MYO18B, a candidate tumor suppressor gene at
chromosome 22q12.1, deleted, mutated, and
methylated in human lung cancer
Michiho Nishioka*†, Takashi Kohno*, Masachika Tani*, Nozomu Yanaihara*, Yoshio Tomizawa*, Ayaka Otsuka*,
Shigeru Sasaki*, Keiko Kobayashi*, Toshiro Niki*, Arafumi Maeshima*, Yoshitaka Sekido‡, John D. Minna§,
Saburo Sone†, and Jun Yokota*¶
*National Cancer Center Research Institute, Tokyo 104-0045, Japan;†Tokushima University School of Medicine, Tokushima 770-8503, Japan;‡Nagoya
University School of Medicine, Nagoya 466-8550, Japan; and§Hamon Center for Therapeutic Oncology Research, University of Texas Southwestern
Medical Center, Dallas, TX 75390-8593
Communicated by Webster K. Cavenee, University of California at San Diego, La Jolla, CA, July 26, 2002 (received for review February 27, 2002)
Loss of heterozygosity on chromosome 22q has been detected in
approximately 60% of advanced nonsmall cell lung carcinoma
(NSCLC) as well as small cell lung carcinoma (SCLC), suggesting the
cancer progression. Here, we isolated a myosin family gene,
MYO18B, located at chromosome 22q12.1 and found that it is
Somatic MYO18B mutations were detected in 19% (14?75) of lung
cancer cell lines and 13% (6?46) of primary lung cancers of both
SCLC and NSCLC types. MYO18B expression was reduced in 88%
(30?34) of NSCLC and 47% (8?17) of SCLC cell lines. Its expression
was restored by treatment with 5-aza-2?-deoxycytidine in 11 of 14
cell lines with reduced MYO18B expression, and the promoter CpG
island of the MYO18B gene was methylated in 17% (8?47) of lung
cancer cell lines and 35% (14?40) of primary lung cancers. Further-
more, restoration of MYO18B expression in lung carcinoma cells
suppressed anchorage-independent growth. These results indicate
that the MYO18B gene is a strong candidate for a novel tumor
suppressor gene whose inactivation is involved in lung cancer
loss of heterozygosity (LOH) analysis (1–3), cytogenetic analysis
(4, 5), and comparative genomic hybridization (6), indicating the
presence of a tumor suppressor gene (TSG) on chr 22q, which is
frequently inactivated in lung cancer. We previously reported
that the incidence of LOH on 22q in advanced-stage non-small
cell lung carcinoma (NSCLC) (?60%) was significantly higher
than that in early-stage NSCLC (?30%) (1, 2). The incidence of
LOH on chr 22q was also high in small cell lung carcinomas
(SCLCs) (?60%) irrespective of clinical stages (3). Thus, it is
likely that inactivation of a TSG on chr 22q plays an important
role in the progression of both SCLC and NSCLC types.
Three known TSGs, SNF5?INI1, NF2, and EP300, have been
mapped to 22q11.2, 22q12.2, and 22q13, respectively. However,
genetic and?or epigenetic alterations of these genes are infre-
quent in both SCLC and NSCLC (7–9). Therefore, a target
gene(s) for 22q LOH in human lung cancer is still unknown.
Recently, we identified a homozygous deletion on chr 22q12.1 in
a SCLC cell line, Lu24 (10). The deleted region included the
genomic structure of SEZ6L and performed a mutational anal-
it was strongly suggested that another unknown gene(s) on chr
22q functioned as a major TSG in lung cancer progression.
Subsequent analyses of the genomic sequence covering the Lu24
deletion identified a novel myosin heavy chain-like gene,
bk125H2.1, whose structure also had been partially determined
llelic losses on chromosome (chr) 22q have been frequently
detected in human lung cancers by several methods, such as
by the chr 22 sequencing project (11). It was recently proposed
that the bk125H2.1 gene be named MYO18B based on the partial
amino acid sequence deduced from the genomic sequence data
(12). Thus, we use the name MYO18B for the bk125H2.1 gene in
Here we isolated the full-length cDNA, determined the
genomic structure of the MYO18B gene, and analyzed it for
deletion, mutation, expression, and methylation in a large num-
ber of lung cancers. The MYO18B gene was frequently altered by
several molecular mechanisms, including homozygous?
hemizygous deletions, intragenic mutations, and hypermethyl-
ation of the CpG island. Restoration of MYO18B expression in
lung cancer cells suppressed colony formation in soft agar. Thus,
it was strongly suggested that the MYO18B gene is a target TSG
of 22q LOH and is involved in the progression of lung cancer.
