Journal of Undergraduate Research
Volume 8, Issue 2
2 - November/December 2006

NF1 Mutations in Children with Learning Disabilities

Astrid Botty Van Den Bruele

ABSTRACT

The neurofibromatosis 1 (NF1) gene is a tumor suppressor gene that encodes neurofibromin. Mutations in the NF1 gene gives rise to neurofibromatosis type 1, a common autosomal dominant tumor condition characterized by localized overgrowth of neural crest derived tissues. The diagnosis of NF1 is based on clinical findings. Learning disabilities (LD) in NF1 patients are frequent. Learning disabilities with or without attention deficit hyperactivity disorder (ADHD) are seen in approximately 40% of NF1-affected individuals.¹ A much smaller percentage experience more significant cognitive difficulties such as mild or moderate mental retardation.¹  There is currently no way to predict which NF1 patients will also suffer from learning disabilities. The goal of my research was to find NF1 mutations in patients with well-characterized cognitive abnormalities and to observe whether certain kinds of mutations correlated with learning disabilities. Since the majority of patients with NF1 have mutations showing  severe clinical effects, the data gathered  suggested that only individuals with severe NF1 mutations correlate with learning disabilities. 

INTRODUCTION

NF1 is a multisystemic disorder affecting many tissues in the body. The NF1 disorder is primarily distinguished by benign Schwann cell tumors, also known as neurofibromas, on peripheral nerves, but other features affect the central nervous system and the peripheral nervous system as well. Patients with NF1 have a germline mutation that is thought to inactivate the gene and protein. The inactivation of neurofibromin has been linked to the initiation of many of the NF1 features, such as neurofibromas. Half of the affected individuals have NF1 as the result of a new NF1 gene mutation. The offspring of an affected individual have a 50% risk of inheriting the altered NF1 gene, but the disease expression is extremely variable, even within a family. NF1 is one of the most common autosomal dominant genetic disorders, with an incidence of 1 case per 3000-4000 people worldwide. The 1 per 10,000 germline mutation rate in the NF1 gene is among the highest known for any single human gene.²  The cause of the unusually high mutation rate in NF1 is unknown.²

Clinical presentation depends on the body system involved. Individuals affected with NF1 usually meet the criteria for diagnosis during childhood.³ In cases in which no family history exists, diagnosis may not occur until early adulthood or when physical symptoms begin to appear. NF1 is known to be a capricious disorder in which its symptoms range from fairly moderate to severe. Why some patients have severe symptoms whereas others have mild symptoms remains unknown. Criteria for the diagnosis of NF1 are met in an individual who shows two or more of the following features: 

• Six or more café au lait macules (size based on age)
• Two or more neurofibromas of any type or one plexiform neurofibroma
• Multiple freckles (Crowe sign) in the axillary or inguinal region
• A distinctive osseous lesion, such as sphenoid dysplasia or thinning of long bone cortex, with or without tibial dysplasia
• Optic glioma
• Two or more iris hamartomas (Lisch nodules)
• A first degree (immediate) relative diagnosed with NF1.¹

Café au lait spots are a classic feature of NF1, as they are found in approximately 95% of all affected individuals.⁴ They are characterized by their flat birthmark appearance with distinct edges that are darker than the surrounding skin. They are macules larger than 5 mm in the greatest diameter in prepubertal children and larger than 1.5 cm in postpubertal individuals.⁵ Most individuals affected with NF1 have 6 or more of these distinctive spots. Neurofibromas are benign tumors of the peripheral nerve sheath and mainly contain Schwann cells. Plexiform neurofibromas are tumors that spread out, can grow in along the nerve, and can be quite large and disfiguring. Plexiform neurofibromas occur in 15-25% of patients, grow diffusely along deeper or larger nerves, and can lead to significant complications.⁶ Optic gliomas can be categorized into 2 groups, benign optic glioma in children and aggressive glioma in adults.⁷ In most instances, optic pathway gliomas arise in children, and in many cases, affected children have NF1. The lesion may or may not result in decreased visual acuity in the affected eye, but is likely to cause visual and additional symptoms when it is large. Optic gliomas occur in roughly 15-20% of patients but only 1-5% develops symptoms.⁷ The symptoms are mostly benign in nature.⁷ Lish nodules, another common sign of NF1, are present in all adults with this disorder and begin to appear in late childhood. These nodules  are tan or brown benign tumors on the iris of the eye and are usually less than two millimeters in diameter.

