BIOLOGY OF REPRODUCTION 73, 773–780 (2005)
Published online before print 22 June 2005.
DOI 10.1095/biolreprod.105.040006
Expression and Localization of Alpha-Smooth Muscle and Gamma-Actins
in the Pregnant Rat Myometrium1
Oksana Shynlova,3 Prudence Tsui,3,4 Anna Dorogin,3 Michelle Chow,3,4 and Stephen J. Lye2,3,4,5
Samuel Lunenfeld Research Institute,3 Mount Sinai Hospital, Toronto, Ontario, Canada M5G 1X5
Departments of Physiology4 and Obstetrics and Gynecology,5 University of Toronto, Toronto, Ontario, Canada M5S 1A1
ABSTRACT
and the cytoskeletal domains of smooth muscle cells [1].
Vertebrates express at least six different actin isoforms in
a tissue-specific manner [2]. Many organisms synthesize
multiple actin isoforms even within the same cell [3]. Differentiated SMCs typically contain a mixture of muscle (a
and g) and cytoplasmic (b and g) actin isoforms [4]. The
muscle isoforms are associated with the contractile filaments, whereas the cytoplasmic isoforms form the noncontractile cytoskeleton. The amount of different actin isoforms present in the muscle cells varies according to the
source of the smooth muscle tissue. The dominant type of
actin in the smooth muscle of mammalian visceral organs
(chicken gizzard, uterus, intestine) is g-actin, whereas the
major actin type in vascular smooth muscle is a-actin [2].
Smooth muscle a-actin is also the first known marker of
differentiated SMCs that is upregulated during vasculogenesis (reviewed in [5]). In contrast, in primary cultured vascular SMCs the expression of this actin isoform decreases
with dedifferentiation [6]. It has been recently shown that
visceral SMCs display an opposite pattern to that in vascular SMCs depending on the differentiation state. In differentiated visceral SMCs, the expression of a-actin is decreased whereas the expression of g-actin increased [7, 8],
as has been shown to occur in the pregnant (compared to
nonpregnant) human myometrium [9].
Uterine smooth muscle undergoes dramatic physiological adaptations during the course of pregnancy. Under the
influence of high progesterone levels, mechanical tension
induces myometrial growth and remodeling, which results
in myocyte hypertrophy and an increased synthesis of interstitial matrix proteins [10]. The onset of labor requires
triggering of the endocrine (the fetal hypothalamic-pituitary-adrenal-placental axis) and mechanical (stretch resulting from the fetal growth during pregnancy) signaling pathways within the myometrium to activate the synchronized
and time-coordinated labor contractions of the uterine
smooth muscle. The involvement of mechanical signals in
the process of myometrial activation at the onset of labor
requires a mechanism by which uterine myocytes can sense
the mechanical signals. One potential mechanosensor of the
hypertrophied myometrial cell is the cytoskeleton, which is
built from three types of biopolymers, including the actin
microfilaments, intermediate filaments, and microtubles
[11]. There are many studies on the physiological changes
in the pregnant uterus [10], however, surprisingly, there is
relatively little information about the expression and localization of contractile proteins.
Our previous data has suggested a role for focal adhesion
kinase (FAK), also known as protein tyrosine kinase 2, in
the onset of labor [12]. Activation of FAK causes the recruitment and formation of the focal adhesion–cytoskeleton
complex. We speculate that myometrial growth during
The myometrium undergoes dramatic changes as pregnancy
progresses through phases of proliferation, hypertrophy, contractile state and labor. In this study, we showed that the composition of the muscle actin isoforms, a major component of the
myometrial contractile apparatus and cytoskeleton, was modified during pregnancy. The expression of smooth muscle alphaactin (Acta2, which we abbreviate as alpha-SM-actin) and gamma-actin mRNAs and proteins was examined by real-time polymerase chain reaction and Western immunoblot, and was localized with immunohistochemistry, in the nonpregnant,
pregnant, and postpartum rat uterus. Both alpha-SM-actin (vascular specific actin isoform) and gamma-actin (predominant in
visceral smooth muscle) were detected in the rat myometrium.
