Edmondson,. D. G., and. Olson ... M., Pierce, J. W., and Baltimore,. D. Protein- ... Baltimore,. D. Multiple nuclear factors interact with the immunoglobulin enhancer.
Vol.
2, 619-629,
December
Differential Xenopus1
Jon B. Scales, Department Texas
Cell
1991
Expression
Eric N. Olson,
of Biochemistry
M. D. Anderson
and Michael
and Molecular
Cancer
Center,
Biology,
Houston,
of Two
of
Texas 77030
Abstract We previously
for
reported
the isolation of several Xenopus Iaevis that encode distinct MyoD proteins. Two of these genes, XImfl and X1mf25, appear to represent a gene duplication as a consequence of the polyploid Xenopus genome. Although both MyoD genes are expressed exclusively in skeletal muscle in adult animals, they have very different temporal patterns of expression in early development. In the present work, we show that XImfl complementary
DNAs
to high
levels
1011/2 fibroblasts In addition, both
to a myogenic proteins directly
shortly
induction
in Xenopus
To whom
Biochemistry
Anderson 77030.
supported
by
NIH
grants
to M.
P. and
E. N. 0.
Cancer
Molecular
Center,
Biology,
1515
The
Holcombe
University
Boulevard,
of
hierarchy,
genes. reported
in embryogenesis,
have
referred
at a position
be-
the
identification
of two
sepa-
prior
to
expression
pseudotetraploidy
of Xenopus
polyploid
species
pairs
functionally
are
Iaevis.
have shown
of
the
differen-
similarity of the Xlmfthat observed between which suggests that they genes resulting from the Studies
with
other
that some duplicated
hemizygous,
whereas
gene
both
genes
in other pairs remain active (19). Furthermore, the expression of duplicated genes in some pairs is coordinately regulated, as seen with the Xenopus skeletal actin genes (2), whereas is independently
the expression of genes in other pairs regulated (20, 21, and references show here that the Xenopus MyoD genes
therein). We are similar to the latter class of duplicated genes based on their temporal pattern of expression. X1mf25 mRNA is a maternal transcript present throughout the early stages of development, whereas Xlmfl mRNA was not detectable until after the midblastula transition. Both XImf proteins induced myogenic conversion and transactivated
fected
a muscle-specific
mouse
regulatory
element
in trans-
fibroblasts.
Results XImfl and X1mf25 Have Different Temporal Patterns of Expression during Embryogenesis. Previous workers detected maternal Xenopus MyoD transcripts by Northern analysis using an XImfl probe (termed XMyoD; 22) and nibonuclease
protection
with
a probe
analogous
to
E. N. 0.
Investigator of the American Heart Association. requests for reprints should be addressed, at Department and
in a regulatory
tiated muscle phenotype. The encoded proteins is greater than other pairs of myogenic factors, may be products of duplicated
by
2
sequence,
genes encoding MyoD-like products, designated X!mfl and Xlmf2S (17). These genes, like myogenin and MyoD in mouse embryos (18), are expressed very early
used a host of markers to characterize at the molecular level the extracellular factors and intracellular signals involved in this process. These markers frequently correspond to genes encoding structural proteins expressed in specific cell types, such as the cardiac and skeletal muscle-specific isoforms of the actin gene family (1, 2). Recent studies of the events that regulate differentiation in cultured muscle cells have illustrated the importance of a family of transcriptional regulatory factors for expression of the skeletal muscle phenotype (reviewed in Ref. 3). Genes in this family, including MyoD (4), myogenin (5, 6), myfs (7), and MRF4 (8), are expressed exclusively
Received 5/3i/9i. 1 This work was is an Established
to a consensus
rate
trans-
embryos
binding
muscle-specific We recently
Introduction of mesoderm
in
tween the extracellular signals that initiate mesoderm induction and the expression of previously characterized
activated reporter genes linked to muscle-specific regulatory elements. XImfl was twice as active in this regard as X1mf25 and required a carboxy-terminal domain for its function. The absence of apparent effect of the maternally expressed myogenic gene in early embryos, but not in transfected fibroblasts, suggests the existence of regulatory mechanisms that repress the function of this gene in cells with nonmuscle fates during early amphibian development.
Studies
efficient
to function
genome at the midblastula transition. In contrast, X1mf25 was expressed as a maternal transcript that was maintained at a relatively constant level throughout early development. X1mf25, like XImfl, was capable of converting phenotype.
Genes
619
to as an E-box (CANNIG), found in the regulatory regions of many muscle-specific genes (10, 12, 13) and in immunoglobulin gene enhancers (14, 15). Forced expression of myogenic factors in many nonmuscle cell types results in their conversion to the muscle phenotype (4, 5, 7, 16). The myogenic regulatory gene family appears
from
transcripts rapidly accumulated after activation of the zygotic
MyoD
& Differentiation
in skeletal muscle in adults. The encoded proteins contam a highly conserved region postulated to adopt a HLH3 conformation that functions as an interface for protein-protein interactions (9, 10). The formation of heterodimers with ubiquitously expressed HLH proteins, such as the products of the E2A gene (1 1), is necessary
Perry2 The University
Distinct
Growth
Texas
Houston,
M.
The abbreviations
of
3
D.
complementary
TX
amphenicol dium
dodecyl
used are HLH, DNA;
MCK,
acetyltransferase; sulfate;
bp,
helix-loop-helix;
muscle-specific
MBS, base
pair(s).
modified
nt, nucleotide;
creatine
Barth’s
kinase;
solution;
cDNA,
CAT,
chlor-
SDS, so-
620
Differential
Expression
of Xenopus
MyoD
Genes
LI,
A
oc
LILt)
N
Flours
0)
6t
L4 9
0
-
Lt 10
-
c’,
I
L(
1
!__-
II2
i i
0.0. o
to
0
I 1
3’fi-;it-i#{244}-T-if4 i 131 MJiTsl
293-
232
#{149}1 -p
Hours
.1..
Stage
L
L-OS L-±J0
N
.
6
9
‘
10
C’J
C’
i
Fig.
_____
Accumulation
RNA from Xlmfl-specitic
12T13I’14J27I41kfi
fl11
1.
using
an
protection
analysis
X1mf25-specific
ribo-
probe. Arrow, the resistant fragments
Xlmf25-
-
WWWWWW
w
specific
I
232-s .
w
expected
hybridization
of
embryo
equivalent
of
loaded
in
lane
stages
20 and
an
used.
