May 1, 1992;
Expression of the myogenic gene MRF4 during Xenopus development.
In a search for myogenic genes in Xenopus, I have cloned homologs of the mammalian myogenic genes MRF4
. The myogenin
clone is a genomic fragment encoding an amino acid sequence with 62% identity to the N-terminal region of rat myogenin
. No myogenin
transcript has been detected and no cDNA has been isolated, suggesting that Xenopus myogenin
, if it is expressed at all, is likely to be expressed at low levels or transiently during development. A Xenopus MRF4
cDNA has been isolated and encodes an amino acid sequence with 72% identity to rat MRF4
. In adult frogs, MRF4
RNA is detectable only in skeletal muscle
, unexpectedly, is also expressed at low levels in the heart). During embryonic development, MRF4
RNA appears later than MyoD
, at a time when the embryonic musculature already shows many differentiated features. This implies that MRF4
is not involved in the commitment or early differentiation of muscle
cells. The accumulation of Xenopus MRF4
RNA overlaps with the formation of neuromuscular connections, suggesting that it may be induced by innervation. Consistent with this possibility, the level of Xenopus MRF4
, but not MyoD
, RNA is reduced in response to denervation of adult frog muscle
[+] show captions
FIG. 1. Partial sequence of Xenopus myogenin-like genomic clone. (A) Nucleotide and deduced amino acid sequence of pGJMGL. The second
methionine codon may be the major initiation site by analogy with mammalian myogenin and because it is closer to the consensus Kozak
sequence. (B) Alignment of Xenopus myogenin-like sequence with amino terminal portion of rat myogenin. Identical residues are shown in the
rat sequence as dashes (-), while differing residues are shown in one-letter code. Spaces in the rat sequence indicate gaps inserted for optimal
alignment. The two sequencea are 62% identical. The B-HLH region is underlined.
FIG. 2. Sequence of Xenopus MRF4 cDNA. (A) Partial nucleotide sequence and deduced amino acid sequence of pCJM4.2. The coding sequence
is complete, with in-frame stop codons both upstream and downstream (underlined). The 3’ untranslated region has not been fully sequenced,
but it is approximately 700-800 bases long and ends with a poly(A) tail (not shown). The 5’NotI cloning site and the G-C tail generated during
library construction are also underlined. (B) Alignment of Xenopus and rat MRF4 amino acid sequences. Identical residues are shown in the rat
sequence as dashes (-), while differing residues are shown in one-letter code. Spaces indicate gaps inserted for optimal alignment. The two
sequences are 72% identical. The B-HLH region is underlined.
FIG. 3. Xenopus MRF4 is detected in skeletal muscle but not in smooth muscle, cardiac muscle, or liver. (a) Total RNA from adult heart (H),
stomach (S), or lower leg muscle (M) was assayed by RNase protection for MRF4 and MyoD RNA. Ten micrograms of total RNA were used for
MRFI, 4.5 rg for MyoD. Equal RNA recovery was confirmed using a control EFla probe (not shown). The MyoD lanes have been overexposed in
order to show the low level of expression in heart; on shorter exposures, the same two bands are visible in the skeletal muscle lane. Five
experiments with different animals gave similar results. (b) Poly(A)-selected RNA from adult lower leg muscle (M), liver (L), heart (H), or
stomach (S) was assayed by Northern blots for MRF4 and MyoD transcripts. Duplicate blots were probed with equivalent amounts of DNA
probe under identical conditions. The relative amounts of mRNA were determined by EFla expression; the absolute amounts of poly(A) RNA
were not measured directly, but the muscle lanes correspond to 120 pg of total RNA. Muscle, liver, and stomach lanes contain equal amounts of
mRNA, heart lanes contain threefold more mRNA, in order to detect low levels of MyoD expression. Individual bands were quantified using a
phosphorImager; MyoD is about IO-fold more abundant than MRF4 in skeletal muscle, and MyoD is about IO-fold more abundant in skeletal
muscle than in heart. The 0.5-kb transcript seen with the MRF4 probe has been detected in various tissues with a full-length cDNA probe, but
not with a coding region probe; it may correspond to a repetitive element within the 3’untranslated region, but it has not been characterized in
detail. The faint 2.3-kb transcript seen with the MyoD probe may be a precursor RNA. Sizes are approximate estimates, based on comparison
with mind111 DNA standards. Two experiments with different animals gave similar results.
FIG. 4. Time course of Xenopus MRF4 expression during early embryogenesis.
Total RNA was extracted from pools of five embryos at
each stage indicated, and assayed by RNase protection for MRFI,
MyoD, and histone H4 RNA expression. A one quarter embryo equivalent
was used for each lane. Based on this and four other experiments
with different batches of embryos, MRF4 RNA is first detected at
stage 18 and becomes maximal by stages 22-23. Based on matched
exposures, MRF4 RNA is about 20-fold less abundant than MyoD at
stage 24 (the exposures shown here are not equivalent).
FIG. 5. Xenopus MRF4 RNA is reduced by denervation of adult muscle.
Lower leg muscles were denervated by cutting the sciatic nerve.
RNA was extracted 10 days later from denervated and contralateral
control muscles, and assayed by RNase protection for MRFI, MyoD,
acetylcholine receptor (a! subunit), and EFla RNA. Amounts used
vary from 2 to 6 pg total RNA per assay, but are equal for innervated
and denervated lanes. I, innervated muscle; D, denervated muscle; C,
control (tRNA). Results from two animals are shown here; three other
animals gave similar results. Bands were quantified by laser densitometry;
MRF4 RNA falls two- to threefold after denervation.
FIG. 6. Rat MRF4 is not localized to the synaptic regioh muscle. Adult rat diaphragms were dissected into synaptic
(+) and nonsynaptic (-) regions, and assayed by RNase protection for
MRF4 and acetylcholine receptor (t subunit) expression. Each lane
represents 10 pg of total RNA; exposure times are equal for both
probes. MRF4 is expressed equally in both fractions, whereas AChR is
concentrated in the synaptic fraction. Two experiments are shown,
from pools of two and three animals, respectively.