Download Genomic structure and mutational analysis of the human

INTERNATIONAL JOURNAL OF ONCOLOGY 23: 737-744, 2003
737
Genomic structure and mutational analysis of the human
KIF1B· gene located at 1p36.2 in neuroblastoma
YU YAN CHEN1, JUNKO TAKITA1, YING ZHANG CHEN1,2, HONG WEI YANG1,
RYOJI HANADA3, KEIKO YAMAMOTO3 and YASUHIDE HAYASHI1
1Department
of Pediatrics, Graduate School of Medicine, University of Tokyo, Tokyo 113-8655; 2Gene Bank,
Tsukuba Institute, Institute of Physical and Chemical Research (RIKEN), Tsukuba 305-0074;
3Division
of Hematology/Oncology, Saitama Children's Medical Center, Saitama 339-0077, Japan
Received March 31, 2003; Accepted May 16, 2003
Abstract. KIF1B is a member of the kinesin superfamily
proteins that are microtubule-dependent molecular motors
involved in important intracellular functions such as organelle
transport and cell division. We previously determined the
structure of the human KIF1Bß gene, which was found to be
a homologue of the murine Kif1bß, and demonstrated that the
human KIF1Bß is a causative gene of Charcot-Marie-Tooth
disease type 2A although we did not prove that it is a tumor
suppressor gene of neuroblastoma. Here, we identified another
isoform of the human KIF1B gene, KIF1B·. The KIF1B· and
KIF1Bß are alternative splicing products of the KIF1B gene
located on 1p36.2. The KIF1B· is distinct from KIF1Bß in
the C-terminal cargo-binding domain; however, they have the
same N-terminal motor domain. We found that the transcript
of approximately 7.8 kb of KIF1B· was expressed in several
tissues, especially in skeletal muscle, by Northern blot analysis.
To determine whether this gene is one of the candidate
tumor suppressor genes for neuroblastoma (NB) or other
pediatric solid tumors, we performed mutational screening of
KIF1B· in 25 NB, 9 rhabdomyosarcoma, 12 Ewing sarcoma
and 24 other pediatric solid tumor cell lines. Using RT-PCR
single-strand conformation polymorphism analysis and direct
sequencing we detected a missense mutation (M807I) in 1 NB
cell line (SK-N-SH), 3 silent mutations in 2 NB cell lines and
1 primitive neuroectodermal tumor cell line, respectively. RTPCR analysis revealed that KIF1B· was obviously expressed
in almost all of the tumor cell lines examined except NB-1.
Furthermore, real-time quantitative RT-PCR showed that
there was no significant difference in KIF1B· expression
between 14 early-stage (stage I and II) and 14 advanced-stage
(stage III and IV) NB fresh tumor specimens. These results
_________________________________________
Correspondence to: Dr Yasuhide Hayashi, Department of
Pediatrics, Graduate School of Medicine, University of Tokyo, 7-3-1
Hongo, Bunkyo-ku, Tokyo 113-8655, Japan
E-mail: [email protected]
Key words: neuroblastoma, KIF1B gene, kinesin superfamily
suggest that KIF1Ba in addition to KIF1Bß may not be a
candidate tumor suppressor gene for NB.
Introduction
Neuroblastoma (NB) is an embryonal tumor derived from
neural crest cells that comprises about 10% of childhood
malignancies and displays clinical, biological and genetic
heterogeneity (1). Cytogenetic studies have suggested that
deletion of the short arm of chromosome 1 (1p) occurs
frequently in NB and is associated with a poor prognosis (2-5).
Recent molecular studies have shown that a relatively high
rate of loss of heterozygosity (LOH) was observed in 1p as
well as in 2q, 9p, 11q, 14q, and 18q in NB (5-14), and it is
widely assumed that 1p36.2-36.3 contains 2 NB suppressor
genes (11,15-18). However, no creditable suppressor genes
have been identified in NBs to date. In addition, the distal
part of 1p displays frequent nonrandom deletions and translocations not only in NB, but also in many other human
malignancies, including melanoma, hepatocellular carcinoma,
breast cancer, lung cancer, gastric cancer and colorectal
cancer (11,19,20), suggesting that several tumor suppressor
genes for various human cancers are located in this region.
We previously reported that 1 NB cell line (NB-1)
showed ~480 kb homozygous deletion in 1p36.2 according to
a high-density sequence tagged site (STS)-content map
aligned on a 20 Mb BAC contig spanning human chromosome band 1p36 (21,22). This homozygous deletion region
contains 7 genes (E4, SCYA5, PGD, Cortistatin, DFF45,
PEX14 and KIF1B) (22-24).
KIF1B is a member of the kinesin superfamily motor
proteins that play important roles in intracellular organelle
transport and cell division (25-27). We previously determined
the genomic structure of KIF1Bß, which was found to be
responsible for the anterograde transport of synaptic vesicle
precursors along the axon, and detected a missense mutation
(Q98L) in the motor domain of the KIF1Bß gene in a Japanese
pedigree of Charcot-Marie-Tooth (CMT) disease type 2A,
a major autosomal dominant hereditary peripheral neuropathy. However, the KIF1Bß gene might not be a candidate
tumor suppressor gene for NB (24,28). In this study we
determined the genomic structure and expression of the
738
CHEN et al: KIF1B· GENE IN NEUROBLASTOMA
human KIF1B· gene, another major isoform of KIF1B gene,
and also performed mutational analysis of the KIF1B· gene
in pediatric solid tumors including NB.