Transcript
ribonuclease-
for
probes
to homologous transcripts. Lanes are indicated by stages (25) or hours after fertilization as tollows: 0, oocyte; FE, fertilized egg; 6-43, stages; M, adult muscle (5 pg); 5, synthetic RNA (iO or 50 pg, as indicated). A single
of
C
Xlmfl
embryos using an riboprobe. B, ri-
bonuclease 298.-
220-
of
and X1m125 transcripts during early development. A, ribonuclease protection analysis of total
Numbers
scripts
bar,
14-
each
one-half
equivalent
C,
quantitation
in
millions/embryo.
was
except
27, where
embryo
Xlmfi
RNA
was
bar,
IranSolid X1mf25).
fragment
was
; hatched
of
12. (0
2io
x 0. 0
U
Xnf1
0
Xfrnf2S
VS 0
I-
Stage
llours
(termed MyoDa; 23). In our previous studies, we did not detect maternal Xlmftranscripts usinga full-length Xlmfl probe under high-stringency hybridization condi-
X1m125
tions (17). To investigate nibonuclease protection that distinguish between
Fig. 1A shows specific
probe
the result and
total
this in more detail, we used a assay using gene-specific probes XImIl and X1m125 transcripts. of such an assay using the XlmflRNA from oocytes, eggs, staged
embryos,
and
protected
from
adult
leg
nuclease
muscle.
A 230-nt
digestion
after
hybridization
of
this probe with Xlmfl transcripts but not after hybnidization with X1mf25 transcripts. Notably, Xlmfl transcripts were not detected in oocytes, eggs, or preblastula embryos but were first observed at stage 10, approximately
9 h after
fertilization,
tion
initiation
and
shortly of zygotic
after
the midblastula
transcription.
The
transinumber
Cell
of transcripts
present
was quantitated
by comparison
was approximately equal embryo (stage 20).
to that
present
4IIII:III:Iiii:I-----------I#{149}---------9BS XlmfI -
in a late
B
‘
a result
contains
including
the
binding
and
of aberrant
amino
basic
and
splicing
acids HLH
(1 7). The
1 through regions
heterodimerization.
No
transcripts
for
for
competition
of the
longer,
more
of the
abundant
message with the shorter, presumably less abundant XImfl 1 transcripts for hybridization to the riboprobe. Hybridization of the niboprobe to synthetic XImfl 1 RNA was not affected by an excess of Xlmfl RNA. Since approximately iO copies (1 pg) of synthetic XImfll RNA
was
less than iO copies/embryo. Is Myogenic and Activates Endogenous Myogenic Regulatory Genes in Mouse Fibroblasts whereas XImfl 1 Cannot. Constitutive expression of XImfl in stably
X!mf25
transfected mouse fibroblasts resulted in the formation of multinucleated cells expressing endogenous musclespecific genes (17). To investigate the myogenic activity of Xlmfl 1 and X1mf25, we assayed their ability to convert fibroblasts to a myogenic phenotype. Xlm125 and Xlmfl were inserted into an expression vector as described previously for XImfl (1 7) to create peMX25 and peMXl respectively, and these plasmids were stably transfected
C3H1OT1/2
cells. expressed
These
plasmids
neomycin
also resistance
#{149}.
186nt
‘y .q
.
J
J
J
0
(12
O
#{149}.v #{231}a
Sy
4-
Xlmf
11
I,
I
contain gene,
Fig.
2.
Analysis
that stained
were easily detected with this assay, we estimate that the abundance of Xlmfl 1 transcripts, if they were present,
into
.
11
2
3
4
5
6
7
8
9
1011
of
Xlmfl
1 expression.
A,
diagram
of
XImfl
and
X/mfl
1
thereby allowing for selection of transfectants containing stably integrated plasmids in G418-containing medium. Several cell lines expressing stably integrated X1mf25 were converted to a myogenic phenotype after differentiation in mitogen-depleted medium. These cells were characterized by the appearance of multinucleated cells
XImfl
constitutively
4.i CO
XImf Probe 336nt
cDNAs indicating the riboprobe, 9BS, used to distinguish between the two transcripts. Boxed areas, coding regions; heavy line, noncoding sequence. Solid boxes, conserved carboxy-terminal regions. The probe was produced from a cloned BamHI-Saul restriction fragment (B, BamHI; 5, Saul). A 336-nt fragment is protected by Xlmfl transcripts, whereas XImfl 1 RNA protects a 186-nt fragment. B, ribonuclease protection analysis of oocyte and embryo RNA with the 9BS probe. The Xlmfl(upper arross’) and Xlmfl 1-specific(/owerarrow)fragmentsare indicated. Synthetic Xlmul (Lane 8, 10 pg) and Xlmfl 1 (Lane 9, 1 pg) RNAs were used as positive controls. Lane 10 contains a mixture of 10 pg Xlmfl and 1 pg Xlmfl 1 RNA. Lane 1 shows nonspecific fragments arising from incomplete digestion of the probe. The bands visible between those indicated as Xlmul and Xlmfl 1 arise from incomplete digestion products, as does the larger band (396 nt) present in every lane.
DNA
size expected from XImfl 1 were previously detected by Northern analysis. We therefore assayed for the presence ofXlmfl 1 transcripts using nibonuclease protection assays with RNA from oocytes, gastrulae, neurulae, and swimming tadpoles. A niboprobe was used that spans the 3’ breakpoint between the XIm(l and XImfl 1 cDNAs, thereby allowing the specific detection of each transcript (Fig. 2A). Hybridization of XImfl 1 RNA to this probe should protect a 186-nt fragment from subsequent ribonuclease digestion. This fragment was not observed with any of the endogenous RNA samples even though hybridization to Xlmfl transcripts was apparent (Fig. 2B, Lanes 3-7). Synthetic XImfl and Xlmfl 1 RNAs were analyzed separately (Lanes 8 and 9) and together (Lane 10)
to test
0
(5,
I
encoded
178 of Xlmf 1,
necessary
.
0
k
stage 24 cDNA library (17). We previously reported the nucleotide sequence of a third cDNA, Xlmul 1, that is identical in sequence to XImfl except for the deletion of nucleotides 658 through 1009, protein
(
F.’
