Molecular genetic diagnosis of hereditary neuropathies

(1)

This chapter is divided into two parts. The first provides a link between phenotypic characteristics and molecular genetic tests, allowing the clinician to order the appropriate genetic tests. This part will largely be presented as a decision tree (Fig. 11.1) for HMSN and a table (Table 11.1) listing characteristic and peculiar clinical features of the different forms of hereditary neuropathies and the corresponding most appropriate genetic tests.

The second part presents the different molecular genetic testing methods themselves with special emphasis on the different principles used to detect the chromosome 17p CMT1A duplication and HNPP deletion.

Ethical issues of molecular genetic testing are dealt with in a separate chapter.

Centers offering molecular genetic tests are listed in the appendix of this book.

11.1 Molecular genetic testing strategies

If the diagnosis hereditary neuropathy is suspected on the basis of typical clinical features and/or a positive family history, the following questions have to be answered before proceeding to molecular genetic tests:

z Is the disease a pure peripheral neuropathy or do patients have additional symptoms or signs? Do they have features of CNS involvement? If CNS symptoms are present, or even prominent, a broad spectrum of other neurological diseases has to be considered, which includes numerous diseases not covered in this book.

z Does the neuropathy involve the motor and sensory system (HMSN), is it a pure motor neuropathy (distal HMN) or a predominantly sensory and/or autonomic neuropathy (HSAN)? Hereditary neuropathy with liability to pressure palsy (HNPP) and hereditary neuralgic amyotrophy (HNA) can often be diagnosed on clinical grounds alone. The most important differential diagnosis of HNA is the much more common sporadic form of neuralgic amyotrophy most often manifesting as a monophasic plexopathy. At present, these diseases can only be differentiated by the family history and the usually monophasic course of the sporadic

of hereditary neuropathies

G. Kuhlenbåumer

(2)

Fig.11.1.Orderingtheappropriatemoleculargenetictest:adecisiontreebasedapproach

11.111.1 (allelictoHMNV) dHMNI dHMNII

(3)

Table 11.1. Characteristic clinical features and genetic tests of non-HMSN hereditaryperipheral neuropathies

Clinical picture/other characteristic features IH Form Gene/Test Prominent/exclusive sensory symptoms

z Prominent sensorysymptoms, but also motor,

earlyonset, sometimes focallyfolded myelin AR CMT4F PRX z Prominent sensorysymptoms, but also motor,

acral ulcerations AD CMT2B RAB7

z Lancinating pain, painless acral ulcerations, sensorydeficit affects mainlypain and temperature

AD HSAN1 SPTLC1

z Neonatal onset, sweating, high blood pressure,

sensory+motor, Ashkenazi Jews AR HSAN3 IKBKAP

z ªCongenital insensitivityto painº (CIPA), fever,

anhidrosis, mental retardation, reduced lifespan AR HSAN4/5 NTRK1 Prominent/exclusive motor symptoms

z Upper limb preponderance, maybe onlymotor

or motor and sensoryin the same family AD CMT2D/HMN V GARS z Onset before 20 y, distal weakness, often scolio-

sis, slow progression AD dHMN I HSPB1/HSP27

z Adult onset, fast progression of distal weakness

over *5 years AD dHMN II HSPB8/HSP22

z Congenital/veryearlyonset, diaphragmatic/vocal

cord involvement, reduced life expectancy AR dHMN VI IGHMBP2 z Earlyadult onset, prominent hand involvement,

vocal cord/diaphragmatic involvement with breathing problems

AD dHMN VII DCTN

Predominant upper limb symptoms or upper limb symptoms earlier than lower limb symptoms z Upper limb preponderance, maybe onlymotor

or motor and sensoryin the same family AD CMT2D/HMN5 GARS z Earlyadult onset, prominent hand involvement,

AD dHMN VII DCTN

Prominent proximal muscle involvement z Prominent proximal muscle involvement,

North African families, severe neuropathy AR CMT4C1 LMNA Sensorineural hearing loss or abnormal acousti-

cally evoked potentials

z Childhood onset, severe neuropathy, Bulgarian

Gypsies AR CMT4D/HMSN-L NDRG1

z CMT with additional CNS signs: AEP abnormal, VEP abnormal, intellectual imp./cerebellar sy. X-

linked CMTX GJB1

(4)