Myometrial expression of alpha-SM-actin mRNA and protein
was high throughout pregnancy. The transcript and protein levels
of gamma-actin were increased significantly in the second part
of gestation (31.8-fold increase for mRNA and 16.7-fold increase
for protein relative to nonpregnant). The localization of gammaactin was markedly altered during pregnancy. In early gestation,
myometria from empty and gravid uterine horns of the unilaterally pregnant rats showed abundant gamma-actin immunostaining in the longitudinal layer but weak staining in the circular layer. Gamma-actin immunostaining increased in only the
circular layer of the gravid horn after midgestation and remained low in the empty one. Gamma-actin protein translocated to the membranous region of uterine myocytes at late gestation. The temporal alteration in gamma-actin expression and
localization at late gestation suggested that this change in myometrial composition of contractile proteins is important to adequately prepare the myometrium for the development of optimal contractions during labor.
parturition, pregnancy, uterus
INTRODUCTION
The molecular structure and organization of the contractile apparatus in smooth muscle enables force generation
through the relative sliding of adjacent actin and myosin
filaments. Smooth muscle cells (SMCs) also contain an extensive cytoskeleton formed by multiple sets of intermediate filaments that interact with the contractile domains of
the cell. Actin is a major component of both the contractile
Supported by a grant from the Canadian Institutes of Health Research.
Correspondence: Stephen J. Lye, Samuel Lunenfeld Research Institute at
Mount Sinai, 600 University Avenue, Suite 870, Toronto, Ontario, Canada
M5G 1X5. FAX: 416 586 8740; e-mail: [email protected]
1
2
Received: 19 January 2005.
First decision: 3 February 2005.
Accepted: 20 June 2005.
Q 2005 by the Society for the Study of Reproduction, Inc.
ISSN: 0006-3363. http://www.biolreprod.org
773
774
SHYNLOVA ET AL.
pregnancy is associated with alteration in the structure and
contractile apparatus characteristic of the uterine smooth
muscles. Because actin filaments terminate at focal adhesions and because the actin cytoskeleton is thought to play
a role in mechanotransduction, we sought to determine
whether pregnancy affects the expression and localization
of different actin isoforms in the myometrium. In the present study we used a reverse transcription-polymerase chain
reaction (RT-PCR) technique, immunoblotting and standard
immunohistological staining to investigate changes in two
major smooth muscle actin isoforms (a and g) in the nonpregnant, pregnant, and postpartum rat myometrium. In addition, the effect of gravidity was investigated on the expression and localization of actin using a unilateral tuballigation rat model.
According to the Rat Genome Database, the standard
genetic symbols for rat smooth muscle alpha-actin and
gamma 2-actin (smooth muscle) are Acta2 and Actg2, respectively. Because of the similarity of these two symbols,
we will use a-SM-actin and g-SM-actin throughout this paper to avoid confusion to readers.
MATERIALS AND METHODS
Animals
Wistar rats (Charles River Co., St. Constance, PQ, Canada) were
housed individually under standard environmental conditions (12L:12D
cycle) and fed Purina Rat Chow (Ralston Purina, St. Louis, MO) and water
ad libitum. Female virgin rats were mated with male Wistar rats. Day 1
of gestation was designated as the day a vaginal plug was observed. The
average time of delivery was the morning of Day 23. The Samuel Lunenfeld Research Institute Animal Care Committee approved all animal experiments.
Experimental Design
Normal pregnancy and term labor. Animals were killed by carbon
dioxide inhalation and myometrial samples were collected on Gestational
Days 0 (nonpregnant [NP]), 6, 8, 10, 12, 14, 15, 17, 19, 21, 22, and 23
(day of labor), and on Days 1 and 4 Postpartum (PP). Tissue was collected
at 1200 h on all days with the exception of the labor (d23L) sample, which
was collected once the animals had delivered at least one pup (n 5 4).
