GASTROENTEROLOGY 2003;125:708 –715
Inactivating Mutations of Caspase-8 Gene in Colorectal
Carcinomas
HONG SUG KIM,* JONG WOO LEE,* YOUNG HWA SOUNG,* WON SANG PARK,*
SU YOUNG KIM,* JONG HEUN LEE,* JIK YOUNG PARK,* YOUG GU CHO,* CHANG JAE KIM,*
SEONG WHAN JEONG,‡ SUK WOO NAM,* SANG HO KIM,* JUNG YOUNG LEE,* NAM JIN YOO,*
and SUG HYUNG LEE*
Departments of *Pathology and ‡Biochemistry, College of Medicine, The Catholic University of Korea, Seoul, Korea
Background & Aims: There has been evidence that dys-
regulation of apoptosis is involved in the pathogenesis
of cancer development. Caspase-8 is an initiation
caspase that activates the caspase cascade during ap-
optosis. The aim of this study was to explore the possi-
bility that mutation of the caspase-8 gene might be
involved in the development of colorectal cancer.
Methods: We analyzed the entire coding region of the
caspase-8 gene for the detection of somatic mutations
in 180 colorectal tumors (98 invasive carcinomas and
82 adenomas) by polymerase chain reaction, single-
strand conformation polymorphism, and DNA sequenc-
ing. Results: Overall, we detected a total of 5 somatic
mutations in 98 invasive carcinomas (5.1%), but no
mutations were detected in 82 adenomas (0%). The
frequency of caspase-8 mutation in the carcinomas was
significantly higher than that in adenomas (P < 0.05).
The 5 mutations consisted of 1 frameshift, 1 nonsense
mutation, and 3 missense mutations. We expressed the
5 tumor-derived caspase-8 mutants and found that 3 of
the 5 mutations markedly decreased apoptosis activity
of caspase-8. Furthermore, expression of the inactivat-
ing caspase-8 mutants interfered with apoptosis by
death receptor overexpression, indicating that these mu-
tants have dominant-negative inhibition of the death
receptor–induced apoptosis. Conclusions: The presence
of caspase-8 mutation in colon carcinomas suggests
that caspase-8 gene mutation might lead to the loss of
its apoptotic function and contribute to the pathogene-
sis of colorectal carcinomas, especially at the late stage
of colorectal carcinogenesis.
poptosis is a fundamental biochemical cell-death
A pathway that is essential for normal tissue ho-
meostasis and cellular differentiation and development.1
Proteases cause the morphological changes that are rec-
ognized as apoptosis and the biochemical changes asso-
ciated with apoptosis.1– 4 Among the proteases, a family
of cysteine proteases that cleave the substrates at aspar-
tate residues, known as caspases, plays a main role during
apoptosis.3,4 The caspase gene family thus far contains at
least 14 mammalian members, of which 12 human en-
zymes are known.3,4 From a functional perspective, the
caspase gene family has been classified into either initi-
ator caspases or effector caspases. The initiator caspases
would seem to be caspase-2, -8, -9, and -10, and the
effector caspases would seem to be caspase-3, -6, and
-7.3,4 The death signals originating from the death re-
ceptors, such as tumor necrosis factor receptors, Fas, and
tumor necrosis factor–related apoptosis-inducing ligand
(TRAIL) receptors, are transduced through the recruit-
ment of pro– caspase-8 to the death-induced signaling
complex (DISC) by the adaptor molecule Fas-associated
death domain protein (FADD).2 The local aggregation of
pro– caspase-8 is sufficient to allow autoprocessing or
transprocessing to produce active caspase-8, which can
subsequently activate downstream executioners, such as
caspase-3 and -7.3,4
The death receptors are widely expressed in normal
and neoplastic cells,5,6 but the expression of these pro-
teins does not necessarily predict susceptibility to kill-
ing.7,8 This can reflect the presence of inhibiting mech-
anisms of the death receptor–mediated apoptosis. The
death receptor–mediated apoptosis can be blocked by
several mechanisms, including mutations of the genes
involved in the death receptor–mediated apoptosis.9 –12
It is now believed that clonal expansion and tumor
growth are the results of the deregulation of intrinsic
proliferation (cell division) and cell death (apoptosis).13
Failure of apoptosis could allow the survival of trans-
Abbreviations used in this paper: DISC, death-induced signaling
complex; FADD, Fas-associated death domain protein; 5-FU, 5-fluoro-
uracil; PARP, poly-(ADP-ribose) polymerase; PCR, polymerase chain
reaction; SSCP, single strand conformation polymorphism; TRAIL,
tumor necrosis factor–related apoptosis-inducing ligand; TRAIL-R2,
TRAIL-receptor 2.