Table 11.1 (continued)

Clinical picture/other characteristic features IH Form Gene/Test z Severe, childhood onset, AEP prolonged I±III

latency, focally folded myelin AR CMT4B1 MTMR2

Central nervous system abnormalities z AEP, VEP, intellectual impairment, cerebellar

symptoms, MRI-T2 transient hyperint. possible X-

linked CMTX GJB1

z Cranial nerve involvement (not known whether

of peripheral or central origin) AD/AR CMT1D/CMT4E EGR2 Vocal cord paralysis and/or respiratory problems

z Earlyonset, severe neuropathywith vocal cord/

diaphragmatic involvement and wheelchair dependent in second decade

AD/AR CMT4A/CMT4C1 GDAP1

z Congenital/veryearlyonset, diaphragmatic/vocal

cord involvement, reduced life expectancy AR dHMN6 IGHMBP2 z Earlyadult onset, prominent hand involvement,

AD dHMN7 DCTN

ªRestrictedº to certain ethnic groups z Childhood onset, severe neuropathy, Bulgarian

Gypsies AR CMT4D/HMSN-L NDRG1

z Neonatal onset, sweating, high blood pressure,

sensory+ motor, Ashkenazi Jews AR HSAN3 IKBKAP

z CCFDN ± congenital cataracts, facial dysmorphism

and neuropathy, Wallachian Gypsies AR CCFDN CTDP1

Peculiar myelin pathology

z Focallyfolded myelin, severe, childhood onset,

AEP prolonged I±III latency AR CMT4B1 MTMR2

z Focallyfolded myelin, first decade onset AR CMT4B2 MTMR13=SBF2 z Sometimes focallyfolded myelin, earlyonset,

prominent sensorysymptoms, but also motor AR CMT4F PRX z Large cytoplasmic Schwann cell extensions,

scoliosis in manypatients AR CMT4C KIAA1985

Miscellaneous

z No clinical symptoms, mild NCV slowing only

(mNCV *37±42 m/s for median nerve) AD DI-CMTD ARHGEF10 z Recurrent painless pressure palsies, entrapment

syndromes AD HNPP Chr.17p

deletion

(5)

form. A molecular genetic test is not available, because the genetic de- fect causing HNA is unknown. HNPP can also present with a CMT1-like clinical picture.

z Which findings are revealed by the neurophysiological examination? In HMSN: mNCVs above 38±40 m/s argue strongly against the diagnosis CMT1, while mNCVs well below 38±40 m/s support this diagnosis. Inter- mediate to normal mNCVs are often found in CMTX patients with GJB1 mutations and in patients with certain MPZ mutations. The CMAPs are predominantly reduced in CMT2 but also often in CMT1, particularly in nerves with very low mNCVs. Patients with distal HMN have normal SNAPs and, in most cases, normal mNCVs and CMAPs, despite severe atrophy and signs of denervation in distal muscles on needle EMG examination.

z What is the mode of inheritance? If half of the family members in more than one generation are affected, autosomal dominant inheritance with high penetrance is most likely, but X-chromosomal dominant inheritance can only be excluded if male-to-male transmission is present in the family. In some families and at first glance, CMTX might appear to be an X-chromosomal recessive disease, which means that female family members are not obviously affected, but half of their male offspring are.

Closer clinical and electrophysiological examination of these female family members usually reveals signs of a mild subclinical neuropathy in CMTX families. Are the parents of the affected persons healthy? If a single child in an otherwise healthy family is affected, three possibilities have to be taken into account: (1) the disease is not heritable, (2) the mode of inheritance is autosomal recessive, (3) the disease is caused by a de-novo mutation. It is important to look for consanguinity of the parents. In our experience, sporadic cases with non-consanguineous parents are more likely to represent de novo mutations than true recessive cases.

