited neuropathies due to metabolic disorders. Molecular genetics techniques have enabled the identification of the mutations associated with neuropathies due to metabolic disorders such as porphyrias, leukodystrophies, and other storage diseases. Inherited amyloid neuropathies may be caused by mutations in 3 different genes: those for gelsolin, apolipoprotein A1, and, most frequently, transthyretin. These advances have improved the diagnosis and genetic counseling of these patients.

Parkinson disease. Mutations associated with Parkinson disease have been reported in at least 10 genes of which the most frequent are parkin and LRKK2. Tests are available for these genes but are useful only in subjects who develop the disease before the age of 30 years. The G2019S mutation in the LRRK2 gene, the most common known cause of Parkinson disease, is widely available as a molecular clinical test for this disease.

Occurrence of 22q11.2 deletions represent a novel genetic risk factor for early-onset Parkinson disease. It is recommended that individuals with classic features of 22q11.2DS should be considered for genetic testing, and those with a known 22q11.2 deletion should be monitored for the development of symptoms of Parkinson disease (08).

Whole-exome sequencing has been successful in identifying genetic factors contributing to familial or sporadic Parkinson disease. A study has identified 12 genes with de novo mutations (MAD1L1, NUP98, PPP2CB, PKMYT1, TRIM24, CEP131, CTTNBP2, NUS1, SMPD3, MGRN1, IFI35, and RUSC2) that could be functionally relevant to pathogenesis of Parkinson disease and point to NUS1 as a candidate gene (20).

Detection of risk genes for attention deficit hyperactivity disorder (ADHD). Genotyping microarrays have been used to identify rare copy number variations in brain-expressed genes, some of which overlap with those found in autism spectrum disorder as well as other neuropsychiatric disorders indicating common underlying risk genes in these conditions (27).

Autism spectrum disorder. Hundreds of genes are implicated in autism spectrum disorder. A study has analyzed gene expression within the blood of toddlers, both with and without autism. Combining the data collected from each childs blood sample with neuron models, the authors discovered an important gene network containing key signaling pathways for fetal brain development that is disrupted in cases of autism spectrum disorder (16). The findings indicate that the main cause of autism is likely a faulty version of this gene network affecting prenatal brain development. They also found that the worse the regulation of this network, the more severe the autism spectrum disorder symptoms experienced later on. No practical molecular diagnostic tests have yet been developed for autism spectrum disorder.

Molecular diagnosis of craniosynostosis. Current diagnostic methods include a thorough clinical evaluation with an adjunctive imaging confirmation. However, genetic factors play an important role in the pathogenesis of craniosynostosis, which has been associated with more than180 syndromes, including Apert and Crouzon syndromes, and can have single- or multisuture involvement. In about 75% to 85% of these syndromic patients, monogenic mutations or chromosomal defects have been identified. In a study, patients with craniosynostosis were tested by an in-solution hybrid capture method using a customized panel covering 34 craniosynostosis-related genes and showed the wide genomic landscape of craniosynostosis that revealed various genetic factors for craniosynostosis pathogenesis (44). In addition, diagnostic workup using target panel sequencing was found to be helpful in the molecular diagnosis of craniosynostosis.

Results

At a tertiary care center, neurogenetic tests yielded a positive result in 21.5% of patients without previously identified familial mutations (12). These results can be used to help establish utilization guidelines for neurogenetic testing.

The advantages of molecular diagnosis in neurogenetic disorders are as follows:

	• High specificity of molecular diagnostics enables screening of large populations for carriers.
	• In contrast to other applications, molecular genetic tests are not usually directed at any anatomical lesion because DNA obtained from any site in the body is equally valid.
	• As a basis for gene therapy. If a normal gene has been cloned to be used as a probe for disease-causing mutations, the same normal sequence can be used to replace the mutated sequence in the patient.
	• By making a definite diagnosis and eliminating some of the alternative possibilities, molecular diagnosis enables a more precise determination of the prognosis. An example of this is Kennedy syndrome, where a positive DNA diagnosis would eliminate all other possibilities including amyotrophic lateral sclerosis, a condition with a worse prognosis.

