815
Development 102, 815-821 (1988)
Printed in Great Britain © The Company of Biologists Limited 1988
A critical period for formation of secondary myotubes defined by
prenatal undernourishment in rats
S. J. WILSON, J. J. ROSS and A. J. HARRIS*
The Neuroscience Centre and Department of Physiology, University of Otago Medical School, PO Box 913, Dunedin, New Zealand
* To whom reprint requests should be sent
Summary
Rats fed a restricted diet during gestation and lactation gave birth to pups with about 60 % the normal
birth weight. Maintaining the undernutrition after
birth reduced the rate of growth of the pups so that
their body weights were only 40 % of control at PN7.
Soleus and lumbrical muscles in these animals had
reduced numbers of muscle fibres, and quantitative
examination of embryonic muscles revealed that this
was due solely to a decreased formation of secondary
myotubes; the number of primary myotubes remained
normal. Undernutrition did not affect the number of
motoneurones surviving normal developmental death.
Restoration of normal dietary intake on E21, one day
Introduction
The number of fibres in a muscle depends both on
genetic (Byrne etal. 1973; Stickland & Handel, 1986)
and environmental circumstances. Muscle fibre numbers in young mammals are reduced following undernutrition of the mother during gestation and lactation
(humans: Montgomery, 1962; sheep: Swatland &
Cassens, 1973; rats: Bedi et al. 1982) and also in the
smallest animals within a litfer of young (pigs:
Hegarty & Allen, 1978; Wigmore & Stickland, 1983).
This effect cannot be rectified by subsequent nutritional rehabilitation (Bedi etal. 1982). Mammalian
muscle fibre numbers are not permanently affected by
periods of undernutrition after weaning (pigs: Stickland et al. 1975; mice and hamsters: Goldspink &
Ward, 1979; rats: Bedi etal. 1982).
Muscle development is biphasic (Kelly & Zacks,
1969; Ontell & Dunn, 1978). Myotubes form by
fusion of mononucleate myoblasts, with primary
myotubes forming first and providing a 'cellular
skeleton' (Kelly, 1983) for the later development of
before birth, did not correct the deficit in muscle fibre
numbers in soleus muscles examined when the animals
reached one month of age. Development of the lumbrical muscle lags behind the soleus and unrestricted
feeding from E21 onwards allowed a normal number
of fibres to develop from this time on, although the
initial deficit was never restored. These experiments
define a critical period in muscle development during
which the potential maximum number of secondary
myotubes is determined.
Key words: prenatal undernutrition, skeletal muscle,
secondary myotubes, critical period, lumbrical muscle,
soleus muscle, rat embryos.
secondary myotubes. Secondary myotubes initially
form near the midpoint of a primary myotube,
beneath the basal lamina (Ontell & Dunn, 1978).
They then grow longitudinally and eventually connect
with the muscle tendons and separate from the
primary myotube. Secondary myotube numbers are
sensitive to environmental variables such as paralysis
or denervation (Harris, 1981; McLennan, 1983). Primary myotube formation, by contrast, has so far
proved refractory to experimental manipulation
(Harris, 1981; Wigmore & Stickland, 1983; Ross etal.
1987 b).
In this study, we ask whether the reduction in
muscle fibre number that follows maternal undernutrition is due solely to modulation of the number of
secondary myotubes that form. To gain further
understanding of the events involved in initiation of
formation of secondary myotubes, we also assess the
capacity of muscles at different stages of development
to 'catch up' after the undernutrition-imposed slowing of the rate of myotube formation. This is done by
restoring a normal diet to the mothers one day prior
816
S. J. Wilson, J. J. Ross and A. J. Harris
to birth, a time when 40 % of the secondary myotubes
would normally have been generated in soleus and
20 % in lumbrical muscles. Our results enable us to
define a critical period in muscle development during
which secondary myotubes must be determined to
form if they are ever to be present in the muscle.