In contrast, if two or more children of healthy parents are the only affected family members, autosomal recessive mode of inheritance is more likely.

z As soon as these questions are answered, focused molecular genetic diagnostic procedures should be applied. The autosomal dominant forms of HMSN are most common. The CMT1A duplication and the HNPP deletion account for the majority of all HMSN patients. Quantitative PCR methods make the molecular diagnosis of the duplication/deletion easier and cost effective. In most cases, the duplication/deletion analysis should be performed first. If no duplication or deletion is found, we proceed in most cases with sequence analysis of the GJB1 and MPZ genes. Only if X-chromosomal inheritance can reliably be excluded on the basis of male-to-male inheritance, the mutation analysis of GJB1 can be skipped.

Figure 11.1 presents a more detailed decision tree explaining the molecular diagnostic proceeding for HMSNs. After a series of dichotomous decisions, a series of possible genetic defects/test (in italics and bold print) is pre-

(6)

sented along with peculiar clinical features that can be caused by mutations in the corresponding gene. These features are intended to help choose the appropriate genetic tests. Table 11.1 presents a different view on the same problem. The subheadings (in bold print) present peculiar clinical features followed by a list of neuropathies and their corresponding genetic defects.

This table is intended to help the clinician who has seen a patient with a possibly hereditary neuropathy, distinguished by a peculiar symptom or sign, e.g., vocal cord paralysis. All abbreviations, with the exception of the chromosome CMT1A duplication/HNPP deletion, correspond to genes that have to be analyzed for mutations.

Nevertheless, the diagnostic and decision making aids presented here can only assist in arriving at the correct diagnosis. One always has to bear in mind that our knowledge about the molecular genetic basis of inherited disease is growing daily and that genotype/phenotype correlations will be subject to permanent refinement and change. In addition to phenotypic variation caused by mutations in different genes and different mutations in the same gene, large phenotypic variation is often even encountered in different patients with the same mutation. This is due to different individual genetic backgrounds and environmental influences. These factors, especially the influence of individual genetic background, are still largely unknown. Known environmental influences are mainly other factors, which might cause or aggravate peripheral neuropathies like diabetes mellitus or certain drugs like vincristin [12].

Applying all the phenotypical and molecular genetic knowledge available today, approximately 75±90% of all hereditary neuropathies can be reliably diagnosed. If despite all efforts a molecular diagnosis cannot be established, this can be due to a number of reasons. The two most important are (1) the defective gene causing the particular phenotype is not yet known. This will most likely hold true if the disease is clearly inherited. (2) The neuropathy is not heritable. This is most likely true in sporadic cases, especially if they present with a pure neuropathy without distinguishing additional features re- lated to a particular genetic disease. Therefore, negative molecular genetic testing results in sporadic cases should prompt an intensive search for other, possibly treatable causes of the neuropathy.

If genetic tests are ordered, it is usually desirable for the testing laboratory to have as much clinical information as possible, especially if the re- ferring physician has not decided on a particular set of tests to be performed. The cost of genetic testing, a matter of much concern, is often not as high as suspected by many physicians. The chromosome CMT1A duplication/deletion test is less expensive than, e.g., the costs for a single MRI examination. For these reasons we think that the threshold, especially for chromosome CMT1A duplication/deletion testing should be low, because the incidence of CMT is rather high and this single test will be positive in

*70% of all heritable cases with motor and sensory involvement [10].

(7)

z Which kind of sample is required for molecular genetic testing?

Most laboratories perform all genetic tests for inherited peripheral neuropathies on genomic DNA. Genomic DNA is usually isolated from peripheral blood leukocytes. In most cases, a 10±20 ml EDTA anticoagulated blood sample is appropriate. The blood sample must not be frozen nor refriger- ated at any time because many methods are not able to isolate DNA from previously frozen samples. Refrigeration might cause undesirable clotting due to cold agglutinins. The blood sample can be sent to the laboratory by ordinary mail as long as it arrives within approximately four days. A 10 ml blood sample contains enough DNA for PCR based methods as well as Southern-blotting methods. Different sampling methods may be necessary for RNA-based analysis and fluorescence in situ hybridization (FISH). If in doubt, it is always advisable to contact the diagnostic lab before collecting the specimen. Prenatal testing and testing using archival paraffin em- bedded nerve biopsy material is possible [2, 3, 9].