© 2003 by the American Gastroenterological Association
0016-5085/03/$30.00
doi:10.1016/S0016-5085(03)01059-X
September 2003
formed cells that are prone to undergo further genetic
damage and play an important role in the pathogenesis of
tumors. Either inactivation of a proapoptotic pathway or
activation of an antiapoptotic pathway results in failure
of apoptosis, thereby promoting tumor cell survival.
There has been evidence that alterations of caspase par-
ticipate in tumor development.12,14 –20
Somatic mutations of the genes involved in apoptosis,
including the death receptors and caspases, have been
reported in human cancers,9 –12,14,16 –19 implying that the
somatic mutations might not be a rare event in the
apoptosis pathway. Among the caspases, mutations of
caspase-3, -5, -8, and -10 have been reported in human
cancer tissues and cell lines.12–14,16 –19 Caspase-8 muta-
tions have been reported in some cancer cell lines,16 –18
suggesting the presence of caspase-8 mutation in human
cancer tissues. However, to date, data on caspase-8 mutation
in cancer tissues are lacking. In this study, to explore the
possibility that alterations of the caspase-8 gene might
play a role in colorectal cancer development, we analyzed
180 colon tumors for detection of caspase-8 mutation.
Materials and Methods
Tissue Samples, Microdissection, and
Plasmids
Paraffin-embedded histological sections of 82 colorec-
tal adenomas (40 low-grade dysplasias and 42 high-grade
dysplasias) and 98 invasive colorectal carcinomas were ob-
tained from the Catholic University of Korea–affiliated hospi-
tals (Seoul, Korea). The term invasive carcinoma means, strictly,
a cancer that has invaded beyond the muscularis mucosa. The
carcinomas originated from cecum (n ⫽ 2), ascending colon
(n ⫽ 18), transverse colon (n ⫽ 6), descending colon (n ⫽ 3),
sigmoid colon (n ⫽ 27), and rectum (n ⫽ 42). Approval for
this study was obtained from the Catholic University of Korea,
College of Medicine’s Institutional Review Board. Informed
consent was provided according to the Declaration of Helsinki.
Malignant and normal cells were selectively procured from
H&E-stained slides by using a 30-gauge 1/2 hypodermic
needle (Becton Dickinson, Franklin Lakes, NJ) affixed to a
micromanipulator, as described previously.21 DNA extraction
was performed by a modified single-step DNA extraction
method, as described previously.21 The plasmid constructs
encoding Fas, FADD, caspase-8, and TRAIL-receptor 2
(TRAIL-R2) were described previously.12
Single-Strand Conformation Polymorphism
Analysis
Genomic DNAs from tumor cells and corresponding
normal cells were amplified with primer pairs covering the
entire coding region (8 exons) of the human caspase-8␤ gene.
Oligonucleotide primers were designed with the program
Oligo (National Biosciences, Plymouth, MN) by using se-
CASPASE– 8 MUTATION IN COLON CANCER 709
quences obtained from Genbank (accession No. NT_005403).
Numbering of complementary DNA of caspase-8␤ was per-
formed with respect to the ATG start codon (NM_033355).
Radioisotope ([32P]deoxycytidine triphosphate) was incorpo-
rated into the polymerase chain reaction (PCR) products for
detection by autoradiogram. The PCR reaction mixture was
denatured for 1 minute at 94°C and incubated for 30 cycles
(denaturing for 30 seconds at 94°C, annealing for 30 seconds
at 53°C– 66°C, and extending for 30 seconds at 72°C). Other
procedures of PCR and single-strand conformation polymorphism
(SSCP) analysis were performed as described previously.9 –12
After SSCP, DNAs showing mobility shifts were cut out from
the dried gel and reamplified for 30 cycles by using the same
primer sets. Sequencing of the PCR products was performed
with the cyclic sequencing kit (Perkin-Elmer, Foster City, CA)
according to the manufacturer’s recommendations.
To analyze the allelic status of caspase-8 in the tumors, we used
6 intragenic polymorphisms. Because biallelic polymorphisms at
the coding sequence (960 A/G, dbSNP: 1045487) and the intron
sequence (IVS9-19A/G, dbSNP: 376918) have been found in the
Genbank database and because an additional biallelic polymor-
phism at the nucleotide position 789 A/G was detected in this
study, SSCP analysis at these polymorphic sites was used for the
detection of loss of heterozygosity, as well as for the detection of
mutations. Complete or nearly complete absence of 1 allele in
tumor DNA of informative cases, as defined by direct visualiza-
tion, was considered as loss of heterozygosity.