Materials and Methods
Samples. Seventy-six lung cancer cell lines, consisting of 26
SCLCs and 50 NSCLCs, were used in this study (13). In 10
SCLCs and 15 NSCLCs, corresponding lymphoblast cells were
available (14) (details are available on request). Forty-six sur-
gical specimens (16 SCLCs and 30 NSCLCs) and adjacent
noncancerous tissues were obtained at surgery. DNA was pre-
pared as described (10). Allelic imbalance (AI) at two micro-
satellite loci, D22S429 and D22S538, in the MYO18B locus (Fig.
1A) was examined in these 46 cases by the method described (1).
Forty cases were informative for either or both of the loci, and
AI was detected in 25 (63%) of the cases. Noncancerous tissue
DNA of 50 unrelated individuals was also used in this study.
Randomly primed cDNAs for lung cancer cell lines, SAEC
(Clonetics, Walkersville, MD), NHBE (Clonetics), WI-38, seven
NSCLCs, and human normal tissues were prepared as described
(10). Alveolar type II cells and bronchiolar epithelial cells were
microdissected by using the LM200 LCM System (Arcturus
Engineering, Mountain View, CA), and total RNA was prepared
with a Micro RNA Isolation kit (Stratagene) (15). Randomly
primed cDNA was reverse-transcribed from total RNA by using
SensiScript reverse transcriptase (Qiagen, Tokyo).
Isolation of Full-Length MYO18B cDNA. Genome DNA sequences
covering the Lu24 deletion region (GenBank accession nos.
AL022329, AL080245, Z98949, AL079300, AL022337,
AL080273, and AL023513) were used to identify exons by the
GENSCAN (16) and BLAST programs (17) after elimination of
SCLC, small cell lung carcinoma; NSCLC, non-small cell lung carcinoma; 5-aza-dC, 5-aza-2?-
deoxycytidine; TSA, trichostatin A; RTQ-PCR, real-time quantitative PCR.
Data deposition: The sequence reported in this paper has been deposited in the GenBank
database (accession no. AB075376).
¶To whom reprint requests should be addressed. E-mail: firstname.lastname@example.org.
www.pnas.org?cgi?doi?10.1073?pnas.192445899 PNAS ?
September 17, 2002 ?
vol. 99 ?
no. 19 ?
repetitive elements with the REPEATMASKER program (18). The
algorithm at www.ebi.ac.uk?emboss?cpgplot?was used to detect
potential CpG islands. Predicted exons were connected by
exon-connection PCR with human skeletal muscle cDNA as a
template with nine primer sets (primers and conditions are
available on request). Amplified cDNA fragments were directly
sequenced as described (10).
Real-Time Quantitative PCR (RTQ-PCR) Analysis. Expression of
MYO18B was measured by RTQ-PCR using ABI Prism 7900HT
(Applied Biosystems). Probe and primers were designed by using
PRIMER EXPRESS software (Applied Biosystems). Expression of
MYO18B was normalized to RNA content for each sample by
using GAPDH as an internal control (primers, probe, and
conditions are available on request). The relative expression was
calculated as the ratio of the average expression levels for all
samples compared with adult human lung mRNA. Levels of
relative expression of ?0.5 were considered as being reduced.
Mutation Analysis. For single-stranded conformation polymor-
phism (SSCP) and WAVE analyses, 42 coding exons of the
MYO18B gene were amplified by PCR using 50 ng of genomic
DNA in each of 51 sets of primers. SSCP analysis was performed
as described (12). For WAVE analysis, the PCR products from
tumor DNA were mixed with those from normal lung tissue
DNA, denatured, reannealed, and analyzed by WAVE DNA
Fragment Analysis and WAVEMAKER software 4.0 (Trans-
genomic, Omaha, NE). PCR products with different mobilities
in either SSCP or WAVE analysis were purified and directly
sequenced. (Primers and conditions are available on request.)
5-Aza-2?-deoxycytidine (5-aza-dC) and?or Trichostatin A (TSA) Treat-
ment. For the first 48 h, cells were incubated with medium
containing 1.0 ?M 5-aza-dC (Sigma), and then for another 24 h
then isolated with a RNeasy minikit (Qiagen), RTQ-PCR was
performed as described above, and a relative expression of ?2
compared with basal levels was considered as being induced.