Additional complications with NF1 include macrocephaly (45%), short stature (25-35%), scoliosis (12-20%), tibial displasia (3%), renal artery stenosis (1-2%), and increased risks for malignancies (>5%).⁸ Individuals affected with NF1 were found to have a below-average life expectancy. The life expectancy is 54.4 for males and 59 for females.⁸  Additionally, malignancies were the cause of death in NF1 patients 1.2 times more often than the general population, and vascular disease was recorded more often in NF1 patients who died before the age of 30.⁸

The NF1 gene is located on chromosome 17q11.2,⁹  spanning 280kb of genomic DNA and including over 60 exons. NF1 mutations of germline DNA consist of deletions, insertions, point substitutions and splicing errors; hundreds of known mutations are spread across the gene.¹⁰ Direct repeats, palindromes and symmetrical elements also contribute to deletions and insertions in NF1.¹¹ Unequal mitotic and meiotic recombination between large repetitive sequences flanking NF1 can lead to microdeletions and subsequent hemizygosity for NF1 in about 5-10% of cases.¹²  The NF1 disorder has a very high mutation rate, estimated at 1 per 10,000 alleles per generation, approximately 10-fold higher than most genes.¹³  It is likely that this high mutation rate contributes greatly to the frequency of NF1 cases. In approximately half of all patients, the disease is the result of a new germline mutation. Given the gene’s instability, large size and likelihood for variation within the NF1 transcript, inactivation of the resulting protein is predicted to occur for most mutations. Few mutations are recurrent, thus mutation screening is very difficult.

The middle eighth of the gene (encoded by exons 21-27b) shows strong homology to GTPase-activating proteins (GAPs).^14,15 GAPs are at least partially responsible for keeping ras proteins, which play a crucial role in cell proliferation, in a GDP-bound, inactive state.¹⁶Neurofibromin’s exact role in the signal transduction pathway is complex, as indicated by studies which have shown that when neurofibromin is bound to tubulin it has diminished GAP activity.¹⁷  Downregulating ras is thought to be neurofibromin’s pertinent role in tumor suppression.

Learning disabilities in NF1 has been studied extensively by many groups. Disorders of reading, attention and executive function are frequent obstacles to learning in NF1 patients, and each of these disorders has been previously linked in some way to deficits of the frontal lobe function. The etiology of these deficits remains unknown but may be related to increased ras activity. Learning disabilities occur in 30% to 45% of patients with NF1, even in the absence of any apparent neural pathology. Individuals affected with NF1 can have learning disabilities that are mild and nonprogressive, impairment of motor coordination and cognitive and physical manifestations that can affect self-esteem.¹⁹ Brain MRI studies showed high-T2-signal-intensity lesions in 43-79% of NF1 patients.¹⁹ These lesions (also called “unidentified bright objects”) are most common in the basal ganglia, internal capsule, thalamus, cerebellum and brain stem.⁶ The lesions do not have mass affect, are likely attributable to unusual water concentration and decrease with age. Some studies have shown an association with lower IQ and decreased neuromotor functions in patients with these lesions, but not all studies found the same correlation and a positive link remains to be identified.

This research tests whether certain types or locations of NF1 mutations correlate with learning disabilities or other cognitive and/or anatomic abnormalities. The samples and clinical data were provided by Dr. Robert Greenwood, Pediatric Neurologist at University of North Carolina.  To date,  of 40 affected subjects, 18 mutations have been found and categorized. 

MATERIALS AND METHODS

DNA and RNA samples were extracted from patient blood from both affected subjects and un-affected individual siblings. The samples came from the University of North Carolina and were extracted at the University of Florida. The whole project involved 40 patients plus unaffected siblings. I studied 22 patients and 4 unaffected siblings for particular regions of the gene. Dr. Greenwood measured 60 physical variables, over a dozen cognitive measures and several MRI variables.

Polymerase chain reactions (PCRs) were performed to amplify exons 2, 4a-c, 5, 6, 7, 35, 36, 38, 39, 40, 41 and 43 for certain affected and unaffected patient samples. These particular exons and samples were specially selected from previous data which had already implicated these portions of the NF1 gene. Once specific PCR products were obtained, single strand conformational polymorphism (SSCP) was performed using native 0.8mm, 10% polyacrylamide gel electrophoresis (PAGE). SSCP and silver staining were used to identify if and where a mutation existed.SSCP is the electrophoretic separation of single-stranded nucleic acids based on fait differences in sequence.¹⁸ The mobility of double-stranded DNA in gel electrophoresis is dependent on both the size of the strand and length of the product (i.e., how many base pairs). The mobility of single strands, however, is noticeably affected by very small changes in sequence due to the unstable nature of single-stranded DNA. In the absence of a complementary strand, the single strand may experience intra-strand base pairing which results in various changes in its overall conformation (such as loops and hairpins which may function in the termination of transcription). The run time for SSCP is highly variable. Gels are run at a constant 9 watts with varying volt/hours depending on the size of the PCR product. Three microliters of PCR product + three microliters of denaturing stop buffer dye were heated at 95ºC for 5 minutes and then loaded onto the gel. A molecular weight ladder was also used, not as a size indicator but rather as a control to determine whether the gel ran properly. Once the gels were run for the appropriate amount of V/hr, they were silver stained.²⁰ In an effort to identify whether the mutations existed, comparisons of the banding patterns were made. If an aberrant pattern was found, further investigation, including a purification step and big dye reaction, were done to prepare the sample for direct cycle sequencing. Once the sample was sequenced, the mutation, if present, could be identified.²⁰

