Download Frameshift mutations of RIZ, but no point mutations in RIZ1

ã
Oncogene (2002) 21, 3038 ± 3042
2002 Nature Publishing Group All rights reserved 0950 ± 9232/02 $25.00
www.nature.com/onc
Frameshift mutations of RIZ, but no point mutations in RIZ1 exons in
malignant melanomas with deletions in 1p36
Micaela Poetsch*,1, Thomas Dittberner2 and Christian Woenckhaus3
1
Institute of Forensic Medicine, University of Greifswald, Germany; 2Department of Dermatology, University of Greifswald,
Germany; 3Institute of Pathology, University of Greifswald, Germany
Recently, the retinoblastoma protein interacting zinc
®nger gene RIZ has been proposed as a candidate for the
tumor suppressor locus on 1p36, because of the common
loss of RIZ1 RNA in human tumors. In addition,
frameshift mutations of this gene have been demonstrated in a variety of tumors with microsatellite
instability. Since alterations in this region have been
described in malignant melanomas, we investigated DNA
of paran-embedded sections from 16 typical naevi, 19
atypical naevi, 33 primary melanoma lesions and 25
metastases and DNA from four melanoma cell lines by
PCR and direct sequencing analysis of RIZ. Frameshift
mutations in the RIZ gene were found in 17% of
melanoma samples and 8.6% of naevi, but we could not
demonstrate any missense mutations in the exons of
RIZ1. No LOH of the RIZ gene nor any microsatellite
instability in six dinucleotide markers or in the
mononucleotide repeats IGRIIR, hMSH3, and hMSH6
could be demonstrated in the samples with RIZ frameshift mutations. Although our results do not explain the
high rate of deletions in 1p36 found in this tumor, they
assign RIZ a potential role in the multi-step tumor
forming process of malignant melanoma of the skin.
Oncogene (2002) 21, 3038 ± 3042. DOI: 10.1038/sj/
onc/1205457
Keywords: melanoma; RIZ; frameshift mutations
Introduction
The short arm of chromosome 1, especially the region
1p36, is deleted or rearranged in many human cancers
including neuroblastoma, breast cancer and malignant
melanoma of the skin (Heim and Mitelman, 1995). A
deletion in 1p36 in malignant melanoma has been
shown by a variety of techniques including cytogenetic
analysis (Thompson et al., 1995), loss of heterozygosity
(LOH) analysis (Dracopoli et al., 1989; BoÈni et al.,
1998) and ¯uorescence in situ hybridization analysis
(FISH) (Poetsch et al., 1998, 1999). These data indicate
that the region 1p36 may harbor one or more tumor
suppressor genes with relevance in malignant melanoma. A possible candidate for such a tumor suppressor
gene may be the retinoblastoma protein-interacting
zinc ®nger gene RIZ which belongs to the PR domain
family (Liu et al., 1997). Members of this PR domain
family are expressed in an unusual yin-yang fashion
with two products di€ering in the presence or absence
of the PR domain. The PR+ product is underexpressed
in cancer cells, whereas the PR7 product is present or
overexpressed. In case of RIZ, a loss of RIZ1 (PR+
product) expression has been shown for example in
hepatocellular carcinoma, breast cancer and neuroblastoma (He et al., 1998; Jiang et al., 1999). Since RIZ1
can induce G2/M cell cycle arrest and apoptosis in
breast cancer and hepatocellular carcinoma cell lines, it
has been supposed as a candidate tumor suppressor
gene (He et al., 1998; Jiang et al., 1999; Jiang and
Huang, 2000). Recently, RIZ has been shown to
demonstrate a high rate of frameshift mutations in
colorectal, gastrointestinal, and endometrial cancer
(Chadwick et al., 2000; Piao et al., 2000; Sakurada et
al., 2001) and forced expression of RIZ1 in colorectal
xenograft tumors with RIZ1 mutations resulted in cells
undergoing apoptosis (Jiang and Huang, 2001). Therefore, we determined the percentage of frameshift
mutations and searched for point mutations in the
exons of RIZ1 by direct sequencing analysis of 35
typical and atypical naevi and 58 malignant melanoma
lesions which ± with two exceptions (case No. NM3
and MM4) ± have been previously analysed by FISH
for deletions in 1p36 (Poetsch et al., 1999). Frameshift
positive cases were also investigated for other microsatellite instability with non-coding dinucleotide repeats and coding mononucleotide repeats.