621
Xlmf
Xlmfll
4) O”vO
V
In contrast to the expression of Xlmfl, X1mf25 transcripts were present in oocytes, in fertilized eggs, and in early cleavage stages (Fig. 1B) well before the detection of Xlmfl transcripts. The number of X1mf25 transcripts was relatively constant (approximately 1 x 106 copies/ embryo) from fertilization through late gastrula (stage 1 1 Y2), increased 2-3-fold by stage 20, and then decreased (Fig. 1C). The relative amount of the Xlmf transcripts between stages 20 and 27 corresponded approximately with the relative abundance of these cDNAs in a
potentially
& Differentiation
A
of
the amount of probe protected in each sample to that protected by predetermined amounts of synthetic sense RNA. An appreciable amount (1 0 copies/embryo) of the Xlmfl mRNA was detected in midgastrula-stage embryos (stages 1 0-1 0Y2). The number of Xlmfl transcripts steadily increased, reaching a maximum of 1.4 x iO copies by stage 20, and then subsequently decreased. The amount ofXImfl mRNA in 5 jzg of cellular RNA from adult skeletal
muscle neurula
Growth
1
1, a
with
a myosin-specific
monoclonal
antibody
(Fig. 3). We obtained no cell lines stably expressing the XImul 1 cDNA that also converted to a myogenic phenotype even after prolonged culture in differentiation medium. Transcripts of the size expected for expression of the exogenous X1mf25 and XImfl 1 genes were present in the corresponding cell lines (Fig. 4). The myogenic conversion of fibroblasts by X1mf25 and apparent lack of conversion by Xlmul 1 was further characterized at the molecular level by examining the comesponding cell lines for expression of the endogenous myogenic regulatory genes MyoD and myogenin. RNA obtained from representative cell lines expressing XImfl (TX.2), X1mf25 (TX25.8), and XImfl 1 (TX.1), cultured under proliferative and differentiation conditions, was subjected to Northern analysis. The forced expression of XImul and X1mf25 was clearly capable of activating expression of both endogenous myogenic factors when
622
Differential
Expression
of Xenopus
MyoD
Genes
fig. .3. Conversion of 1OT1/2 fibroblasts after stable transfection with Xlmf25. A, phase-contrast photomicrograph of difterentiated TX25.8 cells stably expressing Xlmf2S showing the formation of myotubes. B, immunofluorescence photomicrograph of same field as A, showing expression of skeletal muscle-specific genes as demonstrated by staining with an anti-skeletal myosin heavy-chain antibody.
cells were allowed to differentiate (Fig. 4, Lanes 1-4). We did not detect expression of endogenous muscle regulatory genes in proliferating TX2 or TX2.8 cells or in cells expressing Xlmfl 1 after exposure to differentiation medium for up to 8 days (Fig. 4, Lanes 5-7). XImfl 1 contains the regions (basic and HLH) shown to be sufficient for myogenic conversion by MyoD (24); however, it lacks the carboxy-terminal 1 1 0 amino acids of Xlmfl , containing instead 1 1 amino acids that are not present in either Xlmfl or Xlmf25. The present results suggest that the camboxy terminus of Xlmfl is necessary for myogenic conversion of mouse fibroblasts, perhaps reflecting a required interaction of this region of the amphibian protein with the mouse transcriptional apparatus. Studies using fusions between the Gal 4 DNA-binding domain
and
4
regions
1. ShwartL
from
myf5
and E. Olson,
(25) and myogenin4
nianuscript
in preparation.
have
led to
the conclusion that a transcriptional activation domain is present in the carboxy termini of these muscle factors. Interestingly, two conserved regions rich in senine and
threonine and HLH
residues are present downstream of the basic regions in myf5, MyoD, Xlmfl, and X1mf25. The
latter region of homology is also present in myogenin (26) and MRF4 but is absent from XImfll; the absence of this evolutionarily conserved region may contribute to the inactivity of XImfl 1 . We did not independently assay the expression of the Xlmfl 1 protein. It is possible that this protein is susceptible to rapid degradation under
differentiation inactivity. Xlmfl
thereby
contributing
to
its
XImf25 Can Bind and Activate the Murine The previous experiments showed that XImfl and X1mf25 both convert nonmuscle cells to a myogenic phenotype. In each case, myogenic conversion was accompanied by activation of the endogenous myogenic factors MyoD and myogenin. To determine whether XImfl, XImfll, and X1mf25 were capable of
MCK
and
conditions,
Enhancer.
Cell Growth
Tx1_.2
Tx25__8
DG
G
Tx11-3
D
623
C2
00
G
& Differentiation
GD
XLMF
18S>
MoD
18S>
2\..
.
,
,
-.
MYOGENIN
18S>
1
activating
regulatory
the
element
trans-activation
transiently
and
expression
potentials, XImf reporter
well-characterized sensitive
MCK assay
myogenic factors vate cotransfected muscle-specific
of
a muscle-specific
to quantitatively
expressed
of a cotransfected This
5
4
their
we measured the ability of genes to activate expression gene under the control of the
promoter
has been
compare
used
and enhancer to show
that
(27, 28). different
vary remarkably in their ability to actireporter genes under the control of regulatory
elements
(29)
and
to
define
positive control mutations in myogenin that prevent myogenesis without affecting DNA binding (30). In the absence of a cotransfected muscle regulatory factor, the MCK transcriptional regulatory elements were inactive in fibroblasts,
and
and B). Expression cDNAs
6
7
of endogenous myogenic factors in C3H1OT1/2 cells stably expressing medium for 4 days (D) were analyzed for expression of the transfected from cells expressing XlmIl (TX,.2, Lanes 1 and 2), Xlmf2S (TX25.8, Lanes RNA in Lane 7 was isolated from TX,, , cells maintained in differentiation
Fig. 4. Activation in differentiation RNA was isolated (Lanes 8 and 9).
directly
23
resulted
CAT
activity
was
not
of either the XImfl, in the transcriptional
detected X1mf25, activation
(Fig.
5, A
or MyoD of the
89
Xenopus MyoD cDNAs. Proliferating cells (C) and cells placed cDNA (XImf( and endogenous factors (MyoD and myogenin(. 3 and 4), and Xlmfl 1 (TX,,.3, Lanes 5-7), and from C2 myoblasts medium for 8 days.
MCK enhancer as demonstrated by an increase in CAT activity. XImfl reproducibly showed approximately 2-fold greater trans-activation of the reporter gene than either X1m125 or MyoD. Analysis of RNA from parallel transfections showed that XImfl and Xlmf25 transcripts were present at similar levels (data not shown). In contrast, expression of Xlmul 1 did not activate transcription of the
MCK-CAT
reporter
of its inability
gene;
to induce
this was not surprising myogenic
fibroblasts. The expression
of the
ogenin genes was tures by Northern
examined analysis
conversion
endogenous in parallel (Fig. SC).
in light of mouse
MyoD and mytransfected culSince transcripts
from neither gene were detected, we concluded that the amphibian myogenic factors XImfl and X1mf25 most likely
activated
The possibility
the
MCK
that Xlmfl
regulatory
1 might
elements
negatively
directly.
regulate
the
624
Differential
Expression
of Xenopus
MyoD
Genes
ability
A
tested
1.2
of 1.0
0.8
0 V > 0 V
XImf25
to activate
the
MCK
enhan-
for cotransfecting
the
MCK-CAT
reporter
the XImfl 1 expression vector and either XImfl In these tests, XImfl 1 did not affect the ability
either
Xlmfl
enhancer
(data
or
X1mf25
to
trans-activate
the
MCK
not shown).
