1 Introduction

Neurofibromatosis type 1 (NF1) is one of the most common dominantly inherited neurogenetic disorders, affecting about 1 in every 4000 individuals worldwide. It is fully penetrant by the age of five. The condition is characterized by multiple café-au-lait spots, benign neurofibromas, and Lisch nodules (1). Other clinical manifestations include abnormalities of the cardiovascular, gastrointestinal, renal, and endocrine systems; major orthopedic problems; facial and body disfigurement; and cognitive deficit and malignancy, with tumors of the peripheral nerve sheath and central nervous system (CNS). About one-quarter of NF1 patients develop one or more of these clinical complications, demonstrating the significant morbidity and mortality associated with this disorder. However, the clinical expression varies greatly from one patient to another, between families, and even within a given family carrying the same underlying gene lesion. The establishment of diagnostic criteria for NF1 (2) has potentiated the clinical diagnosis of NF1 in most patients. Neurofibromas are a hallmark feature of NF1. Each neurofibroma is composed of Schwann cells (60%–80%), fibroblasts, perineurial cells, axons, and mast cells. The cellular sources of the various neurofibroma-associated mitogens, the occurrence of specific receptors for each of these factors, and their functional role in tumor development remain unclear. Indeed, the characterization of novel tumor growth-promoting factors and their receptors in both benign and malignant cancers may help to define potential treatment strategies for NF1.

Children with NF1 have an increased risk of developing myeloid disease, particularly juvenile myeloid leukemia and myelodysplastic syndrome. In practice, the overall frequency of cancers in NF1 patients is about twice that observed in the general population. Currently, about 22,000 individuals in UK are affected by NF1 and because there is still no treatment, this presents a major health burden to the community.

1.1 The NF1 Gene as Tumor-Suppressor

There is strong evidence that NF1 gene is a tumor-supressor, a gene whose normal function is to inhibit or control cell division. The Knudson “two-hit” tumor-suppressor model predicts that all somatic cells of an affected person harbor a constitutional (germline) NF1 mutation. The occurrence of a second NF1 mutation involving the normal allele is then predicted to result in the loss of cell-cycle control, leading to tumor formation (3–5). The observation of multiple distinct NF1 somatic mutations in neurofibromas from the same NF1 patients (6,7) reveals that each second hit is an independent event. This in turn suggests that the somatic mutational spectrum of the NF1 gene may be an important contributor to the variable clinical expression of NF1.

1.2 NF1 Gene

The NF1 gene located at 17q11.2 spans 350 kb of genomic DNA, contains 60 exons and encodes a 12 kb mRNA transcript (8).Three of the exons are alternatively spliced. The stop codon lies in exon 49, which also includes a 3′ untranslated region (3′ UTR). Neurofibromin, the NF1 gene product, is ubiquitously expressed, and exhibits structural and sequence similarity to an evolutionarily conserved family of proteins, the mammalian GTPase-activating protein (GAP)-related proteins. The most highly conserved region of the protein is the NF1 GAP-related domain (GRD), encoded by exons 21–27a. To date, this is the only domain of neurofibromin to which a function has been ascribed. Activated p21-ras (ras with bound GTP) can transform fibroblasts and increases cell proliferation in vitro. Neurofibromin has been shown to downregulate p21-ras (ras-GDP). Thus, inactivation of neurofibromin has been theorized to be involved directly in tumorigenesis.

Within the NF1 gene, the large intron 27b contains three embedded genes (OMGP, EVI2B, and EVI2A) that are transcribed in the opposite orientation to the NF1 gene. Each of the three embedded genes comprises two exons. EVI2A, a small gene encoding a 232-amino acid polypeptide, is expressed in the brain and the bone marrow; sequence analysis suggests that it has a transmembrane domain (9). EVI2B is also a small gene that encodes a 448-amino-acid protein with unknown function, and it is expressed exclusively in the bone marrow (10). OMGP encodes a 416-amino acid cell-adhesion molecule, and is expressed primarily in oligodendrocytes (11). It is unknown whether embedded genes play a role in the expression of NF1.

Fluorescence in situ hybridization (FISH) analysis and genomic sequence homology of amplified polymerase chain reaction (PCR) products have revealed that there are numerous NF1 homologous loci spread across the genome (12,13).