Unilaterally pregnant rats. Under general anesthesia, virgin female rats
underwent tubal ligation through a flank incision to ensure that they subsequently became pregnant in only one horn [13]. Animals were allowed
to recover from surgery for at least 7 days before mating. Pregnant myometrial samples from empty and gravid horns were collected on Days 15,
17, 19, 21, 22, 23 or 1PP (n 5 3).
Tissue Collection
For RNA and protein extraction the uterine horns were placed into icecold PBS, bisected longitudinally, and dissected away from both pups and
placentas. The endometrium was carefully removed from the myometrial
tissue by mechanical scraping on ice, which we have previously shown to
remove the entire luminal epithelium and the majority of the uterine stroma [14]. The myometrial tissue was flash-frozen in liquid nitrogen and
stored at 2708C. For immunohistochemical studies the intact uterine horns
were placed in ice-cold PBS buffer, cut into 3–10-mm segments using a
scalpel blade, and fixed immediately in 4% paraformaldehyde solution at
48C for 48 h. These segments were further cross-sectioned or sectioned
longitudinally. For each day of gestation, tissue was collected from 3–4
different animals and processed for different experimental purposes.
Real-Time PCR Analysis
Total RNA was extracted from the frozen tissues using TRIZOL (Gibco BRL, Burlington, ON, Canada) according to the manufacturer’s instructions. RNA samples were column-purified using RNeasy Mini Kit (Qiagen, Mississauga, ON, Canada), and treated with 2.5 ml DNase I (2.73
Kunitz unit/ml; Qiagen) to remove genomic DNA contamination. Reverse
transcription (RT) and real-time PCR was performed to detect the mRNA
expression of a-SM-actin and g-SM-actin in rat myometrium. Two mg of
total RNA was primed with random hexamers to synthesize single-strand
cDNAs in a total reaction volume of 100 ml using the TaqMan Reverse
Transcription Kit (Applied Biosystems, Foster City, CA). The thermal cycling parameters of real-time were modified according to the Applied Biosystems manual. Hexamer incubation at 258C for 10 min and RT at 428C
for 30 min was followed by reverse transcriptase inactivation at 958C for
5 min. Twenty mg of cDNA from the previous step was subjected to realtime PCR using specific sets of primers (see the legends to Figs. 1 and 2)
in a total reaction volume of 25 ml (Applied Biosystems). RT-PCR was
performed in an optical 96-well plate with an ABI PRISM 7900 HT Sequence Detection System (Applied Biosystems), using the SYBR Green
detection chemistry. The run protocol was as follows: initial denaturation
stage at 958C for 10 min, 40 cycles of amplification at 958C for 15 sec
and 608C for 1 min. After PCR, a dissociation curve was constructed by
increasing temperature from 658C to 958C for detection of PCR product
specificity. In addition, a no-template control (H2O control) was analyzed
for possible contamination in the master mix. A cycle threshold (Ct) value
was recorded for each sample. PCR reactions were set up in triplicates
and the mean of the 3 Cts was calculated. Relative quantitation of gene
expression was the approach used to compare differences of gene expression across gestation. An arithmetic formula from the comparative Ct
method (see ABI User Bulletin No. 2 available at http://dna-9.intmed.uiowa.edu/RealtimePCRdocs/Compar Anal Bulletin2.pdf) was applied to the raw Ct values to extract relative gene expression data. The
mRNA level from each sample was normalized to ribosomal 18S mRNA.
Validation experiments were performed to ensure that the PCR efficiencies
between the target genes and 18S were approximately equal.