11.2 Molecular genetic tests

11.2.1 Methods for the detection of the chromosome CMT1A duplication/HNPP deletion

In most cases, this duplication encompasses around 1.5´10⁶ bp and it is therefore too large to be detected with most methods employed to detect point mutations as well as small deletions/duplications and other types of mutations. Methods to detect the latter type of mutations are dealt with in the next section.

The CMT1A duplication/HNPP deletion and the molecular architecture of the chromosomal region involved is described in detail in the chapter on ªCMT1Aº. For the understanding of the following methods, it is important to remember that duplication and deletion of the 17p11 region are re- ciprocal events during meiosis, creating one germ cell with a deletion and the other with a duplication of the 17p11 region. In the next generation, the duplication or deletion are stably inherited. Together with one copy of the unaffected parent, the affected child will have either three copies (CMT1A duplication) or one copy (HNPP deletion) of the duplicated/deleted region. The duplication/deletion arises due to unequal crossing over caused by misalignment of two highly homologous, but not identical re- gions ± the CMT1A distal and proximal repeat sequences (CMT1A-REPs).

The CMT1A-REPs flank the duplicated/deleted chromosomal segment [14].

In some cases, smaller duplications/deletions have been found, but all of them contain the PMP22 gene [27]. Most (*80%) cross-over breakpoints are located within a 3.2 kb hotspot of recombination, which can be further

(8)

refined to an interval of 1.7 kb and to a 741 bp sequence harboring approximately 75% and 60% of all recombinations [6, 17, 26].

Only the principles of the different genetic testing methods will be described, because the methods themselves are too numerous to be described in detail. Further detail can be found in the references or at the website of the European CMT consortium (http://molgen-www.uia.ac.be/CMT/Proto- cols/DefaultProtocols.cfm).

All methods are based on three different principles:

z Demonstration of DNA-sequence features unique to the duplication or deletion, e.g., additional restriction fragments. These methods are quali- tative, meaning that the result is in most cases the presence or absence of an additional band after DNA electrophoresis. The main advantage of these methods is the unequivocal interpretation of the results. Most of these tests have the disadvantage that not all duplications/deletions are detected. With many of them rare, atypical smaller duplications or deletions are not detectable, because the specific DNA-sequence features, on which a particular test is based, may be absent in smaller duplications/

deletions. In addition, most of these tests require much more DNA than PCR-based quantitative tests and are time consuming and therefore expensive.

z Demonstration of a gene dosage difference: These methods are quantitative. They measure the amount of genomic DNA at a given locus within the duplicated/deleted region in comparison to a non-duplicated locus outside the region. CMT1A patients harbor three copies of the PMP22 gene. Therefore, the gene dosage of PMP22 is increased by 50% compared to two copies in normal controls. HNPP patients harbor only one copy. The gene dosage is therefore reduced to one half of that of normal controls. The major disadvantage of all quantitative methods are the problems inherent to quantitative DNA measurement, sometimes resulting in equivocal test results. Today, most of these problems have been overcome by technical advances.

z Direct visualization of three copies of the PMP22 gene in CMT1A patients and one copy in HNPP patients by FISH analysis.

The CMT1A duplication was originally discovered using Southern blotting and pulsed field gel electrophoresis (PFGE) of genomic DNA as well as short-tandem-repeat (STR) analysis.

z Detection of the chromosome CMT1A duplication/HNPP deletion by pulsed field gel electrophoresis (PFGE) (e.g., [8, 25])

PFGE is a method to separate large DNA fragments (>100 kb to more than 1´10⁶bp). Genomic DNA is digested with rare cutting restriction enzymes and after PFGE and Southern blotting hybridized with specific DNA probes. Methods to detect the CMT1A duplication/HNPP deletion by PFGE

(9)

are based on the appearance of specific so called ªjunctionº fragments which are not present in normal controls. These junction fragments are created by the misalignment of the proximal CMT1A-REP of one chromosome with the distal CMT1A-REP of the other chromosome and subsequent crossing over during meiosis. The restriction sites flanking this junction fragment are also rearranged, enabling the detection of novel restriction fragments unique for the duplication or deletion. The main disadvantages of these very reliable methods are that they are time consuming, cumbersome and therefore expensive. They are still widely used in the United States of America by the company Athena diagnostics (Athena diagnostics, Worcester, MA, USA).