Results
Frameshift mutations of RIZ
*Correspondence: M Poetsch, Institute of Forensic Medicine, Ernst
Moritz Arndt-University, Kuhstrasse 30, D-17489 Greifswald,
Germany; E-mail: [email protected]
Received 17 July 2001; revised 15 November 2001; accepted 26
November 2001
Only frameshift mutations in the (A)9 track could be
detected in our melanoma samples (Figure 1), but not
in the investigated cell lines. They were found in seven
metastases of six patients (28%), in two nodular
melanoma lesions (11%), in one super®cial spreading
Frameshift mutations of RIZ in malignant melanoma
M Poetsch et al
3039
LOH analysis of RIZ
Of the 49 patients analysed 26 were informative for the
studied polymorphism. Only in two cases (one nodular
melanoma and one super®cial spreading melanoma)
loss of heterozygosity could be demonstrated (data not
shown).
Microsatellite instability analysis
We could not detect any microsatellite instability with
the microsatellite markers D2S123, D3S1611, D5S107,
D17S787, D18S34 or D18S58 in our analysed
melanoma samples. In addition, no frameshift mutations were found in the analysed repetitive mononucleotide tracts of IGFIIR, hMSH3, and hMSH6
(data not shown).
Discussion
Figure 1 Sequence pro®le of the A(9) tract in malignant melanomas. Examples of frameshift mutations in comparison to the
wildtype (above). NM46 (center) has a 1-bp deletion (heterozygous)
and MM41 (below) has a 2-bp deletion (heterozygous)
melanoma lesion (7%), in two atypical naevi (15%)
and in one typical naevus (5%) (Table 1). All but one
of these tumor samples had a 1-bp deletion, all of them
heterozygous. The exception is one metastatic melanoma lesion (M41) which displayed a 2-bp deletion in the
(A)9 track.
Sequencing analysis of RIZ1
Only two point mutations were found in the sequencing analysis of the exons of RIZ1. First, a G?C
change at position 361 could be detected in a part of
the cells from the cell line DX3 resulting in an alanine
to proline exchange in codon 121. Second, a C?T
change at position exons 6 ± 10 (intron 5) could be
found in a metastatic melanoma displaying no frameshift mutations (Figure 2). This mutation was not
found in two other metastases from the same patient
which were also investigated.
The retinoblastoma protein-interacting zinc ®nger gene
RIZ is one candidate for the tumor suppressor gene
locus in the terminal region of chromosome 1 which is
often deleted in malignant melanoma of the skin
(Dracopoli et al., 1989; Poetsch et al., 1998, 1999).
Due to the yin-yang nature of this gene, changes in the
structure of RIZ1 are to be expected for a tumor
promoting e€ect. The frameshift mutation in the (A)9
tract of exon 8 is expected to cause a truncated RIZ
protein lacking part of the C-terminal region involved
in the PR binding function. Therefore, we investigated
super®cial spreading melanoma, nodular melanoma
and metastatic melanoma as well as atypical and
typical naevi for changes in the RIZ gene and loss of
heterozygosity of this gene.
At the moment we are unable to determine the
importance of the two point mutations in RIZ1 found
in this study. The missense mutation in exon 5 in one
cell line concerns the PR domain thus it may disturb
protein function. However, this mutation occurred only
in a part of the cells and no other mutations could be
found in any cell line. The C?T change in intron 5 in
a metastatic melanoma may disturb the splicing of
exon 6. Since only archived paran-embedded material
up to 10 years old was available for our analysis, we
were not able to investigate possible changes of the
mRNA.
The occurrence of frameshift mutations in the (A)9
tract of exon 8 in two SSM, two NM and six
metastases prompted us to investigate three other
repetitive mononucleotide tracts (IGFIIR, hMSH3,
hMSH6) and six dinucleotide markers in these patients
and melanoma lesions looking for additional microsatellite instability (MSI) events. We did not ®nd any
frameshift mutations in the mononucleotide repeats
which is in line with results from other studies
(Richetta et al., 2001). Alterations at dinucleotide
repeats have been reported in malignant melanoma,
but only in 2 ± 30% of primary melanoma lesions
(Richetta et al., 2001; Talwalkar et al., 1998). ThereOncogene
Frameshift mutations of RIZ in malignant melanoma
M Poetsch et al
3040
Table 1 Frameshift mutations of RIZ in tumor samples and naevi
Tumor
samplea
Mutation
del(1)(p36)b
MM15
MM27
MM20
MM25
MM40
MM41
MM44
NM46
NM50
SSM37
AN6
AN12
TN15
(A)9DA
(A)9DA
(A)9DA
(A)9DA
(A)9DA
(A)9DAA
(A)9DA
(A)9DA
(A)9DA
(A)9DA
(A)9DA
(A)9DA
(A)9DA
+
+
+
+
7
+
+
+
+
7
n.d.
n.d.
n.d.