Promoter Methylation Analysis. Genomic DNA was treated with
sodium bisulfite as described (19). PCR was performed by
mixing DNA (?200 ng) with 50 pM of the CpG-F (5?-
AAGGTATGTTTATATGTATT-3?) and CpG-R (5?-CAG-
3?) primers (Fig. 5, which is published as supporting information
on the PNAS web site, www.pnas.org) in a reaction mixture (40
?l) containing dNTPs (200 ?M each) and 1 unit of HotstarTaq
DNA polymerase (Qiagen) at 95°C (1 min), 50°C (1 min), and
72°C (4 min) for 40 cycles. The CpG-R primer contained the
M13-RV sequence (underlined) as a site to initiate sequencing.
PCR products were directly sequenced to obtain average meth-
ylation levels. PCR products of five cell lines and two normal
lung tissues were subcloned into the pGEM T-Easy vector
(Promega) and sequenced.
Definition of Methylation. Plasmid DNA with or without con-
verted cytosines at CpG sites 10 and 11 was amplified by the SP6
prepared by mixing each of the PCR products in an equal ratio
and sequenced by using the M13-RV primer. The percentage of
methylation was calculated by the formula of the peak height of
G divided by the sum of peak heights G and A. The mean values
of the seven samples were 50.5% at both CpG sites 10 and 11,
and small differences in the percentage of methylation were
11 was 2.0% and 1.9%, respectively). When the percentage of
methylation was higher than the mean plus 3 SD (56.5% and
56.2% at sites 10 and 11, respectively), we considered there to be
cells with biallelic methylation.
Cell Proliferation and Soft Agar Growth Assays. FLAG-tagged
MYO18B cDNA was ligated into the pcDNA3.1(?) plasmid
vector (Invitrogen). The cDNA-containing or empty vector was
transfected by using Lipofectamine 2000 (Invitrogen). G418-
resistant colonies were picked up, and MYO18B expression was
confirmed by Western blot analysis with anti-FLAG M2 Ab
(Sigma). For cell proliferation assay, cells were seeded in a
96-well plate at a density of 1 ? 104per well, and the number of
cells was estimated by using the TetraColor ONE Cell Prolifer-
ation Assay System (Seikagaku, Tokyo). Colony formation in
soft agar was measured as described (20). Three weeks after
plating, cells were stained with methylene blue, and the number
of visible colonies was counted.
Sequence Analysis. Sequence similarity and motif searches were
performed by using BLAST and PROFILESCAN (www.ch.embnet.
org?software?PFSCAN?form.html) programs. The coiled–coil
region was predicted with MACSTRIPE 2.0 software (21).
Results and Discussion
The sequence data of chr 22 indicated that the Lu24 deletion
contains the MYO18B gene, which encodes a novel myosin heavy
chain-like protein, consisting of 25 exons (GenBank accession
no. Z98949). However, the amino acid sequence of a deduced
and transcriptional maps of the region containing the MYO18B gene. STS
markers are marked on the top. Locations of CpG islands and three genes are
indicated. The region of homozygous deletion in a SCLC cell line, Lu24, is
indicated by a dashed horizontal line. Exon-intron organization of the
MYO18B gene is depicted as vertical bars. Predicted transcriptional start and
stop sites are indicated by vertical bars under exons 2 and 43, respectively.
Closed arrowheads, somatic mutations; open arrowheads, base substitutions
detected only in cell lines but not in 50 unrelated individuals. Putative func-
tional domains are indicated at the bottom.
Structures of the MYO18B gene and MYO18B protein. (A) Physical
www.pnas.org?cgi?doi?10.1073?pnas.192445899Nishioka et al.
MYO18B protein contained only the C-terminal portion of a
myosin head domain, a neck domain, and a tail domain with a
short coiled–coil. Therefore, it was suggested that the size of the
MYO18B gene is larger than predicted from the sequence data.