Primer Sequences and Temperatures

PCRs were performed to amplify the DNA gene segments.. Two different single-stranded oligonucleotide primers were used  for the PCR reactions, which are flanking the 5’ and 3’ ends of the following exons: 2, 4a, 4b, 4c, 5, 6, 7, 35, 36, 38, 39, 40 and 43. These samples were then amplified under the following conditions: 94ºC for 1 minute, 60ºC (or 65ºC depending on the exon) for 1 minute, 72ºC for 1 minute for 37 cycles with 20 minute 72ºC final extension step. Hotstar and Roche Taq Polymerase were used in these reactions.

Sequencing

After SSCP gel analysis, purification of the possibly mutated product must be completed before the sample could be run on the either the ABI 310 or 377 sequencer. This involves purifying the PCR product, subjecting it to cycle sequencing using the exon primers (ABI Big Dye 2 Kit) and a final purification before being run on the machine. Data was analyzed using SeqEdv.1.0.3 program on a MacIntosh computer.

WAVE analysis (Transgenomic WAVE dHPLC machine) is used to screen for heterozygous changes in several exons that have not yet been examined in a set of patients whose mutations remain unknown. WAVE analysis involves PCR amplification of the exon using the Discoverase Taq polymerase, followed by analysis on the WAVE machine. The WAVE machine partially denatures the products and separates them on the column. Products eluting from the column will then be detected with a fluorescent lamp. The resulting chromatograms show peaks in abnormal places if there are any mutations, because those heteroduplexes melt differently than a normal homoduplex. Abnormal samples will then be sequenced.

RESULTS

This work involved two parts. The first part involved SSCP of 6 exons in the 5’ end of the gene, to follow up on samples that showed abnormal RNA patterns (Table 1).  One positive was found out of seven sequenced exons, and that mutation displayed an inframe three base pair deletion (Figure 1). Mutations predicted to result in loss of function included 14 patients out of the 18 known mutations. Eleven mutations prior to the GAP domain were found, all predicted to cause loss of GAP function (Table 2). The three other mutations that occurred after the GAP domain were also not likely to result in a protein that has function (Table 2). Four patients showed less severe mutations, non-null/non-truncation. These mutations are not expected to affect GAP function with the exception of the missense mutations in the GAP domain.

The other part of this project involves WAVE analysis of exons 35, 36, 38, 39, 40, 41 and 43, underway with no mutations so far.  Four patients with less-severe mutations were less likely to have skeletal anomalies and plexiform neurofibromas than patients with more severe mutations. There were no other obvious differences, but the numbers are too small for statistical significance.

Figure 1. CCChromatogram of sequencing results for exon 2 of patient 98-1007. The sequence shows overlapping peaks where the three base pair deletion occurs as it is reading the normal and deleted alleles at the same time.

DISCUSSION

The in-frame deletion in exon 2 patient 98-1007 falls into the more common mutation type found among individuals affected with NF1 as it is the result of deletion of direct 3-bp repeat. Reviewing the clinical symptoms of this  patient,  the individual did not display anything strikingly abnormal. The patient had a slightly unusual dermatoglyphic pattern (finer/palm print pattern), but the significance of this is not clear. Since the majority of patients with NF1 have mutations that yield a severe clinical affect, the data suggested that only individuals with severe NF1 mutations correlate with learning disabilities.  As displayed in Figure 1, the 5’ direction of the sequence of exon 2 reveals a clean beginning, and as one moves further to the right the peaks become overlapped and a noticeable shift can be seen. From the 3’ direction, the pattern occurs in the reverse direction (from right to left). The mutation, which caused the deletion of a codon never before  reported in NF1, is one of the most common 5’ mutations reported. Since this part of neurofibromin is completely conserved at the protein level in other species, it is reasonable to predict some altered protein formation, but it is not clear how this would affect the GAP activity, if at all.  Despite the fact that this mutation would not have been predicted to affect GAP activity upfront, the patient did not have a mild form of learning disabilities (LD), or any other features that seemed usually mild.

The majority of the mutations revealed during this investigation are predicted to result in a loss of function for the protein (Table 2), particularly of the GAP domain. Only 3 out-of-frame frameshifts located distal to the GAP domain were predicted to be severe enough to result in the inhibition of the tumor suppressor protein. This could be due to the degradation of the truncated protein. The data agrees with current thinking that mutations reducing GAP activity are associated with more complications, such as LD.