Western Immunoblot Analysis
Frozen myometrial tissue was crushed under liquid nitrogen using a
mortar and pestle. Crushed tissue was homogenized for 1 min in RIPA
lysis buffer [50 mM Tris-HCl, pH 7.5, 150 mM NaCl, 1% (vol/vol) Triton
X-100, 1% (vol/vol) sodium deoxycholate, and 0.1% (wt/vol) SDS, supplemented with 100 mM sodium orthovanadate and protease inhibitor
cocktail tablets (CompleteTM Mini; Roche, Laval, QC, Canada). Samples
were spun at 12 000 3 g for 15 min at 48C and the supernatant was
transferred to a fresh tube to obtain a crude protein lysate. Protein concentrations were determined using the BioRad protein assay buffer
(BioRad, Hercules, CA). Protein samples (40–50 mg) were resolved by
electrophoresis on a 12–15% SDS-polyacrylamide gel. Proteins were
transferred onto polyvinylidene difluoride (PVDF) membrane (Millipore,
Bedford, MA) in 25 mM Tris-HCl, 250 mM glycine, 0.1% (wt/vol) SDS,
pH 8.3, for 18 h at 30 V at 48C. Protein expression levels of a and g
actins were measured by Western analysis using anti-SM a-actin (1:1000,
Clone 1A4; DAKO Diagnostics, Mississauga, ON, Canada) mouse and
anti-g-actin (1:1000; Chemicon International, Temecula, CA) sheep primary antibody. PVDF membranes were stripped and reprobed with antismooth muscle calponin (1:3000, clone hCP; Sigma-Aldrich, Oakville,
ON, Canada) mouse primary antibodies to control the loading variations.
Probed membranes were exposed to x-ray film (Kodak XAR; Eastman
Kodak, Rochester, NY) and analyzed by densitometry. Calponin is constitutively expressed in nonpregnant and pregnant rat tissue under the same
protein extraction conditions [15].
Immunohistochemistry
The fixed myometrial tissues were gradually dehydrated in ethanol and
embedded in paraffin. Sections of 5 mm thickness were collected on Superfrost Plus slides (Fisher Scientific Ltd., Nepean, ON, Canada). Paraffin
sections were deparaffinized and rehydrated. After immersion in 3% hydrogen peroxide (Fisher Scientific, Fair Lawn, NJ) the antigen retrieval
was performed using 0.125% trypsin solution at RT for 15 min, followed
by blocking with 5% normal horse serum (for a-SM actin) or DAKO
Protein Serum-Free Blocking solution (for g-actin; DAKO Corporation,
Carpinteria, CA) and incubation with primary antibodies overnight at 48C.
Primary antibodies used to label different actin isoforms were: sheep antisera against g actin (1:6000; Chemicon International) and mouse antisera against a-SM actin (1:50, Clone 1A4; Dako Diagnostics). For the
negative controls, ChromPure nonspecific mouse or sheep IgGs (Jackson
ImmunoResearch Laboratories, Inc., West Grove, PA) were used at the
same concentration, and sections were also incubated with secondary antibodies in the absence of primary antibodies. Secondary antibodies used
for detection were biotinylated anti-mouse (1:200; Vector, Burlingame,
CA) and donkey anti-sheep-HRP-conjugated (1:1000; Serotec, Oxford,
UK). Final visualization was achieved using Vectastain Elite Kit (Vector,
Burlingame, CA). Counterstaining with Harris’ Hematoxylin (Sigma Di-
ACTIN ISOFORM PATTERN IN PREGNANT MYOMETRIUM
FIG. 1. The effect of gestational age on the expression of g-actin (A) and
SM-a-actin (B) in pregnant rat myometrium. Total RNAs were extracted
from frozen myometrial tissues, single-strand cDNAs were synthesized as
described in Materials and Methods, and actin mRNA levels were analyzed on the indicated days of gestation by real-time PCR. Specific forward and reverse primers were designed using Primer Express software,
version 2.0.0 (Applied Biosystems), as follows: g-actin mRNA, 59-GGCATGGAGTCAGCTGGAATT-39 (sense primer) and 59-TCTGCATCCTGTCA
GCAATCC-39 (antisense primer) (GenBank accession no. NM 012893);
a-actin mRNA, 59-CGGGCTTTGCTGGTGATG-39 (sense primer) and 59CCCTCGATGGATGGGAAA-39 (antisense primer) (GenBank accession
no. J02781 M16997); 18S, 59-GCGAAAGCATTTGCCAGAA-39 (sense
primer) and 5-9GGCATCGTTTATGGTCGGAAC-39 (antisense primer)
(GenBank accession no. X01117 K01593). Levels of g-actin (A) and SMa-actin (B) mRNA (in relative fold changes) were normalized to 18S
mRNAs. Bars represent mean 6 SEM (n 5 4 at each time point). Data
labeled with different letters are significantly different from each other (P
, 0.05).