Clark level
Breslow index
Only primary tumors
Survival (months
after diagnosis)c
13, DOD
IV
IV
III
0.3 mm
0.25 mm
1.9 mm
48, DOD
96, AWD
34, DOD
13, DOD
87, DOD
108, NED
82, NED
LTF
a
SSM, super®cial spreading melanoma; NM, nodular melanoma; MM, metastatic melanoma; AN, atypical naevus;
TN, typical naevus; bIndicating deletions in 1p36 in a prior FISH study (Poetsch et al., 1998, 1999); n.d. ± not
determined; cDOD, died of disease, NED, no evidence of disease; AWD, alive with disease; LTF, lost to follow-up
fore, the absence of changes in the dinucleotide
markers tested here is explainable by the small number
of lesions investigated. In gastrointestinal and endometrial carcinomas, MSI has been an important other
criterion in lesions with RIZ frameshift mutations
(Piao et al., 2000; Chadwick et al., 2000). Therefore,
there are two possible reasons for the occurrence of
frameshift mutations in our melanoma samples. First,
the alterations could have been developed by chance.
This is supported by the fact that in two patients with
frameshift mutations we could analyse two or more
metastases, but the mutations were restricted to one of
the metastases. In addition, we could not correlate our
previous FISH results concerning deletions in 1p36 in
malignant melanoma (Poetsch et al., 1998, 1999) with
the occurrence of the frameshift mutations in this
study. We found deletions in 1p36 only in nodular
melanoma lesions or in metastases, but not in super®cial spreading melanoma lesions, whereas two SSM
displayed a frameshift mutation. All mutations here
manifest themselves heterozygous, which is veri®ed by
the fact that no LOH of the RIZ gene could be
detected. In contrast, the same nodular and metastatic
melanoma samples displayed a deletion in 1p36 as
determined by FISH. Therefore, the described deletion
in our tumor samples did not stretch over the RIZ
gene. In addition, a correlation of our results with
survival or recurrence did not reveal any connections.
Second, the RIZ gene could be the one of very few
targets of inactivation by MSI pathways in malignant
melanoma. Loss of DNA-mismatch repair gene
expression has already been demonstrated in melanoma
lesions (Korabiowska et al., 1999; Hussein et al., 2001)
and in other tumors MSI is strongly associated with a
diminished expression of hMLH1, one of the DNA
mismatch repair genes (Fleisher et al., 2000). In
melanoma, however, ®rst studies could not correlate
MSI in microsatellite markers especially from chromosome 1p and 9p with reduction of hMLH1, hMSH2 of
hMSH6 protein expression (Hussein et al., 2001). Since
the authors of this study did not investigate the status
Oncogene
of the RIZ gene, it may well be altered in their
melanoma lesions with decreased mismatch repair
protein expression. Whether the RIZ frameshift
mutations in our melanoma lesions re¯ect a defect in
DNA mismatch repair genes is still to be determined.
Further studies are necessary to elucidate the correlation between the RIZ gene and the DNA mismatch
repair genes.
Nevertheless, since tumorigenesis is assumed to be a
multistep process in which many mutations lead in the
end to the formation of a tumor and since the
percentage of frameshift mutations was higher in
metastases, alterations in RIZ may be important in
the progression of a subset of malignant melanoma of
the skin. Further studies including expression analysis
of RIZ1 and RIZ2 to evaluate the potential role of this
candidate tumor suppressor gene in melanoma are in
preparation. In addition, since deletions in 1p36 are a
characteristic of this tumor (Dracopoli et al., 1989;
Poetsch et al., 1998, 1999) and could not be explained
by alterations of the RIZ gene, the existence of other
tumor suppressor genes in 1p36 with a greater impact
in malignant melanoma of the skin is highly probable.
Materials and methods
Tumors and patients
Fifty-eight non-selected malignant melanoma lesions comprising 14 super®cial spreading melanoma lesions (SSM), 19
nodular malignant melanoma lesions (NM) and 25 metastatic
melanoma lesions comprising a wide range of thickness
(Breslow index) and Clark levels which were staged and
classi®ed as stated elsewhere (Poetsch et al., 1998) were
studied. The samples belonged to 49 patients, whose
corresponding blood samples or normal tissues were also
investigated. Histologic diagnosis of malignant melanoma
was documented for all cases. All specimens underwent
additional independent histopathologic review (C Woenckhaus). In addition, 16 typical and 19 atypical naevi and the
four melanoma cell lines M19, DX3 (Albino et al., 1981),
Frameshift mutations of RIZ in malignant melanoma
M Poetsch et al
Template Preparation Kit (Roche Molecular Biochemicals,
Mannheim, Germany) was used for the DNA extraction.