Xlmfl-E12 and X1mf25-E12 Heterodimers Have Different Abilities to Interact with DNA. The differential trans-activation potentials shown by XImfl and X1mf25 with respect to the MCK enhancer could be based on several properties, including differences in their ability to
U I-
and
as above
gene with or Xlmf25.
>‘
>
of XImfl
cer, by competing either for oligomenization with a cellular partner or for binding to the MCK enhancer, was
0.6
form
0.4
complexes
elements 0.2
in
binding None
XImfl
Xtmfl
1
X1mf25
MyoD
.
properties
of
the
C...
E-box
synthetic
DNA-binding
however,
all three
0
5
1234
6
‘;j’,
less efficiently
the MCK
proteins
complexes
with
capable
of binding
de-
this site (Fig. 6A); this
sequence
and Xlmf25-E12
of the amount
of probe
bound
in the
absence of competitor suggests that Xlmfl-E12 complexes bind more efficiently to the MCK E-box than the corresponding complexes containing X1mf25 (Fig. 6B, Lanes 1 and 5). In the absence of competitor, the amount
of probe
bound
by the complexes
containing
XImfl
was
for
the
observed
differences
in trans-
ence of a vast excess (l000x) of the homologous unlabeled oligonucleotide (Fig. 7A). This kinetic property can be determined independently of the ability of the MyoD proteins required complexes
plexes
to interact with E12. We found that the time for dissociation of 50% ofthe X1mf25-containing (1 .7 mm) was slightly faster than for the com-
with
XImfl
(2.6
mm).
These
from the slope of the linear fit through Fig. 7A, are similar to results obtained
C2 myohlasts
might
as a I)ositive
(28).
formed
reduced binding may contribute to 1 to function in the previous assays.
11g. 5. Trans-activation of MCK-CAT reporter constructs by transient expression of XIrnl genes in C3H1OT1/2 cells. A, graph of CAT activities normalized with respect to pSV2CAT activity. B, autoradiograph of a representative CAT assay. Lane 1 , no trans-activator; Lane 2, XImtl ; Lane 3, Xlnifi 1 ; Lane 4, XImf25; Lane 5, moose MyoD; and Lane 6, pSV2CAT. C, Northern analysis (if RNA from parallel transtections of peMXi , peMX1 1, PeMX25, and peM34O (MyoD) analyzed for expression of the endogenous Slyol) and myogenin genes. Total RNA from differentiated luded
enhancer
alone
of
activation. Altered DNA binding could be a consequence of a difference in the ability of the Xlmf proteins to form heterodimens with E12, an altered affinity of Xlmf-E12 complexes for the binding site, or both. We first determined the relative affinity of both XImf-E12 oligomeric complexes for the MCK E-box by measuring the rates of dissociation of the protein-DNA complexes in the pres-
‘1i.;;;
ifl(
DNA-
performed
amounts corresponding
than the XImfl-E12
This of Xlmfl
be responsible
.
MYOGENIN
was
we
approximately 50% greater than the amount bound by those containing X1mf25, which is consistent with their relative abilities to trans-activate the MCK-CAT reporter gene. This similarity suggested that a difference in the DNA-binding abilities between Xlmfl and X1mf25 could
#{149}ya?
it.
from
XImf
were
A comparison
.,
proteins,
the labeled probe was inhibited by increasing amounts of the unlabeled homologous oligonucleotide [shown for Xlmfl (Lanes 2-4) and X1mf25 (Lanes 6-8)]. The XImfl 1E12 heterodimers interacted with the MCK E-box motif heterodimers. the inability
C
Xlmf
regulatory
investigate
assays using equimolar and an oligonucleotide
tectable
much
C
To
upon the addition of synthetic E12, presumably as a consequence of the formation of heterodimers (Fig. 6B, Lanes 1, 5, and 9). The binding of these complexes to
S..
0
to muscle-specific
enhancer.
of the
to the downstream None
bind
MCK
gel mobility shift synthetic proteins
Transoctivator
B
that the
control
(Lane
5).
(31) slightly
and
indicate greater
MCK downstream To investigate have
that affinity
XImfl-E12 (approximately
values,
calculated
the data points in for mouse MyoD
complexes 1 .5-fold)
have for
a the
E-box than XImf25-E12 complexes. the possibility that Xlmfl and X1mf25
different
interaction
efficiencies
with
the
Cell
A
Fig. 6. Electrophoretic mobility shift assays using synthetic proteins and a MCK E-box binding site. A, synthetic Xlmfi, XImf25, XImfll, and E12 proteins (Lanes 2-5, respectively) were incubated individually with a labeled oligonucleotide containing the downstream E-box motif from the murine MCK enhancer. Synthetic MyoD and E12 incubated together after translation generated a complex that binds to the E-box probe Lane 6). Proteins synthesized from BMV RNA did not bind to the probe (Lane 1).
B
Growth
& Differentiation
625
t)
Ic.’I
ZYc’%d
5’7
5’;7
.#{231}*
p
31030
0
31030
:C
B, synthetic Xlmfl, X1mt25, and Xlmfl 1 proteins were incubated with synthetic E12 and the MCK E-box probe (Lanes 1-9). Increasing amounts of the homologous unlabeled oligonucleotide were added to the reactions contaming Xlmfl Lanes 2-4) and X1mf25 (Lanes 6-8). The mass ratio of competitor to probe is indicated above each lane (:C).
123456
heterodimer
partner
E12, we
measured
123456789
the amount
of
DNA-binding complex formed at various concentrations of XImf and a constant amount of E12. Since Xlmfl and Xlmf25 do not bind the MCK E-box motif in the absence of E12, the amount of DNA-binding complexes at different concentrations of XImf should be proportional to the ability of each protein to interact with El 2. A protein that interacts strongly with E12 should form a DNA-binding
complex at a lower concentration than a protein that interacts with El2 less efficiently. The amount of probe bound in each reaction was normalized to the maximum amount bound and plotted as a function of the amount of synthetic Xlmf protein added (Fig. 7B). When analyzed
in this
manner,
XImfl
and Xlmf25
in their ability to form heterodimers The maximum amount of probe was 1.4 times the amount bound ing the relative affinities of these (Fig. 7A). Taken together, the results of gest that the observed difference E12 complexes to the E-box motif
in protein-DNA
interactions
interactions.