Materials and methods
Adult male and virgin female white Wistar rats (200-250 g)
were mated in wire-bottomed cages. The day a copulation
plug was found was designated day zero of the pregnancy
and the female was caged singly from that time.
Pregnant females were allowed free food for the first 2
days of gestation, and then assigned to control, undernourishment and rehabilitation groups. Control females
were allowed food (standard rat pellets) ad lib throughout
gestation and lactation. Undernourished rats were kept on
a restricted ration of about 30 % the control food intake.
Rats in the rehabilitation group were kept on this regime
until day 21 of gestation (E21), one day before parturition,
from which time they were allowed food ad lib. All animals
had free access to water. Body weights of mothers and
young were recorded each morning. Litter sizes were
standardized to eight pups at birth.
Control animals were taken at E17 to count soleus muscle
primary myotubes, at E20 and at postnatal days 7 (PN7)
and 28 (PN28). Undernourished animals were examined at
E20 and PN7, and rehabilitated animals at PN28. To obtain
the fetuses, the mothers were heavily anaesthetized with
ether and fetuses from near the bottom of the uterine horns
removed, placed on ice to maintain anaesthesia and perfused through the heart with warmed fixative. The fixative
contained 1% paraformaldehyde, 1-25% glutaraldehyde
and 0-5mM-CaCl2 in 0-14M-Hepes, with SOi.u.ml"1
heparin. Two embryos, of undetermined sex, were normally taken from each litter. Postnatal tissue was taken
from male rats anaesthetized with ether and perfused
through the heart with fixative containing 1 % paraformaldehyde, 1 % glutaraldehyde, 0-4mM-CaCl2, 0-03M-glucose
in 0-lM-phosphate buffer with 50i.u. ml"1 heparin. Up to
three siblings were taken from each litter.
Following perfusion, the soleus and FVth lumbrical
muscles and the L4 ventral roots were dissected and
immersed in fixative for a total of 4h. Tissues from both
sides of the animal were normally retained. The tissues
were postfixed in osmium tetroxide and stained en bloc in
uranyl acetate, dehydrated and embedded in TAAB epoxy
resin. Ultrathin transverse sections (~90nm) were cut
through E20 soleus and lumbrical and PN7 lumbrical
muscles at the midbelly endplate-containing region and
through the ventral nerve roots just proximal to the dorsal
root ganglion. The sections were mounted on single-slot
Formvar-coated copper grids, stained with uranyl acetate
and lead citrate and viewed and photographed with a
Philips 410 electron microscope. Semithin sections (1 ism)
from PN7 soleus and PN28 lumbrical and soleus muscles
were mounted on glass slides, stained with methylene blue
and azur II, and viewed and photographed with light
microscopy. Photomontages of the entire muscle and nerve
sections were produced, and all of the fibres counted.
Results
Undernutrition
Body weights
At birth, the undernourished pups had a 42 % deficit
in body weight compared to controls. Continued
undernutrition of the mother to PN7 resulted in a
61 % deficit in the body weights of the pups compared
to age-matched controls.
Musclefibrenumbers
Soleus muscles from control animals were examined
at E17 to count primary myotubes. There were
84 ± 1-7 (n = 4) primary myotubes in this muscle. At
E20 we did not attempt to distinguish primary and
secondary myotubes. At this time, there were
914 ±38 (n = 4) muscle fibres in the control soleus
and 649 ±25 (n = 7) muscle fibres in the undernourished soleus muscles (Fig. 1). This is a 29%
reduction in soleus muscle fibre number as a result of
prenatal undernutrition.
By E20, the IVth lumbrical muscle normally has its
full complement of primary myotubes and has been
generating secondary myotubes for approximately
1 day (Ross et al. 1987a), and these two populations
are readily distinguishable with electron microscopy
(Fig. 2A). Primary myotubes are large cells containing abundant well-organized myofilaments. Secondary myotubes are small myofilament-containing cells
3000
2000
1000
10
20
30
Gestational age
40
50
Fig. 1. Muscle fibre numbers in soleus muscles from
control ( • ) , underfed ( ^ ) and rehabilitated (O) rats.