3041
PCR and sequencing analysis of RIZ
The following PCR conditions were used: 100 ng template
DNA were ampli®ed in a 25 ml reaction volume containing
200 mM dNTPs, 2 mM MgCl2, 15 mM Tris/HCl, 50 mM KCl,
0.3 mM primers, and 1 Unit AmpliTaq Gold (Applied
Biosystems, Weiterstadt, Germany) by denaturing at 958C
for 10 min, followed by 30 cycles of 958C for 30 s, 52 ± 608C
for 90 s, 728C for 120 s, and additional 30 min at 728C with
the Trio Thermocycler (Biometra, GoÈttingen, Germany).
Detection of the frameshift mutations was done with the
primers described by Piao et al. (2000) and the sequencing of
each exon was performed with primers as published by Fang
et al. (2000). For the sequencing PCR the BigDye Ready
Reaction Kit from Applied Biosystems was used with the
same primers as above according to the manufacturer's
instructions with the following PCR conditions: 25 cycles of
958C for 30 s, 608C for 4 min. The PCR products were
sequenced with the ABI 310 DNA sequencer (Applied
Biosystems). Each mutation was veri®ed by a second
sequencing reaction of an independent ampli®cation product.
Assessment of LOH
The RIZ Pro704 deletion polymorphism was assayed by PCR
and analysed by electrophoresis on denaturing gel as
described by Fang et al. (2000).
Microsatellite assay
Figure 2 Sequence pro®le of part of intron 5 and exon 6 of
RIZ1. MM10 (above) displayed a C?T change at position 710
(arrow), whereas in two other metastases of the same patient no
such mutation could be detected (center and below)
MEWO (Siracky et al., 1982), and LT5.1 have been included
in this study.
DNA isolation from peripheral blood and paraffin embedded
tissue
DNA isolation from peripheral blood was performed with
the Blood Extraction Kit (Qiagen, Hilden, Germany)
according to the manufacturer's instructions. DNA isolation
from paran embedded tissue was done as previously
described (Poetsch et al., 2001). Brie¯y, hematoxylin eosin
stained slides were carefully inspected by light microscopy to
identify areas which carry a sucient amount (at least 3
mm2) of tumor tissue measured by a scaled optical
adjustment. This same area was then identi®ed on the
unstained 10 mm dewaxed, rehydrated and airdried tissue
section which was ®xed in an optical installation allowing the
isolation of tumor tissue without adherent non-neoplastic
structures under microscopical control. The High Pure PCR
We studied six dinucleotide microsatellite markers located on
chromosome 2 (D2S123), chromosome 3 (D3S1611), chromosome 5 (D5S107), chromosome 17 (D17S787), and chromosome 18 (D18S34 and D18S58) in those nine patients and 10
tumor samples with frameshift mutations and in additional
®ve matched patient/tumor samples as controls. Primer
sequences were obtained from the Genome Data Base
(http://gdbwww.gdb.org). The PCR was performed in two
multiplex analyses each comprising three di€erent labeled
primers in a ®nal volume of 12.5 ml containing 50 ± 80 ng
DNA, 0.3 mM of each primer, 200 mM of each dNTP, 1.5 mM
MgCl2, 15 mM Tris/HCl, 50 mM KCl, and 1 Unit AmpliTaq
Gold (Applied Biosystems, Weiterstadt, Germany) by
denaturing at 958C for 10 min, followed by 35 cycles of
958C for 30 s, 558C for 90 s, 728C for 120 s, and additional
10 min at 728C with the Trio Thermocycler (Biometra,
GoÈttingen, Germany). 1.5 ml PCR product was mixed with
12 ml deionized formamide and 0.5 ml ROX-labeled internal
lane standard, denatured at 908C for 120 s, snap-cooled on
ice and electrophoresed with the ABI 310 DNA sequencer
(Applied Biosystems).
Frameshift mutation analysis of IGFIIR, hMSH3, and hMSH6
We searched for frameshift mutations in the coding
mononucleotide repeats IGRIIR(G)8, hMSH3(A)8, and
hMSH6(C)8 by PCR and direct sequencing analysis. Primer
sequences and PCR conditions were as described by Catasus
et al. (2000). Sequencing was done as described above.
Acknowledgments
We thank Dr S Burchill for providing the cell lines DX3,
MEWO, and LT5.1, and Prof Dr K Schallreuther and Dr
Oncogene
Frameshift mutations of RIZ in malignant melanoma
M Poetsch et al
3042
N Hibberts for providing the cell line M19. We thank S
Seefeldt and M Maschke for excellent technical assistance.