Site-Directed Mutagenesis
Site-directed mutagenesis was performed by using a
Quick Change Site-Directed Mutagenesis kit (Stratagene, La
Jolla, CA) according to the manufacturer’s instructions. To
change a base, a plasmid that contained the caspase-8␤ gene in
pcDNA3.1-Flag was used as a template. The nucleotide se-
quences of the mutagenized plasmids were confirmed by DNA
sequencing.
Coimmunoprecipitations and
Immunoblotting Assay
Human embryonic kidney 293T cells in log phase
were transfected in 6-well plates with expression plasmids by
using Lipofectamine 2000 reagent (Invitrogen, Carlsbad, CA).
Cells were harvested 1 day later and lysed in ice-cold Nonidet
P-40 lysis buffer supplemented with 1⫻ protease inhibitor
mix (Roche Molecular Biochemicals, Mannheim, Germany).
Cell lysates were subjected to immunoprecipitation with aga-
rose-conjugated anti–Flag M2 antibodies (Sigma, St. Louis,
MO). Immunocomplexes were fractionated by sodium dodecyl
sulfate polyacrylamide gel electrophoresis and transferred to
nitrocellulose membranes. The resulting blots were incubated
with anti-hemagglutinin (1:1000 vol/vol; Roche Molecular
Biochemicals) followed by horseradish peroxidase– conjugated
secondary antibodies and detection by an enhanced chemilu-
minescence method (Amersham Pharmacia Biotech, Bucking-
hamshire, UK). Alternatively, lysates were analyzed directly by
immunoblotting after normalization for total protein content.
710 KIM ET AL. GASTROENTEROLOGY Vol. 125, No. 3

Apoptosis Assay Figure 1) but none in 82 adenomas (0%). None of the

The 293T cells and human colorectal adenocarcinoma
cells (DLD-1) in 6-well plates were transfected by Lipo-
fectamine 2000 reagent with various combinations of expres-
sion plasmids (total, 1.3 ␮g) and 0.2 ␮g of the green fluores-
cent protein marker plasmid pEGFP N2 (Clontect, Palo Alto,
CA). Twenty-four hours after the transfection, the percentage
of green fluorescent protein–positive cells with nuclear apo-
ptotic morphology was determined by fixing in 10% methanol
for 15 minutes and staining with 4⬘,6-diamidino-2-phenylin-
dole 1 ␮g/mL for 15 minutes (mean ⫾ SD; n ⫽ 4). For the
influence of caspase-8 mutations on anticancer drug resistance,
100 to 600 ␮mol/L 5-fluorouracil (5-FU, Sigma, St. Louis,
MO) was added to DLD-1 cells 3 hours after transfection of the
caspase-8 mutants. Twenty-four, 48, and 72 hours later cell
death was determined by the same method described above.
Immunohistochemistry
Antibody for human caspase-8 (dilution 1:100; Santa
Cruz Biotechnology, Santa Cruz, CA) was used to detect
caspase-8 protein on tissue sections. Immunohistochemical
procedures were performed as described previously.9 Tumors
were interpreted as positive by immunohistochemistry when at
least weak to moderate cytosolic staining was seen in ⬎30% of
the neoplastic cells. The results were reviewed independently
by 2 pathologists. As a negative control, a slide was treated by
replacement of primary antibody with nonimmune sera.
Results
Mutations and Allelic Status of the
Caspase-8 Gene
Through the microdissection technique, we selec-
tively procured tumor cells and normal cells from histo-
logical sections of 180 colon tumors. Genomic DNA was
isolated and analyzed for the potential mutations in the
coding region of the caspase-8 gene by PCR-SSCP anal-
ysis. Enrichment and direct DNA sequence analysis of
aberrantly migrating bands led to the identification of 5
mutations in 98 invasive carcinomas (5.1%) (Table 1;
corresponding normal samples showed evidence of mu-
tations by SSCP (Figure 1), indicating that the mutations
had arisen somatically. The frequency of caspase-8 mu-
tation in the colorectal carcinomas was significantly
higher than in the adenomas (Fisher exact test; P ⬍
0.05). As for the relationship between stages and muta-
tions, the 5 cases with mutations consisted of 3 stage II
(tumor-node-metastasis staging), 1 stage III, and 1 stage
IV cancers, but this was not statistically significant (Ta-
ble 1; Fisher exact test; P ⬎ 0.05).
The mutations consisted of 3 missense mutations, 1
frameshift mutation, and 1 nonsense mutation. The mis-
sense mutations would result in the substitution of
amino acids in the second death effector domain, the p20
large protease subunit, and the p10 small protease sub-
unit. The nonsense mutation and the frameshift muta-
tion would result in premature terminations of amino
acid synthesis at the p10 small protease subunit. The
mutation data are summarized in Table 1. We repeated
the experiments 3 times, including tissue microdissec-
tion, PCR, SSCP, and sequencing analysis, to ensure the
specificity of the results, and we found that the data were
consistent (data not shown).