Mothers of undernourished animals were on a restricted
diet from the 3rd day of gestation (E2) continuing
through lactation; mothers of rehabilitated animals were
restored to a normal diet on E21. Parturition occurred on
E22. Error bars, ±S.E.M.
Critical period in skeletal muscle development
Fig. 2. Electron micrographs of E20 lumbrical muscles from control and undernourished rat fetuses. (A) Control
muscle. Clusters of myotubes including a primary myotube (7°), one or two secondary myotubes (2°), and
mononucleate cells (mn) are enclosed within a basal lamina. (B) Undernourished muscle. Primary myotubes
predominate with only an occasional secondary myotube present. Calibration bars, 2/im.
817
818
S. J. Wilson, J. J. Ross and A. J. Harris
1000
5 800
100-
r
er of myotul
2
£ 80-
|
T
.(>
I 600-
400-
40-
B
3 200
20-
(i-
10
\
Control
20
30
40
Gestational age
50
Underfed
Treatment
Fig. 3. Histogram to illustrate counts of primary (1°) and
secondary (2°) myotubes, and mononucleate cells
(presumed myoblasts) (mn) in E20 control and
undernourished lumbrical muscles. Error bars, ±S.E.M.
lying under the same basal lamina and sometimes
interdigitating with primary myotubes. The numbers
of primary myotubes were the same in undernourished (110 ± 2 , /i = 7) and control (109 ± 2,
n = l) lumbrical muscles (P>0-7). There were,
however, very few secondary myotubes in the undernourished (13 ± 1) as compared to the control lumbricals (78 ± 5) (P < 0-001) (Figs 2B, 3), so that the total
number of muscle fibres was 34 % less than controls
(Fig. 4).
The undernourished E20 lumbrical muscles were
less mature than the controls. In the controls, primary
myotubes were separate and most had one or more
associated secondary myotubes, in varying stages of
maturation (Fig. 2A). In the undernourished
muscles, by contrast, the primary myotubes often had
not separated and remained in groups of two or
occasionally three, and were sometimes linked by gap
junctions. Secondary myotubes, in addition to being
reduced in number, were usually small and closely
interdigitated with the primary myotube. There was
rarely more than one secondary myotube associated
with each primary myotube (Fig. 2B). This level of
maturity is similar to that of normal lumbrical muscle
at E19 (Ross et al. 1987a).
Mononucleate cells lying within the basal lamina,
which may include presumptive fibroblasts as well as
myoblasts, were counted in the single midbelly crosssections of soleus and lumbrical muscles at E20.
There was no difference between the number of these
cells in the undernourished and control lumbrical
muscles (Fig. 3). In soleus muscles, there was a 27 %
Fig. 4. Muscle fibre numbers in lumbrical muscles from
control ( • ) , underfed ( • ) and rehabilitated « » rats.
Mothers of undernourished animals were on a restricted
diet from the 3rd day of gestation (E2) continuing
through lactation; mothers of rehabilitated animals were
restored to a normal diet on E21. Parturition occurred on
E22. Error bars, ±S.E.M.
diminution in mononucleate cell number in the
muscles from undernourished animals, but this was
barely statistically significant (0-05<P<0-1) and
matched the reduction in muscle fibre number.
When the restricted diet was continued during the
first week after birth there was a persisting deficit in
muscle fibre number in the pups (Fig. 1). At PN7,
control soleus muscles had 2789 ± 15 muscle fibres
(n = 3) and undernourished muscles had 2255 ± 54
(n = 5), a 19% reduction ( P < 0-001).
Undernourished lumbrical muscles at PN7 had
25 % fewer fibres than controls (727 ± 12, n = 5 versus
974 ±18, n = 8; P<0-001) (Fig. 4), and 28% fewer
mononucleate cells, matching the reduction in fibre
number.