The NF1 promoter is located in a CpG island and exhibits a high degree of sequence conservation with other species (14). The 3′ UTR of the NF1 gene extends 3.5 kb downstream of the translation stop signal in exon 49 (15).

The application of reverse transcriptase-PCR (RT-PCR) reveals NF1 mRNA in almost all tissues. Indeed, lymphocytes express enough NF1 mRNA to provide a ready source of RNA for mutational analysis. Three alternatively spliced isoforms with significantly different expression patterns have been identified.

1.3 NF1 Gene Mutations

The mutation rate for NF1 is 1 in 10,000 gametes/cell/generation, approx 10-fold higher than that found for most other inherited disease genes. The germline mutational spectrum of the NF1 gene has been well-characterized (16), with new mutations occurring in about one-half of all NF1 individuals. To date, some 440 NF1 gene lesions have been logged in either the Human Gene Mutation Database (http://www.hgmd.org) or the NNFF (National Neurofibromatosis Foundation) Mutation Database (http://www.nf.org/nf1). However, no clustering of mutations within the NF1 gene is apparent. Among germline mutations, 80% are truncating, and approx 30% involve abnormalities of splicing (6,16–18). By contrast, data on the tumor-associated somatic mutations of this gene are still sparse.

Characterizing mutations in the NF1 gene has presented a considerable challenge because of the combination of the size of the gene, the absence of any obvious mutational clustering, and the wide diversity of mutations (16). The existence of numerous NF1 pseudogenes adds to the complexity of PCR-based mutation analysis (12,13,19). FISH has been used to identify very large gene deletions or rearrangements that include the NF1 gene (20,21). Pulsed-field gel electrophoresis has been used to identify gross deletions and insertions within the NF1 gene (22–24). Conventional Southern blot analysis has been employed for the study of medium-size gene rearrangements (24–26).

Single-strand conformational polymorphism (SSCP) (24,25,27,28) and heteroduplex analysis (HA) often have been used in the past to detect microlesions (25,29,30). Other methods—including denaturing gradient gel electrophoresis (DGGE) (31), temperature gradient gel electrophoresis (17), the protein truncation test (PTT) (3,17,26,32–34), the long reverse transcriptase-polymerase chain reaction (LRT-PCR) (35), and direct sequencing (17) have all been used for mutation detection; no single technique has been completely effective.

1.4 Mutations in the Non-Coding Regions of the NF1 Gene

The 3′ untranslated region (UTR) of the human NF1 gene is 3.5 kb in length and highly conserved during evolution (36), indicating that this region may be important for mRNA stability or translational efficiency. Although there may be mutations in the 3′ UTR that are of pathological significance, only one 3′ UTR mutation has been reported to date (37,38).

Although the NF1 gene-promoter sequence is highly conserved evolution-arily (14), no pathological mutations in this region have yet been detected (39). Their eventual identification will help to define the mechanisms that regulate neurofibromin expression.

1.5 Molecular Testing

NF1 is usually diagnosed on the basis of clinical and ophthalmic examination, especially in those patients who are more than 5 yr of age. However, in many pediatric situations, a sufficiently precise clinical diagnosis may not be possible, and these families may be greatly aided by the identification of mutations. Some clinicians feel that it is important to make diagnosis early so that patients can be offered counseling and guidance, and that children can be monitored for such complications as optic glioma and hypertension. Since preemptive prophylactic clinical intervention is not yet possible, it is difficult to justify presymptomatic testing. Population and newborn screening are not yet available for NF1 worldwide. Inherited mutations in the NF1 gene have been identified in some families who do not meet the NIH diagnostic criteria for NF1; these include those with segmental NF1, Watson syndrome, or familial caféau-lait spots only (40,41). Detection of NF1-specific mutations in such families would help to confirm the clinical diagnosis.

Although the demand for prenatal diagnosis is not high, it is occasionally sought. A low demand for prenatal testing could be a result of the inability of this test to predict the severity of the condition. Prenatal diagnosis can be performed by direct gene analysis if a specific mutation has already been identified in the family. Alternatively, prenatal diagnosis can be based on genetic linkage if samples are available from enough key family members (42–44). The identification of specific mutations will permit molecular diagnosis independent of family structure, and will be particularly helpful for large numbers of isolated cases.

Fetal DNA specimens may be obtained by amniocentesis or chorionic villus sampling (CVS). The recent application of high-throughput mutation detection techniques has substantially improved the NF1 mutation detection efficiency and may have an impact on molecular testing for this condition.

