The Protectors - PDF eBooks Free | Page 1

2021
Development 126, 2021-2031 (1999)
Printed in Great Britain © The Company of Biologists Limited 1999
DEV3958
C. elegans MAC-1, an essential member of the AAA family of ATPases, can
bind CED-4 and prevent cell death
Dayang Wu1, Pei-Jiun Chen2, Shu Chen1, Yuanming Hu1, Gabriel Nuñez1,* and Ronald E. Ellis2,*,§
1Department of Pathology and Comprehensive Cancer Center, University of Michigan Medical School, and 2Department of
Biology, The University of Michigan, Ann Arbor, Michigan 48109, USA
*Joint senior authors
§Author for correspondence (e-mail: [email protected])
Accepted 9 February; published on WWW 6 April 1999
SUMMARY
In the nematode Caenorhabditis elegans, CED-4 plays a
central role in the regulation of programmed cell death. To
identify proteins with essential or pleiotropic activities
that might also regulate cell death, we used the yeast twohybrid system to screen for CED-4-binding proteins. We
identified MAC-1, a member of the AAA family of ATPases
that is similar to Smallminded of Drosophila.
Immunoprecipitation studies confirm that MAC-1 interacts
with CED-4, and also with Apaf-1, the mammalian
homologue of CED-4. Furthermore, MAC-1 can form a
multi-protein complex that also includes CED-3 or CED-9.
A MAC-1 transgene under the control of a heat shock
promoter prevents some natural cell deaths in C. elegans,
and this protection is enhanced in a ced-9(n1950sd)/+
genetic background. We observe a similar effect in
mammalian cells, where expression of MAC-1 can prevent
CED-4 and CED-3 from inducing apoptosis. Finally, mac1 is an essential gene, since inactivation by RNA-mediated
interference causes worms to arrest early in larval
development. This arrest is similar to that observed in
Smallminded mutants, but is not related to the ability of
MAC-1 to bind CED-4, since it still occurs in ced-3 or ced4 null mutants. These results suggest that MAC-1 identifies
a new class of proteins that are essential for development,
and which might regulate cell death in specific
circumstances.
INTRODUCTION
proteins, which also promote survival (Hengartner et al., 1992;
Hengartner and Horvitz, 1994). Genetic experiments indicate
that ced-9 prevents cells from undergoing programmed deaths
by regulating the activities of ced-3 and ced-4 (Hengartner et
al., 1992; Shaham and Horvitz, 1996). In addition, ectopic
expression of these genes in the touch neurons of nematodes
suggests that ced-4 works by activating ced-3, and that ced-9
somehow blocks this activation (Shaham and Horvitz, 1996).
What factors determine which cells will be protected by ced9, and which will die? Recent results indicate that CED-9 is
inactivated by the product of the egl-1 gene (Conradt and
Horvitz, 1998). EGL-1 binds CED-9, and this binding
apparently allows CED-4 and CED-3 to kill cells. EGL-1
contains a BH3 domain, like some related proteins that induce
cell death; these include Bik, Bid, Harakiri and Bad. The
expression of egl-1 is probably controlled by proteins that
regulate the deaths of specific cells. These regulatory genes
might include ces-1 and ces-2, which control specific neuronal
deaths in the pharynx (Ellis and Horvitz, 1991; Metzstein et
al., 1996).
The discovery of physical interactions between CED-3,
CED-4 and CED-9 helped elucidate the molecular basis of cell
death. When expressed in mammalian or yeast cells, wild-type
Programmed cell death is a conserved mechanism for killing
cells, both during development and for tissue homeostasis
(Ellis et al., 1991). The nematode C. elegans is a leading model
for genetic and molecular analysis of cell death. In C. elegans
hermaphrodites, 131 of the 1090 somatic cells produced during
development undergo programmed deaths (Sulston and
Horvitz, 1977; Sulston et al., 1983). Loss-of-function
mutations in either ced-3 or ced-4 cause all of these cells to
survive (Ellis and Horvitz, 1986), showing that both genes are
essential for cell death to occur. These deaths are likely to be
‘suicides’, since genetic mosaic analysis indicates that ced-3
and ced-4 act within dying cells (Yuan and Horvitz, 1990).
CED-3 belongs to a conserved family of intracellular cysteine
proteases, termed caspases, that are homologous to mammalian
interleukin-1β-converting enzyme (Miura et al., 1993; Yuan et
al., 1993). In both nematodes and mammals, CED-3 and
related caspases are thought to kill cells by cleaving specific
targets (Cohen, 1997).
The nematode gene ced-9 is required to protect cells that
normally survive from undergoing programmed deaths; its
product is homologous to the mammalian Bcl-2 and Bcl-XL
Key words: Apoptosis, Programmed cell death, CED-4,
Smallminded, AAA proteins
2022 D. Wu and others
CED-9 interacts with CED-4 and prevents cell death, whereas
mutant CED-9 neither binds CED-4 nor prevents death
(Chinnaiyan et al., 1997; James et al., 1997; Spector et al.,
1997; Wu et al., 1997a,b). Thus, CED-9 might function by
binding and inactivating CED-4. When expressed in
mammalian cells, CED-3 also binds CED-4, forming a multiprotein complex that includes CED-9 (Chinnaiyan et al., 1997;
Wu et al., 1997a). Thus, CED-4 regulates cell death by
bringing together molecules that promote apoptosis with those
that inhibit it.
CED-4 is homologous to the mammalian protein Apaf-1,
which also promotes cell death (Zou et al., 1997). This
similarity is restricted to their amino-termini, which each
contain a caspase recruitment domain similar to that found in
the prodomains of CED-3 and mammalian caspases like
caspase-9 (Hofmann et al., 1997). Thus, homophilic
interactions between caspase recruitment domains might allow
Apaf-1 to interact with caspase-9, and CED-4 with CED-3 (Li
et al., 1997; Hu et al., 1998; Pan et al., 1998). Once bound,
Apaf-1 promotes the processing and activation of caspase-9,
just as CED-4 promotes the activation of CED-3 (Li et al.,
1997; Hu et al., 1998).
In C. elegans, the genes of the cell death pathway were each
identified by viable mutations (Trent et al., 1983; Ellis and
Horvitz, 1986; Hengartner et al., 1992). Because these genetic
screens might have missed essential genes that act in the cell
death process, or which induce cell death in specific
circumstances, we used the yeast two-hybrid system to find
proteins that bind CED-4. We identified MAC-1, which is a
member of a family of ATPases associated with a variety of
cellular activities, known as AAA ATPases (Confalonieri and
Duguet, 1995). In the presence of CED-4, MAC-1 can form a
multi-protein complex with CED-3 or CED-9. Furthermore,
expression of MAC-1 prevents cell deaths in nematodes and in
mammalian cells. Finally, we show that inactivation of mac-1
blocks larval development in C. elegans, a phenotype similar
to that of the AAA family member Smallminded of
Drosophila.
MATERIALS AND METHODS
Reagents and cell lines
The 293T cell line was described by Wu et al. (1997b). Mouse
monoclonal antibodies against Flag epitope (M2; Kodak/IBI), HA
epitope (12CA5; Boehringer Mannheim, Indianapolis IN) and Myc
epitope (9E10; Santa Cruz Biotech) were obtained from the indicated
sources. Rabbit antibodies against Flag, HA and Myc were from Santa
Cruz Biotech. The enhanced chemiluminescence (ECL) system was
from Amersham. Protein A- or Protein G-Sepharose 4B were from
Zymed Laboratories. Other chemicals were from Sigma, unless
indicated otherwise.
Identification and cloning of the MAC-1 cDNA
A mixed stage C. elegans cDNA library fused to the GAL4 DNAactivation domain of pACT vector (a gift from Robert Barstead) was
screened with a CED-4S bait using the yeast HF7c strain as described
by Wu et al. (1997b). Positive library clones were identified by βgalactosidase staining (Wu et al., 1997b). Plasmids were recovered by
transformation into bacteria and growth in M9 medium lacking
leucine. The cDNA inserts were characterized by restriction enzyme
mapping and nucleotide sequence analysis on an automated DNA
sequencer (Applied Biosystems). False-positive clones were
eliminated by testing the interaction with empty vector and irrelevant
‘bait’. A positive clone cDNA corresponding to MAC-1 was further
characterized. Plasmids that express GAL-4-CED-3, GAL4-CED-9 in
pACT and GAL-4 in pGBT8 have been described (Wu et al., 1997b).
Two hybrid yeast growth assays were performed as described by Wu
et al. (1997b).