Ventral root axon numbers
In order to see whether the reduction in secondary
myotube formation might be secondary to an increase
in motoneurone death, L4 ventral root axons were
counted at E20, PN7 and PN28. Promyelinated fibres
were counted at E20 and, with the data from one
animal omitted, there was no difference between
control and undernourished animals (2380 ± 30, n = 1
versus 2553 ± 26, n = 6; P > 0-25). The exception was
an undernourished fetus whose ventral roots on both
sides had about 30% more axons than any other,
including its siblings. These high numbers are comparable with those found in normal E18 fetuses (Ross
et al. 1987a) and may be due to delayed motoneurone
death in this animal. The muscle fibre counts from
Critical period in skeletal muscle development
this animal were not different from others in the
undernourished group.
At PN7, both myelinated and unmyelinated axons
were counted. The number of axons in undernourished animals (myelinated: 1192 ± 60, unmyelinated: 718 ±40, « = 5) was not different (/>>0-25)
from that in control animals (myelinated: 1160 ± 13,
unmyelinated: 810 ± 28, n = 6).
Nutritional rehabilitation
Body weights
When rats were undernourished during pregnancy
but given free access to food from E21 their young
had birth weights 31 % less than controls (19 %
greater than the continually undernourished group).
This deficit persisted through lactation.
Muscle fibre numbers
Muscles from animals in the rehabilitation group
were examined at PN28. Soleus muscles had the same
number of fibres as in muscles from underfed animals
examined at PN7, significantly less than controls
(Fig. 1). Thus, although rehabilitation was effective
in increasing body weight, it did not increase the
number of soleus muscle fibres able to be generated.
In PN28 lumbrical muscles, however, the number
of muscle fibres in muscles from rehabilitated animals, while still less than control (863 ±11, n = 8
versus 979 ±23, n = 6: P< 0-001), was significantly
greater than in the undernourished animals at PN7
(P< 0-001) (Fig. 4). Since the normal number of
primary myotubes had already been generated at E20
in both control and undernourished lumbrical
muscles, these extra fibres must have all been of
secondary myotube origin. During the period E20 to
PN28, the rehabilitated animals generated 740 secondary myotubes and the controls 792. Thus the
number of secondary myotubes generated during this
period was not substantially less in rehabilitated than
in control animals, but the initial fibre deficit was not
made up.
Ventral root axons
At PN28 the control L4 ventral roots contained
1950 ±112 myelinated axons (n = 6) and ventral roots
from rehabilitated animals contained 1778 ± 78 axons
(n = 10). These numbers are not significantly different (F>0-2). We also assayed subclasses of axon
according to fibre size and found no significant
differences.
Discussion
Our results show that production of secondary myotubes, but not of primary myotubes, is sensitive to
819
prenatal undernutrition. In addition, an experiment
with nutritional rehabilitation revealed that secondary myotube numbers could be recovered, but success
in this manoeuvre depended on the stage of development of the muscle. From these results, we suggest
that there is a critical period in development during
which the maximum number of adult muscle fibres is
determined.
These results confirm and extend previous observations. Although there have been many studies of
the effects of nutritional deprivation on muscle
growth and muscle fibre number, few of these have
involved undernourishment during the period of
muscle fibre generation. We severely undernourished
pregnant rats (30 % of normal food intake) from day
2 of gestation onwards, and examined their offspring
2 days before and one week after birth. The muscles
we studied, the soleus and IVth lumbrical, had
19-34 % deficits in fibre number, as determined by
counting every fibre in the muscle cross-section. This
deficit is comparable to that found by Bedi et al.
(1982) who counted soleus and extensor digitorum
longus muscle fibres (19% and 2 1 % reductions,
respectively) in the adult offspring of does undernourished during gestation and lactation.