the
basic
XImfl
and
and
HLH
and heterodimer nation for their E12. Differences
appeared
these experiments sugin binding of the Xlmfis due to a difference
rather
X1mf25
domains
equivalent
with synthetic E12. bound by Xlmfl-E12 by X1mf25-E12, reflectcomplexes for this site
then
are
that
formation. This ability to interact in DNA binding
protein-protein
highly
mediate
conserved
DNA
in
binding
provides a facile explaindistinguishably with may be due to amino
Discussion The present study shows that two distinct genes encoding similar MyoD-like proteins are unexpectedly differentially expressed during development. X1m125 transcripts were present in oocytes, eggs, and early embryos, whereas Xlmfl transcripts did not accumulate until after
activation
of zygotic
transcription
at the midblastula
tran-
sition. The biphasic expression of the Xenopus MyoD genes in early development is the consequence of the overlapping patterns of expression of XImfl and XIm 125. These results reconcile our previous studies (17) with those of Hopwood et a!. (22) and Harvey (23), who also showed the presence of a low level of maternal MyoD transcripts in oocytes and early embryos. A largely random distribution of maternal MyoD transcripts to daughter cells could account for the unlocalized distribution of this mRNA in blastula embryos (23). In addition, these groups demonstrated that the initiation of zygotic MyoD transcription requires the induction of mesoderm and is restricted to cells that will subsequently form somites. Maternal MyoD transcripts and multiple MyoD genes have not been found in other animals; however, there is precedent for differential regulation of apparently duplicate genes in Xenopus (32). It is likely that the Xenopus MyoD genes are a consequence of a genomic duplication (20) rather than products of variant alleles.
The mechanism
underlying
the differential
expression
to the F-box motif than either XImf-E12 or X1mf25-E12 oligomers. Although it is not known whether this difference is due to altered protein-protein interactions or protein-DNA interactions, the fact that XImfl and XImfl 1
of the Xenopus MyoD genes is not known. Activation of zygotic MyoD transcription occurs at the right time and in the appropriate cells for its participation in the activation of muscle genes in developing somites. The kinetics of accumulation of zygotic MyoD transcripts suggest that their synthesis is initiated with, or very soon after, activation of zygotic transcription at the midblastula transition. Therefore, the temporal and spatial regulation of MyoD transcription in developing embryos appears to be
are identical
throughout
their
a consequence
suggests that ability to bind
the XImfl DNA.
carboxy
acid and
acid
differences X1mf25
region).
that
(four
exist
in the
nonidentical
XImfll-E12
basic
residues
oligomems
region in this
bind
basic
and
terminus
of XImf 1 20-amino
less efficiently
HLH
domains
enhances
its
ably and
include nonmuscle
of localized muscle-specific trans-acting
maternal
factors
that presum-
transcriptional mepressors. These
activators observa-
626
Differential
Expression
of Xenopus
Myot)
Genes
A 0
C 0 fig. 7. Analysis of DNA-binding properties of Xlmfi-E12 and Xlmt25-E12 complexes. A, rates of dissociation in the presence of 1000-fold molar excess of oligo-
10
.0
0 L
0 >
nucleotide
were
0 0
5.0
Time
(mm)
competitor.
removed
reaction
Xlmfl
0
Xlmf25
70
ucts
after
normalizing
electrophoresis
on
gels
and
to methionine
con-
tent. The perentage of probe bound relative to the 1:1 ratio for XImf:E12 was plotted versus the proportion of Xlmt in the reaction. 0, Xlmtl; #{149}, Xlmf25.
II,
I0 >
2 sl of syn0.5, 1, or or Xlmf25, giving 0, 0.25,
SDS-polyacrylamide
60
50 0
0
and
relative molar ratios of 0, 0.125:1, 0.25:1, 0.5:1, and 1:1, respectively. The relative amounts of each protein were determined iy measuring the amount of [‘5Slmethionine incorporated into translation prod-
80 C
contained Ei2
2 MI of Xlmti
go
-U
1, 1.5, 2,
2.5, and 5 mm after addition of the homologous oligonucleotide competitor and analyzed with the gel mobility shift assay (inset). The percentage of probe bound at each time point relative to the zero time point was plotted versus time. 0, XImfl; 0, XImf25. B, efficiencies of heterodimerization of Xenopuc MyoD proteins with E12. Each binding thetic
100
Aliquots
0, 0.5,
30
o.0
0.2
Ratio XImf/E12
tions suggest a scenario that might account for the differential expression of the Xenopus MyoD genes. We suggest that the maternally expressed gene has escaped the negative control that normally prevents MyoD transcrip-
tion in nonmuscle cells, perhaps of a negative regulatory element. acting factor that by studies with
however, MyoD
cannot
represses somatic
by deletion or mutation The existence of a trans-
MyoD expression cell hybrids (33).
be sufficient
to account
is suggested This alone,
for maternal
transcription because of the strict muscle-specific of Xlm125 in adult tissues. Therefore, we furthem suggest that a maternal transcriptional activator, required for MyoD expression in embryos, may accu-
expression
mulate scniption
during oogenesis and result in the selective tranof Xlm125 in oocytes. Zygotic MyoD transcnip-
tion would presumably require the activity ofthe positive maternal activator and the removal of negative control. Forced
expression
of mammalian
a variety of nonmuscle battery of muscle-specific myogenic
factors
(34).
myogenic
factors
in
cells induces expression of a genes, including endogenous It has been
autoregulation of MyoD expression mitment to the myogenic pathway
suggested
that
positive
is important by allowing
for commainteof MyoD and MyoD genes
nance and amplification of the expression other myogenic factors (34). Both Xenopus display the functions generally described for their
mam-
Cell Growth
malian counterparts. Expression of the Xenopus MyoD cDNAs in stably transfected mouse fibroblasts results in expression of the muscle phenotype upon differentiation, including the formation of multinucleated myotubes, expression of muscle-specific gene products, and activation of endogenous myogenic regulatory genes. In transient assays, the Xenopus MyoD genes are capable of directly trans-activating the murine MCK enhancer. Although X!mfl displayed a slightly greater trans-activation potential than XImf2S, suggesting that there may be functional differences between these proteins, the major conclusion from these studies is that both Xenopus genes are clearly capable of initiating myogenesis. Because XImf2S is myogenic when expressed
muscle
mouse
suggests
that
cells,
its presence
regulatory
mechanisms
MyoD
as a maternal prevent
RNA
MyoD
ac-
tivity in the precursors of cells with nonmuscle fates during early development. Recent studies provide evidence for positive regulation of MyoD activity in developing embryos. Hopwood and Gurdon (35) showed that ectopic expression of MyoD (X!mfl ) in Xenopus embryos activated muscle-specific actin transcription in ectodermal cells that do not normally form muscle but did not induce myogenesis or otherwise alter embryonic development.