Transfection, immunoprecipitation and western blot
analysis
Either 3-5×106 in 100 mm or 1-2×106 in 60 mm of human embryonic
kidney 293T cells in tissue culture dishes were transfected with the
indicated amount (see figure legends) of plasmid DNA by the calcium
phosphate method. pcDNA3-ced-4-Myc, -ced-3-HA and -HA-ced-9
constructs were described by Wu et al. (1997a). For
immunoprecipitations, cells from each dish were lysed in 0.5-1 ml of
isotonic lysis buffer containing 0.2% Nonidet P-40 and proteinase
inhibitors at 18-22 hours after transfection, and the soluble lysates
were incubated with different antibodies overnight at 4°C with 5%
(v/v) of protein A- or Protein G-Sepharose 4B beads. Complexes were
centrifuged, washed with lysis buffer at least four times, separated by
10-15% SDS-PAGE and immunoblotted. Signals were detected by
enhanced chemiluminescence system (Amersham). About 5% of the
total lysate was used for immunoblotting to assess the expression of
the transfected constructs.
Apoptosis assay
1×105 293T cells in each well of 12-well plates were transiently cotransfected in triplicate with 0.2 µg of a reporter plasmid pcDNA3-βgal with or without 0.2 µg pcDNA3-ced-3-HA, 0.05 µg pcDNA3-ced4-Myc, 1 µg pcDNA3-HA-ced-9, 1 µg pcDNA3-Flag-MAC-1, 1 µg
pcDNA3-Flag-MAC-1 (1-239) and 1 µg pcDNA3-Flag-MAC-1 (234813) by the calcium phosphate method. The total amount of
transfected plasmid DNA was always 2 µg per 3 wells and was
adjusted by adding vector pcDNA3 plasmid. The percentage of
apoptotic cells was determined 18 hours after transfection in triplicate
cultures by analysis of at least 300 cells expressing β-galactosidase
(Wu et al., 1997a).
Construction and analysis of transgenic nematodes
Nematodes were raised at 20°C as described by Brenner (1974). We
generated transgenic animals by co-injection of 50 ng/µl of the
plasmid pRF4, which carries the rol-6(su1006sd) mutation, with 100
ng/µl of an hsp16-2::mac-1 plasmid, an hsp16-41::mac-1 plasmid, or
an empty hsp16-2 vector (pPD49.78). In each case, the injected DNA
formed an extra chromosomal array, which we followed by scoring
the Rol-6 phenotype (Mello et al., 1991). Individual unc-29(e1072)
fog-3(q504); vEx8[hsp16-2::mac-1 rol-6(su1006sd)] females were
prepared by standard crosses, and mated with ced-9(n1950gf); him5(e1490) males. We transferred the fertilized animals to fresh plates,
and subjected them to a 1-hour heat shock at 31°C. The plates were
returned to 20°C for an additional 90 minutes, after which the parents
were removed. Once the embryos had matured, we scored the
pharynges of Rol larvae and young adults for alterations in the normal
pattern of programmed cell death (Ellis and Horvitz, 1991; Hengartner
et al., 1992). Similar crosses were carried out in parallel to study two
other transgenes, vEx9[hsp16-41::mac-1 rol-6(su1006sd)] and
vEx10[hsp16-2 vector pPD49.78 rol-6(su1006sd)]. We used heat
shocks of a 1-hour duration at 31°C because higher temperatures or
longer exposures led to significant lethality in both the experimental
and the control animals. In these experiments, we usually scored 14
cells in the anterior pharynx of each animal (Ellis and Horvitz, 1991).
However, when we examined the effects of an empty vector in wildtype animals, the dorsal pharynx from one of the nine animals could
not be scored, so three cells from this region were not included in the
data set. The hsp16-41::mac-1 transgene also prevented programmed
deaths, but its effect was weaker than that of the hsp16-2::mac-1
transgene (data not shown).
MAC-1 is essential and can interact with CED-4 2023
RNA-mediated interference
We generated RNA complementary to the mac-1 gene by using in
vitro transcription from T7 promoters. The RNAs were precipitated
and resuspended in 1× injection buffer (Fire, 1986), allowed to anneal,
and their concentration estimated by ethidium staining. The injections
were carried out as described by Guo and Kemphues (1995) and Fire
et al. (1998). The double-stranded RNA for the unc-4 control
corresponded to the entire predicted unc-4 transcript.
RT-PCR and northern blot analysis
Nematodes for RNA preparation were grown in liquid culture, as
described by Sulston and Brenner (1974). RNA from mixed stage
worms was used as a template for RT-PCR analysis. The RT-PCR
reactions were performed using a cDNA cycle kit (Invitrogen). The
products of reverse transcription were amplified with either the
SL1 primer (GGTTTAATTACCCAAGTTTGAG) or SL2 primer
(GGTTTTAACCCAGTTACTCAAG), and primers complementary to
the 5′ region of the MAC-1 coding sequence. The products were
purified and sequenced using an automated DNA sequencer (Applied
Biosystems model 373A). We used four strains for northern analysis:
N2 XX hermaphrodites, which were harvested as a mixed population,
him-5(e1490) XX and XO animals (Hodgkin et al., 1979), which were
harvested as a mixed population containing 30% males and 70%
hermaphrodites, fem-3(q96gf,ts) XX animals, which were harvested as
adults after shifting the cultures to the restrictive temperature of 25°C,
and fem-1(hc17ts) XX animals, which were also harvested as adults
after shifting the cultures to 25°C. The fem-3(q96gf,ts) mutants
produce only sperm (Barton et al., 1987) whereas the fem-1(hc17ts)
animals produce only oocytes (Kimble et al., 1984). For northern
analysis, mac-1 antisense RNA fragment B (Fig. 9) was radio-labeled
by incorporation of [32P]UTP and hybridized to blots containing
nematode RNAs.
SL2 primer in combination with any of several primers from
the 5′ region of mac-1. The sequence of these PCR products
overlaps with that of the original cDNA, and defines the 5′-end
of the message. A BLAST search identified two partial EST
sequences from C. elegans that correspond to the mac-1
message (Kohara, 1996).
Our composite cDNA contains an ORF encoding a protein
of 813 amino acids, with a predicted molecular mass of 89 kDa
(Fig. 2A). A search of protein databases revealed that MAC-1
contains two conserved ATPase domains (CADs); each domain
includes Walker A and B signature sequences and an AAA
minimal consensus motif (Fig. 3A; Swaffield and Purugganan,
1997; Patel and Latterich, 1998). Thus, we named this protein
MAC-1, for member of the AAA family that binds CED-4.
Although MAC-1 shares amino acid and structural similarity
with many AAA proteins, it most resembles mouse VCP
(Egerton et al., 1992), yeast CDC48 (Fröhlich et al., 1991) and
Smallminded, a Drosophila protein required for larval
development (Long et al., 1998). The percentage identity
between these proteins is particularly high within two regions
of MAC-1, amino acids 204-455 and 507-757; these regions
correspond to the conserved AAA modules characteristic of
this family of ATPases (Fig. 3A,C).
MAC-1 interacts with CED-4 in mammalian cells
To verify that MAC-1 associates with CED-4, 293T human
RESULTS
Identification of MAC-1
To search for proteins that bind CED-4, we screened a C.
elegans cDNA library using GAL4-CED-4 as ‘bait’ in the yeast
two-hybrid assay. From a screen of 2× 106 clones, four cDNAs
were found whose products interact with GAL4-CED-4, but
not with control baits (Wu et al., 1997b). The products of two
of these four cDNAs bind CED-4 in other assays. One contains
the ced-9 coding region, fused to the GAL4 transcriptional
activation domain (Wu et al., 1997b). Here we describe the
other, whose product we call MAC-1 (see below). In the twohybrid system, CED-4 interacts with MAC-1, CED-3 and
CED-9 but not with the control vector, as expected from
previous results (Fig. 1A). Using GAL4-MAC-1 as bait, MAC1 associates with CED-4 but not with CED-3 or CED-9, which
shows the specificity of the CED-4-MAC-1 interaction (Fig.
1B).
MAC-1 is a novel member of the AAA family of
ATPases
Sequence analysis revealed that our cDNA contained an insert
of approx. 2.4 kb fused to the GAL4-DNA-transcription
activation domain. Because it was unclear if the cDNA
contained the entire mac-1 coding region, we used the reverse
transcriptase-polymerase chain reaction (RT-PCR) to amplify
mixed stage C. elegans RNA, using forward primers
corresponding to either the SL1 or SL2 trans-spliced leaders
(Krause and Hirsh, 1987) and reverse primers from the 5′
region of our cDNA. We could amplify fragments using the
Fig. 1. Interaction of MAC-1, CED-3, CED-4 and CED-9 in yeast
cells. (A) A plasmid expressing CED-4S fused to the GAL4 DNA
binding domain was co-transfected with indicated plasmids
expressing MAC-1, CED-3 or CED-9 fused to the GAL4 DNAactivation domain. Growth of yeast cells in the absence of leucine
(Leu−), tryptophan (Trp−) and histidine (His−) is indicative of
protein-protein interaction. Growth in the absence of leucine and
tryptophan is shown as control. (B) A plasmid expressing MAC-1
fused to the GAL4 DNA transcription activation domain was cotransfected with indicated plasmids expressing CED-3, CED-4 or
CED-9 fused to the GAL4 DNA-binding domain.
