L M.,. Abdailah, B., Hassan, A., Levi, G., de Luze, A., Reed,. J. C., Demeneix, B. A. Apoptosis in Xenopus tadpole tail muscles involves Bax-dependent pathways.
Apoptosis in Xenopus tadpole dependent pathways
tail muscles
LAURENT M. SACHS,* BASSIMA ABDALLAH,* AHMED AMAURY DE LUZE,* JOHN C. REED,t AND BARBARA *Laboratoire de Physiologie G#{233}n#{233}rale Ct Compar#{233}e, Museum CNRS 90, 75231 Paris, cedex 5, France; tLabtratoiy tThe Burnham Institute, La Jolla, California 92037 Apoptosis is a fundamental mediaimplicated in normal development. One of the most spectacular developmental events involving apoptosis is tail regression during amphibian metamorphosis. We analyzed how thyroid hormone (3,5, 3’-triiodothyronine, ‘F3), the orchestrator of metamorphosis, affects expression and function of the proapoptotic gene Bax in the tail muscle of free-living Xenopus tadpoles. During natural metamorphosis Bax mRNA was expressed in tail muscles and was spatially correlated with apoptosis. Precocious treatment of tadpoles with T3 induced Bax expression and apoptosis. To verify that Bax expression was causally related to apoptosis we used a naked DNA gene transfer method to express Bax in the dorsal tail muscle. This induced apoptosis, and the process was exacerbated by T3 treatment. To determine whether T3 effects on Bax expression involved transcriptional regulation, we injected a Bax promoter sequence into dorsal and caudal tail muscles. In the dorsal muscle, ‘F3 treatment did not affect transcription from the Bax promoter. However, in the caudal muscle, T3 treatment significantly increased Bax transcription. We conclude that Trinduced apoptosis in Xenopus tadpole tail muscles involves Bax.activating and Bax-synergistic mechanisms. These programs are induced in spatially and temporally distinct manners.-Sachs, L M., Abdailah, B., Hassan, A., Levi, G., de Luze, A., Reed, J. C., Demeneix, B. A. Apoptosis in Xenopus tadpole ABSTRACT
nism
tail muscles involves Bax-dependent 801-808 (1997)
pathways. FASEB
J. 11,
Key Words: transfer
thyroid honnone
metamorphosis
in vivo gene
IS A HIGHLY CONSERVED, genetically governed response for cells to commit suicide (for a recent review, see ref 1). Regulation of cell death is as critical to development as regulation of cell proliferation; in the mature animal, maintenance of homeostasis implies balancing cell production and cell elimination (2). Deregulation of apoptosis can lead to impaired development and pathologies such as tumorigenesis.
ApOPTOSIS
0892-6638/97/001
1-0801/$02.25
C FASEB
involves
Bax-
HASSAN,* GIOVANNI LEVI, A. DEMENEIX*.’ National d’Histoire Naturelle, URA
of Molecular USA
Biology,
CBA-IST, Genoa,
Italy; and
One of the first systems in which apoptosis was described was the regressing tail of metamorphosing tadpoles (3). Indeed, amphibian metamorphosis, which is orchestrated by thyroid hormone (3, 5, 3’triiodothyronine or T3) ,2 is one of the best-studied hormone-regulated developmental processes (4, 5), and it provides an excellent experimental model for studying apoptosis (6). First, the free-living animal can be easily manipulated; second, the process can be induced (or blocked) by modif’ing the thyroid status of the tadpoles. Moreover, in vivo gene transfer can be applied to the tail muscle tissue to study gene regulation and function (7, 8). Apoptosis proceeds in three stages: triggering, judgment, and execution. Numerous genes play key roles at each stage. Triggers vary widely according to cell type and developmental stage. In the judgment or decision phase, the Bcl-2 family of genes is seen as providing central checkpoints (9). Last, the cysteine proteases or caspases are key players in execution. Our interest focused
on the checkpoint
process and
genes related to the proto-oncogene Bc12. Proteins encoded by these genes can act as repressors (Bcl-2, Bcl-xL, Mci-i) or effectors (Bax, Bak, Bad, Bcl-xs) of apoptosis (10). Each protein has a highly conserved dimerization domain that enables the proteins to interact by forming homodimers or heterodimers, or both. Current thinking holds that, since in most experimental situations Bcl-2 protects cells and Bax induces apoptosis, then Bcl-2 and other anti-apoptotic members of the family compete for the binding of Bax in order to inactivate it (10). Bax has been described as an effector of apoptosis in many cellular systems (ii). In keeping with this hypothesis, Bax null mice display lymphoid hyperplasia (12). However, little in vivo data are available about the involvement of the protein in apoptosis in ‘Correspondence: Laboratoire de Physiologie G#{233}n#{233}rale et Compar#{233}e, Museum National d’Histoire Naturelle, URA CNRS 90, 7, rue Cuvier, 75231 Paris, cedex 5, France Abbreviations: T,, 3,5, 3’-triiodothyronine; GFP, green fluorescent protein; ISH, in situ hybridization; p.i., postinjection; ECM,
extracellular
matrix;
deoxytransferase-mediated
TR, T5 receptor;
dUTP-biotin
TUNEL,
nickend
terminal
labeling.