These
absence
from
essential
for complete
regulation
results
were
ectodermal
myogenesis.
of myogenic
studies with cultured or certain peptide
interpreted
to
cells of a positive Evidence
factor
cells. growth
indicate
factor
activity
the
that is
for negative
has come
High concentrations factors, including
from
of serum fibroblast
growth factor and transforming growth factor 13, can block muscle cell differentiation and the activity of constitutively expressed myogenic factors in cultured cells through
a mechanism
that
may
involve
the
dominant
(17).
initial
ume
of water-saturated
and
transforming
growth
Xenopus
dichotomy positive
embryos
(38, reviewed
suggests signals
for
that, inducing
in Ref.
in addition a subset
role in the muscle in
39). This
apparent
to functioning of ectodermal
as cells
to become mesoderm, peptide growth factors might inhibit MyoD activity in the early embryo and thereby allow the continued proliferation of nonmuscle lineages. It should
be
pointed
out
that
the
timing
and
sites
of
MyoD protein synthesis during amphibian development are unknown; it is possible that MyoD synthesis is translationally regulated so that the maternal MyoD mRNA is not expressed before mesoderm induction. In view of the ability of MyoD to activate its own expression, it will be interesting to investigate whether translation of the maternal MyoD mRNA precedes zygotic activation of MyoD transcription.
extraction
by the addition
T3 RNA transcripts
factor
and Methods
Embryo Studies. Adult Xenopus females (from Xenopus!, Ann Arbor, Ml) were primed for ovulation by injection with 500-100 units of pregnant mare serum (Sigma Chemical Co., St. Louis, MO) into the dorsal lymph sac followed 1 day later by injection with 500 units of human chorionic gonadotropin (Sigma). Eggs were manually stripped on the third day and fertilized with a suspension
was
isolated
from
performed
phenol
of 0.5 volume
X/mf25-specific
cloned initiation
oocytes
and
by adding
with
vortexing,
of chloroform
and
0.5
vol-
followed a second
probe
(25Sty)
was
prepared
from
a
genomic fragment that spans the transcription site by digestion with Styl and transcription with polymerase. from the
This probe hybridizes to X!m125 transcription initiation site to nucleo-
tide 163 of the sequence reported previously for this clone (17). The 9BS probe used to distinguish between XImfl and XImfl 1 transcripts was made by subcloning a fragment of XImfl between the BamHl site and Saul sites into
pBluescript.
This
template
was
digested
with
Xbal
and transcribed with T7 RNA polymerase. Protection of this probe by XImfl RNA generates a 336-nt fragment, whereas the XImfl 1 transcript protects a 186-nt fragment. Antisense RNA probe synthesis and ribonuclease protection assays were performed essentially as described
by Krieg (42). Briefly, probes were synthesized in a meaction containing 1 zg of template DNA-10 mi dithiothreitol-0.1 mg/mi bovine serum albumin-40 mt’i TrisHCI, pH 7.5-6 mM MgCl2-2 m’i spemmidine-40 units RNAsin-500 MM each GTP, CTP, ATP-10 fLM UTP-35 zCi [a-32P]UTP-20 units of T3 or T7 bacteriophage RNA polymerase. for 1 h at
Materials
was
vortexing. A phenol:chloroform (1:1) solution was used for subsequent extractions. Nucleic acids were precipitated by the addition of 0.1 volume of 3 M sodium acetate, pH 5.2, and 2.5 volumes of 95% ethanol. The resulting pellet was resuspended in homogenization buffer without proteinase K and extracted once with phenol-chloroform and once with chloroform. After precipitation as described above, the resulting nucleic acid pellet was washed in 70% ethanol and resuspended in sterile water. Ribonuclease Protection Analysis. The plasmid 9Apa, containing nucleotides 98-331 of the XImfl cDNA, was used to prepare an X/mIl-specific riboprobe by digestion with EcoRl and transcription with T3 RNA polymerase.
of the
growth
RNA
The
The
fibroblast
Total
staged embryos by homogenization in 50 m&i Tris-HCI, pH 7.5-50 mM NaCl-5 mi EDTA-0.5% SDS-400 ig/ml proteinase K. The homogenate was extracted three times with phenol:chloroform (1:1), and once with chloroform.
negative inhibitor of differentiation, Id (36), or postsynthetic modifications that inhibit trans-activation without affecting DNA binding (37). Interestingly, growth factors factor f families are thought to play a pivotal specification of mesoderm and, subsequently,
627
of frog sperm in MBS (40). Fertilized eggs were dejellied in 2.5% i-cysteine-0.6% Tris, pH 7.9, and then washed extensively in MBS. Approximately 5 h after fertilization, embryos were placed in 0.lx MBS. Developmental stages were determined according to Nieuwkoop and Faber (41). Embryos at desired developmental stages were transferred to plastic microcentrifuge tubes, and the excess fluid was removed; they were then quickly frozen on dry ice. Frozen embryos were stored at -80#{176} C until further use. RNA Isolation. Total cellular RNA was isolated from adult frog tissues and cultured cells as described previously
in non-
& Differentiation
Synthesis 4#{176}C.The
reactions template
were allowed to proceed DNA was removed by
treatment with DNase I, and the full-length probe was obtained by isolation from a 5% acrylamide-50% urea gel. For each RNA sample, 2-5 X iO cpm of probe were used. For all assays involving embryo RNA, one embryo equivalent of RNA was used unless otherwise indicated. Cell Culture and DNA Transfection. C3H 1OT1 /2 fibroblasts (43) and C2 cells (44) were maintained in Dulbecco’s
modified
Eagle’s
medium
supplemented
with
20%
628
Differential
Expression
fetal bovine cloned into vided
of Xenopus
Genes
MyoD
serum. The X!mfl 1 and X!mf2S the expression vector peM33O
by S. Pearson-White
and
C. Emerson)
cDNAs (kindly
were pro-
as previously
described for Xlmf (17). These vectors, designated peMXl peMXl 1 and peMX25, were transfected separately into 1OT1/23 cells, and clones harboring stably integrated plasmids were selected by their resistance to G4l8. Duplicate 15-cm dishes were transfected; clones were isolated from one dish while its duplicate was placed in differentiation medium (Dulbecco’s modified Eagle’s medium with 2% horse serum) and later immunostained for the presence of myosin heavy chain as described previously (1 7). The expanded clones obtained for pmMXl 1 and pmMX25 were also immunostained for ,
myosin
,
heavy
chain
after
exposure
to differentiation
me-
dium. Trans-activation of MCK gene regulatory elements was performed with pCKCATE4, a CAT reporter gene fused to 246 bp of the MCK promoter with a 300-bp region of the MCK enhancer, from -1351 to -1050, inserted into the BamHl site downstream of the CAT gene (15). The expression plasmids peMXl, peMX25, peMXll, and peM34O (MyoD) were cotransfected with pCKCATE4 into duplicate cultures of lOTl/2 cells, and the cells were placed in differentiation medium 16-20 h after transfection. Forty-eight h later, the cells were hamvested; extracts were prepared from one set of dishes for measurement of CAT activity (1 5), and RNA was purified from
the
other
set of dishes
to analyze
expression
of the
transfected In Vitro
and endogenous myogenic factors. Transcription and Translation. The plasmids pBS/XImul, pBS/XImfl 1, and pBS/X1m125 were previously described (17). The E12-containing plasmid was a gift of C. Murre and D. Baltimore (9). RNA was synthesized in a l00-zl reaction volume for 1 h at 37#{176}C using 1 zg of linearized DNA and T3 or T7 RNA polymerase, as appropriate, in the presence of RNAsin and mRNA cap analogue (New England Biolaboratories, Beverly, MA). Proteins were produced by in vitro translation of synthetic RNA using a rabbit reticulocyte lysate (Promega, Madison, WI) in a 50-zl reaction. A 5-zl aliquot of each translation reaction was incubated with 10 zCi of [35S] methionine. The synthesis of radiolabeled proteins was analyzed by electrophoresis on 10% acrylamide-SDS gels and subsequently quantitated by scanning the dried gel with a Betascope model 603 blot analyzer (Betagen, Waltham, MA) to determine the relative amounts of protein produced in each translation reaction. The unlabeled portion of each translation reaction was used in DNA-binding
assays.