Accordingly, we determined the structure of full-length
MYO18B cDNA by GENSCAN gene prediction program analysis
of the genomic sequence covering the Lu24 deletion in conjunc-
tion with reverse transcription–PCR analysis. Then, the full-
length cDNA sequence was compared with the genomic se-
quence of this locus. By this approach, 42 coding exons and a 5?
noncoding exon were identified (Fig. 1A). The 5? noncoding
exon was confirmed as being the initial exon (exon 1) by primer
extension analysis (data not shown). The region of 424 bp
(22) (Fig. 5). The full-length MYO18B cDNA fragment was 8,051
bp in size, contained an ORF of 7,701 bp, and encoded 2,567 aa
with a predicted Mrof 285,000. An ATG in exon 2, located 260
bp downstream of the predicted transcription start site, was
speculated to be a translation start codon (Fig. 1A). The
sequence context of this ATG was identical to the optimal
translation initiation signal (23), whereas an in-frame termina-
tion codon was not found upstream of this ATG. Thus, the
MYO18B gene consisted of 43 exons and was dispersed within a
286-kb region between the ADRBK2 gene and the SEZ6L gene
on chr 22q (Fig. 1A).
Searches of public nucleotide and protein databases with the
protein (GenBank accession no. AK016515) and shares 40%
identity with human and mouse MysPDZ proteins (GenBank
accession nos. D86970 and AB026497, respectively). A sche-
matic diagram of the domain structure of MYO18B protein is
shown in Fig. 1B. A head (motor) domain was identified from
codons 573 to 1321. Alignments with other myosins revealed a
consensus ATP binding site and part of an actin binding region
(Fig. 1B). A GPA motif, which is very rich in Gly, Pro, and Ala
residues and is believed to interact with F actin (24), was
identified at the C terminus of the actin binding region. An IQ
motif was found between the head and tail domains (Fig. 1B). To
date, at least 18 myosin subfamilies have been identified (12). A
phylogenetic analysis revealed that MYO18B, human and mouse
MysPDZ, and Drosophila alt1 constituted a new class (XVIII) of
myosin family proteins (data not shown), as described (12).
MYO18B has an ATP binding site and a myosin light chain
binding IQ motif, suggesting that this molecule acts as a motor
protein regulated by light chains in the presence of ATP.
MYO18B also contains a short coiled–coil domain, which
probably allows for dimerization to form a two-headed structure
and a globular structure at the end of the tail.
Northern blot analysis indicated that MYO18B transcripts of
?8.0 kb in size were expressed in skeletal muscle and heart but
not in other tissues, including lung (data not shown). However,
RTQ-PCR analysis revealed that the MYO18B gene is expressed
in diverse tissues, including adult and fetal lungs (Fig. 2A).
Furthermore, MYO18B expression was also detected in two
primary cultured bronchial epithelial cells, SAEC and NHBE, a
lung fibroblast cell line, WI-38, and laser capture microdissected
alveolar type II cells and bronchiolar epithelial cells (Fig. 2A).
Thus, the MYO18B gene is expressed in both lung epithelial cells
and fibroblasts. In particular, its expression was detected in
possible precursor cells for lung adenocarcinoma and squamous
cell carcinoma, alveolar type II cells, and bronchiolar epithelial
protein may have a common function in various types of human
cells, and that MYO18B expression has some physiological
function for lung epithelial cells.
Genomic PCR analysis revealed that exons 34–43 of the
MYO18B gene were homozygously deleted, whereas exons 1–33
were retained, in the Lu24 cell line (Fig. 1A). To clarify whether
the MYO18B gene is genetically altered in other lung cancers, 75
lung cancer cell lines were subjected to single-stranded confor-
mation polymorphism and?or WAVE analyses followed by
direct sequencing. No homozygous deletions were detected
except in Lu24. However, 51 different types of nucleotide
substitutions were detected among the 75 cell lines. Among
them, six different types of substitutions detected in six cell lines,
H209, H128, H2347, H2009, H2107, and H1607, were confirmed
as being somatic mutations, because these substitutions were not
detected in the corresponding lymphoblast cell lines (Table 1).
Another 10 different types of substitutions detected in eight cell
lines, H69, Ma29, Ma2, H23, N417, VMRC-LCD, H526, and
A549, were likely to be somatic mutations and not genetic
polymorphisms because these substitutions were not detected in
50 noncancerous tissues (Table 1). Fourteen of these 16 muta-
tions were missense, and the remaining two were silent and
intronic. Among the 14 cell lines with MYO18B mutations, four
cell lines were homozygous for the mutant alleles, whereas the
remaining 10 were heterozygous with retention of the wild-type
were genetic polymorphisms, because these sequence variants
were also detected in noncancerous tissues. In total, mutations
of the MYO18B gene were detected in 14 of the 75 (19%) lung
cancer cell lines. We next examined 46 primary lung cancers, and
six (13%) were concluded as having somatic mutations, because
these sequence variants were detected only in cancerous tissues
but not in the corresponding noncancerous tissues. Four muta-
adenocarcinoma; AdSqC, adenosquamous carcinoma; SqC, squamous cell car-
cinoma; LCC, large cell carcinoma. (A and C) Expression in normal human
and cancerous tissue (43T) were obtained from the same patient. (B) Expres-
sion in lung cancer cell lines.