We examined the allelic status of caspase-8 in the tumors
by using 3 intragenic polymorphisms. In the tumors ana-
lyzed, heterozygosity rates of the polymorphisms at nucle-
otides 789, 960, and IVS9-19 were 11.2%, 33.7%, and
43.8%, respectively. All 5 cases with the caspase-8 muta-
tion were heterozygous for 1 or more polymorphisms, but
none showed evidence of allelic loss (Table 1).
Abrogation of Apoptosis Activities by
Caspase-8 Mutations
To determine whether the mutant forms of
caspase-8 are functionally defective, we generated
caspase-8 mammalian expression constructs containing
the mutations found in this study (Flag-caspase-8-

Table 1. Summary of Caspase-8 Mutations Identified in Colon Carcinomas

LOH analysis
Case Anatomic TNM Caspase-8 Mutation Mutation site Nucleotide Predicted amino acid
no. site stage expression 789a 960a IVS9-19b type (domain) change change
5 Rectum III Positive NI NI HET Missense Exon 7 (p10) 1238 G to A 413R3G
17 Ascending IV Positive NI NI HET Nonsense Exon 7 (p10) 1237 C to T 413R3stop
21 Ascending II Positive NI HET NI Frameshift Exon 8 (p10) 1381 insertion G Frameshift after codon 461
and stop at codon 479
41 Rectum II Positive NI HET NI Missense Exon 7 (p20) 1098 G to T 366Q3H
70 Descending II Positive HET HET NI Missense Exon 3 (DED) 485 G to T 162R3I

NI, not informative (homozygosity); HET, retention heterozygosity; TNM, tumor-node-metastasis; DED, death effector domain; p20, p20 large
subunit; p10, p10 small subunit.
aPolymorphisms at nucleotides 789 and 960 of caspase-8␤ complementary DNA.
bThe polymorphism at intron 7 of caspase-8.
September 2003
Figure 1. Mutations of the
caspase-8 gene in colon can-
cers. (A) SSCP and (B) se-
quencing analysis of caspase-8
DNA from tumors (lane T) and
normal tissues (lane N). (A) Ar-
rows (lane T) indicate aberrant
bands as compared with SSCP
from normal tissue (N). (B) Ar-
rows indicate nucleotide sub-
stitutions in tumor tissue as
compared with normal tissue.
R162I, Flag-caspase-8-G366H, Flag-caspase-8-R413G,
Flag-caspase-8-R413 stop, and Flag-caspase-8-frame-
shift) by using site-directed mutagenesis. On their trans-
fection into 293T cells, we found that the mutants
R413G, R413 stop, and frameshift showed significant
defects in apoptosis function compared with the wild-
type caspase-8 (Figure 2A; Dunnett multiple comparison
test; P ⬍ 0.001).
Furthermore, to explore how the caspase-8 mutants
impair apoptosis, we coexpressed each inactivating
caspase-8 mutant (R413G, R413 stop, and frameshift)
either with Fas or TRAIL-R2 in 293T cells. We found
that these mutants interfered with the death induction
by Fas or TRAIL-R2 overexpression, indicating that
these mutants have dominant-negative inhibition of the
death receptor–induced apoptosis (Figure 2B; Dunnett
multiple comparison test; P ⬍ 0.001). Also, to deter-
mine if the caspase-8 mutants are functional, we trans-
fected the caspase-8 mutants or wild-type caspase-8 in
CASPASE– 8 MUTATION IN COLON CANCER 711
293T cells, and compared the poly-(ADP-ribose) poly-
merase (PARP) cleavage in the lysates by immunoblot-
ting. Overexpression of the mutants R413G, R413stop
and frameshift showed absence of PARP cleavage,
whereas overexpression of wild-type caspase-8, caspase-
8-R162I and Flag-caspase-8-G366H showed the discern-
able cleavage of PARP (Figure 3).
Finally, we determined whether the inactivating muta-
tions of caspase-8 render colorectal cancer cells resistant to
the induction of apoptosis by drugs commonly used to treat
colorectal cancer. To this purpose, transient transfection of
DLD-1 cells with the caspase-8 mutants or wild-type
caspase-8 was performed. After 3 hours of the transfection,
100 – 600 ␮mol/L 5-FU was added. Figure 4 shows de-
creased cell death by the caspase-8-R413 stop and caspase-
8-frameshift compared to the wild-type caspase-8 and other
caspase-8 mutants after 5-FU (600 ␮mol/L) treatment. At
other ranges of 5-FU concentration (100 –500 ␮mol/L),
similar differences were observed (data not shown).
712 KIM ET AL.
Figure 2. Defective apoptotic activities of tumor-derived caspase-8
mutants. (A) 293T cells were transfected with 1.3 ␮g of wild-type (WT)
caspase-8 or each tumor-derived caspase-8 mutant together with 0.2
␮g of pEGEF. Twenty-four hours after transfection, cells were fixed in
10% methanol for 15 minutes and stained with 4⬘,6-diamidino-2-
phenylindole 1 ␮g/mL for 15 minutes, and the nuclei were examined
by fluorescence microscopy (mean ⫾ SD; n ⫽ 4). (B) 293T cells were
co-transfected with Fas or TRAIL-R2 construct and a 4-fold excess of
each mutant caspase-8. Cell death was analyzed as described in (B).
Inactivating Caspase-8 Mutants Bind With
Fas-Associated Death Domain Protein
The interference in the death pathway caused by
the caspase-8 mutants (Figure 2) led us to hypothesize
that the inactivating caspase-8 mutants might be re-
cruited into the DISC. To see this, we coexpressed FADD
either with a dominant-negative caspase-8 (an artificial
caspase-8 C360A mutant that has a mutation at the
activation site) or with the mutant R413G, R413 stop,
or frameshift. The coimmunoprecipitation analysis
showed that FADD coimmunoprecipitates with the in-
activating mutants identified in this study (R413G,
R413 stop, or frameshift; Figure 5), suggesting that
these mutants are recruited in the DISC. Figure 3 also
shows that FADD, as a positive control, coimmunopre-
Figure 3. Impaired PARP cleavage by caspase-8 mutations. 293T
cells were transfected with 1.3 ␮g of wild-type (WT) caspase-8 or each
tumor-derived caspase-8 mutant. Ten hours after transfection, cell
lysates were normalized for total protein content and then used by
direct immunoblot analysis by anti-PARP antibodies (Pharmingen, San
Diego, CA).
GASTROENTEROLOGY Vol. 125, No. 3
cipitates with caspase-8 dominant-negative. In contrast,
FADD did not coimmunoprecipitate with caspase-7
(data not shown), thus showing the specificity of these
results.
Expression of Caspase-8
In addition to genetic alterations, epigenetic
modification of caspase-8 by DNA methylation has re-
cently been implied in loss of caspase-8 function in
several solid tumors of childhood.17,20 To see whether
caspase-8 expression is lost in colon carcinoma samples,
we performed caspase-8 immunohistochemistry. We ob-
served that all of the 98 colon carcinoma samples showed
detectable caspase-8 expression in the cytosol of the
tumor cells (Figure 6A; Table 1). Negative controls
using nonimmune serum as primary antibodies showed
no signal (Figure 6B).
Discussion
The aim of this study was to address whether
human colon carcinoma has caspase-8 gene mutation. If
so, the next aim was to address whether these mutations
inactivated the proapoptotic activity of caspase-8. We
found that in colon carcinoma, but not in colon adenoma,
the caspase-8 gene is somatically mutated. Moreover, the
functional study showed that the cancer-derived
caspase-8 mutants showed losses of apoptotic function.
These data, together with the earlier reports on the
inactivating mutations of proapoptosis genes in human
cancers,9 –12,14,16 –19 indicate that somatic mutation of
proapoptosis genes, including caspase-8, may be a fre-
quent event in the pathogenesis of human cancers and
suggest that these mutations may play an important role
in the apoptosis resistance of cancer cells.
Figure 4. Decreased 5-fluorouracil (5-FU)-induced cell death by
caspase-8 mutations. 5-FU (600 ␮mol/L) was added to DLD-1 cells 3
hours after transfection of the caspase-8 mutants; 24, 48, and 72
hours later, cell death was determined by 4⬘,6-diamidino-2-phenylin-
dole staining.
September 2003
Figure 5. Binding of caspase-8
mutants to FADD. 293T cells
were cotransfected with expres-
sion constructs for human HA-
FADD either with the Flag-
caspase-8 dominant-negative or
Flag-caspase-8 R413Q, Flag-
caspase-8 R413 stop, or Flag-
caspase-8 frameshift mutant as
indicated. The transfected cells
were harvested and lysed 1 day
later, and immunoprecipitated
with anti-Flag antibodies, and se-
quentially immunoblotted with
anti-HA antibody as indicated.
Aliquots of the same lysates
(normalized for total protein con-
tent) were also analyzed directly
by SDS-PAGE/immunoblotting
as indicated. IP, immunoprecipi-
tation; WB, Western blotting.