DNA-binding taming
the
Assays. downstream
A 30-bp E-box
oligonucleotide in
the
MCK
conenhancer
(underlined below) was formed by annealing two singlestranded 26-bp oligonucleotides and filling the ends with Klenow fragment of DNA polymerase I and [a-32P]dATP. One strand of this oligonucleotide has sequence 5’AGCTTCCCCCAACACCTGCTGCCTGAAGCT. Binding reactions were performed in a total volume of 20 zl consisting of 10 mM Tris-HCI, pH 7.5-50 mti NaCI-1 m&i dithiothreitol-1 mi EDTA-5% glycerol-i ag of polydeoxyinosinic-deoxycytidylic
acid
(Pharmacia,
Piscataway,
NJ)-l-3 MI of synthetic protein-i-2 X i0 cpm oligonucleotide probe. Components were incubated together for 5 mm, probe was added, and the incubation continued for an additional 20 mm. Reticulocyte lysates contaming Xlmfs and E12 were premixed for 5-10 mm at
room temperature before adding to the binding reaction. The Teactions were loaded onto 6% acrylamide-0.3% bis-acrylamide-0.5x TBE (0.5x TBE is 45 mti Tris-45 mri boric acid-l mM EDTA) gels and electrophoresed at 150 V. For the kinetic determinations, DNA binding reactions were assembled as above except that 1000-fold molar excess of unlabeled oligonucleotide was added to the reaction after the 20-mm incubation with labeled DNA. Aliquots were removed at timed intervals and immediately loaded onto a running gel. Quantitation was performed on dried gels as described above. Acknowledgments We thank D. Lu for isolating and sequencing Xlmf2S genomic sequences, S. Jassar for performing tissue culture, and Bill Klein and craig Hinkley for comments on the manuscript. Oligonucleotides were synthesized in the Macromolecular Synthesis Facility at M. D. Anderson cancer center.
References 1. Mohun, 1. J., Brennan, S., Dathan, N., Fairman, S., and Gurdon, J. B. Cell type-specific activation of actin genes in the early amphibian embryo. Nature (Lond.), 31 1: 716-721, 1984. 2.
Mohun,
actin
1. J., Garret,
N., Stutz,
F., and
gene
Xenopus
is expressed during laevis. J. Mol. Biol., 202:
3. Olson,
E. N. MyoD
Dev., 4: 1454-1461,
4. Davis, transfected
family:
early 67-76,
Spohm,
G. A third
development 1988.
a paradigm
striated
in the
muscle
amphibian
for development?
Genes
&
1990.
R. L., Weintraub, cDNA converts
H., and Lassar, A. B. Expression of a single fibroblasts to myoblasts. Cell, 51: 987-1000,
1987. 5.
Edmondson,
D.
myc similarity region sufficient to activate 3: 628-640, i989.
G.,
and
Olson,
of MyoDl the muscle
E. N.
A gene
is expressed differentiation
with
Wright, W. E., Sassoon, D., and Lin, V. K. Myogenin, myogenesis, has a domain homologous to MyoD. i989.
6.
7.
Braun,
1.,
Buschhausen-Denker,
C.,
Arnold, H. H. A novel human muscle MyoDi induces myogenic conversion 709, 1989. 8. Rhodes, S. J., and Konieczny, member of the muscle regulatory 2050-206i,
homology
to
during myogenesis program. Genes
Bober,
E.,
the
and is & Dev.,
a factor regulating Cell, 56: 607-617, Tannisch,
E.,
and
factor related to but distinct from in 1011/2 cells. EMBO J., 8: 701 -
S. F. Identification factor gene family.
of MRF4: a new Genes & Dev., 2:
1989.
9. Murre, C., McCaw, C., and Baltimore, D. A new DNA binding and dimerization motif in immunoglobulin enhancer binding, daughterless, MyoD, and myc proteins. Cell, 56: 777-783, 1989. 10. Murre, C., McCaw, P. 5., Vaessin, H., Caudy, M., Jan, L. Y., Jan, Y. N., Cabrera, C. V., Buskin, J. N., Hauschka, S. D., Lassar, A. B., Weintraub, H., and Baltimore, D. Interactions between heterologous helix-loop-helix proteins generate complexes that bind specifically to a common DNA sequence. Cell, 58: 537-544, 1989. 11. tion
Sun, X-H., and Baltimore, D. An inhibitory domain factor prevents DNA binding in E12 homodimers
heterodimers. 12.
Brennan,
Cell,
64: 459-470,
1. 1. and
Olson,
of Ei2 transcripbut
not
in
E12
1991. E. N. Myogenin
acquires high affinity for a conserved ization. Genes & Dev., 4: 582-595,
enhancer 1990.
resides
in the
element
nucleus
and
on hetemodimer-
i3. Lassar, A. B., Buskin, J. N., Lockshon, D., Davis, R. L., Apone, S., Hauschka, S. D., and Weintraub, H. MyoD is a sequence-specific DNA binding protein requiring a region of myc homology to bind to the muscle creatine kinase enhancer. Cell, 58: 823-831, 1988. 14. Lenardo, M., Pierce, J. W., and Baltimore, D. Protein-binding sites in Ig gene enhancers determine transcriptional activity and inducibility. Science (Washington DC), 236: 1573-1577, 1987. 1 5.
Sen.
R., and
immunoglobulin
Baltimore,
enhancer
D. Multiple
sequences.
nuclear
factors
Cell, 46: 705-716,
interact
with
the
1986.