agnostics, St. Louis, MO) was carried out before slides were mounted with
Cytoseal XYL (Ricard-Allan Scientific, Kalamazoo, MI). Myometrial cells
from each of the three tissue sets were observed on a Leica DMRXE
microscope (Leica Microsystems, Richmond Hill, ON, Canada). A minimum of five fields were examined for each gestational day and uterine
horn for each set of tissue, and representative tissue sections were photographed with Sony DXC-970 MD (Sony Ltd., Toronto, ON, Canada)
3CCD color video camera.
Statistical Analysis
Gestational profiles were subjected to a one-way analysis of variance
(ANOVA) followed by pairwise multiple comparison procedures (StudentNewman-Keuls method) to determine differences between groups. Tubal
ligation data were analyzed by two-way ANOVA followed by pairwise
multiple comparison procedures as described above. Where required the
data was transformed by the appropriate method to obtain a normal distribution. Statistical analysis was carried out using SigmaStat version 2.03
(Jandel Corp., San Rafael, CA) with the level of significance for comparison set at P , 0.05.
775
FIG. 2. The effects of gestational age on the expression of g -actin and
SM-a-actin proteins in the pregnant rat myometrium. Representative
Western blots and densitometric analysis of g-actin (A) and a-SM-actin
(B) protein levels throughout normal gestation and postpartum are shown.
Actin protein expression levels were normalized to calponin. Bar graphs
show the mean 6 SEM relative optical density (n 5 3–4 at each time
point). Data labelled with different letters are significantly different from
each other (P , 0.05).
RESULTS
The abundance of the g-smooth muscle actin (g-SMactin) and a-SM-actin gene transcripts in the myometrium
throughout pregnancy relative to NP is shown in Figure 1,
A and B. Pregnancy was associated with a significant increase in g-actin mRNA abundance (8.83 fold increase at
Day 6, P , 0.001) as compared to NP mRNA levels. There
was a further significant (P , 0.001) increase in g-actin
gene expression in the second half of gestation (14.89-fold
increase at Day 14 relative to NP), which peaked at Day
19 (31.78-fold increase relative to NP, P , 0.001). On the
day of labor (Day 23) g-actin levels were slightly decreased
(15.21-fold increase relative to NP, P , 0.001). During the
postpartum period (1PP), g-actin gene expression returned
to nonpregnant levels (Fig. 1A). In contrast, no statistical
difference was found in mRNA levels for a-SM-actin gene
among gestational samples (Fig. 1B).
Western immunoblot analysis was performed using antig-actin antibody that recognized protein of both forms of
g-actin (smooth muscle and cytoplasmic) and anti-a-actin
antibody that recognized only smooth muscle-specific aactin isoform (Fig. 2, A and B). Both proteins were detected
as single bands of the same size (42 kDa) on the Western
blots; therefore, analysis of g- and a-actin isoforms was
performed using different PVDF membranes to avoid crossreactivity between antibodies. Although the expression of
a-SM-actin did not change across gestation (Fig. 2B), statistically significant changes were detected in the expression of g-actin protein in pregnant myometrium. After nor-
776
SHYNLOVA ET AL.
FIG. 3. Expression of g-actin gene in the myometrium of unilaterally
pregnant rats during late gestation. Levels of g-actin mRNA were analyzed
on the indicated days of gestation by real-time PCR. Specific forward and
reverse primers were designed using Primer Express software, version
2.0.0 (Applied Biosystems), as follows: g-actin mRNA, 59-CCTTCCTTCA
TTGGCATGGA-39 (sense primer) and 59-TCTGCATCCTGTCAGCAATCC-39
(antisense primer) (GenBank accession no. NM 012893); 18S—see Figure 1 legend. Levels of g-actin gene expression in empty (white bars) and
gravid (black bars) uterine horns (in relative fold changes) were normalized to 18S mRNAs. Bars represent mean 6 SD (n 5 3 at each time point).