1.6 Current Molecular Methodology

The authors currently use both FISH and DHPLC (denaturing high-performance liquid chromatography) for NF1 molecular diagnosis, and have found these methodologies to be both efficient and highly complementary. A significant minority (5%) of patients with NF1 are heterozygous for a germline 1.5-Mb deletion at 17q11.2, which includes the NFl gene (45). The deletion is associated with a typical phenotype of minor facial abnormalities, mental retardation, learning disabilities, and early or excessive burden of cutaneous or plexiform neurofibromas (46). At present, the most effective method to detect large deletions is FISH performed on standard metaphase spreads using probes that are specific for the NF1 gene region. If the patient does not have a deletion of the relevant probe, a signal will be seen on both chromosome 17 homologs at 17q11.2. A single signal in each case examined is indicative of a chromosomal deletion ( Fig. 1 ).

Fig. 1.
figure 1

FISH studies showing a deletion of the NF1 gene.

Full size image

1.7 Overall Approach

1.7.1 FISH

In the authors’ laboratory, two probes have proven to be reliable: P1-9, which is 65 kb is size and maps to exons 2 to 11, and P1-12, which is 55 kb in size and maps to intron 27b. These were kindly supplied by Dr. Bruce Korf (21). An indirectly labeled hapten-conjugated nucleotide such as biotin-16-dUTP is then incorporated into the DNA by nick translation. This is then detected via a fluorescein isothiocyanate (FITC)-labeled avidin. For best results with FISH, labeled probe fragments of approx 200–500 basepairs (bp) are required. Longer sequences result in decreased probe penetration and increased nonspecific binding (leading to a higher background signal).

1.7.2 DHPLC

DHPLC is a modification of the basic gel-based heteroduplex mutation analysis method (47), and is applied to those samples without a FISH-detectable deletion. This methodology relies upon the rapid separation and visualization of homo- and heteroduplex DNA molecules using an ion-pair reverse-phase liquid chromatography system and has proven to be both rapid and sensitive. The sensitivity of the analysis is maximized by maintaining the liquid chromatography column at a temperature that favors partial strand denaturation in the presence of bp mismatching. Preliminary local assessment of the DHPLC system has successfully identified every lesion within exon 16 of the NF1 gene (48). Subsequently, DHPLC conditions were further optimized to screen the entire coding region of the NF1 gene; the sensitivity of this approach for NF1 mutational analysis is now 95% (Han et al., 2001) confirming DHPLC to be a rapid and efficient tool for NF1 mutational analysis (30).

The authors carry out DHPLC on a WAVE™ DNA fragment analysis system using a DNASep® column (Transgenomic, Inc., Omaha, NE 68107) (48). DNASep® columns contain non-porous alkylated polystyrene-divinylbenzene particles that are both electrically neutral and hydrophobic. The successful resolution of heteroduplexes from homoduplexes requires an elution gradient at a partially denaturing temperature. At this temperature, only heteroduplexes are destabilized by the mismatched bases, and are therefore slightly more melted than the homoduplexes, resulting in earlier elution than the homoduplexes. This special resolution temperature (Tr) can be predicted by use of the DHPLCMelt software (http://insertion.stanford.edu/melt.html).

The DHPLCMelt software successfully determined the optimal Tr for each of the 60 fragments containing individual exons of the NF1 gene. Because of the unavailability of complete sequence for intron 1 of the NF1 gene in GenBank, empirical determination of the Tr and elution gradient of the intron 1 fragment was performed (48).

2 Materials

2.1 FISH

2.1.1 Nick Translation

  1. 1.

    Nick translation system. The authors use the Invitrogen (Paisley, Scotland) nick translation system. Detailed instructions for use are supplied with the kit. Store at −20°C.

  2. 2.

    Indirectly labeled hapten. The authors use biotin-16-dUTP (Boehringer, Mannheim, Germany) for detection using avidin and biotinylated anti-avidin. Digoxygenin-11-dUTP (Boehringer Mannheim) can be used as an alternative, with detection via fluorescein-labeled antibodies. Store at 4°C.

  3. 3.