801
other vertebrate systems. We chose to follow endogenous Bax expression and then examine the consequences of expressing Bax in the tail muscle of premetamorphic tadpoles. We found that it did induce apoptosis and that the apoptotic effect was exacerbated by T3. We conclude that 1) Bax expression is induced in the tail during metamorphosis, 2) this induction is correlated with apoptosis, and 3) T3 induces apoptosis by activating Bax-dependent pathways.
MATERIALS
AND
METHODS
pure
DNA
(between
1.1 and
2.6 jig)
in 0.07
M NaCI
colored
with fast green (Sigma, St Quentin Fallavier, France). For studies involving quantification of promoter activity, transfection efficiency was normalized using a constituitive construct,
pcDNA3-LUC. Reporter
gene assays
Luciferase
and CAT assays were carried out according to de Luze et al. (7). In order to follow GFP expression, the live tadpole, anesthetized in MS222 (0.1%, Sandoz, Basel, Switzerland), was placed directly under an Olympus fluorescent microscope illuminated with a 395 nm ultraviolet excitation beam, and emission was observed at 509 nm.
Anhnak In situ hybridization Xenopus laevis tadpoles were raised and maintained as previously described (7) and staged according to Nieuwkoop and Faber (13). At stages 50-51, before prometamorphosis, they
were transferred to 0.1% reactifs, Nanterre, France) hybridization and TUNEL ated dUTP-biotin nickend obtained from the Service National de Ia Recherche Metamorphosis was induced Quentin Fallavier, France) changed daily. Plasmids
and
in situ
sodium
perchlorate
to block
development.
(Carlo Erba For in situ
(terminal deoxytransferase-medilabeling), we used albino tadpoles d’Elevage de X#{233}nope du Centre Scientifique (Rennes, France). by adding 10 nM T, (Sigma, St. to the aquarium water that was
hydridization
(ISH) probes
Plasmids were propagated and purified by standard CsC1 techniques (14). pcDNA3-LUG (kindly provided by Dr. M. Schleef, Quiagen), the firefly luciferase activity, is driven by the cytomegalovirus (CMV) promoter of pcDNA3 (Invitrogen, San Diego, Calif.). PcDNA3-Bax was constructed by subcloning the EcoRI-EcoRI fragment of mouse Bax cDNA in the EcoRI site of pcDNAS. The mouse Bax cDNA (70Z7) (15) was a gift from Dr. S. J. Korsmeyer (Howard Hughes Medical Institute, St. Louis, Mo.). PcDNA3-GFP was constructed by subcloning the EcoRI-XhoI fragment of pRSET B-S65T containing the S65T green fluorescent protein (GFP) mutant cDNA (16) in the EcoRl-XhoI sites of pcDNAS. The GFP mutant clone was pro-
vided by Dr. R. Y. Tsien (Howard Hughes Medical Institute, La Jolla, Calif.). pcDNA3-xp53 was obtained by subcloning the EcoRl-EcoRI fragment of Xenopus p53 B2 clone (17) into the EcoRI
site of pcDNA3.
The
pBluescript-xp53
clone
was
a gift
from Dr. M. Mechali (InstitutJacques Monod, Paris, France). pSV4OxTF43: The full-length cDNA of Xenapus TRI3 was cloned into the pTL1 expression vector (18). This plasmid was kindly provided by Dr. R. Tata (Medical Research Council, London, U.K.). pTM667-3: The 371 bp fragment of the human bax promoter between -318 bp and -687 bp was subcloned into the Hindlil site of pUCSVOCAT (19). pBluescript-Bar.The first 502 bp of the fragment EcoRI-BamHI of mouse BaxcDNA was subcloned into the EcoRI-BamHI site of pBluescript (SK+). This construction provides an antisense probe with T3 RNA polymera.se and a sense probe with T7 RNA polymerase.
In situ hybridization (ISH) probes were synthetized using T3 (for Bax antisense probe) and T7 (for Bax sense probe) polymerase (Boehringer, Mannheim, Germany), as previously described (8). Before the last step of purification by ammonium acetate, the probe was fragmented by alkaline lysis in bicarbonate buffer (NaHCO, 0.2 M, Na2CO, 0.2 M, pH 10). Localization of Bax mRNA transcripts was established on longitudinal cryostat sections by in situ hybridization, as previously described by Whitfield and co-workers (20). Slides were dipped in NTB2 photographic emulsion (Kodak), left for 2 wk, and then developed. After dehydration, slides were stained with hematoxylin/eosin, photographed, and the grain density quantified by computerized videomicroscopy using a digital image processor (Optilab program from Graftek running on a Macintosh Power computer). The selected microscopic image was digitalized under optimal conditions of illumination and contrast. All values were corrected for background with sense probe.