2024 D. Wu and others
A
MPGGMGFPSDPALLPRVQAHIRKFPGTKYFKPELVAYDLQQEHPEYQRKNHKVFMGMV
REALERIQLVAKEENDEKMEEKEAMDDVQEIPIVKALETRKRKAPAAGRKSTGQAAAA
KEVVLSDDSEDERAARQLEKQIESLKTNRANKTVLNLYTKKSAPSTPVSTPKNQATKK
PPGASAAPPALPRGLGAVSDTISPRESHVKFEHIGGADRQFLEVCRLAMHLKRPKTFA
TLGVDPPRGFIVHGPPGCGKTMFAQAVAGELAIPMLQLAATELVSGVSGETEEKIRRL
FDTAKQNSPCILILDDIDAIAPRRETAQREMERRVVSQLCSSLDELVLPPREKPLKDQ
LTFGDDGSVAIIGDSPTAAGAGVLVIGTTSRPDAVDGRLRRAGRFENEISLGIPDETA
REKILEKICKVNLAGDVTLKQIAKLTPGYVGADLQALIREAAKVAIDRVFDTIVVKNE
GHKNLTVEQIKEELDRVLAWLQGDDDPSALSELNGGLQISFEDFERALSTIQPAAKRE
GFATVPDVSWDDIGALVEVRKQLEWSILYPIKRADDFAALGIDCRPQGILLCGPPGCG
KTLLAKAVANETGMNFFSVKGPELLNMYVGESERAVRTVFQRARDSQPCVIFFDEIDA
LVPKRSHGESSGGARLVNQLLTEMDGVEGRQKVFLIGATNRPDIVDAAILRPGRLDKI
LFVDFPSVEDRVDILRKSTKNGTRPMLGEDIDFHEIAQLPELAGFTGADLAALIHESS
LLALQARVLENDESVKGVGMRHFREAASRIRPSVTEADRKKYEHMKKIYGLKQATPPS
V
Fig. 2. Deduced amino acid sequence and structure of MAC1. (A) Amino acid sequence (single letter amino acid code)
predicted from the MAC-1 cDNA. (B) Schematic structure
of MAC-1. Conserved ATPase Domains (CADs) are
indicated by gray boxes.
58
116
174
232
290
348
406
464
522
580
638
696
754
812
813
B
1
204
455
507
757 813
MAC-1
kidney cells were transiently co-transfected with expression
plasmids producing Flag-tagged MAC-1 and Myc-epitopetagged CED-4. We prepared immunoprecipitates using a
monoclonal antibody to Flag, and separated them by SDS
polyacrylamide gel electrophoresis. Immunoblotting with a
rabbit antibody to Myc showed that CED-4 coimmunoprecipitates with MAC-1 (Fig. 4A). The MAC-1CED-4 interaction is specific, because it requires coexpression of both MAC-1 and CED-4, and was not detected
in precipitates formed using a control antibody (Fig. 4A). We
also measured the ability of MAC-1 to associate with three
mutant CED-4 proteins, each of which causes a loss-offunction in C. elegans. The mutation n1948 causes an amino
acid change (Ile to Asn) at position 258, the mutation n1894
creates a stop codon at position W401 (Yuan and Horvitz,
1992), and the mutation G328 creates a stop codon at amino
acid 328 (Wu et al., 1997a). MAC-1 co-immunoprecipitates
CED-4 I258N and CED-4 W401, but not CED-4 G328 (Fig.
4A). Thus, amino acids 328 to 401 of CED-4 are required for
association with MAC-1.
We engineered two mutants, MAC-1-N (amino acids 1-239)
and MAC-1-C (amino acids 234-813) to determine which
portions of MAC-1 were required for its interaction with CED4 (Fig. 4B). We then co-transfected 293T cells with expression
plasmids for both Myc-tagged CED-4 and either wild-type or
mutant Flag-tagged MAC-1. Our results show that the Nterminal and C-terminal regions of MAC-1 can interact
independently with CED-4, although the binding ability of
each mutant protein is significantly lower than that of wild-type
MAC-1 (Fig. 4C). To confirm these results, we performed
reciprocal
experiments
in
which
CED-4
was
immunoprecipitated
using
anti-Myc
antibodies.
Immunoblotting with anti-Flag antibodies confirmed that
CED-4 can associate with the wild-type and both mutant MAC1 proteins (Fig. 4D).
MAC-1 associates with CED-3, CED-9 and Apaf-1 in
mammalian cells
Because CED-4 interacts with both CED-3 and CED-9 in
mammalian cells (Chinnaiyan et al., 1997; Wu et al., 1997a),
we assessed the ability of MAC-1 to associate with CED-3 and
CED-9. A plasmid producing Flag-tagged MAC-1 was cotransfected with plasmids expressing HA-tagged CED-3 or
CED-9, in the presence or absence of CED-4.
Immunoprecipitation of MAC-1 complexes with anti-Flag
antibodies revealed that CED-3 and CED-9 each coimmunoprecipitate with MAC-1, even in the absence of CED4 (Fig. 5A). However, expression of CED-4 significantly
increased the amounts of CED-3 and CED-9 that coimmunoprecipitate with MAC-1 (Fig. 5A). We verified these
results in reciprocal experiments, in which CED-3 or CED-9
were immunoprecipitated using anti-HA antibodies. In these
experiments, either CED-3 or CED-9 could coimmunoprecipitate MAC-1 (Fig. 5B). Because MAC-1 does
not interact with CED-3 or CED-9 in the yeast two-hybrid
system (Fig. 1), this co-immunoprecipitation might involve an
adapter protein present in 293T cells. One candidate is Apaf1, a homologue of CED-4 that is expressed in 293T cells (Li
et al., 1997). To see if MAC-1 can bind Apaf-1, we coexpressed Myc-tagged Apaf-1 and Flag-tagged MAC-1, and
immunoprecipitated MAC-1 with anti-Flag antibodies.
Immunoblotting revealed that Apaf-1 co-immunoprecipitates
with MAC-1 (Fig. 5C).
Genomic organization and expression of mac-1 in C.
elegans
A BLAST search using the cDNA sequence showed that mac1 is located on the YAC clones Y48C3 and Y81G3, which are
being sequenced by the C. elegans Genome Sequencing
Consortium (Sulston et al., 1992). This result places mac-1 on
the right arm of chromosome II, between the genes rtw-3 and
rsn-2 (Fig. 6A,B). Comparison of the genomic and cDNA
sequences shows that mac-1 contains seven exons, and is
trans-spliced to the SL2 leader sequence (Fig. 6C). We used
northern analysis to measure the size of the mac-1 transcript,
and study its expression pattern. RNA from a mixed
population of males and hermaphrodites of all ages contains
a single mac-1 transcript of approx. 2.4 kb; this size matches
the length of our cDNA. The mac-1 mRNA is found
throughout larval development (data not shown), but is
expressed at highest levels in hermaphrodites that are
producing oocytes (Fig. 6D).
MAC-1 is essential and can interact with CED-4 2025
Elevated expression of MAC-1 in C. elegans
prevents programmed cell deaths
To determine if MAC-1 could regulate programmed cell deaths
in C. elegans, we created a transgenic strain of animals in
which expression of mac-1 could be driven by the heat shock
promoter hsp16-2 (Candido et al., 1989). Initial observation of
13 transgenic animals subjected to a heat shock during early
embryogenesis revealed that 10 pharyngeal cells (out of a
possible 182) had failed to undergo programmed deaths,
whereas only 1 cell failed to die (out of a possible 123) in
control animals that had also been heat shocked. Because this
effect was small, we next studied the expression of transgenic
mac-1 in cells that were genetically predisposed towards
survival. In ced-9(n1950gf)/+ progeny of wild-type mothers,
one third of the cells that would normally die instead survive
(Hengartner et al., 1992). We observed this same frequency
following heat shock of ced-9(n1950gf)/+ transgenic animals
that carried an empty hsp16-2 vector (Fig. 7). By contrast, heat
shock of similar animals carrying the hsp16-2::mac-1
transgene doubles the number of surviving cells (Fig. 7). The
extra cells we observe in these animals are in the same
positions, and share the same morphologies, we have observed
for surviving cells in ced-3 or ced-4 mutants. Thus, expression
of MAC-1 can prevent programmed cell deaths in C. elegans.