The main limitations of molecular diagnostics for genetic disorders are as follows:

	Heterogeneity of genetic changes that underlie inherited disorders. The mutations can be so diverse that no 2 persons will have the same change. This hinders the construction of a molecular diagnostic test that is applicable to all patients with a certain disease. For example, 2 distinct genes have been implicated in tuberous sclerosis in different families.
	Masking of the gene deletion by a normal counterpart. Non-sex-linked genes come in pairs so that the normal gene can mask loss of part or all of the other copy. If a molecular probe for a gene is hybridized to a Southern blot of DNA from a person with deletion of that gene, the unaffected copy will produce a normal hybridization band.
	Screening the population at large for the FMR1 gene is a controversial issue both practically and ethically. This is a situation where molecular technology and its potential applications have leapt far ahead of informed medical and public debate. Some of the concerns that have been raised include the difficulty of phenotype prediction in a female fetus with a full mutation; mosaicism for permutation in a male fetus; and the lack of accurate information about the risk of a female carrier of a permutation having an affected child. If population screening were restricted to a search for full mutation in the newborn male, accurate prognosis could be given supplemented by an offer of investigation of the extended family.
	Advances in genetics and genomics have resulted in a dramatic growth of genetic testing in the health care system. The rapid development of new technologies, however, has also brought challenges, including the need for evidence-based evaluation of the validity and utility of genetic tests, questions regarding the best ways to incorporate them into medical practice, and how to weigh their cost against potential short- and long-term benefits. A report on evidence framework for genetic testing has been published (35).

Adverse effects

No adverse effects have been reported to be associated with molecular diagnostics.

Special considerations

Prenatal diagnosis by molecular methods is available for the following 5 neurogenetic disorders: (1) Down syndrome, (2) fragile X syndrome, (3) Huntington disease, (4) muscular dystrophy, and (5) adrenoleukodystrophy.

Technical advances have enabled in vitro fertilization and blastomere biopsy of the 6 to 8 cell embryo and single cell DNA analysis. X-linked diseases can be diagnosed at this stage. This makes it possible for couples to exclude the risk of transmission of an X-linked disease, and to start a pregnancy knowing that their child will not be affected. A rapid polymerase chain reaction-based technique, when applied to the embryo diagnosis, allows the avoidance of offspring affected with an X-linked disease by transferring only female embryos for implantation and ensuing pregnancy. Preimplantation diagnosis can be made of several neurologic disorders that include the following:

Charcot-Marie-Tooth disease (type 1A) Down syndrome Duchenne muscular dystrophy Fragile X syndrome Huntington chorea Myotonic dystrophy Spinal muscular atrophy TaySachs disease

Scientific basis

	Molecular diagnosis of neurogenetic disorders involves several techniques, biomarkers, and next-generation sequencing.
	Application of bioinformatics will not only facilitate diagnosis but will also improve the understanding of gene disease associations.
	Cytogenetics includes techniques for chromosomal analysis.

Molecular diagnosis of neurogenetic disorders involves several techniques, which have been described in detail elsewhere (24). Apart from advances in molecular diagnostic techniques, biomarkers, and next-generation sequencing, application of bioinformatics will not only facilitate diagnosis, but will also improve the understanding of gene disease associations. Cytogenetics, which includes techniques for chromosomal analysis, and sequencing will be mentioned here briefly.

Chromosome analysis. This requires the provision of a karyotype (individual's chromosome constitution), a DNA probe, and a cell fusion procedure using human and nonhuman cells. Various procedures for chromosome analysis are:

	Somatic cell hybrids (human cells, such as fibroblasts, fused with tumor cells from other animals, such as rodents, using a viral agent) have been used for chromosome analysis. These cell lines differ from each other in the number of human chromosomes that they have retained and that can be used for mapping purposes. A DNA probe can be hybridized to a panel of somatic cell hybrids and enable determination of chromosomal origin for the DNA probe.
	Flow cytometry can be used to analyze and sort large numbers of chromosomes in a short period of time with high reproducibility. In contrast to conventional slide-based microscopy, chromosomes are isolated as a suspension, stained with 1 or more fluorescent DNA dyes, and passed singly through 1 or 2 laser beams. The intensity of the fluorescent signal from each chromosome is recorded, the value being dependent on the DNA content of each chromosome. The data are presented in the form of histograms. This technique provides the possibility of physically separating individual chromosomes.
	Polymerase chain reaction has enabled molecular studies on a much smaller number of chromosomes. A library of chromosome-specific probes can be generated from as few as 500 flow-sorted chromosomes. In another polymerase chain reaction approach, the amplified DNA fragments are biotinylated and used to perform in situ hybridization and the fluorescent signals are shown to be restricted to the original sorted chromosomes. This technique can show changes that are not detectable by conventional or flow cytogenetics. The presence or absence of a DNA sequence can also be assessed by polymerase chain reaction with primers specific to that sequence. A repeat-primed polymerase chain reaction has been used to test for a large hexanucleotide repeat expansion (GGGGCC) within C9ORF72 on chromosome 9p21, which accounts for approximately 40% of cases of familial amyotrophic lateral sclerosis and 30% of cases of familial frontotemporal dementia (39).The hexanucleotide repeat expansion is not found in persons with Alzheimer disease, and its absence serves to differentiate it from frontotemporal dementia, which is often misclassified as Alzheimer disease on basis of clinical examination (30).
	Fluorescent in situ hybridization has an important role in mapping and ordering DNA probes along a chromosomal segment. Fluorescent in situ hybridization has been used to detect expanded trinucleotide repeats in diseases, such as fragile X syndrome, using a microscope unaided by image analysis.
	Comparative genomic hybridization is a modified in situ hybridization technique, which allows detection and mapping of DNA sequence copy differences between 2 genomes in a single experiment. In comparative genomic hybridization analysis, 2 differentially labeled genomic DNA (study and reference) are cohybridized to normal metaphase spreads. Chromosomal locations of copy number changes in the DNA segments of the study genome are revealed by a variable fluorescence intensity ratio along each target chromosome. Array comparative genomic hybridization has much greater multiplexing capabilities than fluorescent in situ hybridization. This technology has the potential to examine many regions of the genome simultaneously for changes in DNA copy number and identify complex patterns of gains and losses within the genome. This method enables determination of trisomy 13, 18, 21, and monosomy X, as well as normal ploidy levels of all other chromosomes from single fibroblasts within a day.

Sequencing. It is difficult to find the location of a gene buried in the tangle of chromosomal DNA in the nucleus. Sequencing of individual nucleotide bases may be required. A candidate gene sequence analysis involves analyzing the sequence of a gene for the protein plausibly involved in the pathogenesis of a disease. DNA sequence analysis is a multistep process comprising sample preparation, generation of labeled fragments by sequencing reactions, electrophoretic separation of fragments, data acquisition, assembly into a finished sequence, and most importantly, functional interpretation. Automated sequencers are now available.

Types of mutations associated with genetic disease that involve some defect at DNA level include point mutations, deletions, insertions, and expansion of unstable repeat sequences. There are several mechanisms by which mutations can give rise to an inherited disease. These may lead either to a loss of gene function or a gain of gene function with synthesis of a gene product that is abnormal.

Direct diagnosis of monogenic neurologic diseases is possible if the genes are known. Indirect diagnosis can be made in cases where the disease gene has not yet been identified but the disease locus has been assigned to a chromosomal locus. Indirect diagnosis is based on the analysis of segregation of polymorphic DNA markers linked to the disease locus within a family.

DNA sequencing now plays an important role in diagnosis of genetic disorders as the cost is becoming more affordable. Next generation sequencing has opened a new era in neurogenetics and successfully demonstrated the feasibility of using this platform for multiplex genetic tests for several rare diseases along with the use of cloud computing for bioinformatics analysis of data in clinical laboratories (43). Massive parallel sequencing enables the screening of a large number of genes, or of the entire genome, or all of the coding regions as well as rapid identification of causative genes and a more precise genetic diagnosis of many neurologic disorders (11). Thus, genetic testing could be considered earlier in the diagnostic approach to a patient.

Whole exome sequencing is sequencing of the coding regions, or exons, of an entire genome from a single individual. Whole exome sequencing is shown to be useful in differentiating between mitochondrial leukoencephalopathies and Alexander disease, an astrogliopathy (37).

Exome and whole-genome sequencing are becoming increasingly routine approaches in Mendelian disease diagnosis. Application of transcriptome sequencing (RNA sequencing) as a complementary diagnostic tool enabled identification of intronic mutation in COL6A1 in genetically unsolved patients in an external collagen VI-like dystrophy cohort, thus explaining approximately 25% of patients who were clinically suspected of having this rare muscle disorder, but previous genetic analysis was negative (10).

Genomic DNA methylation signatures. Some neurodevelopmental syndromes present with systemic, complex, and often overlapping clinical features due to Mendelian inheritance of mutations in genes involved in DNA methylation and histone modifications, with specific DNA methylation signatures. A machine learning tool can be built by combining these epi-signatures to concurrently screen for multiple syndromes with high sensitivity and specificity, which is useful for solving ambiguous cases and subjects presenting with the overlapping clinical and molecular features associated with the disruption of the epigenetic machinery (01).