Materials and methods
Tumor cell lines. We used 25 NB cell lines (NB-1, NB-16,
NB-19, NB-69, GOTO, NH-12, CHP-134, IMR-32, SK-NSH, TGW, NBTU1, SJNB-1-SJNB-8, LAN-1, LAN-2,
LAN-5, SCMC-N2, SCMC-N4 and SCMC-N5) (10,23,24,29),
9 rhabdomyosarcoma cell lines (SJRH-1, SJRH-4, SJRH-18,
SJRH-30, RD, RMS, SCMC-RM2, SCMC-MM1 and KYM-1)
and 12 Ewing sarcoma cell lines (SJES-1-SJES-3, SJES-5SJES-8, SK-ES, RD-ES, SCMC-ES1, UTP-ES-1 and ES-1-OT)
(30,31). Twenty-four other solid tumor cell lines were also
examined including 3 melanoma cell lines (C32-TG, MeWo
and VMRC-MELG), 3 osteosarcoma cell lines (H003N1,
H009 and JCRB-614-NY), 4 primitive neuroectodermal
tumor cell lines (SK-N-MC, SK-N-P1, SK-N-L0 and KPPNTTBm2), 7 malignant rhabdoid tumor cell lines [MRTK(KA),
STM-91-01, TTC-1240, TTC-642, TTC-549, YAM-RTI and
TM87-16] (31), and 7 medulloblastoma cell lines (T98G,
U87, U251, U343, SF126, SF188 and NMCG-1). All cell
lines were cultured in RPMI-1640 (Gibco RLB) medium
supplemented with 10% fetal bovine serum in a humidified
atmosphere containing 5% CO2 at 37˚C.
Primary tumor specimens. Twenty-eight tumor specimens
were randomly obtained from patients with NB at the time of
initial surgery or biopsy, mainly in Saitama Children's Medical
Center and the Affiliated Hospital of the University of Tokyo
(9,10). Informed consent was obtained from the parents of
each patient. Of the 28 cases, 7 were classified as stage I,
7 as stage II, 6 as stage III and 8 as stage IV. Nineteen patients
were infants under 1 year of age at diagnosis and 9 patients
were over 1 year old. Nineteen of the 28 cases (68%) were
diagnosed by a mass screening program. MYCN amplification
was detected in only 1 patient (3.6%) who was classified as
stage IV. Patients with stage I or II were treated with either
surgery alone or surgery plus chemotherapy, mainly consisting
of vincristine and cyclophosphamide without radiotherapy.
Patients with stage III or IV were treated with multidrug
chemotherapy consisting of cyclophospamide, adriamycin,
cisplatin and etoposide with or without surgery, radiotherapy
and hematopoietic stem cell transplantation.
Normal controls. Fifty peripheral blood samples from healthy
volunteers were used as normal controls after informed consent
was obtained.
Total RNA and DNA extraction. Total RNA was extracted
from all specimens using the acid guanidine thiocyanatephenol chloroform method. Randomly primed cDNA was
synthesized from total RNA using a cDNA synthesis kit as
previously described (10,29,32). High molecular weight
DNA was extracted from all specimens by proteinase K
digestion and phenol/chroloform extraction (29,32).
Determination of the genomic structure of the human KIF1B·
gene. NCBI database analysis and RT-PCR were used to
identify the full coding region of the KIF1B· gene. Primers
BBS and BR (Table I) were derived from exon 20 of KIF1Bß
and the human homologue of the C-terminal sequence of the
murine Kif1b·, respectively. Using this primer set, RT-PCR
was carried out to determine whether this human homologue
shares the same N-terminal motor domain sequence of KIF1Bß
as the murine Kif1b gene (24,25,33,34). The reaction mixture
of RT-PCR was as follows: 1 µl of cDNA, 1X PCR reaction
buffer, 0.1 mM dNTP, 0.25 µM of each primer, 0.5 U of
Gold Taq Polymerase (Perkin-Elmer, NJ, USA) in a final
volume of 10 µl. RT-PCR was performed in a GeneAmp
PCR system-9700 (Perkin-Elmer, Norwalk, CT, USA) under
the following conditions (29,32): denatured at 95˚C for 9 min
followed by 35 cycles amplification (95˚C for 30 sec, 55˚C
for 30 sec, and 72˚C for 30 sec) and 7 min extension at 72˚C.
The PCR products were electrophoresed on a 2% agarose gel,
stained with ethidium bromide, and photographed under UV.
Northern blot analysis. Northern blot analysis was used to
study the expression pattern of the KIF1B · transcript in
normal tissues. Human multiple tissue Northern blots
(Clontech) were hybridized with 32P-labeled cDNA probes
derived from the motor domain and the cargo-binding domain,
respectively.