MAC-1 prevents CED-3 and CED-4 from causing
apoptosis in mammalian cells
To determine if MAC-1 could prevent cell death in mammalian
cells, we co-transfected 293T cells with expression constructs
producing either wild-type or mutant MAC-1, and plasmids
expressing CED-3 and CED-4, and then assayed the cells for
apoptosis. Expression of either CED-3, or of both CED-3 and
CED-4, induces apoptosis in 293T cells (Wu et al., 1997a).
This effect can be blocked by co-expression of CED-9 (Wu et
al., 1997a). Significantly, the expression of either wild-type
MAC-1, or of the MAC-1-C (amino acids 234-813) mutant,
can prevent the induction of cell death by either CED-3, or by
CED-3 plus CED-4 (Fig. 8A). However, the MAC-1-N (amino
acids 1-239) mutant had no effect on cell death. Further
experiments showed that apoptosis induced by CED-3, or by
A
MAC-1
mVCP
CDC48
smallminded
MP GG- MGFP S D P A L L P RV QA HI RK FP GTK Y FK P E L V A Y DL QQE H P E Y QRK NHK V F MGMV RE A L E R I QL V A K E E NDE K ME E K E A MDDV QE I P I V K A L E T RK RK A P A A GR K S T GQA A A A K E V V L S - - DDS E D E RA A RQL E K M- - - - - - - - - A S GA DS K G- - - - DD- L S TA I L K QK NRP NRL I V DE A I NE DNS V V S L S QP K MDE L QL F RGD TV L L K GK K RRE A V CI - - - - - V L S DD T CS D E K I R MNRV V R NNL - RV RL GDV I S I QP CP DV K Y GK RI HV L P I D
MGE E HK P L L DA S GV DP RE - - - - E DK TA TA I L RR K K K DN ML L V DD A I NDDNS V I A I NS N T MDK L E L F RGD TV L V K GK K RK D TV L I - - - - - V L I DDE L E D GA CRI NRV V R NNL - RI RL GDL V T I HP CP DI K Y A T RI S V L P I A
MK K A - K P L L HD HL I T I RV K K Y L E E HI GE TY L DV K QMT RE L MQK Y P E Y S RRK F GP F RQL V HQA FS I I S E S Y NL DK V S S S E E DCV S E DS E P P P T NS V MNN MMNS L Y S QP R K P L A P K P I S E P I DI S S GDE NE D DS N TK T T NGD
136
120
130
139
MAC-1
mVCP
CDC48
smallminded
- - - - - - - - - - - - - - - QI E S L K T NRA NK TV L NL Y TK K S A P S TP V S TP K NQA TK K P P GA S A A P P A L P - - - - - - - - - - R GL GA V S D T I S P RE S HV K - - - - - - - - - - - - - - - - FE HI GGA DRQ F L E V CRL A - MH L K RP K T FA T L
D TV E GI T GNL F E V Y L K P Y F L E A Y RP I RK GDI F L V RGGMRA V E FK V V E T DP S P Y C I V A P D TV I HCE GE P I K RE DE E E S L NE V - - - - - - - - - - - - - - - - - - - - - - - - - - - GY DDV GGCRK Q L A QI K E MV E L P L RHP A L FK A I
D T I E GI T GNL F DV F L K P Y FV E A Y RP V RK GDH FV V RGGMRQV E FK V V DV E P E E Y A V V A QD T I I HWE GE P I NRE DE E N N MNE V - - - - - - - - - - - - - - - - - - - - - - - - - - - GY DDI GGCRK Q MA QI RE MV E L P L RHP QL FK A I
GV A A A A A P P P P TP A V QGS A L K R L ME E V P E I A V A A K K A K P N T I HV S S S E A I QK L H QV V GNRA K NL S E DA V P RS K DHR NV P GL Y QQL HQNQS RDRL RK FK RDL E V QHP TE S F RDI GGMDS T L K E L CE ML - I H I K S P E FY F QL
234
233
243
278
MAC-1
mVCP
CDC48
smallminded
GV DP P RGF I V H GP P GCGK T MFA QA V A GE L A I P ML QL A A TE L V S GV S GE TE E K I R RL F D TA K QNS P CI L I L DDI DA I
GV K P P RGI L L Y GP P GT GK T L I A RA V A NE T GA F F F L I NGP E I MS K L A GE S E S NL R K A FE E A E K NA P A I I F I DE L DA I
GI K P P RGV L MY GP P GT GK T L MA RA V A NE T GA F F F L I NGP E V MS K MA GE S E S NL R K A FE E A E K NA P A I I F I DE I DS I
GL L P S RGL L L H GP P GCGK T F L A RA I S GQL K MP L ME I P A TE L I GGI S GE S E E RI R E V F DQA I GY S P CV L F I DE I DA I
GDS P TA A GA GV L V I
- - - - - - - - A HV I V M
- - - - - - - - S NV V V I
- - - - - - - - - SVVVI
374
344
354
392
MAC-1
mVCP
CDC48
smallminded
GT TS RP DA V DGRL RRA GR FE NE I S L GI
A A T NRP NS I DP A L RR F GR F DRE V DI GI
A A T NRP NS I DP A L RR F GR F DRE V DI GI
A A T T RP DV L DP GL RRI GR F DHE I A I HI
P DE TA R E K I L E K I CK - V NL A GDV T L K QI A K L TP GY V GA DL QA L I RE A A K V A I DRV - - - - - - - - - - - - - - - F D T I V V - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - K NE GHK P DA T GR L E I L QI H TK N MK L A DDV DL E QV A NE T HGHV GA DL A A L CS E A A L QA I RK K - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - P DA T GR L E V L RI H TK N MK L A DDV DL E A L A A E T HGY V GA DI A S L CS E A A MQQI RE K - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - P S RK E R RE I L RI QCE GL S V DP K L NY DK I A E L TP GY V GA DL MA L V S RA A S V A V K RRS MK K F RE L HA A S E K N MT TV T L DDDE P S E DA GE TP V P D S K GE E TA K DA E A E QK V DGDK E
467
426
436
532
MAC-1
mVCP
CDC48
smallminded
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - TS A K DK S E GDS P NI
- - - - - - - - - - - - - - - - - - - - - - NL TV E QI K E E L DRV L A WL QGDDD- - P S A L S E L NGGL Q- - - - - - - - - - - - I S FE D FE RA L S T I QP A A K RE G FA TV P DV S WDD I GA L V E V RK Q
- - - - - - - - - - - - - - - - - - MDL I DL E DE T I DA E V MNS L A - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - V T MDD F RWA L S QS NP S A L RE T V V E V P QV T WE D I GGL E DV K RE
- - - - - - - - - - - - - - - - - - MDL I DL DE D E I DA E V L DS L G - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - V T MDN F R FA L GNS NP S A L RE T V V E S V NV T WDD V GGL DE I K E E
K S P QK T P K K S A E K P T DA A MDV DNV A P E E P K K A V E QE V D S S S S NDE Y Y E P T L A E L T N F L DN P P E E FA DP N F C L T L I D FV DA I K V MQP S A K RE G F I TV P D T T WDD I GA L E K I RE E
544
488
498
672
MAC-1
mVCP
CDC48
smallminded
L E WS I L Y P I K R A DD FA A L GI DC RP QGI L L CGP P GCGK T L L A K A V A NE T GMN F FS V K GP E L L N MY V GE S E RA V R TV F QRA RDS QP CV I
L QE L V QY P V E H P DK F L K F GMT - P S K GV L FY GP P GCGK T L L A K A I A NE CQA N F I S I K GP E L L T MWF GE S E A NV RE I F DK A RQA A P CV L
L K E TV E Y P V L H P DQY TK F GL S - P S K GV L FY GP P GT GK T L L A K A V A TE V S A N F I S V K GP E L L S MWY GE S E S NI RDI F DK A RA A A P TV V
L K L A V L A P V K Y P E ML E RL GL TA - P S GV L L CGP P GCGK T L L A K A I A NE A GI N F I S V K GP E L MN MY V GE S E RA V RA C F QRA RNS A P CV I
MAC-1
mVCP
CDC48
smallminded
I V DA A I
I I DP A I
QI DP A I
I I DP A I
MAC-1
mVCP
CDC48
smallminded
DRK K Y E H MK K - - - - - - - - - - - - I Y GL - K QA TP P S - - - - - - - - - - - - - - - - - - - - - - - - - - - - V
DI RK Y E MFA QT L QQS RG- F GS F R FP S GNQGGA GP - - - - S QGS GG GT GGS V Y TE - - DNDDDL Y G
E L RRY E A Y S QQ MK A S RGQFS N F N F NDA P L GT TA T DNA NS NNS A P S GA GA A F GS N A E E DDDL Y S
DRK I Y DK L RL - - - - - - - - - - - - K Y A A P RV P T L N D- - - - - - - - - - - - - - - - - - - - - - - - - - - - K
************
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - E T P K KA T N GNSS I
A P RRE TA QRE ME RRV V S QL CS S L DE L V L P P RE K P L K DQL T F GD DGS V A I
A P K RE K T HGE V E RRI V S QL L T L MDGL - - - - K QR- - - - - - - - - - - - - - - A P K RDK T NGE V E RRV V S QL L T L MDGM- - - - K A R- - - - - - - - - - - - - - - GGNRQWA S K D ME RRI V S QL I S S L DNL - - - - - - - - - - K A NE F GQ - - - - - -