In this study, only 3 cases were documented in which in-frame shift mutations that would not be predicted to inhibit the GAP activity. This,suggests that the GAP domain may not, in fact, display a conclusive link with learning disabilities. No direct relationship currently exists between the severity of the mutation and  learning disability.  More research is needed to generate evidence of this relationship, including protein-level studies.

This study's sample size is too small for statistical significance. However, the data did lead me to hypothesize other possibilities concerning  the correlation between NF1 mutations affecting the GAP domain and its effect on learning disabilities. Since not all children affected with the NF1 disorder display cognitive problems, the real cause could involve  other phenomena. Perhaps the cause for the cognitive problems could lie within the brain cells, in an area that could fall victim to additional mutations by random events. Thiscould  explain why some patients fall into both categories. Or, perhaps other genetic background or environmental effects are responsible for the development of LD.

---

REFERENCES

1. Pletcher, B A. Neurofibromatosis Type 1. AAP Committee on Genetics: Health supervision for children with neurofibromatosis. Pediatrics 1995 Aug; 96 (2 Pt 1): 368-72
2. Friedman JM, Birch PH: Type 1 neurofibromatosis: a descriptive analysis of the disorder in 1,728 patients. Am J Med Genet 1997 May 16; 70(2): 138-43
3. DeBella K, Szudek J, Friedman JM: Use of the national institutes of health criteria for diagnosis of neurofibromatosis 1 in children. Pediatrics 2000 Mar; 105(3 Pt 1): 608-14
4. Sunnucks, P., et al.  SSCP Is Not So Difficult: The Application and Utility of Single-Stranded Conformation Polymorphism in Evolutionary Biology and Molecular Ecology.  Molecular Ecology.
5. Khan, N.A. 2005. “Neurofibromatosis Type 1”. E-medicine. http://www.emedicine.com/RADIO/topic474.htm
6. Conologue , T. 2005. “Dermatologic Manifestations of Neurologic Disease.” E-medicine. http://www.emedicine.com/DERM/topic549.htm
7. Woodcock, R. “Optic Nerve Glioma.” 2005. E-medicine. http://www.emedicine.com/RADIO/topic486.htm
8. Ward K, O'Connell P, Carey JC, et al: Diagnosis of neurofibromatosis I by using tightly linked, flanking DNA markers. Am J Hum Genet 1990 May; 46(5): 943-9
9. Goldgar DE, Green P, Parry DM, Mulvihill JJ. Multipoint linkage analysis in neurofibromatosis type 1: an international collaboration. Am F Human Genet 1989, 44, 6-12.
10. NF1 International Mutation Database
11. Bailey, A. D., and C. K. Shen. 1993. Sequential insertion of Alu family repeats into specific genomic sites of higher primates. Proc Natl Acad Sci U S A. 90(15):7205-9.
12. Lazaro, C., A. Gaona, A. Ravella, V. Volpini, T. Casals, J. J. Fuentes, and X. Estivill. 1993. Novel alleles, hemizygosity and deletions at an Alu-repeat within the neurofibromatosis type 1 (NF1) gene. Hum Mol Genet. 2(6):725-30.
13. Crowe FW, Schull WJ, Neel JV. A Clinical, Pathological and Genetic Stufy of Multiple Neurofibromas. Springfield, Illinois, Carles C. Thomas Publishers, 1956.
14. Buchberg AM, Cleveland LS, Jenkins NA, Copeland NG. Sequence homology shared by neurofibromatosis type-1 gene and IRA-1 and IRA-2 negative regulators of the ras cyclic AMP pathway. Nature 1990, 347, 291-294.
15. Xu G, O’Connell P, Viskochil D, et al. The neurofibromatosis type 1 gene encodes a protein related to GAP. Cell 1990, 62, 599-608.
16. Downward R. Regualtion of p21ras by GTPase activating proteins and guanine nucleotide exchange proteins. Curr Opin Genet Devel 1992, 2, 13-18.
17. Bollag G, McCormick F, Clark R. Characterization of full-length neurofibromin: tubulin inhibits ras activity. EMBO F 1993, 12, 1923-1927.
18. Abernathy et al., 1997.
19. Riccardi VM (1999) Historical background and introduction. In: Friedman JM, Gutmann DH, MacCollin M, Riccardi VM (eds) Neurofibromatosis phenotype, natural history, and pathogenesis. Johns Hopkins University Press, Baltimore, pp 1-25.
20. Sunnucks, P., et al.  SSCP Is Not So Difficult: The Application and Utility of Single-Stranded Conformation Polymorphism in Evolutionary Biology and Molecular Ecology.  Molecular Ecology; 2000, (9): 1699-710.
21. Gutmann DH: Recent insights into neurofibromatosis type 1: clear genetic progress. Arch Neurol 1998 Jun; 55(6): 778-80.

---

--top--

Back to the Journal of Undergraduate Research