Real-time quantitative RT-PCR. To quantify more accurately
the levels of KIF1B· transcript in 28 primary tumors of NB,
we used real-time quantitative RT-PCR analysis with an
iCycler iQ real-time PCR detection system (Bio-Rad, USA)
using QuantiTectTM SYBR Green PCR Kit (Qiagen). Using
primers BBS and BR (Table I), the reaction mixture was
prepared as follows: 1 µl of cDNA, 1X QuantiTect SYBR
Green PCR Master Mix (Qiagen), 0.3 µM of each primer,
0.5 unit of Uracil-N-glycosylase (Applied Bio-systems) in a
final volume of 50 µl. The amplification conditions for
quantitation were an initial 2 min of incubation at 50˚C
(to allow Uracil N-glycosylase to degrade any carryover
contamination), 15 min at 95˚C (to activate the enzyme),
followed by 45 cycles amplification (denaturation at 95˚C for
30 sec, annealing at 60˚C for 30 sec, extension at 72˚C for
30 sec) (35). The reactions for quantifying the ß-actin copy
number were also performed exactly as described above
alongside the KIF1B · reactions. Serially diluted cDNA
(1X, 0.1X, 0.01X, 0.001X, 0.0001X) were used to generate
the standard curve with correlation coefficients greater than
0.995. All of the reactions were run in duplicate and were
performed 3 times. The mean copy number values of KIF1B·
were corrected for the mean copy number values obtained for
ß-actin from the same cDNA samples to obtain the values
reported as KIF1B·/ß-actin normalized.
RT-PCR-SSCP analysis. Since the first 20 exons of the
KIF1B· are identical to the KIF1Bß gene, which has been
screened for mutations in NB cell lines (24), we only screened
exon 21 of the KIF1B· in all of the tumor cell lines using
RT-PCR single-strand conformation polymorphism analysis
(RT-PCR-SSCP). Exon 21 was amplified in seven overlapping fragments of 200-400 bp to cover the entire 1.5 kb
exon and part of exon 20. The 7 sets of primers used in this
study are listed in Table I. cDNA (1 µl) was suspended in a
INTERNATIONAL JOURNAL OF ONCOLOGY 23: 737-744, 2003
739
Table I. Primers of KIF1B· gene for RT-PCR-SSCP.
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
Primer name
Primer sequence (5'→3')
Location of RT-PCR producta
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
Sense primer BBSb
GGCTACAGGAAATGGAGATC
nt 2057-nt 2286
Antisense primer BR
TGGTTCACGTTTCTTCCCACTG
Sense primer DF
GTGTGAAGAGAGCTGGAAACTG
nt 2172-nt 2561
Antisense primer DR
ATCTTCTCATCCCCAACACCGA
Sense primer EF
TGGCAAGAAAGACCCCAATGAG
nt 2475-nt 2771
Antisense primer ER
TTGTGGCTTCCTGGGCTTTTCT
Sense primer CF
AGCACTGATGTAGATGACCTC
nt 2590-nt 2933
Antisense primer CR
ATTTGCTCTTGCCTCATCCAG
Sense primer FF
CTGTTGGTGCTGGTGTTAGTAG
nt 2783-nt 3178
Antisense primer FR
CCCCTTTTTCTTGGCTCTCTTC
Sense primer GF
GCAACCCTAAACACAGAAACTC
nt 3044-nt 3453
Antisense primer GR
GTAAGACTGACGGTGTTGATGA
Sense primer HF
TAATCAGCAACAGCCACCTCAACT
nt 3270-nt 3663
Antisense primer HR
CACTGTCTGTTTCTTCCACCATGA
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
aThe location is based on the sequence of KIF1B· mRNA (GenBank accession number AY139835). bPrimer BBS is derived from exon 20,
and the others are derived from exon 21.
––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
total of 10 µl of PCR buffer containing 10 mM Tris-HCl
(pH 8.3), 50 mM KCl, 1.5 mM MgCl2, 100 µM of each deoxynucleotide triphosphate, 0.25 µM of each primer, 1.14 µCi of
[·-32P]dCTP and 0.5 U of Taq DNA polymerase (PerkinElmer, NJ, USA). PCR amplification was performed with a
GeneAmp PCR system-9700 (Perkin-Elmer, Norwalk, CT,
USA) as follows: 30 sec at 95˚C, 30 sec at the respective
annealing temperature, and 30 sec at 72˚C for 35 cycles,
followed by 7 min at 72˚C. The RT-PCR-SSCP analysis was
performed in a low pH buffer system that showed improved
separation of long mutant fragments of up to 800 bp (36).
The labeled PCR products were mixed with 45 µl of formamide
dye (95% formamide, 0.05% xylene cyanol, 0.05% bromophenol blue and 20 mM EDTA) and denatured for 10 min at
95˚C. Each sample (2 µl) was applied to nondenaturing
polyacrylamide gel containing 5% polyacrylamide (99:1
acrylamide to bisacrylamide) and TME (30 mM Tris, 35 mM
MES (2-[N-Morpholine]ethanesulfonic acid, Dojin Chemicals)
and 1 mM Na 2EDTA, pH 6.8), electrophoresed in TME
buffer at 25˚C. The gels were dried and exposed to Kodak
X-ray film for 12-24 h at 25˚C or -80˚C (24,29,32). The PCR
products, which showed differing mobility, were purified using
QIAquick PCR Purification Kit (Qiagen) and sequenced
directly in both directions using the BigDye Terminator
Cycle Sequencing Ready Reaction Kit (Perkin-Elmer, USA)
with the ABI PRISM Genetic Analyzer (Perkin-Elmer,
USA).
Statistical analysis. The mean value and standard deviation
of KIF1B·/ß-actin normalized from early-stage (stage I and II)
and advanced-stage (stage III and IV) NBs were calculated,
respectively. The significance of the difference in KIF1B·
expression between early-stage and advanced-stage NBs was
evaluated by Student's t-test.