I
-
●●●●●●
L RP GR L DK I L FV D FP S V E DRV DI
L RP GR L DQL I Y I P L P D E K S RV A I
L RP GR L DQL I Y V P L P D E NA RL S I
L RP GR L D T I L Y V GFP E QS E R TE I
F F DE I DA L V P K RS HG- E - - S S GGA RL V NQL L T E MDGV E GRQK V F L I GA T NRP D
F F DE L DS I A K A RGGNI GDGGGA A DRV I NQI L T E MDGMS TK K NV F I I GA T NRP D
F L DE L DS I A K A RGGS L GDA GGA S DRV V NQL L T E MDGMNA K K NV FV I GA T NRP D
F F DE F DS L CP K RS DG- GDGNN S GT RI V NQL L T E MDGV E E RK GV Y I L A A T NRP D
●●●●●●
************
L RK S TK NGT RP ML GE DI D F HE I A QL P E L A GF T GA DL A A L I HE S S L L A L QA RV L - - - - - - - E NDE S V K GV GM- - - - - - - - - - - - - - - - - - - - L K A N L R- - - K S P V A K DV DL E F L A K MT N- - GFS GA DL TE I CQRA CK L A I RE S I E S E I RR E RE RQT NP S A ME V E - - - - - - - - - - - - E DDP V P E I
L NA Q L R- - - K TP L E P GL E L TA I A K A T Q- - GFS GA DL L Y I V QRA A K Y A I K DS I E A HRQH E A E K E V K V E GE DV E MT DE GA K A E QE P E V DP V P Y I
L K A T TK NGK RP V L A D DV DL DE I A A Q TE - - GY T GA DL A GL V K QA S MFS L RQS L N- - - - - - - NGD T NL DDL CV - - - - - - - - - - - - - - - - - - - - -
R- - H F RE A A S RI RP S V TE A
RRDH FE E A MR FA RRS V S DN
TK E H FA E A MK TA K RS V S DA
RS QH F QE A L QQL RP S V NE Q
681
627
637
810
791
750
772
920
813
806
835
943
C
B
Percent Identity
1
1
2
3
4
2
23.2
3
4
21.1 42.3
69.4 20.7
20.3
1
1
2
3
4
MAC-1
mVCP
757 813
MAC-1
1
12.3%
201
43.9%
427 451
47.6%
698 806
1
8.2%
212
41.7%
436 461
44.3%
708 835
1
18.5%
232
49.1%
474 635
62.7%
884 943
CDC48
Smallminded
AAAAAA
AAAAAA
AAAAAA
AAAAAA
AAAAAA
AAAAAA
AAAAAA
455 507
204
mVCP
CDC48
Smallminded
Fig. 3. MAC-1 is a member of the AAA family of ATPases. (A) Alignment of MAC-1 with mouse VCP (accession number 400712), yeast
CDC48 (accession number 1705679) and Drosophila Smallminded (accession number 1770214). Positions in which a majority of family
members have identical residues are shown by black boxes. Two conserved ATPase domains or CADs are boxed. A set of Walker A (P-loop)
and B signature sequences, and the AAA minimal consensus motif (Swaffield and Purugganan, 1997; Patel and Latterich, 1998) within each
CAD are indicated by underlined stars, dots and solid lines, respectively. (B) Percentage amino acid identity between MAC-1, Smallminded,
VCP and CDC48. (C) Percentage amino acid identity by regions of the proteins. CADs are indicated by gray boxes. The percentage amino acid
identity in the aligned regions is indicated. The number above boxes refers to amino acid residue number.
2026 D. Wu and others
Fig. 4. Interaction of MAC-1 with CED-4 in mammalian cells. (A) Interaction of MAC-1 with wt (wild-type) and mutant CED-4. Human
embryonic kidney 293T cells were transiently co-transfected with 5 µg of plasmids expressing the indicated proteins. In the case of transfection
with one plasmid, cells were co-transfected with empty plasmid so that the total amount of transfected plasmid DNA was always 10 µg.
Cellular lysates were immunoprecipitated with mAb anti-Flag or isotype-matched control (indicated by cont). Immunoprecipitates were
immunoblotted with rabbit anti-Myc or anti-Flag antibody. Expression of wt and mutant CED-4 protein in total lysates is shown in lower
panels. (B) Schematic structure of wt and mutant MAC-1 proteins. Grey areas denote ATPase domains. (C,D) Interaction of CED-4 with wt
MAC-1 (WT), MAC-1-N (N) and MAC-1-C (C). Transfection and immunoprecipitation analysis were performed as indicated in A. Expression
of CED-4 in total lysates is shown in lower part of C.
CED-3 plus CED-4, was inhibited by MAC-1 in a dosedependent manner (Fig. 8B). Finally, co-expression of MAC1 and CED-9 inhibits apoptosis of 293T cells more effectively
than expression of either CED-9 or MAC-1 alone (Fig. 8B).
MAC-1 is essential for larval development in C.
elegans
To study the effect of reducing MAC-1 activity in C. elegans,
we used RNA-mediated interference (Guo and Kemphues,
1995; Fire et al., 1998). This procedure involves injecting RNA
corresponding to a target gene into the germ line of an adult
hermaphrodite, causing the inactivation of that gene in its
progeny. For many targets, this method reproduces the
phenotype of known loss-of-function mutations (Guo and
Kemphues, 1995; Lin et al., 1995; Hunter and Kenyon, 1996;
Rocheleau et al., 1997). Because double-stranded RNA is far
more effective than antisense RNA (Fire et al., 1998), we used
double-stranded RNA in most of our experiments. We refer to
the progeny of animals injected with double-stranded RNA
complementary to mac-1 as mac-1(dsRNAi) animals.
We first observed mac-1(dsRNAi) animals produced using
construct A (Fig. 9A). These animals complete embryogenesis
and hatch, but most stop growing early in the second stage of
larval development, and a few arrest later, in the L3 stage; all
of these animals continue to move normally (Fig. 9B). Most of
the mac-1(dsRNAi) animals remain arrested indefinitely, but
some slowly continue development and eventually reach
adulthood. In addition to this arrest, we observed other defects
in mac-1(dsRNAi) animals. Even in the presence of food, the
pharynges of young individuals did not pump normally. After
several days, the intestines of most animals had large numbers
of vacuoles, and some sections of the intestine were almost
completely degraded (Fig. 9C). Finally, among the mac1(dsRNAi) animals that eventually reached adulthood, we
observed defects in vulval and gonadal morphogenesis. These
defects included the formation of a protruding vulva, and
aberrant uterine structure.
To confirm that these results were caused by inactivation of
mac-1, and not of another nematode gene containing the CAD
domain characteristic of AAA proteins, we repeated these
injections using constructs B and C, each of which corresponds
to a region of the transcript unique to mac-1. The mac1(dsRNAi) animals produced using constructs B and C
resemble those produced using construct A, although the
severity of the effect was slightly weaker (Fig. 9B). To
determine if this phenotype required injection of double-
MAC-1 is essential and can interact with CED-4 2027
stranded RNA, we also used antisense RNA from construct B
in a series of control experiments; these mac-1(RNAi) animals
showed the same phenotype of larval arrest (data not shown).
Finally, we used a construct corresponding to the unc-4
transcript in a series of control injections. Among the
unc-4(dsRNAi) animals, 51% were unable to move backwards,
and 39% backed poorly, but none showed a developmental
arrest (n=57). This inability to move backwards is
characteristic of unc-4(lf) mutants (Miller et al., 1992; White
et al., 1992), so our results support the conclusion of Fire et al.
(1998) that injection of double-stranded RNA can specifically
reproduce the phenotype caused by null mutations in a gene.
The larval arrest of mac-1(dsRNAi) animals is not
suppressed by ced-3 or ced-4
To determine if this larval arrest was caused by inappropriate
activation of the programmed cell death pathway, we carried
out a series of injections using ced-3(n718) and ced-4(n1416)
mutants. Because all of our dsRNA constructs resulted in the
same phenotype, we adopted fragment D for these
experiments, since it is co-extensive with B and C, and does
not include sequences homologous to other nematode AAA
messages. We found that inactivation of mac-1 caused the ced3 and ced-4 animals to arrest at roughly the same point in
development as the wild-type. At 4 days after fertilization, 90%
of the N2 mac-1(dsRNAi) animals were still L2 or L3 larvae
(n=73), whereas untreated N2 animals were mature adults.
Similarly, 24% of the ced-3; mac-1(dsRNAi) animals were L1
larvae, and 73% were L2 or L3 animals (n=143). Finally, 29%
of the ced-4; mac-1(dsRNAi) individuals were L1 larvae, and
68% were L2 or L3 animals (n=66). That some of the ced-3
and ced-4 individuals arrest at an earlier age than their wildtype counterparts is probably due to the fact that untreated ced3 and ced-4 animals occasionally arrest early in development
(Ellis and Horvitz, 1986; Hengartner et al., 1992).
DISCUSSION
We identified MAC-1 through a yeast two-hybrid screen for
Fig. 5. MAC-1 associates with CED-3, CED-9 and Apaf-1 in mammalian cells. (A,B) 293T cells were transiently co-transfected with 5 µg of
plasmids expressing the indicated proteins. Cellular lysates were immunoprecipitated with mAb anti-Flag (A), anti-HA (B) or isotype-matched
control mAb (indicated by C). Immunoprecipitates were immunoblotted with rabbit anti-HA, anti-Myc or anti-Flag antibody (A) or anti-Flag
(B). Expression of MAC-1, CED-3, CED-4 and CED-9 in total lysates is shown in the lower part of the panels. (C) MAC-1 associates with
Apaf-1. 293T cells were transiently co-transfected with 5 µg of plasmids expressing Flag-tagged MAC-1 and Myc-tagged Apaf-1. Cellular
lysates were immunoprecipitated with mAb anti-Flag and immunoblotted with rabbit anti-Myc or anti-Flag antibodies. Expression of Apaf-1 in
total lysates is shown in the lower part of the panel. Non-specific bands are indicated by a star.
2028 D. Wu and others
CED-4-interacting proteins. We used this approach because we
suspected that genes with lethal or pleiotropic phenotypes
might not have been identified in genetic screens for cell death
mutants. We have shown that MAC-1 is essential for larval
development in C. elegans, can bind to CED-4 in both yeast
and mammalian cells, and can regulate programmed cell death
in nematodes and mammalian cells.
1997). Instead, MAC-1 most resembles the Smallminded
protein of Drosophila (Figs. 3B,C; Long et al., 1998). This
similarity is particularly significant in the N-terminal 200
amino acids, since this domain differs greatly in most AAA
family members. Strikingly, smallminded mutant flies arrest
development as second instar larvae, and mac-1(dsRNAi)
worms arrest as L2 larvae. This result suggests that mac-1 and
smallminded identify a new subfamily of AAA proteins that
regulate growth or development in young larvae.
MAC-1, a novel member of the AAA family of
ATPases
What is the function of MAC-1 in larval
The predicted amino acid sequence of MAC-1 suggests that it
development?
is a member of the AAA family of ATPases. To date, more than
100 members of the AAA family have been identified in
Two experiments shed light on the normal function of mac-1
eukaryotes, eubacteria and archaebacteria (Swaffield and
in nematodes. First, elevated (or ectopic) expression of mac-1
Purugganan, 1997). Members of this family have one or two
can prevent programmed cell deaths, which suggests that
copies of a highly conserved region of 230-250 amino acids,
MAC-1 might help regulate cell death. Second, inactivation of
termed the conserved ATPase domain or CAD; this domain
mac-1 by RNA-mediated interference causes animals to arrest
includes the AAA minimal consensus motif and a set of Walker
as L2 larvae. Although most worms arrest permanently, some
A and B signature sequences characteristic of ATPases
eventually complete development, perhaps because the double(Swaffield and Purugganan, 1997; Patel and Latterich, 1998).
stranded RNA that inactivates mac-1 is degraded. These mature
MAC-1 contains two such CAD
domains. Members of the AAA
family participate in a wide
variety of cellular functions,
including control of the cell
cycle, vesicular transport during
protein secretion, biogenesis
of organelles, formation of
protein complexes, regulation
of transcription, mitochondrial
membrane-bound proteolysis,
and proteosomal regulation
(Confalonieri and Duguet,
1995;
Swaffield
and
Purugganan, 1997; Patel and
Latterich, 1998). The precise
biochemical functions of these
ATPases are not understood,
but it has been proposed
that
they
modulate
the
folding, transfer, assembly or
degradation
of
protein
complexes in an ATP-dependent
manner (Confalonieri and
Duguet, 1995).
Mammalian VCP and its
yeast and Xenopus orthologs,
CDC48 and p97, three AAA
family members that share a
high level of amino acid and
structural similarity with MAC1, participate in organelle
membrane-fusion
events
(Confalonieri and Duguet,
1995). However, it is unlikely
that MAC-1 is the C. elegans Fig. 6. Genomic organization and expression of mac-1 in C. elegans. (A) Genetic map position of the
ortholog of VCP/CDC48/p97, mac-1 locus. Genes shown in gray have been positioned only by their location on the physical map, and
since two nematode AAA not by genetic recombination. (B) The position of the mac-1 locus on the YACs Y48C3 and Y81G3 is
members, still uncharacterized, shown, based on preliminary data from the Genomic Sequencing Project (Sulston et al., 1992). (C) The
exhibit even greater homology structure of the mac-1 locus. (D) Expression of mac-1 in wild-type and mutant C. elegans. The femwith this subfamily (Beyer, 3(gf) animals produce only sperm, and the fem-1(lf) animals produce only oocytes.
vector
4
8
4
2
0
2
6
10
12
Number
of Animals
hsp16-2::mac-1
4
2
0
2
4
6
8 10 12
Extra cells in anterior pharynx
14
Fig. 7. MAC-1 promotes cell survival in C. elegans. Histograms
show the number of cells in the anterior pharynx that failed to
undergo programmed deaths. This value is 0 in wild-type animals,
5.3 in ced-9(n1950)/+ controls, and 10.0 in ced-9(n1950)/+; hsp162::mac-1 animals. The bars indicate the mean, ± one standard
deviation.
animals show additional defects, such as the formation of
vacuoles in the intestine, and abnormal vulval and gonadal
development. We do not know if these problems reflect a
widespread requirement for mac-1 in larval development, or
are a side effect of slowed development. However, animals that
fail to develop because of starvation have not been seen to
exhibit these problems (L. Avery, personal communication; our
unpublished observations).
Our results show that the requirement for mac-1 during
larval development is unlikely to involve the regulation of
programmed cell death, since mutations in ced-3 and ced-4 do
not suppress the arrest of mac-1(dsRNAi) animals. Since the
screen that identified mac-1 was capable of finding proteins
that are essential, or which have pleiotropic effects on
development, it should not be surprising that MAC-1 appears
to have more than one function, and that its major function
might be unrelated to programmed cell death. In this respect,
several members of the AAA family of ATPases, including
mammalian VCP, are known to be involved in diverse cellular
functions, including cycle regulation (Madeo et al., 1997b),
endocytosis (Pleasure et al., 1993), membrane transport
(Rabouille et al., 1995) and proteasome regulation (Dai et al.,
1998).
What is the function of MAC-1 in programmed cell
death?
Although it remains possible that MAC-1 does not bind and
regulate CED-4 during normal development, our results show
that it is capable of controlling programmed cell death in
nematodes and mammalian cells. Four models can account for
these observations. First, mac-1 might help protect all cells that
should survive from undergoing programmed deaths. The fact
that increased expression of MAC-1 prevents programmed cell
deaths from occurring, both in mammalian cells and in
transgenic nematodes, suggests that mac-1 could protect cells
from inappropriate activation of CED-4 and CED-3. However,
mac-1(dsRNAi) animals do not resemble ced-9(lf) mutants,
which die early in development from ectopic cell deaths
(Hengartner et al., 1992). Furthermore, mutations in either ced3 or ced-4 suppress the lethality caused by mutations in ced-9
(Hengartner et al., 1992), but not that caused by inactivation of
mac-1. This result shows that mac-1(dsRNAi) worms do not die
because of inappropriate activation of the cell death pathway.
Thus, if MAC-1 plays a general role in preventing programmed
cell deaths, it is likely to act as an accessory to CED-9, rather
than being essential in its own right. The fact that expression
of MAC-1 shows a greater protective effect if CED-9 activity
has also been increased is consistent with this hypothesis.
Genetic studies suggest that the ced-8 gene might also be a
non-essential partner in the cell death process (reviewed by
Hengartner, 1997).