z Detection of the chromosome CMT1A duplication/HNPP deletion by Southern blotting of genomic DNA (e.g., [4, 15, 26])

One commonly used method uses digestion of genomic DNA with the restriction enzyme MSP1 followed by Southern transfer and detection with the probe pVAW409R3a, which is located close to the PMP22 gene and always contained in the duplication/deletion [15]. Three different alleles may be detected by this probe (2.8 kb ± 50%, 2.7 kb ± 44% and 1.9 kb ± 6%).

Control individuals show two of these alleles, which may be the same or different ones (e.g., 2´2.8 kb=one band of 2.8 kb or 1´2.8 kb and 1´2.7 kb=2 bands). Duplicated individuals show three alleles. The diagnosis is straightforward if three different alleles are present. If only two alleles are present, one of them shows a hybridization signal, which is twice as strong as the one of the other allele because it is represented in two copies. If only one allele is present, the hybridization signal is even stronger, but there is no other band to compare it with. In this case, the signal may be compared to that of a second unrelated probe, which recognizes a nonpolymorphic fragment outside the duplicated region. If the HNPP deletion is present, only one allele will be seen and comparison to the signal of a second unrelated probe is also necessary. The disadvantage of this method is that it re- lies in most cases on dosage difference. The analysis of dosage differences are subject to a number of technical influences and may be difficult to detect and interpret.

The formation of unique junction fragments during the formation of the CMT1A duplication/HNPP deletion has been discussed above. This region carries restriction sites not present in the normal controls and can therefore be recognized by the presence of a specific novel fragment on hybridization of Southern blots with a probe derived from the CMT1A-REP region after digestion with appropriate restriction enzymes [26]. Unfortunately, the crossover does not always occur in the same place and depending on the method (enzymes and probes) junction fragments are not always detected. In these cases, the method has to rely again on dosage differences.

(10)

z Detection of the chromosome CMT1A duplication/HNPP deletion by PCR methods employing special characteristics of the CMT1A repeat region (e.g., [22, 23])

These methods use PCR amplification of the 3.2 kb hotspot of recombination either alone or in combination with restriction digestion to demonstrate special features of the junction region on subsequent electrophoretic analysis. The main disadvantage of these methods is that only about 80%

of cross-over breakpoints are located in this region restricting the sensitivity of these assays to a maximum of 80%. Advantages are simplicity and low cost.

z Detection of the chromosome CMT1A duplication/HNPP deletion by short-tandem-repeat (STR) analysis (e.g., [1, 5, 20])

STRs are di-, tri- and tetranucleotide repeats that are polymorphic due to different repeat length. These repeats are often highly polymorphic with 2?20 different alleles. The duplication is detected by the presence of three alleles of a particular STR located in the duplicated region, while the deletion is detected by hemizygosity of a number of markers in the deleted region. The sensitivity of the test depends on the heterozygosity and the number of markers used for the analysis. Recently published STR-marker panels of approximately 15STRs are able to detect >99% of all duplications. The detection of the HNPP deletion has to rely on statistical methods, calculating the likelihood that a certain number of seemingly homozy- gous (in reality hemizygous) STR markers occurred by chance versus the likelihood that they occurred because of the HNPP deletion, because homozygosity and hemizygosity of an STR marker cannot be reliably distinguished. The advantages of these tests are unequivocal readouts of the results and high sensitivity if the recently developed marker panels are used.

z Detection of the chromosome CMT1A duplication/HNPP deletion by fluorescent in situ hybridization on chromosome preparations(e.g., [16, 21]) FISH methods use in most cases hybridization of fluorescently labeled DNA probes encompassing the PMP22 gene. A number of variations exist.

All methods directly visualize the presence of the duplication and deletion by the presence of an additional or missing hybridization signal. The main advantage of this method is the direct and unequivocal visualization of the results. Its main disadvantages are that it is comparatively time consuming, expensive and requires special equipment and expertise not present in many laboratories.

(11)

z Detection of the chromosome CMT1A duplication/HNPP deletion by quantitative PCR assays (e.g., [7, 28, 30])

Quantitative PCR assays show gene dosage differences of the PMP22 gene itself. A number of different quantitative PCR assays exist that demonstrate 150% of the PMP22 gene dosage of normal controls in CMT1A patients and 50% of the normal gene dosage in HNPP patients. Modern methods developed in recent years mostly use so called real-time quantitative PCR methods. While these methods were initially mistrusted by many, they have in the meantime proven their sensitivity, reliability and efficiency. The major disadvantage are ambiguous results in some patients, which have to be reanalyzed by a different method. Major advantages are the high speed and relatively low price of most of these assays.

Currently, the use of two different independent methods is regarded as the gold standard for the diagnosis of the CMT1A duplication/HNPP deletion.

11.2.2 Mutation detection methods for other genetic defects causing hereditary neuropathies

All other genetic defects known to cause hereditary peripheral neuropathies are either point mutations or small rearrangements ± mostly duplications/deletion of one to a few bases ± which can be detected by ªconven- tionalº mutation detection methods which are also applicable to many other heritable diseases. The methodological details will not be discussed, because they are described in detail in standard textbooks of molecular ge- netics and laboratory methods [18].

All the following methods may either be used to analyze genomic DNA or cDNA generated by reverse transcription of the patient's RNA. For all genes known to be defective in hereditary neuropathies, mutation analysis on genomic DNA is usually preferred. Only three out of many mutation detection methods will be described here. They were chosen because they are widely used and are good examples to demonstrate the advantages and disadvantages of different methods. All methods presented here require amplification of the target sequences by polymerase chain reaction (PCR) prior to mutation analysis.

z Direct DNA-sequencing

Today direct DNA-sequencing is (nearly) exclusively performed by the San- ger Dideoxy method [19]. Advantages of direct DNA sequencing are (1) very high sensitivity and (2) direct detection of the causative mutation. All other methods presented below screen only for the presence of a mutation in the target sequence but require subsequent DNA sequencing of positive samples to identify the exact mutation. Disadvantages of direct DNA se-

(12)

quencing are that it is (1) more expensive and (2) slower than screening methods. These disadvantages play an important role if the yield of mutations is small compared to the number of samples. Another disadvantage (3): DNA sequences of patients still have to be compared visually to sequences of normal controls, because highly reliable automated mutation detection algorithms able to detect all visually detectable heterozygous mutations do not yet exist. This work is time-consuming and monotonous but requires at the same time a high amount of concentration, making it prone to examiner mistakes.

PCR amplified DNA fragments of up to 400±600 bp can be reliably analyzed in a single sequencing reaction. It is essential to sequence both strands of the target DNA.

z Single strand conformational polymorphism (SSCP) analysis [13]

Single stranded DNA adopts under nondenaturing conditions unique con- formations dependent on its base sequence. In most cases, even single base changes introduce conformational changes that can be detected in PCR amplified target fragments of up to 200±300 bp by nondenaturing poly- acrylamide (PAA) gel electrophoresis and subsequent silver staining of the gel. The major disadvantages of this so called single strand conformational polymorphism (SSCP) analysis are (1) the sensitivity of SSCP concerning point mutations is not higher than 70±80% at the most. (2) The reliability decreases with increasing size of the target fragment [11]. Small target fragments of not more than 250 bp are optimal. (3) SSCP positive samples require subsequent sequencing to identify the exact mutation. The main advantages of SSCP are its low price, speed and relatively easy handling. In our opinion SSCP should not be used anymore in the diagnostic setting, because negative results do not reliably exclude mutations.

z Denaturing high performance liquid chromatography (DHPLC)

DHPLC is a relatively novel and highly sensitive mutation detection method [29]. It is based on the ability of a novel chromatography matrix to separate heteroduplex DNA molecules (=double stranded DNA with a single base pair mismatch) from homoduplex DNA molecules. PCR amplified DNA is first heat denatured and then slowly cooled to room temperature to allow heteroduplex formation if a mutation in one allele is present. The PCR product is heated again close to its melting temperature and separated on the column by elution with increasing concentrations of organic solvent.

The sensitivity of DHPLC is highly dependent on the optimization of the DHPLC process with as many known mutations as possible, rendering it less suitable if DNAs with known mutations are not available in the lab or if the sample numbers are low. The main advantages are its low operating costs, the high sensitivity (>95%) and the possibility to use fairly large PCR products up to 700±1000 bp [24].

(13)

References

1. Badano JL, Inoue K, Katsanis N, Lupski JR (2001) New polymorphic short tandem repeats for PCR-based Charcot-Marie-Tooth disease type 1A duplication diagnosis. Clin Chem 47:838±843

2. Beckmann A, Schræder JM (2000) Screening for Charcot-Marie-Tooth type 1A and hereditary neuropathy with liability to pressure palsy in archival nerve biopsy samples by direct-double-differential PCR. Acta Neuropathol (Berl) 100:459±463 3. Bernard R, Labelle V, Negre P, Tardieu S, Azulay JP, Malzac P et al (2000) Prenatal

detection of a 17p11.2 duplication resulting from a rare recombination event and novel PCR-based strategy for molecular identification of Charcot-Marie-Tooth disease type 1A. Eur J Hum Genet 8:229±235

4. Chance P, Alderson M, Leppig K, Lensch M, Matsunami N, Smith B et al (1993) DNA deletion associated with hereditary neuropathy with liability to pressure palsies. Cell 72:143±151

5. Latour P, Boutrand L, Levy N, Bernard R, Boyer A, Claustrat F et al (2001) Poly- morphic short tandem repeats for diagnosis of the Charcot-Marie-Tooth 1A duplication. Clin Chem 47:829±837

6. Lopes J, Ravis N, Vandenberghe A, Palau F, Ionasescu V, Mayer M et al (1998) Fine mapping of de novo CMT1A and HNPP rearrangements within CMT1A-REPs evidences two distinct sex-dependent mechanisms and candidate sequences involved in recombination. Hum Mol Genet 7:141±148

7. Lorentzos P, Kaiser T, Kennerson ML, Nicholson GA (2003) A rapid and definitive test for Charcot-Marie-Tooth 1A and hereditary neuropathy with liability to pressure palsies using multiplexed real-time PCR. Genet Test 7:135±138

8. Lupski J, de Oca-Luna R, Slaugenhaupt S, Pentao L, Guzzetta V, Trask B et al (1991) DNA duplication associated with Charcot-Marie-Tooth disease type 1A.

Cell 66:219±232

9. Navon R, Timmerman V, Læfgren A, Liang P, Nelis E, Zeitune M et al (1995) Pre- natal diagnosis of Charcot-Marie-Tooth disease type 1A (CMT1A) using molecular genetic techniques. Prenat Diagn 15:633±640

10. Nelis E, Van Broeckhoven C, De JP, Læfgren A, Vandenberghe A, Latour P et al (1996) Estimation of the mutation frequencies in Charcot-Marie-Tooth disease type 1 and hereditary neuropathy with liability to pressure palsies: a European collaborative study. Eur J Hum Genet 4:25±33

11. Nelis E, Warner L, De Vriendt E, Chance P, Lupski J, Van Broeckhoven C (1996) Comparison of single-strand conformation polymorphism and heteroduplex analysis for the detection of mutations in Charcot-Marie-Tooth type 1 disease and re- lated peripheral neuropathies. Eur J Hum Genet 4:329±333

12. Neumann Y, Toren A, Rechavi G, Seifried B, Shoham NG, Mandel M et al (1996) Vincristine treatment triggering the expression of asymptomatic Charcot-Marie- Tooth disease. Med Pediatr Oncol 26:280±283

13. Orita M, Suzuki Y, Sekiya T, Hayashi K (1989) Rapid and sensitive detection of point mutations and DNA polymorphisms using the polymerase chain reaction.