Results
Genomic structure of the human KIF1B· gene. We initially
found that EST KIAA1448 was mapped into homozygous
deletion region using SGC30747 based on our STS-content
map at 1p36 (22). Furthermore, we found that part of EST
KIAA1448 shared high homology (94%) with the C-terminal
end of the murine Kif1b· (25) (GenBank accession number
D17577) (Fig. 1). Since the 2 isoforms of the murine Kif1b
gene, Kif1b· and Kif1bß, have a 2.1 kb identical N-terminal
sequence (25,33,34), we assumed that KIAA1448 was the
C-terminus of the human KIF1B · and shared the same
N-terminus with KIF1Bß (GenBank accession number
AF257176). To confirm our prediction, we designed a pair of
primers, which were derived from exon 20 of KIF1Bß and
the predicted start of exon 21 of the human KIF1B · ,
respectively (Table I). Using the 2 primers, RT-PCR was
carried out for normal cDNA and all of the tumor cell lines.
The results showed the existence of this 230 bp product in all
samples, suggesting that EST KIAA1448 is located downstream of exon 20, which contained a classical exon-intron
boundary structure and polyA signal. Thus, we concluded that
alternative splicing of the human KIF1B gene also generated
2 major isoforms: KIF1B·‚ and KIF1Bß (Fig. 2). The human
KIF1B· gene has 21 exons and extends over about 95 kb on
1p36.2. The first 20 exons of KIF1B·, which are identical to
KIF1Bß, contain a conserved kinesin-like motor domain and
an ATP/GTP binding motif, whereas exon 21 encodes the
cargo-binding domain. This gene encodes 1153 amino acids.
740
CHEN et al: KIF1B· GENE IN NEUROBLASTOMA
Figure 1. Comparison of human KIF1B· protein (GenBank accession number AY139835) and murine Kif1b· protein (GenBank accession number D17577).
The asterisks indicate conserved amino acids in humans and mice.
Figure 2. Schematic representation of the genomic and protein structures of
human KIF1B· and KIF1Bß. The KIF1B gene consists of 48 exons (shown
in bars) in which the KIF1B· gene comprises exon 1 to exon 21 whereas the
KIF1Bß gene contains 47 exons, which splices out exon 21. The protein
structures of the KIF1B· and KIF1Bß are comprised of identical motor
domain (black box) and different cargo-binding domains (slashed and dotted
boxes).
Expression pattern of the human KIF1B· gene in normal
tissues. When the cDNA probe specific for KIF1B·, which
was derived from the cargo-binding domain, was used in
Northern blot analysis, a band of approximately 7.8 kb was
detected in the fetal brain, lung, kidney, adult heart, placenta,
testis, ovary, and small intestine, and was particularly
abundant in the skeletal muscle (data not shown). When we
amplified the RT-PCR product of the motor domain as a
probe, both the 11 kb band and 7.8 kb band were present in
various tissues as expected, representing KIF1Bß and KIF1B·,
respectively (Fig. 3). These results further confirmed that
KIF1B· and KIF1Bß resulted from alternative splicing of
1 gene. However, 2 smaller bands with estimated sizes of 5 kb
and 2 kb were present in all of the tissues examined when the
probe derived from the motor domain was used.
INTERNATIONAL JOURNAL OF ONCOLOGY 23: 737-744, 2003
741
Figure 3. Northern blot analysis of KIF1B· mRNA. The probe used for hybridization was derived from the motor domain. A, KIF1B· (7.8 kb) and KIF1Bß
(11 kb) expression patterns in various fetal tissues. B, KIF1B· (7.8 kb) and KIF1Bß (11 kb) expression patterns in various adult tissues.
Expression of the human KIF1B · gene in cell lines and
primary tumors of NB. The RT-PCR analysis, using the
primer set covered exon 20 to exon 21 of KIF1B· (BBS and
BR, Table I), revealed an obvious expression in all of the
tumor cell lines except NB-1 (Fig. 4).
With the same primer set, we performed real-time
quantitative RT-PCR analysis to quantify the relative amount
of KIF1B· transcript in 28 primary tumors of NB. We found
that the mean value of KIF1B·/ß-actin normalized (49.5)
obtained from early-stage NBs was slightly higher than that
(40.7) obtained from advanced-stage NBs but both standard
deviations (56.7 and 49.4, respectively) were quite large
(Fig. 5). However, there was no statistical significance
between them using Student's t-test (P>0.05).
Figure 4. RT-PCR analysis of the KIF1B· gene in NB cell lines using BBS
and BR primes. Lane 11, NB-1; lanes 1-10, other NB cell lines; M, marker.
A clear expression was detected in all NB cell lines whereas no expression
was detected in NB-1.
Mutation of the human KIF1B· gene in pediatric solid tumor
cell lines. A total of 70 pediatric solid tumor cell lines including
NB were examined for mutations of exon 21 of KIF1B· using
RT-PCR-SSCP analysis, and abnormally migrating bands were
detected in 3 cell lines. Direct sequence analysis of these
variant SSCP bands revealed a missense mutation with ATG
(Met)→ATT (Ile) at codon 807 in SK-N-SH (NB) (Fig. 6) and 3
silent mutations at codon 938 (ACA→ACG) in SK-N-SH,
codon 748 (GCT→GCC) in SJNB-1 (NB), and codon 756
(CGG→CGA) in SK-N-P1 (PNET), respectively (Table II).
Sequencing analysis of the corresponding genomic sequence
further confirmed these nucleotide changes. None of these
mutations were found in 50 normal samples.