16. Weintraub, H., Tapscott, S. I., Davis, R. L., Thayer, M. J., Adam, M. A., Lassar, A. B., and Miller, D. Activation of muscle-specific genes in pigment, nerve, fat, liver, and fibroblast cell lines by forced expression of MyoD. Proc. NatI. Acad. Sci. USA, 86: 5434-5438, 1989.
Cell Growth
17. Scales, with
J. B., Olson,
homology
E. N., and Perry,
M. Two distinct
to MyoDl
embryogenesis.
Mol.
are expressed before Cell. Biol., 10: 1516-1524,
somite
Xenopus genes formation in early
1990.
18. Sassoon, 0., Lyons, G., Wright, W. E., Lin, v., Lassar, A., Weintraub, H., and Buckingham, M. Expression of two myogenic regulatory factors myogenin and MyoDi during mouse embryogenesis. Nature (Lond.), 341: 303-307, 1989. 19.
Ferris,
idisation.
S. D., and Whitt, Nature
(Lond.),
256:
G. S. Loss of gene 258-260, 1977.
20. Kobel, H. R., and Pasquier, 1. D. Genetics TrendsGenet., 2:310-315, 1986. 21.
expression
after
of polyploid
Whitt,
G. S. Evolution of isozyme loci and their differential and J. L. Reveal (eds.), Evolution Today, of the Second International Congress of Systematic and Biology, pp. 271-289. University of British Columbia, 1981. In: G. G. E. Scudder
22. the
Hopwood, N. D., Pluck, A., and Gurdon, I. B. MyoD forming somites is an early response to mesoderm Xenopus embryos. EMBO I., 8: 3409-3417, 1989.
23. Harvey, R. P. The Xenopus mRNA predates lineage restricted velopment, 108: 669-680, 1990.
32.
in in
MyoD gene: an unlocalised maternal expression in the early embryo. De-
D. G., and Olson,
E. N. corrigendum.
Genes
& Dev.,
Gossett, 1. A., Kelvin, 0. J., Sternberg, E. A., and Olson, E. N. A new myocyte-specific enhancer-binding factor recognizes a conserved element associated with multiple muscle-specific genes. Mol. Cell. Biol., 9: 27.
5022-5033,
1989.
29.
Yutzey,
ogenin
K. E., Rhodes, associated
and NRF4.
with
Mol.
cell.
S. J., and the
Konieczny,
muscle regulatory Biol., 10: 3934-3944,
S. F. Differential factors
MyoDl,
1989.
35. Hopwood, N. D., and Gurdon, J. B. Activation of muscle without myogenesis by ectopic expression of MyoD in frog embryo Nature (Lond.), 347: 197-200, 1990. 36. Benezra, R., Davis, R. 1., Lockshon, H. The protein Id: a negative regulator proteins. Cell, 61: 49-59, 1990.
trans my-
1990.
30. Brennan, 1. J., chakraborty, T., and Olson, E. N. Mutagenesis of the myogenin basic region identifies an ancient protein motif critical for activation of myogenesis. Proc. NatI. Acad. Sci. USA, 88: 5675-5679, 1991.
genes cells.
D., Turner, 0. L., and Weintraub, of helix-loop-helix DNA binding
37. Chakraborty, 1., Brennan, T., and Olson, E. Differential trans-activation of a muscle-specific enhancer by myogenic helix-loop-helix proteins is separable from DNA binding. J. Biol. chem., 266: 2878-2882, 1991.
1., Vale, embryocell, 63:
38. Thomsen, G., Woolf, 1., Whitman, M., Sokol, S., vaughan, W., and Melton, D. A. Activins are expressed early in Xenopus genesis and can induce axial mesoderm and anterior structures. 485-493, 1990. 39. Smith, J. C. Mesoderm induction and mesoderm-inducing early amphibian development. Development, 105: 665-677,
factors 1989.
in
40. Colman, A. Translation of eukamyotic messenger RNA in Xenopus oocytes, In: B. D. Hames and S. J. Higgins (eds.), Transcription and Translation: A Practical Approach. Oxford: IRL Press, 1984. 41. laevis
28. Sternberg, E. A., Spizz, G., Perry, W. M., vizard, D., Weil, T., and Olson, E. N. Identification of upstream and intragenic regulatory elements that confer cell-type-restricted and differentiation-specific expression on the muscle creatine kinase gene. Mol. cell. Biol., 8: 2896-2909, 1988. activation
expression of two EMBO J., 8: 4091 -
34. Thayer, M., Tapscott, S. J., Davis, R. 1., Wright, W. E., Lasser, A. B., and Weintraub, H. Positive autoregulation ofthe myogenic determination gene MyoDl . Cell, 58: 241 -248, 1989.
regulation. Proceedings Evolutionary
25. Braun, T., Winter, B., Bober, E., and Arnold, H. H. Transcriptional activation domain of the muscle-specific gene-regulatory protein myfs. Nature (Lond.), 346: 663-665, 1990. Edmondson, 4: 1450, 1990.
S., Taylor, M., and M#{233}chali,M. Differential proto-oncogenes during development.
c-myc
33. Thayer, M., and Weintraub, H. Activation and repression of myogenesis in somatic cell hybrids: evidence for trans-negative regulation of MyoD in primary fibroblasts. Cell, 63: 23-32, 1990.
24. Tapscott, S. I., Davis, R. 1., Thayer, M. J., cheng, P. F., Weintraub, H., and Lassar, A. B. MyoDl : a nuclear phosphoprotein requiring a myc homology region to convert fibroblasts to myoblasts. Science (Washington DC), 242: 405-411, 1988.
26.
vriz,
4097,
Xenopus.
expression induction
629
31. Weintraub, H., Davis, R., Lockshon, D., and Lassar, A. MyoD binds cooperatively to two sites in a target enhancer sequence: occupancy of two sites is required for activation. Proc. NatI. Acad. Sci. USA, 87: 56235627, 1990. Xenopus
polyplo-
& Differentiation
P. 0., and Faber, J. (eds.). Normal Ed 2. Amsterdam: Elsevier/North-Holland
Nieuwkoop, (Daudin),
table of Xenopus Publishing Co.,
1967.
42. Krieg, polymerase. Technology.
P. A. Synthesis In:
I. W.
London:
Dale
of RNA probe using SP6, 17 and and P. G. Sanders (eds.), Methods
JAI Press,
T3 RNA in Gene
in press.
43. Kimes, B. W., and Brandt, B. L. Characterization of two putative smooth muscle cell lines from rat thoracic aorta. Exp. Cell Res., 98: 1 15133, 1976. 44. Yaffe, D., and Saxel, 0. Serial ogenic cells isolated from dystrophic 725-727, 1977.
passaging and differentiation of mymouse muscle. Nature (Lond.), 270:
