Journ31l oi un.nderr.3adu.3-e :es-.earch
..Oluin.e 4, is.ue ' - June 2i:l:i-
NF2 LOH in NF1 Tumors
Nicole J. Paulson
Neurofibromatosis type one (NF1) is most commonly characterized by the occurrence of multiple
neurofibromas, which are complex benign tumors arising from peripheral nerve sheaths. In addition
to neurofibromas, pigment abnormalities, such as cafe au lait macules, Lisch nodules of the iris, and axillary
freckling, are also hallmark features of the disease (Shen et al. 1996). Within the last sixty years, it has been
found that NF1 is one of the most common autosomal dominant disorders, and is unbiased to ethnicity or
regional location with a population frequency of 1 in 3500 individuals (reviewed in Friedman and Riccardi, 1999).
The NF1 gene is located on chromosome 17 (17q11.2) and encodes a GTPase activating protein, neurofibromin.
A distinct feature of the NF1 gene is the very high spontaneous mutation rate, estimated at 1 per 10,000 alleles
per generation (Shen et al. 1996). This mutation rate is about 100 fold higher than the usual mutation rate for
a single locus, resulting in approximately 50% of all NF1 cases being heterozygous for new mutations (reviewed
in Friedman and Riccardi, 1999).
Mutation in the large NF1 tumor suppressor gene is the first step in neurofibroma development.
Dermal neurofibromas are discrete, cutaneous or subcutaneous lesions (which tend to be small), whereas
plexiform neurofibromas are larger and more complex tumors, developing from larger nerves (Korf, 1999).
Malignant peripheral nerve sheath tumors (MPNSTs) are thought to arise from pre-existing plexiform tumors
and occur in 3-5% of NF1 patients (Wood, 1999). The mechanism in which tumorigenesis occurs in NF1 follows
the two-hit hypothesis for tumor-suppressor genes that was initially postulated by Knudson (1971). According
to Knudson's hypothesis, this model for NF1 would predict that individuals are constitutionally heterozygous,
meaning the mutated allele is inactivated, with tumors such as neurofibromas arising due to loss of the
remaining normal allele due to somatic mutation. One mechanism is deletion of the normal allele, which can often
be detected by a method called loss of heterozygosity (LOH) analysis (Colman et al., 1995).
Somatic mutations that inactivate the remaining NF1 allele may initiate tumorigenesis; however, alterations in
other genes, such as NF2, may be necessary for tumor progression. The NF2 gene, also a tumor suppressor,
is located on chromosome 22 (22q12), and is associated with nervous-system related tumors (Wallace and
MacCollin, 2001). Neurofibromatosis type 2 (NF2), also known as Bilateral Acoustic Neurofibromatosis (BAN)
or Central NF, is less common than NF1, occurring in 1 out of every 40,000 people worldwide. Mutations in the
NF2 gene (with the resultant loss of expression/function of the NF2 tumor suppressor protein) are one of the
most common causes of benign human brain tumors, including acoustic neuromas and meningiomas (Wallace
and MacCollin, 2001). Early symptoms of NF2 are dysfunction of the acoustic (sound) and vestibular
(balance) nerves. These symptoms are related to the hallmark of NF2, bilateral schwannoma arising from
the vestibular branch of the eighth cranial nerve. Over 90% of individuals inheriting NF2 will eventually
develop bilateral vestibular schwannomas as well as also potentially developing schwannomas at other locations
in the peripheral nervous system (Wallace and MacCollin, 2001). Also, the NF2 protein and neurofibromin may
be functionally linked through independent involvement with kinesin complexes (Hakimi et al. 2002). Thus, to
test the involvement of NF2 genetic alterations in NF1 tumors, we screened NF1 tumors, NF1 tumor Schwann
cell cultures and non-NF1 Schwann cell tumors for loss of an allele in the NF2 gene. Our data show LOH in some
of the samples, which could implicate the NF2 gene in NF1 tumorigenesis.
MATERIALS AND METHODS
Samples were obtained by extracting DNA from the patients' blood and tumor specimens. Patients included
those diagnosed with NF1, NF2, or neither in the case or non-NF2 schwannomas. Schwann cell tumor cultures
were prepared and provided through a collaboration with Dr. David Muir (UF Neuroscience).
Loss of heterozygosity (LOH) studies were employed to examine two polymorphisms on chromosome 22:
a tetranucleotide microsatellite polymorphic marker in intron 1 (NF2tet 5'-GAG AAT CGC TTA AAC CTG-3'; 5'-
CCT TAT GCC ATG TTC TTG-3', 550C) and a restriction fragment length polymorphism (RFLP) in intron 12
(NF2Il12F 5'-GGC CTG CAA CCT TCA GAT AAA C-3'; NF2Il12R 5'-ATA AAA CAC CAG TGG GCC TTG T-3', 58UC).
The primers were designed based on NCBI's database of single nucleotide polymorphisms (SNPs) and
Genbank genomic sequence of the NF2 gene. These primers were used for LOH analysis by amplification of
the specific NF2 regions in patient blood and tumor(s) samples using standard polymerase chain reaction
(PCR) buffer and cycling conditions: 1 minute at 94UC, 1 minute at primer specific annealing temperature, and
1 minute at 720C for 35 cycles followed by a 30 minute final extension at 720C. The NF2tet amplified
products (~270bp) were electrophoretically separated on 6% SpreadEx acrylamide gel (Elcrom Scientific),
stained with ethidium bromide, and visualized under a UV light. The NF2112 amplified products (363bp)
were digested using AccI (NEB), fragments separated on an 8% polyacrylamide gel, and also stained with
ethidium bromide to enable visualization with UV light. For individuals found to be heterozygous constitutionally
(in blood), tumor samples were then screened to look for allelic loss (LOH).
If there was a clear loss of one allele in tumor versus blood, the tumor was considered positive for LOH and
the experiment repeated for confirmation. Subtle shifts in intensity were examined in several repetitions
and analyzed through a densitometric analysis (NIH Image freeware). Subsequent chi-square analysis was used
to statically confirm the shift in allele intensity and LOH (p <: 0.05).
RESULTS AND DISCUSSIONS
NF1 dermal, plexiform, and MPNST tumors, including some Schwann cell cultures, were tested for LOH in the
NF2 gene. We are the first to report finding NF2 allele loss in NF1 tumors. Informative samples (samples that
were heterozygous for both NF2tet and/or NF2Il12) were analyzed and the results are shown in Table 1, with
an example shown in Figure 1. The analysis of both markers found LOH in the MPNSTs of NF1 patients, which
was not unexpected given the malignant nature of these cells. However, there was also LOH in some dermal
and plexiform neurofibromas. There was no LOH observed in any of the Schwann cell cultures dermall, MPNST,
or plexiform). Schwannomas and schwannoma cultures from NF2 patients showed LOH for NF2 markers, which
was expected. Additionally, schwannomas from non-NF2 patients showed LOH at both markers.
Results of NF2 LOH Analysis on DNA from informative primary tumors and tumor-derived cultures
Marker NF2tet NF2112 (Accl RFLP)
Sample Total LOH %LOH Total LOH %LOH
Dermal neurofibromas 29 0 0 23 1 4.4
Dermal Schwann cell cultures 1 6 0 0 3 0 0
Plexiform neurofibromas 52 4 7.7 20 4 20
Plexiform Schwann cell cultures 2 12 0 0 10 0 0
MPNST 6 4 66.7 2 1 50
MPNST cultures 3 3 0 0 2 0 0
Schwannomas (NF2) 2 2 100 2 2 100
Schwannoma culture (NF2)4 1 1 100 1 1 100
Schwannomas (non-NF2) 7 4 57.1 5 3 60
1 corresponds to 5 primary samples
2 corresponds to 10 primary samples
3 corresponds to 1 primary sample
4 corresponds to 1 of the primary schwannomas
AIkle I -
AlIkke2 E .,. l
Figure 1. Gel showing genotyping of NF2112 RFLP in two patients' germline DNA (blood samples) and their tumor
samples (T). UF450 and 460 are constitutionally heterozygous at this locus, and the relative intensities of allele 1 (uncut)
versus the two fragments of allele 2 in the blood samples set a normalization for equal quantities of each allele. In 450T,
allele 2 shows decreased intensity relative to allele 1, with residual allele signal due to normal cells within the tumor
In 460T, a complete loss of allele 2 is seen, which is consistent with this being an MPNST (malignancy), a much
purer tissue than the benign tumors
In the NF2tet marker of NF1 patients, 0% of the dermal tumors showed LOH; however, 7.7% of the
plexiform neurofibromas had LOH. A majority of the MPNSTs (4/6) showed LOH at NF2tet. There were no
cultures that showed NF2tet LOH for dermal, plexiform, or MPNST tumors.
For marker NF2112, 1/23 (4.4%) informative dermal neurofibromas showed LOH. In addition, 20% of the
informative primary plexiforms showed LOH. One of the two informative MPNSTs had NF2 LOH at this marker. As
in NF2tet, none of the were positive for NF2I12 LOH.
This NF2 analysis was limited to intron 1 (NF2tet) and intron 12 (NF2I12); therefore, any deletions outside
those regions, but affecting the gene, would not have been detected in this analysis. As a consequence, the
detected frequency of LOH in this study may actually be an under-estimate of the true involvement of NF2
mutations in NF1 tumors. Table 2 indicates previously defined NF1 LOH status for tumors found with NF2 LOH.
Other genetic data for NF1 tumors positive for NF2 LOH
Sample Type NF1 LOH NF2tet NF2112
505Tg Dermal No LOH No LOH LOH
378T1 Plexiform No LOH LOH NI
450T Plexiform No LOH LOH LOH
469T Plexiform No LOH LOH LOH
499T2 Plexiform LOH No LOH7 LOH
573T4 Plexiform LOH LOH LOH
282T MPNST LOH LOH NI
284T MPNST No LOH LOH NI
441T MPNST LOH LOH NI
460T MPNST LOH LOH LOH
ND = Not done
Tc = Tumor derived Schwann cell culture
NI = Not informative (sample was homoztgous at that marker)
As predicted, the NF2-derived schwannoma samples and cultures were all positive for NF2 LOH in both markers
(two independent tumors and one tumor derived Schwann cell culture from the same patient). These data
are consistent with the aforementioned and well-documented two-hit phenomenon occurring in the NF2
gene (Wallace and MacCollin, 2001). The non-NF2 schwannomas for NF2112 were from a same patient who did
not meet the diagnostic criteria for NF2, but had multiple dermal schwannomas in one arm. The NF2tet
marker included two other patients that had schwannomas not associated with NF2. Non-NF2 schwannomas
showed LOH at both markers, 57.1% at NF2tet and 60% at NF2112. This supports the notion of a role for the
NF2 gene in non-NF2 schwannomatosis.
The discovery of NF2 LOH in tumors may suggest that reduction in the NF2 protein contributes to neurofibromas
and MPNST tumorigenesis in some cases. Alterations in the NF2 region have been previously discovered in
NF1 MPNSTs (Koga et al., 2000). Alternatively, it could indicate that the tumors have multiple random
somatic rearrangements, and the NF2 LOH is a merely a coincidence. Or, the NF2 gene could be lost
coincidentally, and/or another nearby locus on 22 could have functionally driven the allelic loss (Bruder et al.,
1999). Of interest, the frequency of NF2 LOH in NF1 tumors increases with complexity of the tumor type,
dermal being most simple and MPNST most complex. This is also true of complexity of karyotype rearrangements
in these tumors (Wallace et al., 2000).
These findings may provide further insight into NF1 tumorigenesis and the development of neurofibromas.
Further investigation into the effects of NF2 gene loss in cells that are already lacking the NF1 gene could help
define the cooperative, functional consequences leading to tumorigenesis.
Bruder CE, Ichimura K, Tingby 0, Hirakawa K, Komatsuzaki A, Tamura A, Yuasa Y, Collins VP, Dumanski JP.
(1999). A group of schwannomas with interstitial deletions on 22q located outside the NF2 locus shows no
detectable mutations in the NF2 gene. Human Genetics 104:418-24.
Colman SD, Williams CA, and Wallace MR. (1995) Benign neurofibromas in type 1 neurofibromatosis (NF1)
show somatic deletions of the NF1 gene. Nature Genetics 11: 90-92.
Friedman JM, and Riccardi VM. (1999) Clinical and Epidemiological Features. In: Neurofibromatosis:
Phenotype, Natural History, and Pathoqenesis (ed. Friedman, J.M., et al.), pp.29-86. Johns Hopkins Press, Baltimore.
Hakimi MA, Speicher DW, Shiekhattar R. (2002) The motor protein kinesin-1 links neurofibromin and merlin in
a common cellular pathway of neurofibromatosis. Journal of Biological Chemistry 277(40):36909-12.
Koga T, Iwasaki H, Ishigure M, Matsuzaki A, and Kikuchi M. (2000). Frequent
genomic imbalanes in chromosomes 17, 19, and 22q in peripheral nerve sheath tumors detected by
comparative genomic hybridization analysis. Journal of Pathology 197:98-107.
Knudson AG. (1971) Mutation and cancer: statistical study of retinoblastoma, Proceedings of the National Academy
of Sciences USA 68:820-823.
Korf BR (1999). Plexiform neurofibromas. American Journal of Medical Genetics
Shen M, Harper P, and Upadhyaya M. (1996). Molecular genetics of neurofibromatosis type 1 (NF1). American
Journal of Medical Genetics 33:2-17.
Wallace M., Rasmussen SA, Lim IT, Gray BA, Zori RT, and Muir D. (2000) Culture of cytogenetically
abnormal Schwann cells from benign and malignant NF1 tumors. Genes Chromosomes and Cancer 27:117-123.
Wallace MR and MacCollin M (2001). Genetics of NF1 and NF2. In "Molecular Genetics of Cancer", Second edition, J.
K. Cowell (ed.), BIOS Scientific Publishers, Oxford, pp. 95-113.
Wood JM. (1999). Pathology of tumors of the peripheral nerve sheath in type 1
neurofibromatosis. American Journal of Medical Genetics 89:23-30.
Back to the Journal of Undergraduate Research
College of Liberal Arts and Sciences I University Scholars Program I University of Florida I
Iy UNIVERSITY of
The 'iKiaKrATTTGir Vtin
ï¿½ University of Florida, Gainesville, FL 32611; (352) 846-2032.