Expression of the MYO18B gene. SCC, small cell carcinoma; AdC,
Nishioka et al. PNAS ?
September 17, 2002 ?
vol. 99 ?
no. 19 ?
tions were missense, and the remaining two were silent and
intronic (Table 1; Fig. 1B). Two of the six cases showed loss of
the wild-type allele.
Among the 18 missense mutations detected in lung cancers,
mutations in codons 590, 835, and 1970 caused substitutions of
evolutionarily conserved amino acids (Thr to Met, Ala to Gly,
and Ala to Glu, respectively). Mutations in codon 661, detected
as the same nucleotide substitution in H23 and N2111T, caused
an amino acid change in the ATP binding site (Arg to Trp).
Meanwhile, mutations in codon 1238, detected as two different
types of nucleotide substitutions in H2009 and 1551T, caused
different types of amino acid changes in the GPA motif (Pro to
Thr or Gln). Thus, it is highly possible that these mutations have
some effects on the physiological function of the MYO18B
We next assessed expression of the MYO18B gene in 51 lung
cancer cell lines and seven primary NSCLCs by RTQ-PCR
analysis. MYO18B expression was reduced in 8?17 (47%) of
SCLC cell lines, 30?34 (88%) of NSCLC cell lines, and 5?7
(71%) of primary NSCLCs (Fig. 2 B and C). Thus, expression of
the MYO18B gene was reduced in approximately 85% of
NSCLCs, including both cell lines and primary tumors, and 50%
of SCLC cell lines.
Aberrant promoter methylation is now acknowledged as a
mechanism for gene inactivation in cancer cells (25, 26). Re-
duced MYO18B expression in lung cancer cells could be caused
by transcriptional silencing of this gene by hypermethylation of
the CpG island in the promoter region. To address this question,
we first screened for a possible restoration of its expression by
treatment of the cells with 5-aza-dC. Among 14 cell lines (three
SCLCs and 11 NSCLCs) with reduced (or without) MYO18B
expression, its expression was restored in 11 cell lines (79%) with
5-aza-dC. In contrast, in two cell lines (H1299 and H82) with
relatively high MYO18B expression, its expression was not
that the MYO18B gene is frequently silenced by hypermethyl-
ation of the promoter region in lung cancer cells.
region containing the CpG island with the transcription start
sites of this gene (Fig. 5). DNAs from two NSCLC cell lines
without MYO18B expression, those from one NSCLC cell line
and two SCLC cell lines with MYO18B expression, and those
from two normal lung tissues were used for the analysis. PCR
products of 748 bp containing 30 CpG sites surrounding the first
exon were then subcloned into a TA-cloning vector, and 10
independent clones were sequenced. Several specific CpG sites
were hypermethylated in the two cell lines without MYO18B
expression but hypomethylated in the three cell lines with
expression and in the normal lung tissues (Fig. 3B). In particular,
CpG sites 10 and 11 (CpG 10?11) in the CpG island appeared
to be consistently hypermethylated in cell lines without MYO18B
expression but hypomethylated in cell lines with MYO18B ex-
pression. Namely, CpG 10?11 were methylated in ?60% of the
clones analyzed for cell lines without MYO18B expression, but in
?50% of the clones analyzed for those with MYO18B expression.