Among the 5 mutations of caspase-8 found in this
study, 3 showed significant defects in inducing apopto-
sis. The frameshift mutation would change the last 20
amino acids in the p10 subunit. Because the active
caspase-8 consists of a (p20/p10)2 tetramer,22 this frame-
shift mutation may also lead to abnormal construction of
the tetramer. Also, 1 nonsense mutation (R413 stop) was
identified in the coding regions of the small p10 subunit.
This mutation is predicted to cause a premature termi-
nation of caspase-8 protein synthesis in the middle of the
p10 and, hence, resembles a typical loss-of-function mu-
tation. Another inactivating mutation of caspase-8
would also change amino acid residue 413 (R413G). The
arginine at amino acid 413 is conserved in other caspases,
such as caspase-1, -2, -3, -4, -5, and -6 (Genbank data-
base). The previous structural analysis of caspases showed
that the residues constituting the binding pocket for P1
aspartic acid are conserved in caspases and are important
for the apoptosis induction of the caspases.23 The argi-
nine 413 of caspase-8 is one of the constituting residues
for the P1 binding pocket,24 and it seems that alteration
of the arginine 413 decreases the protease function of
caspase-7. Because 2 different mutations in unrelated
patients would involve the same amino acid (arginine
413) and both of these mutations resulted in significant
functional losses of caspase-8, this position may represent
a mutational hotspot in the caspase-8 coding sequence.
In contrast to the inactivating mutations, 2 mutations
did not show any significant loss of apoptosis function of
caspase-8 (Figure 2). However, the functional implica-
tions of these mutations are unknown at this stage.
CASPASE– 8 MUTATION IN COLON CANCER 713
There is a death effector domain/caspase gene cluster
on human chromosome 2q33–34 that contains the genes
for FLICE-inhibitory protein, caspase-8, and caspase-10,
suggesting that all of these genes have arisen by tandem
duplication.25,26 Such duplications have been associated
with other chromosome regions that are altered during
tumorigenesis, and this may reflect the relative instabil-
ity of the chromosomal region.27 The occurrences of
caspase-10 gene mutations in non-Hodgkin lymphoma
and the caspase-8 gene mutations in colon cancer found
in this study suggest that caspase-8 and caspase-10
might be candidate tumor-suppressor genes at chromo-
some 2q33–34. Whether caspase-8 is such a tumor sup-
pressor is unknown at this stage, but its function as an
apoptosis initiator would be consistent with tumor sup-
pressor function.
In the colon carcinomas analyzed, all of the 5 muta-
tions of the caspase-8 gene showed evidence of retention
of the wild-type allele (Table 1). Also, 3 of the mutations
(cases 5, 17, and 21) showed a marked decrease of apo-
ptosis induction (Figure 2A). Therefore, in these cases, it
is possible that the heterozygously mutated caspase-8,
which inefficiently autoprocesses, and normal caspase-8
are co-recruited into the DISC and could therefore block
the caspase-8 activation in a dominant-negative fashion
(Figure 2B).
Transcription of caspase-8 can be inhibited by DNA
methylation,20 which, in turn, blocks the production of
caspase-8 protein. To discover whether caspase-8 in colon
cancer could be inactivated by DNA methylation, we
analyzed caspase-8 protein expression by immunohisto-
714 KIM ET AL.
Figure 6. Visualization of caspase-8 expression in the tissues by immu-
nohistochemistry. Antibodies were detected by a diaminobenzidine
method that produces a brown color. Counterstaining of nuclei was
performed with hematoxylin (blue). (A) Primary colon carcinoma cells
show immunoreactivity for caspase-8 in the cytosol. Fibroblasts between
the tumor nests are negative for the immunostaining. (B) Negative con-
trol of caspase-8 immunostaining (original magnification 200⫻).
chemistry. All of the colon cancer samples expressed
caspase-8 protein (Figure 4), indicating that caspase-8 in
colon cancers may not be transcriptionally inactivated by
DNA methylation and suggesting that mutation might
be a main mechanism of caspase-8 inactivation in the
colon cancers.
Colorectal carcinogenesis is a multistep process and
seems to require at least 7 genetic events for comple-
tion.28 However, the whole process is still poorly under-
stood. To date, many mutations of the components of the
death pathway, such as death receptors and caspases, have
been described in human cancers, but few cases of colon
cancers have been reported to have mutations of such
genes.29 The significant difference of caspase-8 mutation
frequency between adenoma and carcinoma of the colon
GASTROENTEROLOGY Vol. 125, No. 3
indicated that caspase-8 mutation is involved in the late
event of colorectal carcinogenesis, but not in the initiat-
ing event of it.