The authors would also like to thank two anonymous
reviewers for their useful comments.
References
Albino AP, Lloyd KO, Houghton AN, Oettgen HF and Old
LJ. (1981). J. Exp. Med., 154, 1764 ± 1778.
BoÈni R, Matt D, Voetmeyer A, Burg G and Zhuang Z.
(1998). J. Invest. Dermatol., 110, 215 ± 217.
Catasus L, Matias-Guiu X, Machin P, Zannoni GF, Scambia
G, Benedetti-Panici P and Prat J. (2000). Cancer, 88,
2290 ± 2297.
Chadwick RB, Jiang G-L, Bennington GA, Yuan B, Johnson
CK, Stevens MW, Niemann TH, Peltomaki P, Huang S
and de la Chapelle A. (2000). Proc. Natl. Acad. Sci. USA,
97, 2662 ± 2667.
Dracopoli NC, Harnett P, Bale SJ, Stanger BZ, Tucker MA,
Housman DE and Ke€ort RF. (1989). Proc. Natl. Acad.
Sci. USA, 86, 4614 ± 4618.
Fang W, Piao Z, Simon D, Sheu J-C and Huang S. (2000).
Genes Chromosomes Cancer, 28, 269 ± 275.
Fleisher AS, Esteller M, Harpaz N, Leytin A, Rashid A, Xu
Y, Liang J, Stine OC, Yin J, Zou TT, Abraham JM, Kong
D, Wilson KT, James SP, Herman JG and Meltzer SJ.
(2000). Cancer Res., 60, 4864 ± 4868.
He L, Yu JX, Liu L, Buyse IM, Wang MS, Yang QC,
Nakagawara A, Brodeur GM, Shi YE and Huang S.
(1998). Cancer Res., 58, 4238 ± 4244.
Heim S and Mitelman F. (ed.). (1995). Cancer Cytogenetics:
Chromosomal and molecular genetic aberrations of tumor
cells. New York: Wiley-Liss.
Hussein MR, Roggero E, Sudilovsky EC, Tuthill RJ, Wood
GS and Sudilovsky O. (2001). Am. J. Dermatopathol., 23,
308 ± 314.
Jiang G-L, Liu L, Buyse IM, Simon D and Huang S. (1999).
Int. J. Cancer, 83, 541 ± 547.
Jiang G-L and Huang S. (2000). Histol. Histopathol., 15,
109 ± 117.
Oncogene
Jiang G-L and Huang S. (2001). Cancer Res., 61, 1796 ± 1798.
Korabiowska M, Brinck U, Ruschenburg I, HoÈnig JF,
Stachura J and Droese M. (1999). Oncology Rep., 6,
921 ± 923.
Liu L, Shao G, Steele-Perkins G and Huang S. (1997). J.
Biol. Chem., 272, 2984 ± 2991.
Piao Z, Fang W, Malkhosyan S, Kim H, Horii A, Perucho M
and Huang S. (2000). Cancer Res., 60, 4701 ± 4704.
Poetsch M, Woenckhaus C, Dittberner T, Pambor M,
Lorenz G and Herrmann FH. (1998). Lab. Invest., 78,
883 ± 888.
Poetsch M, Woenckhaus C, Dittberner T, Pambor M,
Lorenz G and Herrmann FH. (1999). Virch. Archiv, 435,
105 ± 111.
Poetsch M, Dittberner T and Woenckhaus C. (2001). Cancer
Genet. Cytogenet., 125, 21 ± 26.
Richetta A, Ottini L, Falchetti M, Innocenzi D, Bottoni U,
Faiola R, Mariani-Costantini R and Calvieri S. (2001).
Melanoma Res., 11, 283 ± 289.
Sakurada K, Furukawa T, Kato Y, Kayama T, Huang S and
Horii A. (2001). Genes Chromosom. Cancer, 30, 207 ± 211.
Siracky J, Blasko M, Borovansky J, Kovarik J, Svec J and
Vrba M. (1982). Neoplasma, 29, 661 ± 668.
Talwalkar VR, Scheiner M, Hedges LK, Butler MG and
Schwartz HS. (1998). Cancer Genet. Cytogenet., 104, 111 ±
114.
Thompson FH, Emerson J, Olson S, Weinstein R, Leavitt
SA, Leong SPL, Emerson S, Trent JM, Nelson MA,
Salmon SE and Taetle R. (1995). Cancer Genet. Cytogenet., 83, 93 ± 104.