This finding indicates that methylation-associated MYO18B si-
lencing occurs, at least in part, through methylation of CpG
10?11. To further assess the effect of methylation at CpG 10?11,
genomic DNAs from 42 other lung cancer cell lines were treated
with sodium bisulfite, amplified by PCR with the same primers,
and analyzed by direct sequencing. Thus, in total, 47 cell lines
were analyzed for the methylation status of CpG 10?11. Among
showed hypermethylation at both CpG 10?11, and 26 showed
low methylation levels at these sites (Fig. 3C; Table 2). In
particular, none of 13 cell lines with MYO18B expression showed
hypermethylation at CpG 10?11. This result indicated that, in
approximately 17% (8?47) of cell lines, the MYO18B gene is
silenced by methylation of CpG 10?11. We next analyzed 40
primary lung cancers for CpG 10?11 methylation by the same
method. Fourteen (35%) of the 40 primary cancers showed
hypermethylation, but the remaining 26 showed hypomethyla-
tion (Fig. 3C; Table 2). In all six pairs of primary cancer and the
corresponding normal tissue, methylation levels were higher in
cancers than in normal tissues. In one case, the tumor (43T) with
hypermethylation at CpG 10?11 showed reduced MYO18B ex-
pression, but the corresponding normal lung tissue (43N) with
hypomethylation at both sites showed MYO18B expression (Figs.
2C and 3C). Thus, it was suggested that CpG 10?11 are hyper-
methylated in a subset of primary lung cancers, and hypermeth-
ylation of those CpG sites leads to MYO18B gene silencing in
5-aza-dC failed to restore MYO18B gene expression in three
of the 14 cell lines, suggesting that mechanisms other than
hypermethylation, such as histone deacetylation, may also be
involved in the repression of the MYO18B gene. To address this
question, we screened for a possible restoration of its expression
by treatment of the cells with a histone deacetylase inhibitor,
TSA (Fig. 3A). By TSA treatment, MYO18B expression was
restored in all three cell lines, which showed no response to
5-aza-dC. Thus, histone deacetylation is likely to be another
mechanism of MYO18B gene silencing in lung cancer cells.
Restoration of MYO18B expression by TSA in a total of 13 of the
14 cell lines indicates that deacetylation contributes to MYO18B
gene silencing as frequently as DNA methylation, and that
histone deacetylation cooperates with DNA methylation for the
gene silencing. Among the 11 cell lines with restoration of
MYO18B expression by 5-aza-dC, hypermethylation at CpG
H520). Thus, repression of MYO18B expression could also be
caused by methylation of other CpG sites in the MYO18B gene
Table 1. Mutations of the MYO18B gene in human lung cancers
IVS41 ? 21 (C?T)
IVS4 ? 26 (G?A)
SCC, small cell carcinoma; AdC, adenocarcinoma; SqC, squamous cell carci-
noma; LCC, large cell carcinoma. Exp., expression; Met., methylation; ?,
reduced or absent; ND, not determined; M, methylated; U, unmethylated.
†Corresponding noncancerous tissue DNA was not available.
‡Homozygous base substitution.
www.pnas.org?cgi?doi?10.1073?pnas.192445899Nishioka et al.
or methylation of a transactivating regulator for the MYO18B
We next assessed the expression and methylation as well as
deletions of the wild-type allele of the MYO18B gene in 14 cell
lines with mutated MYO18B genes (Table 1). Expression was
reduced or absent in eight of them, thus, repression of MYO18B
expression seemed to occur irrespectively of the mutation status
in this gene. Methylation at CpG 10?11 was detected only in one
of the eight cell lines, Ma2, thus, repression of MYO18B expres-
sion in the other seven cell lines might have occurred by several
other mechanisms, as described above. Among the six cell lines
with MYO18B expression, the wild-type allele was deleted only
in the Ma29 cell line, suggesting that the MYO18B gene is
inactivated by mutation accompanied by loss of the wild-type
allele in this cell line. Two different mutations were detected in
another cell line with MYO18B expression, H69. The mutations
might have occurred in both alleles in this cell line. In the four
other cell lines with MYO18B expression, the wild-type allele was
retained. Thus, it was unclear whether the MYO18B gene is
inactivated in these cell lines. However, it is possible that some
of the mutated MYO18B proteins have dominant negative
effects to the wild-type MYO18B protein. Although we could
not assess the expression status in six primary lung cancers with
mutated MYO18B genes, the wild-type allele was deleted in two
cases, and methylation at CpG 10?11 was detected in one case
Molecular analyses of the MYO18B gene in a large number of
lung cancers suggested that this gene could play an important
role as a tumor suppressor in the development of lung cancer.
Thus, we further assessed whether the restoration of wild-type
MYO18B expression suppresses the growth of lung cancer cells.