Despite increased awareness of colon cancer and im-
proved methods for early detection, many colon cancer
patients die of colon cancer.30 From an understanding of
basic apoptosis mechanisms, a variety of new strategies
for combating cancer are beginning to emerge. For ex-
ample, small molecules as activators of caspases, which
are sufficient for overcoming apoptosis suppression,
could be used for the treatment of cancer.31 Before such
manipulation, however, it is imperative to understand
the genetic alterations of caspases in cancers. In this
study, we found somatic mutations of the caspase-8 gene
in primary human cancer, and these data could be useful
for future antineoplastic therapy with the apoptosis ma-
chineries.
References
1. Reed JC. Mechanisms of apoptosis. Am J Pathol 2000;39:1415–
1430.
2. Nagata S. Apoptosis by death factor. Cell 1997;88:355–365.
3. Nicholson DW. Caspase structure, proteolytic substrates, and
function during apoptotic cell death. Cell Death Differ 1999;6:
1028 –1042.
4. Kumar S. Mechanisms mediating caspase activation in cell
death. Cell Death Differ 1999;6:1060 –1066.
5. Leithäuser F, Dhein J, Mechtersheimer G, Koretz K, Brn̈derlein S,
Henne C, Schmidt A, Debatin K-M, Krammer PH, Möller P. Con-
stitutive and induced expression of APO-1, a new member of the
new growth factor/tumor necrosis receptor superfamily, in nor-
mal and neoplastic cells. Lab Invest 1993;69:415– 429.
6. Schneider P, Thome M, Burns K, Bodmer JL, Hofmann K, Kataoka
T, Holler N, Tschopp J. TRAIL receptors 1 (DR4) and 2 (DR5)
signal FADD-dependent apoptosis and activate NF-kappaB. Immu-
nity 1997;7:831– 836.
7. Owen-Schaub LB, Radinsky R, Kruzel E, Berry K, Yonehara S.
Anti-Fas on nonhematopoietic tumors: levels of Fas/APO-1 and
bcl-2 are not predictive of biological responses. Cancer Res
1994;54:1580 –1586.
8. Sheridan JP, Marsters SA, Pitti RM, Gurney A, Skubatch M,
Baldwin D, Ramakrishnan L, Gray CL, Baker K, Wood WI, Goddard
AD, Godowski P, Ashkenazi A. Control of TRAIL-induced apoptosis
by a family of signaling and decoy receptors. Science 1997;277:
818 – 821.
9. Lee SH, Shin MS, Park WS, Kim SY, Kim HS, Han JY, Park GS,
Dong SM, Pi JH, Kim CS, Kim SH, Lee JY, Yoo NJ. Alterations of
Fas (Apo-1/CD95) gene in non-small cell lung cancer. Oncogene
1999;18:3754 –3760.
10. Lee SH, Shin MS, Kim HS, Lee HK, Park WS, Kim SY, Lee JH, Han
SY, Park JY, Oh RR, Jang JJ, Han JY, Lee JY, Yoo NJ. Alterations
of the DR5/TRAIL receptor 2 gene in non-small cell lung cancers.
Cancer Res 1999;59:5683–5686.
11. Shin MS, Kim HS, Lee SH, Lee JW, Song YH, Kim YS, Park WS,
Kim SY, Lee SN, Park JY, Lee JH, Xiao W, Jo KH, Wang YP, Lee
KY, Park YG, Kim SH, Lee JY, Yoo NJ. Alterations of Fas-pathway
genes associated with nodal metastasis in non-small cell lung
cancer. Oncogene 2002;21:4129 – 4136.
12. Shin MS, Kim HS, Kang CS, Park WS, Kim SY, Lee SN, Lee JH,
Park JY, Jang JJ, Kim CW, Kim SH, Lee JY, Lee SH. Inactivating
September 2003
mutations of CASP10 gene in non-Hodgkin lymphomas. Blood
2002;99:4094 – 4099.
13. Hanahan D, Weinberg RA. The hallmarks of cancer. Cell 2000;
100:57–70.
14. Kurokawa H, Nishio K, Fukumoto H, Tomonari A, Suzuki T, Saijo
N. Alteration of caspase-3 (CPP32/Yama/apopain) in wild-type
MCF-7, breast cancer cells. Oncol Rep 1999;6:33–37.
15. Palmerini F, Devilard E, Jarry A, Birg F, Xerri L. Caspase 7 down-
regulation as an immunohistochemical marker of colonic carci-
noma. Hum Pathol 2001;32:461– 467.
16. Mandruzzato S, Brasseur F, Andry G, Boon T, van der Bruggen P.
A CASP-8 mutation recognized by cytolytic T lymphocytes on a
human head and neck carcinoma. J Exp Med 1997;186:785–
793.