Discussion
We have identified the major isoform of the human KIF1B
gene (KIF1B·) and confirmed that the human KIF1B gene
generates at least 2 isoforms (KIF1B· and ß) according to
Figure 5. Real-time quantitative RT-PCR analysis in 28 primary tumors of
NB. The samples of early-stage NBs are from 1 to 14 and those of
advanced-stage NBs are from 15 to 28. The mean value of KIF1B·/ß-actin
normalized from early-stage NBs is 49.5 with a standard deviation of 56.7,
whereas the mean value of KIF1B·/ß-actin normalized from advanced-stage
NBs is 40.7 with a standard deviation of 49.4.
742
CHEN et al: KIF1B· GENE IN NEUROBLASTOMA
Figure 6. A missense mutation in SK-N-SH (NB). A, Mobility shift in RTPCR-SSCP analysis using EF and ER primers. The arrow indicates the
abnormally migrating band in SK-N-SH. B, Sequence eletropherogram of
SK-N-SH and wild-type in the region where the band shift was shown by
RT-PCR-SSCP. Heterozygous T→C resulted in a Met→Ile substitution within
the cargo-binding domain of KIF1B·.
alternative splicing as in the murine Kif1b gene (25,33,34).
We found that as in the murine Kif1b gene, the 5'-end of the
human KIF1B· (2.1 kb), which contained the motor domain
and ATP/GTP binding motif, was identical to that of KIF1Bß
but the cargo-binding domain was completely different through
RT-PCR and Northern blotting. Northern blot analysis
revealed a 7.8 kb band in the fetal brain, lung, kidney, adult
heart, placenta, testis, ovary, and small intestine, and was
particularly abundant in the skeletal muscle when a unique
probe of KIF1B· was used. However, with a probe from the
sequence of the motor domain, not only the 7.8 kb band was
detected as described above, but also the 11 kb band of
KIF1Bß was detected in the fetal brain, kidney, adult placenta,
liver, kidney, and pancreas, being particularly abundant in
the brain. Nangaku et al firstly reported that the transcript of
8 kb was highly expressed in the mouse heart, as the
transcript of 11 kb was abundant in the mouse brain when
different probes containing part of both the motor domain
and the cargo-binding domain were used (25). It seems likely
that only part of the motor domain and common region were
hybridized in their experiment. Gong et al detected the 7.8 kb
mRNA in the mouse skeletal muscle and heart, and in less
abundance in the ovary, while the 11 kb mRNA was
expressed highly in the brain when specific probes for the
respective 3'-end were used (34). However, Conforti et al
detected the 11 kb transcript in the mouse brain but detected
no Kif1b· signal by Northern blotting (33). We previously
reported that the 11 kb transcript was found in various human
tissues in comparison with the 8 kb transcript only in the
testis (24). Taken together, these results indicate that KIF1B·
seems to have a lower expression level and a more restricted
pattern of expression than that of KIF1Bß. Thus, there was a
little difference among the results of each group. Furthermore, splicing events were found to be generated in 8
different isoforms of Kif1b (33,34). In our study, smaller
bands with estimated sizes of 5 kb and 2 kb were present in
all of the tissues examined when the probe derived from the
motor domain was used. Although the motor domain of
KIF1B has high homology with other kinesin superfamily
members such as KIF1A, we could not determine that the
smaller bands represented other isoforms of KIF1B or other
KIF members.
Table II. Mutations of KIF1B· gene in pediatric solid tumor
cell lines.
–––––––––––––––––––––––––––––––––––––––––––––––––
Cell lines
Codon
Nucleotide
Amino acid
change
change
–––––––––––––––––––––––––––––––––––––––––––––––––
SK-N-SH (NB)
807
ATG→ATT
Met→Ile
SK-N-SH (NB)
938
ACA→ACG
Thr (silent)
SJNB-1 (NB)
748
GCT→GCC
Ala (silent)
SK-N-P1 (PNET)
756
CGG→CGA
Arg (silent)
–––––––––––––––––––––––––––––––––––––––––––––––––
In the kinesin superfamily (Kif), Kif1b was the first to be
proved as a monomeric motor for anterograde transport.
Nangaku et al characterized the binding properties of murine
Kif1b· to transport mitochondria along the axon (25). However, the cargo-binding domain of Kif1bß, which maintains a
high homology with Kif1a and unc-104, has been reported to
be responsible for the anterograde transport of synaptic
vesicle precursors along the axon (28). Knockout mice of
Kif1b died at birth from apnea due to nervous system defects.
The death of knockout neurons in culture can be rescued by
expression of the ß isoform but not the · isoform, which was
explained by the potential deficiency in mitochondrial
transport caused by lack of Kif1b·, apparently compensated
for with redundant mitochondrial motors such as Kif5a, b
and c (37,38). The Kif1b heterozygotes were defective in
transporting synaptic vesicle precursors and suffered from
progressive muscle weakness similar to human neuropathies.
Furthermore, we found that patients of a Japanese CMT2A
pedigree had a loss-of-function mutation (Q98L) in the motor
domain of the KIF1Bß gene (28).
Due to the role of intracellular organelle transport and cell
division of KIF motors, abnormalities in KIF family proteins
may affect the function of organelles and proteins that are
necessary for cell differentiation and apoptosis. The KIF1B
gene, mapped to 1p36.2 within the commonly deleted region
of NB, was homozygously deleted in NB-1, implying that
KIF1B may be a candidate tumor suppressor gene for NB.
However, mutational analysis of KIF1Bß did not show
enough evidence of this (24). As KIF1B· has a different 3'end, resulting in a different function, it is necessary to
investigate whether KIF1B· is one of the candidate tumor
suppressor genes for NB. In this study, we screened 70
pediatric solid tumor cell lines, including 25 NB cell lines for
mutations in the cargo-binding domain. Through RT-PCRSSCP and direct sequencing analysis we detected a missense
mutation (M807I) in SK-N-SH (NB) as well as a silent
mutation in the same cell line and 2 other silent mutations in
SJNB-1 (NB) and SK-N-P1 (PNET). All of these nucleotide
changes were confirmed by the corresponding genomic DNA
and none of them were detected in 50 normal samples. The
amino acid methionine, which was substituted with isoleucine
in SK-N-SH, is conserved in humans and mice. However,
whether this missense mutation, located at the cargo-binding
domain of KIF1B·, affects the function of KIF1B· is unknown.
RT-PCR analysis revealed a clear expression in all of the
tumor cell lines examined except NB-1. Furthermore, when
INTERNATIONAL JOURNAL OF ONCOLOGY 23: 737-744, 2003
we performed real-time quantitative RT-PCR analysis in
28 primary tumors of NB, we obtained a mean value of
KIF1B·/ß-actin normalized to 49.5 with a standard deviation
of 56.7 from early-stage NBs and a mean value of 40.7 with
a standard deviation of 49.4 from advanced-stage NBs.
Although the mean value of early-stage NBs was slightly
higher than that of advanced-stage NBs, there was no
statistically significant difference between them (P>0.05).
Given this, it seems that the KIF1B· gene is significantly
expressed in unfavorable NB as well as in favorable NB.
However, the normalized values of the KIF1B·/ß-actin of all
specimens ranged widely from 0.08 to 172, suggesting that
the expression level of the KIF1B· gene might vary, regardless
of the NB stage. These results were similar to those of the
KIF1Bß (24). A further study is needed to clarify this point.
Recently, it was reported that Kif1B· directly interacts
through its C-terminal postsynaptic density-95 (PSD-95)/
discs large/zona occludens (PZD) domain-binding motif
with PDZ proteins including PSD-95/synapse-associated
protein-90 (SAP90), SAP97, and synaptic scaffolding
molecules (S-SCAM)-90 (SAP90) (39). PDZ proteins contain
various domains for protein interactions, enabling motors to
interact with a large number of proteins, and may help the
cargo-PDZ protein-motor complex dock at its destinations.
They proposed that since PZD proteins serve as Kif1b·
receptors, Kif1b· may play an important role in the axonal
transport of a variety of SAP97- and S-SCAM-associated
cargos more than mitochondria. They also speculated that
Zhao et al (28) could not rescue the cultured Kif1b-/- neurons
by Kif1b· probably because of the blockage of the free Cterminal carboxylate group of the PDZ-binding peptide using
the Kif1b·-EGFP in their rescue experiments. Therefore, the
role of KiF1B· in neuronal survival, differentiation, and
CMT2A peripheral neuropathy remains to be studied further.
In conclusion, our study determined the genomic structure
of the human KIF1B· gene and detected its expression in
several human tissues, especially in the skeletal muscle.
RT-PCR analysis revealed a clear expression in all of the
pediatric solid tumor cell lines examined except NB-1. Realtime quantitative RT-PCR analysis of primary NB tumors
showed no significant difference between early-stage and
advanced-stage NBs. However, we detected a missense
mutation (M807I) in 1 NB cell line, and 3 silent mutations in
2 NB cell lines and 1 in PNET cell line, respectively. Whether
these mutations are functionally significant or not is still
unclear, however, there was only 1 missense mutation,
suggesting that the KIF1B· gene may have less significance
as a candidate tumor suppressor gene for these solid tumors
including NB. Further analysis of KIF1B· in CMT2A to reassess its association with neuropathy is considered to be
needed.
Acknowledgements
We thank Mrs. S. Soma and Mrs. H. Soga for their excellent
technical assistance. This study was supported by a Grant-inAid for Cancer Research from the Ministry of Health, Labour
and Welfare of Japan; a Grant-in-Aid for Scientific Research
on Priority Areas and Grant-in-Aid for Scientific Research
(B) and (C) from the Ministry of Education, Culture, Sports,
Science and Technology of Japan.
743
References
1. Westemann F and Schwab M: Genetic parameters of neuroblastomas. Cancer Lett 184: 127-147, 2002.
2. Brodeur GM, Green AA, Hayes AF, Williams KJ, Williams DL
and Tsiatis AA: Cytogenetic features of human neuroblastoma
and cell lines. Cancer Res 41: 4678-4686, 1981.
3. Brodeur GM, Maris JM, Yamashiro DJ, Hogarty MD and
White PS: Biology and genetics of human neuroblastoma. J
Pediatr Hematol Oncol 19: 93-101, 1997.
4. Hayashi Y, Kanda N, Inaba T, Hanada R, Nagahara N, Muchi H
and Yamamoto K: Cytogenetic findings and prognosis in
neuroblastoma with emphasis on marker chromosome 1. Cancer
63: 126-132, 1989.
5. Caron H, van Sluis P, De Kraker J, Bokkerink J, Egeler M,
Laureys G, Slater R, Westerveld A, Voute PA and Versteeg R:
Alleic loss of chromosome 1p as a predictor of unfavorable outcome
in patients with neuroblasoma. N Engl J Med 334: 225-230, 1996.
6. Cheng NC, van Roy N, Chan A, Beitsma M, Westerveld A,
Speleman F and Versteeg R: Deletion mapping in neuroblastoma cell lines suggests 2 distinct tumor suppressor genes in
the 1p35-36 region, only one of which is associated with N-myc
amplification. Oncogene 10: 291-297, 1995.
7. Takita J, Hayashi Y, Kohno T, Shiseki, M, Yamaguchi N,
Hanada R, Yamamoto K and Yokota J: Allelotype of neuroblastoma. Oncogene 11: 1829-1834, 1995.
8. Takita J, Hayashi Y, Kohno T, Yamaguchi N, Hanada R,
Yamamoto K and Yokota J: Deletion map of chromosome 9 and
p16 (CDKN2A) gene alterations in neuroblastoma. Cancer Res
57: 907-912, 1997.
9. Takita J, Hayashi Y, Takei K, Yamaguchi N, Hanada R,
Yamamoto K and Yokota J: Allelic imbalance on chromosome
18 in neuroblastoma. Eur J Cancer 36: 508-513, 2000.
10. Takita J, Yang HW, Chen YY, Hanada R, Yamamoto K, Teitz T,
Kidd V and Hayashi Y: Allelic imbalance on chromosome 2q
and alterations of the caspase 8 gene in neuroblastoma.
Oncogene 20: 4424-4432, 2001.
11. Schwab M, Praml C and Amler LC: Genomic instability in 1p and
human malignancies. Genes Chromosomes Cancer 16: 211-229,
1996.
12. White PS, Maris JM, Sulman EP, Jensen SJ, Kyemba SM,
Beltinger CP, Allen C, Kramer DL, Biegel JA and Brodeur GM:
Molecular analysis of the region of distal 1p commonly deleted
in neuroblastoma. Eur J Cancer 33: 1957-1961, 1997.
13. Hoshi M, Otagiri N, Shiwaku HO, Asakawa S, Shimizu N,
Kaneko Y, Ohi R, Hayashi Y and Horii A: Detailed deletion
mapping of chromosome band 14q32 in human neuroblastoma
defines a 1.1-Mb region of common allelic loss. Br J Cancer 82:
1801-1807, 2000.
14. Maris JM, Weiss MJ, Guo C, Gerbing RB, Stram DO, White PS,
Hogarty MD, Sulman EP, Thompson PM, Lukens JN,
Matthay KK, Seeger RC and Brodeur GM: Loss of heterozygosity at 1p36 independently predicts for disease progression
but not decreased overall survival probability in neuroblastoma
patients: a Children's Cancer Group Study. J Clin Oncol 18:
1888-1899, 2000.
15. Schleiermacher G, Peter M, Michon J, Hugot JP, Vielh P,
Zucker JM, Magdelenat H, Thomas G and Delattre O: Two
distinct deleted regions on the short arm of chromosome 1 in
neuroblastoma. Genes Chromosomes Cancer 10: 275-281, 1994.
16. Takeda O, Homma C, Maseki N, Sakurai M, Kanda N, Schwab M,
Nakamura Y and Kaneko Y: There may be two tumor suppressor
genes on chromosome 1p closely associated with biologically
distinct subtypes of neuroblastoma. Genes Chromosomes Cancer
10: 30-39, 1994.
17. Caron H, Peter M, van Sluis P, Speleman F, De Kraker J,
Laureys G, Michon J, Brugieres L, Voute PA, Westerveld A,
Slater R, Delattre O and Versteeg R: Evidence for two tumor
suppressor loci on chromosomal bands 1p35-36 involved in
neuroblastoma: one probably imprinted, another associated with
N-myc amplification. Hum Mol Genet 4: 535-539, 1995.
18. Caron H, Spieker N, Godfried M, Veenstra M, van Sluis P,
De Kraker J, Voute P and Versteeg R: Chromosome bands
1p35-36 contain 2 distinct neuroblastoma tumor suppressor loci,
one of which is imprinted. Genes Chromosomes Cancer 30:
168-174, 2001.
19. Ezaki T, Yanagisawa A, Ohta K, Aiso S, Watanabe M, Hibi T,
Kato Y, Nakajima T, Ariyama T, Inazawa J, Nakamura Y and
Horii A: Deletion map on chromosome 1p in well-differentiated
gastric cancer. Br J Cancer 73: 424-428, 1996.
744
CHEN et al: KIF1B· GENE IN NEUROBLASTOMA
20. Nomoto S, Haruki N, Tatematsu Y, Konishi H, Mitsudomi T and
Takahashi T: Frequent allelic imbalance suggests involvement
of tumor suppressor gene at 1p36 in the pathogenesis of human
lung cancers. Genes Chromosomes Cancer 28: 342-346, 2000.
21. Chen YZ, Hayashi Y, Wu JG, Takaoka E, Maekawa K,
Watanabe N, Inazawa J, Hosoda F, Arai Y, Ohiki M,
Mizushima H, Morohashi A, Ohira M, Nakagawara A, Liu SY,
Hosi M, Horii A and Soeda E: A BAC-based STS-content map
spanning a 35-megabase region of human chromosome 1p35-36.
Genomics 74: 55-70, 2001.
22.Chen YZ, Soeda E, Yang HW, Takita J, Chai L, Horii A,
Inazawa J, Ohki M and Hayashi Y: Homozygous deletion in a
neuroblastoma cell line defined by a high-density STS map
spanning human chromosome band 1p36. Genes Chromosomes
Cancer 31: 326-332, 2001.
23. Yang HW, Chen YZ, Piao HY, Takita J, Soeda E and Hayashi Y:
DNA fragmentation factor 45 (DFF45) gene at 1p36.2 is homozygously deleted and encodes variant transcripts in neuroblastoma cell line. Neoplasia 3: 165-169, 2001.
24.Yang HW, Chen YZ, Takita J, Soeda E, Piao HY and Hayashi Y:
Genomic structure and mutational analysis of the human KIF1B
gene which is homozygouly deleted in neuroblastoma at
chromosome 1p36.2. Oncogene 20: 5075-5083, 2001.
25. Nangaku M, Sato-Yoshitake R, Okada Y, Noda Y, Takemura R,
Yamazaki H and Hirokawa N: KIF1B, a novel microtubule plus
end-directed monomeric motor protein for transport of mitochondria. Cell 79: 1209-1220, 1994.
26. Nakagawa T, Tanaka Y, Matsuoka E, Kondo S, Okada Y, Noda Y,
Kanai Y and Hirokawa N: Identification and classification of 16
new kinesin superfamily (KIF) proteins in mouse genome. Proc
Natl Acad Sci USA 94: 9654-9659, 1997.
27. Hirokawa N: Kinesin and Dynein superfamily proteins and the
mechanism of organelle transport. Science 279: 519-526, 1998.
28. Zhao CJ, Takita J, Tanaka Y, Setou M, Nakagawa T, Takeda S,
Yang HW, Terada S, Nakata T, Takei Y, Saito M, Tsuji S,
Hayashi Y and Hirokawa N: Charcot-Marie-Tooth disease
type 2A caused by mutation in a microtubule motor KIF1Bß.
Cell 105: 587-597, 2001.
29. Yang HW, Piao HY, Chen YZ, Takita J, Kobayashi M,
Taniwaki M, Hashizume K, Hanada R, Yamamoto K, Taki T,
Bessho F, Yanagisawa M and Hayashi Y: The p73 gene is less
involved in the development but involved in the progression of
neuroblastoma. Int J Mol Med 5: 379-384, 2000.
30. Choi SH, Kong XT, Taki T, Tsuchida Y, Kawaguchi H, Kato H,
Hanada R, Look AT and Hayashi Y: Reduced or absent
expression and codon 201Gly/Arg polymorphism of DCC gene in
rhabdomyosarcoma and Ewing's sarcoma/PNET family. Int J
Mol Med 6: 463-467, 2000.
31. Uno K, Takita J, Yokomori K, Tanaka Y, Ohta S, Shimada H,
Gilles FH, Sugita K, Abe S, Sako M, Hashizume K and
Hayashi Y: Aberrations of the hSNF5/INII gene are restricted to
malignant rhabdoid tumors or atypical teraroid/rhabdoid tumors
in pediatric solid tumors. Genes Chromosomes Cancer 34: 22-41,
2002.
32. Kong XT, Choi SH, Inoue A, Xu F, Chen T, Takita J, Yokota J,
Bessho F, Yanagisawa M, Hanada R, Yamamoto K and
Hayashi Y: Expression and mutational analysis of the DCC,
DPC4, and MADR2/JV18-1 genes in neuroblastoma. Cancer
Res 57: 3772-3778, 1997.
33. Conforti L, Buckmaster EA, Tarlton A, Brown MC, Lyon MF,
Perry VH and Coleman MP: The major brain isoform of Kif1b
lacks the putative mitochondria-binding domain. Mamm Genome
10: 617-622, 1999.
34. Gong TW, Winnicki RS, Kohrman DC and Lomax MI: A novel
mouse kinesin of the UNC-104/KIF1 subfamily encoded by the
Kif1b gene. Gene 239: 117-127, 1999.
35. Donovan JW, Ladetto M, Zou GY, Neuberg D, Poor C, Bowers D
and Gribben JG: Immunoglobulin heavy-chain consensus
probes for real-time PCR quantification of residual disease in
acute lymphoblastic leukemia. Blood 95: 2651-2658, 2000.
36. Kukita Y, Tahira T, Sommer SS and Hayashi K: SSCP analysis
of long DNA fragment in low pH gel. Hum Mutat 10: 400-407,
1997.
37. Tanaka Y, Kanai Y, Okada Y, Nonaka S, Takeda S, Harada A
and Hirogawa N: Targeted disruption of mouse conventional
kinesin heavy chain, Kif5B, results in abnormal clustering of
mitochondria. Cell 93: 1147-1158, 1998.
38. Kanai Y, Okada Y, Tanaka Y, Harada A, Terada S and
Hirokawa N: Kif5c, a novel neuronal kinesin enriched in motor
neurons. J Neurosci 20: 6374-6484, 2000.
39. Mok H, Shin H, Kim S, Lee JR, Yoon J and Kim E: Association
of the kinesin superfamily motor protein Kif1b · with postsynaptic density-95(PSD-95), synapse-associated protein-97,
and synaptic scaffolding molecule PSD-95/disc large/zona
occludens-1 proteins. J Neurosci 22: 5253-5258, 2002.