By examining lumbrical muscles on E20, we show
that only secondary myotube generation is affected
by prenatal undernutrition. This differential response
was already suggested by Wigmore & Stickland
(1983) who in a semiquantitative light microscope
study of fetal pigs found that semitendinosus muscles
in the smallest animals had fewer secondary myotubes but the same number of primary myotubes as
those in their larger littermates. Fetal growth retardation is probably a consequence of prenatal malnutrition due to the position of the animal within the
uterine horn (McLaren, 1965).
A critical period for secondary myotube formation
Our most interesting finding is that there is a critical
period for initiating formation of secondary myotubes
and that this period may precede the actual generation of the myotube. Although only about 40 % of
soleus muscle secondary myotubes would normally
have formed by E21 (calculated by extrapolation),
refeeding undernourished animals from that time did
not correct the depression in their capacity to generate secondary myotubes. The lumbrical muscle is
developmentally about 1 day behind the soleus
muscle and only 19% of the secondary myotubes
have developed by E21 (Ross et al. 1987a). Refeeding
after previous undernutrition restored almost the full
myogenic capacity of this muscle, but there was no
compensation for myotubes not formed prior to
refeeding.
820
S. J. Wibon, J. J. Ross and A. J. Harris
Role of innervation in regulating muscle fibre
numbers
Evidence from experiments on frogs (Ferns & Lamb,
1987) and rats (Ross et al. 1987£>) shows a stoichiometric relation between muscle innervation and secondary myotube generation. Accordingly, it was
important to ask whether undernutrition affected
muscle innervation. We found no evidence for excess
motoneurone death in undernourished animals. We
cannot, however, comment on whether undernourished motoneurones formed fewer than their
normal number of terminals, which in normal development form in advance of the muscle fibres they will
eventually innervate (Harris & McCaig, 1984).
Mechanism of the effect of undernutrition
Fairly severe undernutrition was required before we
saw any effect on muscle fibre numbers. In a preliminary experiment, holding animals at constant weight
during pregnancy had no effect on muscle fibre
numbers in their offspring. Beerman (1983) found a
reduction to 50 % of normal food intake from the 8th
day of pregnancy onwards in rats did not affect
muscle fibre number in the offspring, although it had
a long-lasting effect on muscle growth.
We also did not see the marked reduction in
mononucleate cell numbers in the midbelly region of
the muscle that accompanies the halt in secondary
myotube production produced by fetal denervation
(Ross et al. 19876). The reduction we saw after fetal
malnutrition was such that the ratio of mononucleate
cells to total fibre number remained constant. This
implies that the rates of myoblast proliferation and
fusion are both reduced in the muscles of the undernourished animals, even if the number of myoblasts
fusing to form each myotube remained constant
(Penney et al. 1983; McLennan, 1987).
As the number of secondary myotubes that form in
a muscle is well regulated (Ross et al. 1987a) and very
much less than the number of fusion-competent cells
that are present throughout development, it appears
that secondary myotube formation is initiated by a
special cell fusion event associated with the presence
of a nerve terminal (Ross et al. 1987i»). Once the
nascent secondary myotube exists, it can grow by
accepting fusion from myoblasts that come in contact
with it. It is likely that the effect of undernutrition is,
directly or indirectly, to reduce the number of the
subpopulation of myogenic cells which participate in
the initiating event, thereby permanently limiting
secondary myotube numbers.
This work was supported by the New Zealand Medical
Research Council, the Vernon Willey Trust and the Muscular Dystrophy Association. We thank K. S. Bedi for advice
on the regime for undernourishment.
References
BEDI, K. S., BIRZGALIS, A. R., MAHON, M., SMART, J. L.
& WAREHAM, A. C. (1982). Early life undernutrition in
rats. 1. Quantitative histology of skeletal muscles from
underfed young and refed adult animals. Br. J. Nutr.
47,417-431.
BEERMAN, D. H. (1983). Effects of maternal dietary
restriction during gestation and lactation, muscle, sex
and age on various indices of skeletal muscle growth in
the rat. J. Anim. Sci. 57, 328-337.
BYRNE, I., HOOPER, J. C. & MCCARTHY, J. C. (1973).
Effects of selection for body size on the weight and
cellular structure of seven mouse muscles. Anim. Prod.
17, 187-1%.
FERNS, M. J. & LAMB, A. H. (1987). Regulation of cell
numbers in the developing neuromuscular system in
Xenopus laevis. Neurosci. Letts Supplement 27, 572.
GOLDSPINK, G. & WARD, P. S. (1979). Changes in rodent
muscle fibre types during postnatal growth,
undernutrition and exercise. J. Physiol. 296, 453-469.
HARRIS, A. J. (1981). Embryonic growth and innervation
of rat skeletal muscles. 1. Neural regulation of muscle
fibre numbers. Phil. Trans. Roy. Soc. Lond. B 293,
257-277.
HARRIS, A. J. & MCCAIG, C. D. (1984). Motoneuron
death and motor unit size during embryonic
development in the rat. J. Neurosci. 4, 13-24.
HEGARTY, P. V. J. & ALLEN, C. E. (1978). Effect of
prenatal runting on the postnatal development of
skeletal muscles in swine and rats. J. Anim. Sci. 46,
1634-1640.
KELLY, A. M. (1983). Emergence of specialization of
skeletal muscle. In Handbook of Physiology (ed. L. D.
Peachey), Section 10, pp. 507-537. Baltimore: Williams
and Wilkins.
KELLY, A. M. & ZACKS, S. I. (1969). The histogenesis of
rat intercostal muscle. /. Cell Biol. 42, 135-153.
MCLAREN, A. (1965). Genetic and environmental effects
on foetal and placental growth in mice. J. Reprod.
Fertil. 9, 79-98.
MCLENNAN, I. S. (1983). Neural dependence and
independence of myotube production in chicken
hindlimb muscles. Devi Biol. 98, 287-294.
MCLENNAN, I. S. (1987). Hormonal regulation of
myoblast proliferation and myotube production in vivo:
Influence of prostaglandins. J. exp. Zool. 241, 237-245.
MONTGOMERY, R. D. (1962). Muscle morphology in
infantile protein malnutrition. J. Clin. Path. 15,
511-521.
ONTELL, M. & DUNN, R. F. (1978). Neonatal muscle
growth: a quantitative study. Am. J. Anal. 152,
539-556.
PENNEY, R. K., PRENTIS, P. F., MARSHALL, P. A. &
GOLDSPINK, G. (1983). Differentiation of muscle and
the determination of ultimate tissue size. Cell Tissue
Res. 228, 375-388.
Ross, J. J., DUXSON, M. J. & HARRIS, A. J. (1987a).
Formation of primary and secondary myotubes in rat
lumbrical muscles. Development 100, 383-394.
Critical period in skeletal muscle development
Ross, J. J., DUXSON, M. J. & HARRIS, A. J. (19876).
Neural determination of muscle fibre numbers in
embryonic rat lumbrical muscles. Development 100,
395-409.
STICKLAND, N. C. & HANDEL, S. E. (1986). The numbers
and types of muscle fibres in large and small breeds of
pigs. J. Anat. 147, 181-189.
821
deficiencies on the fibres and nuclei in skeletal muscle
of pigs. Brit. J. Nutr. 34, 421-428.
SWATLAND, H. J. & CASSENS, R. G. (1973). Inhibition of
muscle growth in foetal sheep. /. Agric. Sci., Camb.
80, 503-509.
WIGMORE, P. M. C. & STICKLAND, N. C. (1983). Muscle
development in large and small pig foetuses. J. Anat.
137, 235-245.
STICKLAND, N. C , WIDDOWSON, E. M. & GOLDSPINK, G.
(1975). Effects of severe energy and protein
{Accepted 19 January 1988)