17. Takita J, Yang HW, Chen YY, Hanada R, Yamamoto K, Teitz T,
Kidd V, Hayashi Y. Allelic imbalance on chromosome 2q and
alterations of the caspase 8 gene in neuroblastoma. Oncogene
2001;20:4424 – 4432.
18. Liu B, Peng D, Lu Y, Jin W, Fan Z. A novel single amino acid
deletion caspase-8 mutant in cancer cells that lost proapoptotic
activity. J Biol Chem 2002;277:30159 –30164.
19. Schwartz S Jr, Yamamoto H, Navarro M, Maestro M, Reventos J,
Perucho M. Frameshift mutations at mononucleotide repeats in
caspase-5 and other target genes in endometrial and gastroin-
testinal cancer of the microsatellite mutator phenotype. Cancer
Res 1999;59:2995–3002.
20. Fulda S, Kufer MU, Meyer E, van Valen F, Dockhorn-Dworniczak B,
Debatin KM. Sensitization for death receptor- or drug-induced
apoptosis by re-expression of caspase-8 through demethylation
or gene transfer. Oncogene 2001;20:5865–5877.
21. Lee JY, Dong SM, Kim SY, Yoo NJ, Lee SH, Park WS. A simple,
precise and economical microdissection technique for analysis
of genomic DNA from archival tissue sections. Virchows Arch
1998;433:305–309.
22. Muzio M, Chinnaiyan AM, Kischkel FC, O’Rourke K, Shevchenko
A, Ni J, Scaffidi C, Bretz JD, Zhang M, Gentz R, Mann M, Krammer
PH, Peter ME, Dixit VM. FLICE, a novel FADD-homologous ICE/
CED-3-like protease, is recruited to the CD95 (Fas/APO-1) death-
inducing signaling complex. Cell 1996;85:817– 827.
23. Boldin MP, Goncharov TM, Goltsev YV, Wallach D. Involvement of
CASPASE– 8 MUTATION IN COLON CANCER 715
MACH, a novel MORT1/FADD-interacting protease, in Fas/APO-1-
and TNF receptor-induced cell death. Cell 1996;85:803– 815.
24. Blanchard H, Kodandapani L, Mittl PR, Marco SD, Krebs JF, Wu
JC, Tomaselli KJ, Grutter MG. The three-dimensional structure of
caspase-8: an initiator enzyme in apoptosis. Structure Fold Des
1999;7:1125–1133.
25. Fernandes-Alnemri T, Armstrong RC, Krebs J, Srinivasula SM,
Wang L, Bullrich F, Fritz LC, Trapani JA, Tomaselli KJ, Litwack G,
Alnemri ES. In vitro activation of CPP32 and Mch3 by Mch4, a
novel human apoptotic cysteine protease containing two FADD-
like domains. Proc Natl Acad Sci U S A 1996;93:7464 –7469.
26. Grenet J, Teitz T, Wei T, Valentine V, Kidd VJ. Structure and
chromosome localization of the human CASP8 gene. Gene 1999;
226:225–232.
27. Baens M, Aerssens J, van Zand K, Van den Berghe H, Marynen P.
Isolation and regional assignment of human chromosome 12p
cDNAs. Genomics 1995;29:44 –52.
28. Kinzler KW, Vogelstein B. Lessons from hereditary colorectal
cancer. Cell 1996;87:159 –170.
29. Yamamoto H, Gil J, Schwartz S Jr, Perucho M. Frameshift muta-
tions in Fas, Apaf-1, and Bcl-10 in gastro-intestinal cancer of the
microsatellite mutator phenotype. Cell Death Differ
2000;7:238 –239.
30. Kim JC, Kim CN, Yu CS, Lee HI, Kim SW, Lee JH, Kim WK, Kang
GH, Lee MK. Treatment of hepatic metastasis of colorectal can-
cer: a retrospective analysis of the outcome in 99 patients. J
Korean Cancer Assoc 1998;30:1175–1183.
31. Jiang X, Kim HE, Shu H, Zhao Y, Zhang H, Kofron J, Donnelly J,
Burns D, Ng SC, Rosenberg S, Wang X. Distinctive roles of PHAP
proteins and prothymosin-alpha in a death regulatory pathway.
Science 2003;299:223–226.
Received February 3, 2003. Accepted June 12, 2003.
Address requests for reprints to: Sug Hyung Lee, M.D., Department
of Pathology, College of Medicine, The Catholic University of Korea,
505 Banpo-dong, Socho-gu, Seoul 137-701, Korea. e-mail:
suhulee@cmc.cuk.ac.kr; fax: (82) 2-537-6586.
Supported by funding from the Molecular Medicine Research Group
program from the Ministry of Science and Technology of Korea (Grant
M10106000075-02B1700-01810) and by funds from POSCO (Grant
20027045).