A FLAG-tagged MYO18B expression vector was transfected
was undetectable and restored by 5-aza-dC. Compared with the
parent (H322) or empty vector transfectants (V1 and V2), the
anchorage-independent growth was markedly suppressed in
clones with MYO18B expression in an expression-dependent
manner (Fig. 4, and Fig. 6, which is published as supporting
information on the PNAS web site). Furthermore, the popula-
tion doubling time was ?10% prolonged in clones with strong
MYO18B expression, 2–1 and 2–20. Expression of exogenous
MYO18B also inhibited the anchorage-independent growth of
5-aza-dC and?or TSA in 14 cell lines with reduced (or without) MYO18B expression. N, no treatment; 1, 5-aza-dC treatment; 2, TSA treatment; 3, 5-aza-dC and
TSA treatment. Each result is the average of two independent tests.*indicates expression levels were more than twice as much as basal levels. (B) Bisulfite
and two normal lung tissues (NL-1 and NL-2). Each circle indicates a CpG site in the primary DNA sequence, and each line of circles represents analysis of a single
cloned allele. The numbers of the CpG site are indicated at the top and are identical to the numbers in Fig. 5. E, unmethylated CpG sites; F, methylated CpG
sites; arrow, transcription start site. (C) Methylation status at CpG 10?11 in normal lung tissues and lung cancer cells. The ranges of 50% methylation (n ? 7,
samples with or with reduced (or without) MYO18B expression, filled in red or blue, respectively. Open squares and circles indicate samples in which MYO18B
expression was not determined. Groups N, P, and C indicate normal lung tissue (n ? 6), primary tumor (n ? 40), and cell line (n ? 47), respectively.
Table 2. Incidence of hypermethylation at CpG sites 10 and 11 in lung cancers
Cell lines (%)Surgical specimens (%)
SCLC (n ? 15) NSCLC (n ? 32) Total (n ? 47) SCLC (n ? 11) NSCLC (n ? 29) Total (n ? 40)
10 and 11
Nishioka et al. PNAS ?
September 17, 2002 ?
vol. 99 ?
no. 19 ?
the H1299 cell line, in which endogenous MYO18B expression Download full-text
was detected and not up-regulated by 5-aza-dC (Figs. 4 and 6).
These results suggest that overexpressed MYO18B has a growth
suppressor activity in lung cancer cells.
We provide here the evidence that MYO18B is a strong
candidate for TSG at chr 22q12.1, which is inactivated in
approximately 50% of lung cancer by deletion, mutation, and
promoter methylation. The activity of anchorage-independent
growth suppression in lung cancer cells supports the fact that
MYO18B acts as a TSG in lung carcinogenesis. Further func-
tional and biological studies of the MYO18B gene will help us
understand the molecular mechanisms of human lung cancer
We thank T. Kamigaito, M. Yoshizumi, and T. Nakano for technical
assistance and T. Akiyama and M. Shiraishi for helpful discussion. We
T. Terasaki, S. Hirohashi, M. Takada, and A. F. Gazdar. Cell lines were
also obtained from the American Type Culture Collection and the
Japanese Collection of Research Bioresources. This work was supported
in part by grants-in-aid from the Ministry of Health, Labor, and Welfare
for the 2nd-term Comprehensive 10-year Strategy for Cancer Control,
the Program for Promotion of Fundamental Studies in Health Sciences
of the Organization for Pharmaceutical Safety and Research, the Min-
istry of Education, Culture, Sports, Science, and Technology of Japan,
and the G. Harold and Leila Y. Mathers Charitable Foundation and
National Cancer Institute Specialized Programs of Research Excellence
Grant P50 CA70907. K.K. is a recipient of the Research Resident
Fellowship from the Foundation for Promotion of Cancer Research.
1. Shiseki, M., Kohno, T., Nishikawa, R., Sameshima, Y., Mizoguchi, H. &
Yokota, J. (1994) Cancer Res. 54, 5643–5648.
2. Kawanishi, M., Kohno, T., Otsuka, T., Adachi, J., Sone, S., Noguchi, M.,
Hirohashi, S. & Yokota, J. (1997) Carcinogenesis 18, 2057–2062.
3. Virmani, A. K., Fong, K. M., Kodagoda, D., McIntire, D., Hung, J., Tonk, V.,
Minna, J. D. & Gazdar, A. F. (1998) Genes Chromosomes Cancer 21, 308–319.
J. Y., Taguchi, T. & Keller, S. M. (1994) Genes Chromosomes Cancer 11,
5. Mertens, F., Johansson, B., Hoglund, M. & Mitelman, F. (1997) Cancer Res. 57,
6. Levin, N. A., Brzoska, P., Gupta, N., Minna, J. D., Gray, J. W. & Christman,
M. F. (1994) Cancer Res. 54, 5086–5091.
7. Sekido, Y., Pass, H. I., Bader, S., Mew, D. J., Christman, M. F., Gazdar, A. F.
& Minna, J. D. (1995) Cancer Res. 55, 1227–1231.
8. Manda, R., Kohno, T., Hamada, K., Takenoshita, S., Kuwano, H. & Yokota,
J. (2000) Cancer Lett. 153, 57–61.
9. Gayther, S. A., Batley, S. J., Linger, L., Bannister, A., Thorpe, K., Chin, S. F.,
Daigo, Y., Russell, P., Wilson, A., Sowter, H. M., et al. (2000) Nat. Genet. 24,
10. Nishioka, M., Kohno, T., Takahashi, M., Niki, T., Yamada, T., Sone, S. &
Yokota, J. (2000) Oncogene 19, 6251–6260.
11. Dunham, I., Shimizu, N., Roe, B. A., Chissoe, S., Hunt, A. R., Collins, J. E.,
Bruskiewich, R., Beare, D. M., Clamp, M., Smink, L. J., et al. (1999) Nature
(London) 402, 489–495.
12. Berg, J. S., Powell, B. C. & Cheney, R. E. (2001) Mol. Biol. Cell 12, 780–794.
13. Hamada, K., Kohno, T., Takahashi, M., Yamazaki, M., Yamazaki, M., Tashiro,
H., Sugawara, C., Ohwada, S., Sekido, Y., Minna, J. D., et al. (2000) Genes
Chromosomes Cancer. 27, 308–318.
14. Phelps, R. M., Johnson, B. E., Ihde, D. C., Gazdar, A. F., Carbone, D. P.,
McClintock, P. R., Linnoila, R. I., Matthews, M. J., Bunn, P. A., Jr., Carney,
D., et al. (1996) J. Cell Biochem., Suppl. 24, 32–91.
15. Fend, F., Emmert-Buck, M. R., Chuaqui, R., Cole, K., Lee, J., Liotta, L. A. &
Raffeld, M. (1999) Am. J. Pathol. 154, 61–66.
16. Burge, C. & Karlin, S. (1997) J. Mol. Biol. 268, 78–94.
17. Altschul, S. F., Gish, W., Miller, W., Myers, E. W. & Lipman, D. J. (1990) J.
Mol. Biol. 215, 403–410.
18. Smit, A. F. (1996) Curr. Opin. Genet. Dev. 6, 743–748.
19. Clark, S. J., Harrison, J., Paul, C. L. & Frommer, M. (1994) Nucleic Acids Res.
20. Kolligs, F. T., Kolligs, B., Hajra, M. H., Hu, G., Tani, M., Cho, K. R. & Fearon
E. R. (2000) Genes Dev. 14, 1319–1331.
21. Lupas, A., Van-Dyke, M. & Stock, J. (1991) Science 252, 1162–1164.
22. Gardiner-Garden, M. & Frommer, M. (1987) J. Mol. Biol. 196, 261–282.
23. Kozak, M. (1987) Nucleic Acids Res. 15, 8125–8148.
24. Urrutia, R. A., Jung, G. & Hammer, J. A., III (1993) Biochim. Biophys. Acta
25. Jones, P. A. & Laird, P. W. (1999) Nat. Genet. 21, 163–167.
26. Baylin, S. B., Herman, J. G., Graff, J. R., Vertino, P. M. & Issa, J. P. (1998)
Adv. Cancer Res. 72, 141–196.
and MYO18B-transfected clones (2–7, 1–8, 2–5, 2–1, 2–20) were analyzed for
population doubling time and colony formation in soft agar (Left). H1299 (P),
vector-transfected controls (V1, V2), and MYO18B-transfected clones (6, 124,
163) were also analyzed (Right). The arrowhead indicates the expression of
FLAG-tagged MYO18B protein (285 kDa) by Western blot analysis. Twenty
micrograms of whole-cell lysate was loaded on each lane. The doubling time
[DT (hr)] was calculated from the logarithmic phase of the growth curve. The
ability to form colonies in soft agar was measured by counting the number of
visible colonies?cm2in triplicate of a 6-well plate. Error bars indicate SD.
Suppression of proliferation and anchorage-independent growth by
www.pnas.org?cgi?doi?10.1073?pnas.192445899Nishioka et al.