    1% agarose check gel. Add 1 g agarose (Roche) to 100 mL TAE working solution, dissolve, and add 5 μL ethidium bromide working solution (10 mg mL−1). To make a stock solution of TAE buffer, add 48.4 g Tris-HCl (USB), 17 mL glacial acetic acid (BDH, Poole, UK), and 3.72 g EDTA (Sigma, St. Louis, MO) to deionized water and make up to 1 L. Store at room temperature. Dilute 1∶10 with deionized water to make a working solution.

  4. 4.

    Ethidium bromide (Sigma Aldrich, Steinheim, Germany). Add 1 mL sterile water to 10 mg ethidium bromide to make the working solution. Store at 4°C.

  5. 5.

    Bromophenol blue (Sigma).

  6. 6.

    1-kb ladder (InVitrogen).

2.1.2 Probe Preparation

  1. 1.

    Cot-1 DNA (InVitrogen).

  2. 2.

    3 M sodium acetate. Add 110.8 g sodium acetate (Sigma) to sterile water and make up to 100 mL. Store at room temperature.

  3. 3.

    Hybridization buffer. The hybridization buffer contains 50% formamide, 10% dextran sulfate, and 2X standard saline citrate (SSC). Dissolve 20% dextran sulfate (Sigma) in molecular-grade formamide (BDH, Poole, UK). Aliquot into microcentrifuge tubes and keep as a stock solution at −20°C. To prepare a working solution, add 600 μL sterile water and 150 μL 20X SSC to 750μLstock solution. Store at −20°C.

  4. 4.

    20X SSC. Add 175.3 g sodium chloride (Sigma) and 88.22 g sodium citrate (Sigma) to sterile water and make up to 1 L. Mix to ensure that all of the chemicals have dissolved. Store at room temperature.

  5. 5.

    Ethanol (BDH).

2.1.3 Hybridization and Probe Detection

  1. 1.

    Denaturation solution. The denaturation solution consists of 70% formamide (BDH) in 2X SSC at pH 7.0–8.0. Prepare by adding 350 mL formamide to 50 mL 20X SSC and 100 mL sterile water; adjust pH to 7.0–8.0 using HCl. The denaturation solution can be used for up to 1 wk. Store at room temperature.

  2. 2.

    Ethanol wash solutions. 500 mL w/v solutions of 70%, 90%, and 100% are prepared using 100% ethanol and sterile water. The ethanol wash solutions can be used for up to 1 wk. Store at room temperature.

  3. 3.

    2X SSC posthybridization wash solution. Prepare by adding 50 mL 20X SSC to 450 mL sterile water, and adjusting the pH to 7.0–7.5 with 1 M NaOH. The wash solution can be used for up to 1 wk.

  4. 4.

    4T buffer. Add 100 mL 20X SSC and 250 μL Tween-20 (Sigma) to 400 mL sterile water. Make up fresh buffer for each hybridization.

  5. 5.

    Blocking solution. Add 250 mg Boehringer blocking agent (Boehringer Mannheim) to 50 mL 4X SSC to make a 0.5% working solution. Store in 2.5-mL aliquots in glass bottles at −20°C. Use a new aliquot for each hybridization.

  6. 6.

    Fluorescein-avidin. To make a working solution, add 1 μL stock fluorescein-avidin solution, as supplied by the manufacturer (Vector Laboratories, Burlingame, CA), to 100 μL blocking solution. Make sufficient for each day’s hybridization (120μL per slide). Keep at 4°C while in use. Discard any excess at the end of the day.

  7. 7.

    Biotinylated anti-avidin. To make a working solution, add 1 μL stock biotinylated anti-avidin solution (Vector Laboratories) to 100 μL blocking solution. Make sufficient for each day’s hybridization (60 μL per slide). Keep at 4°C while in use. Discard any excess at the end of the day. The stock solution is prepared by resuspending the powder in 1 mL sterile water.

  8. 8.

    DAPI and propidium iodide (PI) counterstain. Prepare by adding 5 μL DAPI (Sigma), and 5 μL PI (Sigma) working solution (0.5 mg/mL) to 1 mL of Vectashield mountant (Vector Laboratories). To make the working solution, add 1μL5 mg/mL PI stock solution to 9 μL sterile water.

2.2 DHPLC

  1. 1.

    2 M triethylammonium acetate (TEAA): (cat. no. 553303, Transgenomic) store at 4°C.

  2. 2.

    Acetonitrile (ACN): HPLC-grade, (cat. no. 152857L, Merck), store at room temperature, is toxic and should be used in a chemical fume hood. Wear gloves when handling this solution.

  3. 3.

    0.5 M EDTA: cat. no. 306, Teknova. Store at room temperature.

  4. 4.

    Milli-Q water or HPLC-grade water.

  5. 5.

    DNASep™ column: store at room temperature.

  6. 6.

    Size standard: this standard consists of a restriction digest of plasmid pUC18 by HaeIII and represents a pool of nine DNA fragments with the following sizes 80, 102, 174, 257, 267, 298, 434, 458, 587 bp. The DNA digest standard is used to evaluate instrument performance for size-based DNA fragment separations.

  7. 7.

    Mutation standard: this standard consists of a combination of two analogous 209-bp fragments representing A and G alleles at position 168 of the polymorphic DYS271 locus. Upon heating and renaturation, this fragment mixture forms two homoduplexes and two heteroduplexes that are used to check instrument parameters for heterodupelex-based mutation screening.

  8. 8.

    A thermal cycler with a heated lid.

3 Methods

3.1 FISH

3.1.1 Nick Translation

During the nick translation procedure, all solutions are kept on ice.

  1. 1.

    To a 1.5-mL microcentrifuge tube, add the following; 5 μL nucleotide mix without deoxythymidine triphosphate (dTTP), X μL template DNA (a total amount of 1 μg probe DNA is required), 2.5 μL biotin-16-dUTP, make up to 45 μL with Y μL of distilled water.

  2. 2.

    Mix gently, then add 5 μL DNA polymerase I/DNase I solution. Mix gently but thoroughly, and pulse-spin.

  3. 3.

    Incubate at 16°C until the DNA is fragmented to the appropriate size. For probes P1-9 and P1-12, this will take between 90 and 120 min.

  4. 4.

    Place the solution on ice and run a check gel to ensure that the DNA fragments are 200–500 bp in size.

  5. 5.

    Add 5 μL reaction to 3 μL bromophenol blue and load on the 1% agarose gel.

  6. 6.

    Add 5 μL of the 1-kb ladder to a separate well. Carry out the electrophoresis at 100 V for 30 min and examine the gel on a UV trans-illuminator. If the fragments are too large, incubate for a longer period of time. If the fragments are too small, discard the mixture and start again.

  7. 7.

    When the fragments are of the appropriate size, halt the reaction by adding 5 μL stop solution.

3.1.2 Probe Preparation

It is advisable to test a single aliquot of probe for its hybridization efficiency before using the probe in a diagnostic context. If the hybridization is successful, all of the DNA may be prepared for use.

  1. 1.

    To a microcentrifuge tube, add the following; 4 μL probe solution from the nick translation reaction (this should contain approx 60 ng DNA), 2.5 μL Cot-1 DNA, 0.5 μL 3 M sodium acetate, and 64 μL of ethanol.

  2. 2.

    Mix well and leave at −20°C for 30 min.

  3. 3.

    Centrifuge at 17,000g for 10 min.

  4. 4.

    Remove the supernatant, being careful not to disturb the deposit that should be visible at the bottom of the tube.

  5. 5.

    Dry off the pellet in a 37°C incubator. This should take approx 30–60 min, with occasional gentle mixing by and. If there are problems in resuspending the mixture, a pipet may be used to mix the solution. The probe is now ready for use.

3.1.3 Hybridization and Probe Detection

3.1.3.1 Slide Denaturation

Slides are made by standard cytogenetic laboratory procedures. For best results, make the slides up 2–3 h prior to use. Old or very fresh slides may be used, but they yield poorer hybridization and chromosome morphology, respectively.

  1. 1.

    Place the slides in a Coplin jar containing 50 mL of denaturation solution in a 73°C water bath inside a Class 1 cabinet for 2 min. It is not recommended that more than four slides are used at one time.

  2. 2.

    Transfer the slides to 70% ice-cold ethanol for 2 min, followed by 2 min in 90% and 100% ethanol at room temperature.

  3. 3.

    Remove the slides, drain the excess ethanol by touching the bottom of the slide on tissue, and wipe the back of the slide dry with a tissue. Allow to air-dry in an upright position. This usually takes 3–5 min.

3.1.3.2 Probe Denaturation

10 μL probe mixture per slide is generally used.

  1. 1.

    Place the probe mixture in the 73°C water bath for 10 min.

  2. 2.

    Place the probe in the 37°C water bath for approx 60 min.

3.1.3.3 Hybridization
  1. 1.

    Apply the 10-μL probe mixture to an appropriate area of the slide.

  2. 2.

    Immediately place a 22 mm ×22 mm cover slip over the probe solution and allow the probe solution to spread evenly under the cover slip. Any air bubbles are removed by applying gentle pressure with a pipet tip.

  3. 3.

    Seal with rubber solution.

  4. 4.

    Place the slides in a humidified chamber and allow to hybridize overnight in a 37°C incubator.

3.1.3.4 Posthybridization Washes
  1. 1.

    Remove the rubber solution and then gently remove the cover slip.

  2. 2.

    Immediately place the slides in a Coplin jar containing 50 mL of 2X SSC wash solution in a 73°C water bath inside a Class 1 cabinet for 2 min. Do not process more than four slides at once.

  3. 3.

    Wash in 4T buffer in a Coplin jar for 1 min at room temperature.

  4. 4.

    Add 60 μL blocking solution, pre-warmed to 37°C, to the slide, apply a 64 mm ×22 mm cover slip and incubate in the humidified chamber at 37°C for 10 min.

3.1.3.5 Detection of Probe Signal
  1. 1.

    After incubation in blocking solution, gently remove the cover slip. Add 60 μL avidin-FITC working solution to the slide, seal with a cover slip and incubate in the humidified chamber at 37°C for 15 min.

  2. 2.

    Gently remove the cover slip and wash 3×2 min in 4T buffer in Coplin jars at room temperature.

  3. 3.

    Add 60 μL biotinylated anti-avidin working solution to the slide, seal with a cover slip, and incubate in a humidified chamber at 37°C for 15 min.

  4. 4.

    Gently remove the cover slip and wash 3×2 min in 4T buffer as previously.

  5. 5.

    Repeat steps 1 to 2.

  6. 6.

    Drain the excess 4T buffer by touching the bottom of the slide on tissue and wipe the back of the slide dry with a tissue.

  7. 7.

    Add 30 μL DAPI/PI counterstain to slides and apply a 64 mm×22 mm cover slip. The slides are now ready for analysis. Slides can be stored at 4°C prior to analysis.

3.1.4 Visualization of the Probe Signal

For FISH analysis, a fluorescence microscope with a 50-W (or preferably a 100-W) high-pressure mercury lamp and the appropriate filter set (FITC/PI/DAPI) is needed.

In the authors’ laboratory, 30 cells are checked for the deletion.

3.2 DHPLC

3.2.1 Preparation of Buffers

Use clean glass bottles for the preparation and storage of buffers. Always rinse flasks and bottles three times before buffers are prepared.

3.2.1.1 Buffer A

0.1 M triethylammonium acetate (TEAA), 1 mM EDTA, pH 7.0. Use 50-mL and 1000-mL volumetric flasks to prepare this buffer. Measure out 50 mL 2 M TEAA, and transfer to a 1000-mL volumetric flask. Rinse the 50-mL volumetric flask 3 times with HPLC-grade water and transfer the rinse to the 1000-mL volumetric flask to ensure proper volume is transferred. Add 200μL 0.5 M EDTA into the 1000-mL flask. Adjust the final vol to 1000 mL with HPLC-grade water.

3.2.1.2 Buffer B

0.10 M TEAA, 25% ACN, pH 7.0. Prepare buffer B using the 1000-mL, 250-mL, and 50-mL volumetric flasks. Measure 50 mL 2 M TEAA and 250 mL ACN and transfer to a 1000-mL volumetric flask. Rinse each three times to ensure all the TEAA/ACN is transferred. Bring the volume up to approx 950 mL with HPLC-grade water. Seal the flask to prevent evaporation, and leave at room temperature for at least 10 min. Adjust the final volume to exactly 1000 mL.

3.2.1.3 Buffer C

75% ACN, pH 7.0. Using a 250-mL volumetric flask, transfer 750 mL of acetonitrile to a 1000-mL flask, add 250 mL HPLC-grade water to bring the final vol to 1000 mL.

3.2.1.4 Buffer D

8% ACN, pH 7.0. Prepare this by using a glass graduated-cylinder to measure 80 mL HPLC-grade ACN, transfer into a 1000-mL volumetric flask, and bring the final vol to 1000 mL with HPLC-grade water. This solution is used to rinse the needle and injection valve. It is also used to deliver sample through the injection valve without any losses.

3.2.2 PCR Amplification of 60 Exons of NF1 Gene

Primers and PCR conditions for all 60 exons of the NF1 gene are listed in Table 1 . However, the guidelines for the design of new primers are available from Transgenomic™ (application note No. 101). For PCR amplification, a thermal cycler with a heated lid should be used. The use of mineral oil for the prevention of evaporation of PCR reaction is not recommended.

Table 1 NF1 Primer Sequences and DHPLC Fragments Analysis
Full size table

Heteroduplexes are formed by subjecting amplified PCR product to 95°C for 5 min and allowing it to cool to 25°C at a ramp of 0.02°C per s. This step can be added to a PCR regimen after the final extension.

3.2.3 Selection of Oven Temperature

The optimal melting temperature for the sequence of interest can readily be predicted by appropriate software if the sequence is known. Apart from WAVEMaker™ software, the Stanford University’s DHPLC melt program (available from website http://insertion.stanford.edu/melt.html) is also widely employed. The majority of the sequence of the NF1 gene is available, and can be retrieved from GenBank (Accession nos. AC004022 and AC004562). The complete sequence of intron 1 is still not available, and thus the conditions for amplification of fragment 1 of the NF1 gene are empirical.

3.2.4 Temperature and Gradient Conditions

The temperature and gradient conditions for the analysis of an amplicon constitute the most important part of the procedure. Gradients for mutation detection comprise a DNA-loading step, the linear separation gradient, a clean-off step, and finally, equilibration. Once the method has been tested, it can be copied from one application folder to another, and thus used extensively for future analysis.

NF1 exon 22 fragment is presented as an example. The resolution Temperature (Tr) is predicted by using Stanford University’s DHPLC melt program. The melting temperature of the entire fragment and each portion of fragment is given, as well as gradient conditions corresponding to each temperature.

This sequence should be run at all of the following temperatures to ensure detection of all polymorphisms.

Recommended temperature: 58°C.

The following concentrations are rough recommendations only. Some adjustment may be necessary. Run from B1 to B2 in 0.5 min, then B2 to B3 in 3.0 min.

For setting up a method on the WAVE™ instrument, the recommended temperature predicted from the Stanford melt program should be used.

The start gradient conditions (0.5 min) may be obtained from the percentage of buffer B given in column B3 of Table 2 (58.8°C in the example shown). The Wave machine requires the operator to enter the concentration of buffer A, which may be obtained by subtracting the percentage of buffer B from 100%. In our laboratory, subtracting 2 from the starting percentage of buffer B was found to give a retention time of between 3 and 4 min for DNA fragments analyzed. The starting percentage of A in the example shown would be 43% (100-57). However, it may be necessary to optimize this according to the fragment size.

Table 2
Full size table

The gradient is formed so that the concentration of buffer B increased by 7% at 4 min. Thus, the value of the percentage of buffer A entered at 4 min would be 36% in the example shown. The flow rate was set to 0.9 mL.min−1.

3.2.5 Setting Up Sample Table

Select the right method and the right position for the sample under analysis. The positioning scheme for the Auto Sampler plate is displayed in Table 3 .

Table 3 Sample Positioning Scheme for WAVE™ Auto Sampler Plate
Full size table

When placing samples in the Auto Sampler, make sure that it is in the right position. There is a cooling system in Auto Sampler to prevent evaporation of samples under higher temperatures. This cooling temperature must be adjusted according to environmental temperature. Heavy condensation may interfere with the system.

Click Data Acquisition icon; select the required sample table.

Click Start Series when the oven temperature is ready to start run.

Chromatograms for NF1-specific mutations are included in Fig. 2 .

See also Table 4 for PCR plate settings.

Fig. 2.
figure 2

DHPLC chromatograms for NF1 mutations identified in exons 4b, 22, and 39.

Full size image
Table 4
figure 3figure 3

PCR Plate Setting for the Amplification of 60 Exons of NF1 Gene Based on Six Test and Two Control (Positive and Negative) Samples

Full size image

4 Notes

  1. 1.

    In the authors’ laboratory, we find that the counterstain containing both DAPI and PI produces a banding pattern that allows ready identification of chromosome 17. To facilitate identification of chromosome 17, labeled DNA from a control probe may be added to the hybridization mixture. This could be from the short arm or distal long arm of chromosome 17.

  2. 2.

    For old slides, or slides that have a large amount of cytoplasm around the chromosomes, pretreatment in pepsin or collagenase may improve probe penetration. In addition, the slide denaturation time may be increased from 2–5 min.

  3. 3.

    A high background signal may be caused by a number of factors that include: the post-hybridization wash being of insufficient stringency, the probe fragments being too large (greater than 500 kb), and poor-quality metaphase chromosome spreads with surrounding cytoplasm. The first problem may be overcome by removing the cover slip, washing the counterstain off the slide with 4T buffer and replacing the slide in a higher-stringency solution such as 1X SSC or 0.5X SSC.

  4. 4.

    A weak or absent signal may be caused by a number of factors that include: insufficient denaturation of the probe or slide, a too stringent post-hybridization wash, insufficient probe concentration, and incorrect use of the detection reagents. A weak signal may be amplified by further application of biotinylated anti-avidin and fluorescein-avidin.

  5. 5.

    The Cot-1 DNA in the probe mixture acts as competitor DNA to suppress repetitive sequences that would interfere with the hybridization signal. The formamide reduces the melting temperature (Tm) of the DNA and the dextran sulfate is an inert polymer which greatly increases the rate of hybridization.

  6. 6.

    Instead of incorporating indirectly labeled haptens into the DNA at nick translation, nucleotides directly labeled with a fluorochrome may be used.

  7. 7.

    If a deletion is found in a patient, parental bloods should be requested for analysis, as there are documented cases of asymptomatic mosaic carriers of the deletion passing on the deletion to their children. Analysis of 30 cells rules out a level of 10% mosaicism with 95% confidence.

  8. 8.

    All the buffer solutions should be made to analytical chemistry standards, using quality-assured glassware. For health and safety reasons, always wear gloves when making buffers. Whenever possible, use water of the highest quality—e.g., Millipore water. The minimum standard is HPLC-grade water. The smaller the amount of transition ions in the water, the more prolonged the column’s life will be.

    Before use, the Duran bottles should be washed out with Millipore water to remove any transition ions, especially if the glassware has been autoclaved.

  9. 9.

    Accurate and consistent compositions are critical when preparing buffers A and B. Elution times are dependent on accurately prepared buffers. Incorrectly prepared buffers could also cause samples to elute early or late.

  10. 10.

    Evaporation will lead to an increase in ionic strength and volatility of the ACN. Both conditions will affect the performance of the instrument.

  11. 11.

    Buffers A and B must be freshly prepared within 1 wk. Buffer C must be changed every 3 wk.

  12. 12.

    The column should be well-equilibrated before the initiation of analysis the first injection. It may be necessary to set up atleast one blank before the first injection.

  13. 13.

    A column can be maintained on the instrument for 2–4 d by pumping the column at a sleep condition (flow rate 0.05 mL/min with 50% buffer A and 50% buffer B). Prior to long-term storage, pump the column with 50% ACN for 30 min at 0.75 mL/min to prevent bacterial growth. Disconnect the column and tightly seal it with end fittings to prevent drying of the bed.

  14. 14.

    If the column is new or has been stored off the instrument, flush the column with 50% buffer A, 50% buffer B, at 0.5 mL/min−1 for at least 45 min, then gradually increase the flow rate from 0.5,.75, and finally to 0.9 mL/min−1. Run three blank gradients before the first analytical run. Recalibration may be necessary by using DNA mol-wt markers (size standard), as well as a mutation standard.

  15. 15.

    Do not start the buffer flow until the oven and the column have reached at least 40°C. Once the temperature is reached, the flow may be started at 0.5 mL/min. An additional 10–15 min will be required for the temperature and pressure to stabilize.

  16. 16.

    After fitting a new peak inline filter. Disconnect the column and flush the peak inline filter with 100% buffer B at 0.5 mL/min for at least 10 min before re-fitting the column.

  17. 17.

    Wash the injection valve and needle before starting a series of runs. Press the “wash” key on the key pad on the front of the auto sampler. Washing solution will be taken up by the syringe in the auto sampler. Then push through the injection valve and out through the wasted line. Repeat this step until you see liquid empty through the waste line into the waste bottle.

  18. 18.

    For every 100 injections, to increase the life of the column, it is advisable to wash the column with 75% ACN at 65°C. Before starting a new analysis, an equilibration for at least 10 min is required to balance the column back to a normal condition.