A second possibility is that mac-1 regulates the survival or
A
70
60
% Apoptotic Cells
hsp16-2
50
40
30
20
10
0
+
-
Vector
CED-3
CED-4
CED-9
MAC-1
B
% Apoptotic Cells
Number
of Animals
MAC-1 is essential and can interact with CED-4 2029
+
-
+
-
-
-
WT N
C
+
-
- + +
- + - WT
- + +
- - N
C
+
+
-
+
+
+
-
+
+
-
- + +
+ +
- -
WT N
C
80
60
40
20
0
CED-3
CED-4
CED-9
MAC-1
+
+
-
+ + + +
+ + + +
- - - 0.1 0.2 0.5 1.0
+ + +
+ + +
+ + +
+ + +
0.1 0.2 0.5
0.1 0.1 0.1
0.1 0.2 0.5
-
-
-
Fig. 8. MAC-1 inhibits apoptosis induced by CED-3 and CED-4 in
293T cells. (A) 293T cells were transiently transfected with plasmids
producing the indicated proteins (see Materials and methods), and a
reporter plasmid expressing β-gal (Wu et al., 1997a). 16-20 hours
after transfection, cells were stained with X-gal and examined by
light microscopy. The results represent the percentage of blue cells
that exhibit morphologic features of apoptosis (Wu et al., 1997a) and
are given as the mean ± s.d. of triplicate cultures. WT, wild-type
Mac-1; N, Mac-1-N; C, Mac-1-C (see Fig. 4). (B) MAC-1 enhances
the protective ability of CED-9 against apoptosis.
2030 D. Wu and others
induce chromatin condensation and cell death, even without
co-expression of CED-3 (James et al., 1997). Similarly, when
CED-4 is expressed in the touch neurons of C. elegans ced-3
mutants, some of these cells die (Shaham and Horvitz, 1996),
which is consistent with this model. If CED-4 is able to induce
some deaths in the absence of CED-3, perhaps MAC-1
regulates this activity.
Finally, mac-1 might regulate the survival of cells in
response to conditions not found in the laboratory. In
mammals, programmed cell death plays an important role in
the elimination of cells that have been infected with certain
viruses (Tschopp et al., 1998), or that have become damaged
or cancerous (Ellis et al., 1991). We do not know if these
stimuli induce programmed deaths in nematodes, but if so,
mac-1 might regulate this response. Furthermore, since MAC1 is an essential protein, one exciting possibility is that MAC1 induces death if normal cell growth or differentiation have
been seriously perturbed. One observation supports this
possibility – a missense mutation of the gene CDC48, a
homologue of mac-1 that is required for cell division in yeast,
can induce some features typically found during programmed
cell death, such as membrane staining with annexin V, DNA
fragmentation, and chromatin condensation (Madeo et al.,
1997a). Perhaps MAC-1 regulates CED-4 activity to allow
complete induction of the cell death program in analogous
situations during C. elegans development.
Because these results have broad implications for how cell
death is controlled during normal development, and how
essential genes might influence this process, we have initiated
a search for mutations that inactivate mac-1. Such mutations
would allow more detailed study of its function and behavior
through genetic mosaic analysis or study of mac-1 transgenes.
Fig. 9. RNA-mediated inactivation of mac-1 arrests development of
C. elegans. (A) The location of the three mac-1 fragments used to
produce double stranded or antisense RNAs. Exon boundaries are
shown as black lines, and the regions encoding the CAD domains in
gray. (B) The frequency of larval arrest among mac-1(RNAi) and
unc-4(RNAi) animals. Animals scored as arrested were still young
larvae 6 days after fertilization. mac-1 A, n=85, mac-1 B, n=37, mac1 C, n=42, unc-4, n=43. (C) Nomarski photomicrograph of the
posterior half of a mac-1(dsRNAi) individual, age 9 days. The arrows
indicate vacuoles that have formed in part of the intestine. Anterior is
to the left, dorsal is up.
death of a small number of cells during development. Because
of the defects mac-1(dsRNAi) animals show in feeding, we
examined the anterior pharynges of these animals, but did not
observe any abnormal deaths, or cells that had failed to die
(data not shown). We also examined the heads and ventral
cords of ced-1(e1735); mac-1(dsRNAi) larvae to determine if
extra cell deaths or inappropriate survivals had occurred in
these areas, but observed no differences from the ced-1 controls
(data not shown). However, we could not accurately study cell
deaths that normally occur after the developmental arrest in
mac-1(dsRNAi) worms, so it remains possible that MAC-1
regulates cell death during late larval development or
adulthood.
A third possibility is that MAC-1 regulates the ability of
CED-4 to kill cells by a mechanism that does not require CED3. We know that expression of CED-4 in fission yeast can
We are grateful to M. Benedict, D. Ekhterae, N. Inohara, M.
Gonzalez-Garcia, T. Koseki, and L. del Peso for critical review of the
manuscript. We thank A. Fire, R. Horvitz and X. Wang for generous
supply of plasmids. This research was supported in part by grant CA64556 from the National Institutes of Health. G. N. was supported by
a Research Career Development Award K04 CA64421-01 from the
NIH. R. E. and P. C. were supported by grant RPG-97-172-01-DDC
from the American Cancer Society. Y. H. is supported by Posdoctoral
Training grant 2T32HL07517 from the NIH.
REFERENCES
Barton, M. K., Schedl, T. B. and Kimble, J. (1987). Gain-of-function
mutations of fem-3, a sex-determination gene in Caenorhabditis elegans.
Genetics 115, 107-119.
Beyer, A. (1997). Sequence analysis of the AAA protein family. Protein
Science 6, 2043-58.
Brenner, S. (1974). The genetics of Caenorhabditis elegans. Genetics 77, 7194.
Candido, E. P., Jones, D., Dixon, D. K., Graham, R. W., Russnak, R. H.
and Kay, R. J. (1989). Structure, organization, and expression of the 16kDa heat shock gene family of Caenorhabditis elegans. Genome 31, 690697.
Chinnaiyan, A. M., O’Rourke, K., Lane, B. R. and Dixit, V. M. (1997).
Interaction of CED-4 with CED-3 and CED-9: a molecular framework for
cell death. Science 275, 1122-1126.
Cohen, G. M. (1997). Caspases: the executioners of apoptosis. Biochemical
Journal 326, 1-16.
Confalonieri, F. and Duguet, M. (1995). A 200-amino acid ATPase module
in search of a basic function. Bioessays 17, 639-650.
Conradt, B. and Horvitz, H. R. (1998). The C. elegans protein EGL-1 is
MAC-1 is essential and can interact with CED-4 2031
required for programmed cell death and interacts with the Bcl-2-like protein
CED-9. Cell 93, 519-529.
Dai, R. M., Chen, E., Longo, D. L., Gorbea, C. M. and Li, C. C. (1998).
Involvement of valosin-containing protein, an ATPase Co-purified with
IκBα and 26 S proteasome, in ubiquitin-proteasome-mediated degradation
of IκBα. J. Biol. Chem. 273, 3562-3573.
Egerton, M., Ashe, O. R., Chen, D., Druker, B. J., Burgess, W. H. and
Samelson, L. E. (1992). VCP, the mammalian homolog of cdc48, is tyrosine
phosphorylated in response to T cell antigen receptor activation. EMBO J.
11, 3533-3540.
Ellis, H. M. and Horvitz, H. R. (1986). Genetic control of programmed cell
death in the nematode C. elegans. Cell 44, 817-829.
Ellis, R. E. and Horvitz, H. R. (1991). Two C. elegans genes control the
programmed deaths of specific cells in the pharynx. Development 112, 591603.
Ellis, R. E., Yuan, J. Y. and Horvitz, H. R. (1991). Mechanisms and functions
of cell death. Ann. Rev. Cell Biol. 7, 663-698.
Fire, A. (1986). Integrative transformation of Caenorhabditis elegans. EMBO
J. 5, 2673-2680.
Fire, A., Xu, S., Montgomery, M. K., Kostas, S. A., Driver, S. E. and Mello,
C. C. (1998). Potent and specific genetic interference by double-stranded
RNA in Caenorhabditis elegans. Nature 391, 806-811.
Fröhlich, K. U., Fries, H. W., Rudiger, M., Erdmann, R., Botstein, D. and
Mecke, D. (1991). Yeast cell cycle protein CDC48p shows full-length
homology to the mammalian protein VCP and is a member of a protein
family involved in secretion, peroxisome formation, and gene expression. J.
Cell Biol. 114, 443-453.
Guo, S. and Kemphues, K. J. (1995). par-1, a gene required for establishing
polarity in C. elegans embryos, encodes a putative Ser/Thr kinase that is
asymmetrically distributed. Cell 81, 611-620.