TIJNEL
technique
The TUNEL procedure was performed on sections sequential to those used for ISH. A commercial in situ cell death detection kit (1 684 809, Boehnnger-Mannheim) was used. After labeling, the slides were dehydrated, stained with hematoxyun, and mounted. Quantification of cell death was carried out by counting the red (alkaline hosphatase positive) stained
nuclei and normalized unmunoprecipitation
per mm. and Western
blotting
J.
We used rabbit anti-peptide antisera specific for amino acids 43 to 61 of the mouse Bax protein (21). For immunoprecipitation, 10 dorsal muscles cotransfected with 1 jig of pcDNA3LUG and 1 jig of pcDNA3-Bax were sonicated on ice in 1 ml lysis buffer (10 mM Tris, pH7.5,.150 mM NaCl, 5 mM EDTA, 1% Triton X100, 1 mM phenylmethylsulfonyl fluoride, 5 jig/ ml Aprotinin, 2 jig/mi Leupeptin and I jig/mi Pepstanin). Lysates
In vivo gene transfer
gene transfer in Xenapus tadpole dorsal muscle was performed as described previously (7). The same methodology was used for in vivo gene transfer in the caudal muscle. The injection solution (1 p3) contained various amounts of Somatic
802
Vol.11
August 1997
were
treated
according
to Miyashita
and
collaborators
(22) before PAGE. After transferral to a prewet PVDF membrane (BioRad) and incubation (overnight, 4#{176}C) in blocking buffer (Tris HCI 50 mM, pH 7.6, NaCl 200 mM, gelatine 0.25%), it was incubated with anti-bax (1/3000, overnight, 4#{176}C). Antibodies were detected using an ECL kit from Amersham
The FASEB journal
(Arlington
Heights,
Ill.).
SACHS
ET AL.
Figure apoptosis
1. Bax expression are spatially
and corre-
lated in caudal muscle during natural
metamorphosis.
En-
dogenous Bax mRNA expression (A, B) was followed by ISH, using an antisense 32P-labeled mouse Bax probe. Apoptosis (C, D) was followed by a TUNEL technique with alkaline phosphatase-linked streptavidin to reveal incorporated UTP-biotin. A, C) Sequential sections
of the
caudal
from a tadpole
muscle
in premetamor-
phosis at st54 showing bridization signal (A)
no
hy-
and
no
stained apoptotic nuclei (C). B, D) Sequential sections of a typical area of caudal muscle from a tadpole at metamorphic climax (st6l) showing a dense Bax signal (B) and numerous apoptotic nuclei labeled red by alkaline phosphatase (D, short, thick arrows). X3000. Thin arrows indicate normal nuclei stained with hematoxylin. E) Quantification of ISH (open columns) and TUNEL signal (filled columns) in sections of
E
15
.
caudal muscle from tadpoles at different stages of metamor-
#{149}
phosis. The drawing represents a st54 tadpole and the rectangle indicates the area quantified. The levels of endogenous T3 indicated are adapted from Leloup and Buscaglia (24).
Means
o
#{149} 54 Dev. Stage (NF)
61
SEM
are given; n a 6
63
sections
of
65
TUNEL were carried out on sequential tails from tadpoles in premetamorphosis
RESULTS Endogenous Bax expression metamorphosis
is induced
during
natural
We followed endogenous Xenopus Bax expression and its eventual correlation with apoptosis during metamorphosis. Given the high identities between Bax genes of different species (89.4% between mouse and human), we carried out ISH with a probe derived from the mouse Bax gene. The specificity of the signal obtained was verified with a sense probe (see Materials and Methods). Apoptosis was followed by the TUNEL technique (23). Controls included eliminating the terminal transferase or UTP-biotin, or both. ISH probes for Bax mRNA and
T3, BAX, AND
±
in all cases (three two tadpoles).
APOPTOSIS
IN XENOPUS
TADPOLES
sections of (stage 54, or st54), metamorphic climax (st6l), and during maximal tail regression (st63). st6l corresponds to the T3 peak during metamorphosis (Fig. 1E, data from Leloup and Buscaglia (24)). Endogenous Xenopus Bax mRNA and numbers of apoptotic nuclei were lower during premetamorphosis (Fig. 1A, C) than at st6l (Fig. lB. D) or st63 (Fig. 1E). The greatest increase in numbers of apoptotic nuclei per mm2 occurred between stages 61 and 63 (P