Genomics 5:874±879

14. Pentao L, Wise C, Chinault A, Patel P, Lupski J (1992) Charcot-Marie-Tooth type 1A duplication appears to arise from recombination at repeat sequences flanking the 1.5Mb monomer unit. Nat Genet 2:292±300

15. Raeymaekers P, Timmerman V, Nelis E, De Jonghe P, Hoogendijk JE, Baas F et al (1991) Duplication in chromosome 17p11.2 in Charcot-Marie-Tooth neuropathy type 1a (CMT 1a). The HMSN Collaborative Research Group. Neuromuscul Dis- ord 1:93±97

(14)

16. Rautenstrauss B, Fuchs C, Liehr T, Grehl H, Murakami T, Lupski J (1997) Visuali- zation of the CMT1A Duplication and HNPP Deletion by FISH on Stretched Chro- mosome Fibers. JPNS 2:319±322

17. Reiter L, Murakami T, Koeuth T, Pentao L, Muzny D, Gibbs R et al (1996) A recombination hotspot responsible for two inherited peripheral neuropathies is located near a mariner transposon-like element. Nat Genet 12:288±297

18. Sambrook J, Fritsch EF, Maniatis T (2000) Molecular cloning: a laboratory man- ual, 3rd edn. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY 19. Sanger F, Nicklen S, Coulson AR (1977) DNA sequencing with chain-terminating

inhibitors. Proc Natl Acad Sci 74:5463±5467

20. Seeman P, Mazanec R, Zidar J, Hrusakova S, Ctvrteckova M, Rautenstrauss B (2000) Charcot-Marie-Tooth disease type 1A (CMT1A) and hereditary neuropathy with liability to pressure palsies (HNPP): reliable detection of the CMT1A duplication and HNPP deletion using 8 microsatellite markers in 2 multiplex PCRs. Int J Mol Med 6:421±426

21. Shaffer LG, Kennedy GM, Spikes AS, Lupski JR (1997) Diagnosis of CMT1A duplications and HNPP deletions by interphase FISH: implications for testing in the cytogenetics laboratory. Am J Med Genet 69:325±331

22. Stronach EA, Clark C, Bell C, Læfgren A, McKay NG, Timmerman V et al (1999) Novel PCR-based diagnostic tools for Charcot-Marie-Tooth type 1A and hereditary neuropathy with liability to pressure palsies. J Peripher Nerv Syst 4:117±122 23. Tachi N, Kozuka N, Ohya K, Chiba S (1999) Charcot-Marie-Tooth disease type 1A

duplication by PCR analysis. Pediatr Neurol 20:360±363

24. Takashima H, Boerkoel CF, Lupski JR (2001) Screening for mutations in a geneti- cally heterogeneous disorder: DHPLC versus DNA sequence for mutation detection in multiple genes causing Charcot-Marie-Tooth neuropathy. Genet Med 3:335±342

25. Timmerman V, Læfgren A, Le Guern E, Liang P, De Jonghe P, Martin J et al (1996) Molecular genetic analysis of the 17p11.2 region in patients with hereditary neuropathy with liability to pressure palsies (HNPP). Hum Genet 97:26±34

26. Timmerman V, Rautenstrauss B, Reiter L, Koeuth T, Læfgren A, Liehr T et al (1997) Detection of the CMT1A/HNPP recombination hotspot in unrelated patients of European descent. J Med Genet 34:43±49

27. Valentijn LJ, Baas F, Zorn I, Hensels GW, de Visser M, Bolhuis PA (1993) Alterna- tively sized duplication in Charcot-Marie-Tooth disease type 1A. Hum Mol Genet 2:2143±2146

28. Wilke K, Duman B, Horst J (2000) Diagnosis of haploidy and triploidy based on measurement of gene copy number by real-time PCR. Hum Mutat 16:431±436 29. Xiao W, Oefner PJ (2001) Denaturing high-performance liquid chromatography:

A review. Hum Mutat 17:439±474

30. Young P, Stægbauer F, Wiebusch H, Læfgren A, Timmerman V, Van Broeckhoven C et al (1998) PCR-based strategy for the diagnosis of hereditary neuropathy with liability to pressure palsies and Charcot-Marie-Tooth disease type 1A. Neurology 50:760±763