Hengartner, M. O. (1997). Cell death. In C. elegans II, (ed,. D. L. Riddle, T.
Blumenthal, B. J. Meyer and J. R. Priess), pp. 383-416. Plainview NY: Cold
Spring Harbor Laboratory Press.
Hengartner, M. O., Ellis, R. E. and Horvitz, H. R. (1992). Caenorhabditis
elegans gene ced-9 protects cells from programmed cell death. Nature 356,
494-499.
Hengartner, M. O. and Horvitz, H. R. (1994). C. elegans cell survival gene
ced-9 encodes a functional homolog of the mammalian proto-oncogene bcl2. Cell 76, 665-676.
Hodgkin, J., Horvitz, H. R. and Brenner, S. (1979). Nondisjunction mutants
of the nematode Caenorhabditis elegans. Genetics 91, 67-94.
Hofmann, K., Bucher, P. and Tschopp, J. (1997). The CARD domain: a new
apoptotic signalling motif. Trends Biochem. Sci. 22, 155-156.
Hu, Y., Benedict, M. A., Wu, D., Inohara, N. and Nunez, G. (1998). BclXL interacts with Apaf-1 and inhibits Apaf-1-dependent caspase-9
activation. Proc. Natl. Acad. Sci. USA 95, 4386-4391.
Hunter, C. P. and Kenyon, C. (1996). Spatial and temporal controls target
pal-1 blastomere-specification activity to a single blastomere lineage in C.
elegans embryos. Cell 87, 217-226.
James, C., Gschmeissner, S., Fraser, A. and Evan, G. I. (1997). CED-4
induces chromatin condensation in Schizosaccharomyces pombe and is
inhibited by direct physical association with CED-9. Curr. Biol. 7, 246-52.
Kohara, Y. (1996). [Large scale analysis of C. elegans cDNA]. Tanpakushitsu
Kakusan Koso – Protein, Nucleic Acid, Enzyme 41, 715-720.
Krause, M. and Hirsh, D. (1987). A trans-spliced leader sequence on actin
mRNA in C. elegans. Cell 49, 753-761.
Li, P., Nijhawan, D., Budihardjo, I., Srinivasula, S. M., Ahmad, M.,
Alnemri, E. S., and Wang, X. (1997). Cytochrome c and dATP-dependent
formation of Apaf-1/caspase-9 complex initiates an apoptotic protease
cascade. Cell 91, 479-489.
Lin, R., Thompson, S. and Priess, J. R. (1995). pop-1 encodes an HMG box
protein required for the specification of a mesoderm precursor in early C.
elegans embryos. Cell 83, 599-609.
Long, A. R., Yang, M., Kaiser, K. and Shepherd, D. (1998). Isolation and
characterisation of smallminded, a Drosophila gene encoding a new member
of the Cdc48p/VCP subfamily of AAA proteins. Gene 208, 191-199.
Madeo, F., Fröhlich, E. and Fröhlich, K. U. (1997a). A yeast mutant showing
diagnostic markers of early and late apoptosis. J. Cell Biol. 139, 729-734.
Madeo, F., Schlauer, J. and Fröhlich, K. U. (1997b). Identification of the
regions of porcine VCP preventing its function in Saccharomyces cerevisiae.
Gene 204, 145-151.
Mello, C. C., Kramer, J. M., Stinchcomb, D. and Ambros, V. (1991).
Efficient gene transfer in C.elegans: extrachromosomal maintenance and
integration of transforming sequences. EMBO J. 10, 3959-3970.
Metzstein, M. M., Hengartner, M. O., Tsung, N., Ellis, R. E. and Horvitz,
H. R. (1996). Transcriptional regulator of programmed cell death encoded
by Caenorhabditis elegans gene ces-2. Nature 382, 545-547.
Miller, D. M., Shen, M. M., Shamu, C. E., Burglin, T. R., Ruvkun, G.,
Dubois, M. L., Ghee, M. and Wilson, L. (1992). C. elegans unc-4 gene
encodes a homeodomain protein that determines the pattern of synaptic
input to specific motor neurons. Nature 355, 841-845.
Miura, M., Zhu, H., Rotello, R., Hartwieg, E. A. and Yuan, J. (1993).
Induction of apoptosis in fibroblasts by IL-1 beta-converting enzyme, a
mammalian homolog of the C. elegans cell death gene ced-3. Cell 75, 653660.
Pan, G., O’Rourke, K. and Dixit, V. M. (1998). Caspase-9, Bcl-XL, and
Apaf-1 form a ternary complex. J. Biol. Chem. 273, 5841-5845.
Patel, S. and Latterich, M. (1998). The AAA team: related ATPases with
diverse functions. Trends Cell Biol. 18, 65-71.
Pleasure, I. T., Black, M. M. and Keen, J. H. (1993). Valosin-containing
protein, VCP, is a ubiquitous clathrin-binding protein. Nature 365, 459462.
Rabouille, C., Levine, T. P., Peters, J. M. and Warren, G. (1995). An NSFlike ATPase, p97, and NSF mediate cisternal regrowth from mitotic Golgi
fragments. Cell 82, 905-914.
Rocheleau, C. E., Downs, W. D., Lin, R., Wittmann, C., Bei, Y., Cha, Y.
H., Ali, M., Priess, J. R. and Mello, C. C. (1997). Wnt signaling and an
APC-related gene specify endoderm in early C. elegans embryos. Cell 90,
707-716.
Shaham, S., and Horvitz, H. R. (1996). Developing Caenorhabditis elegans
neurons may contain both cell-death protective and killer activities. Genes
Dev. 10, 578-591.
Spector, M. S., Desnoyers, S., Hoeppner, D. J. and Hengartner, M. O.
(1997). Interaction between the C. elegans cell-death regulators CED-9 and
CED-4. Nature 385, 653-656.
Sulston, J. E. and Brenner, S. (1974). The DNA of Caenorhabditis elegans.
Genetics 77, 95-104.
Sulston, J. E. and Horvitz, H. R. (1977). Post-embryonic cell lineages of the
nematode, Caenorhabditis elegans. Dev. Biol. 56, 110-156.
Sulston, J., Du, Z., Thomas, K., Wilson, R., Hillier, L., Staden, R.,
Halloran, N., Green, P., Thierry-Mieg, J., Qiu, L. and et al. (1992). The
C. elegans genome sequencing project: a beginning. Nature 356, 37-41.
Sulston, J. E., Schierenberg, E., White, J. G. and Thomson, J. N. (1983).
The embryonic cell lineage of the nematode Caenorhabditis elegans. Dev.
Biol. 100, 64-119.
Swaffield, J. C. and Purugganan, M. D. (1997). The evolution of the
conserved ATPase domain (CAD): reconstructing the history of an ancient
protein module. J. Mol. Evol. 45, 549-563.
Trent, C., Tsung, N. and Horvitz, H. R. (1983). Egg-laying defective mutants
of the nematode Caenorhabditis elegans. Genetics 104, 619-647.
Tschopp, J., Thome, M., Hofmann, K. and Meinl, E. (1998). The fight of
viruses against apoptosis. Curr. Opin. Genet. Dev. 8, 82-87.
White, J. G., Southgate, E. and Thomson, J. N. (1992). Mutations in the
Caenorhabditis elegans unc-4 gene alter the synaptic input to ventral cord
motor neurons. Nature 355, 838-841.
Wu, D., Wallen, H. D., Inohara, N. and Nunez, G. (1997a). Interaction and
regulation of the Caenorhabditis elegans death protease CED-3 by CED-4
and CED-9. J. Biol. Chem. 272, 21449-21454.
Wu, D., Wallen, H. D. and Nunez, G. (1997b). Interaction and regulation of
subcellular localization of CED-4 by CED-9. Science 275, 1126-1129.
Yuan, J. and Horvitz, H. R. (1992). The Caenorhabditis elegans cell death
gene ced-4 encodes a novel protein and is expressed during the period of
extensive programmed cell death. Development 116, 309-320.
Yuan, J., Shaham, S., Ledoux, S., Ellis, H. M. and Horvitz, H. R. (1993).
The C. elegans cell death gene ced-3 encodes a protein similar to
mammalian interleukin-1 beta-converting enzyme. Cell 75, 641-652.
Yuan, J. Y. and Horvitz, H. R. (1990). The Caenorhabditis elegans genes
ced-3 and ced-4 act cell autonomously to cause programmed cell death. Dev.
Biol. 138, 33-41.
Zou, H., Henzel, W. J., Liu, X., Lutschg, A. and Wang, X. (1997). Apaf-1,
a human protein homologous to C. elegans CED-4, participates in
cytochrome c-dependent activation of caspase-3. Cell 90, 405-413.
`