A significant difference between gravid and empty horn of the same gestational day is indicated by * (P , 0.05).
malization to calponin, g-actin protein levels increased
10.2-fold starting from Day 17 (P , 0.001) until Day 23
of gestation (16.7-fold increase relative to NP, P , 0.001),
followed by a decrease postpartum (Fig. 2A). Thus, the
pattern of g-actin protein expression measured by immunoblotting was similar to the pattern of mRNA expression
measured by real-time PCR (Figs. 1A and 2A).
To determine whether g-actin gene expression was affected by the stretch of the myometrium during late pregnancy, real-time PCR reaction was performed using RNA
isolated from empty and gravid uterine horns of unilaterally
pregnant rats. Expression of the g-actin gene in the empty
horn remained relatively constant at late gestation. Our data
(Fig. 3) indicated that the increase in g-actin gene expression during late pregnancy was significantly greater in gravid horns than in nongravid horns (P , 0.05). Interestingly,
the expression of g-actin mRNA was dramatically decreased on Day 1 PP in both empty and gravid horns, compared to midpregnant levels. The gravid horn mRNA profile
of unilaterally pregnant rats agreed with that of normal
pregnant rats at late gestation (Figs. 1A and 3). In contrast
to the significantly elevated expression of g-actin mRNA
in the gravid horn, protein levels were only slightly increased and did not reach statistical significance (data not
shown).
Immunohistochemical staining for g-actin and a-SM actin revealed changes in the localization and the temporal
distribution of actins in rat myometrium at various stages
of pregnancy (Figs. 4–7). Immunostaining of g- and a-SM
actin was detected in both (longitudinal and circular) muscle layers of the gravid horn myometrium. Strong immunoreactivity of g-actin in SMCs in the longitudinal layer
was observed throughout gestation in all myometrial samples examined and no marked changes were detected. In
contrast, in the circular muscle layer of gravid rat uterus
we observed a temporal alteration of g-actin staining. Immunostaining of g-actin in circular uterine muscle layer of
nonpregnant and early pregnant (Days 6–10) animals was
extremely weak compared with that in the longitudinal layer (Fig. 4, A and B). However, starting from midgestation,
FIG. 4. Temporal alterations of g-actin immunolocalization in pregnant
rat myometrium during gestation. Immunohistochemical examination was
performed on sections of uterus from nonpregnant (A), Day 8 (B), Day 14
(C), Day 18 (D), and Day 21 (E) of pregnancy and Day 4 PP (F) animals.
Intense immunoreactivity of g-actin was detected in longitudinal SMCs
throughout gestation (A–F, arrowheads). Relatively weak immunostaining
was present in circular myometrial layers of NP (A) and early pregnant
(B) animals. Staining of g-actin in circular myometrial layer was significantly elevated during late gestation (Days 18–21, D–E, large arrows) but
by Day 4 PP (F) it had begun to disappear, with expression of g-actin
being lower than in the longitudinal muscle layer. Note the lack of staining after incubation of myometrial tissue with nonspecific sheep IgGs
(Day 14, G) or with anti-sheep secondary antibodies in the absence of
primary antibody (Day 8, H). Original magnification 3200. Bar 5 50 mm
(A–E).
g-actin levels in the circular myometrial layer were dramatically upregulated (Fig. 4, C–E). This high level of protein expression in circular uterine muscle was maintained
until labor (Day 23; Fig. 4E), and then declined by 4 days
postpartum to almost undetectable levels (Fig. 4F). As with
the PCR and Western blotting data, these changes were primarily limited to the gravid horns of unilaterally pregnant
rats. Expression of g-actin was very low in the circular
muscle layers of empty uterine horns at early pregnancy
and no increase was observed at mid- to late gestation (Fig.
5). These data demonstrate differential regulation of g-actin
expression between the muscle layers that would result in
ACTIN ISOFORM PATTERN IN PREGNANT MYOMETRIUM
777
FIG. 6. Spatial distribution of g-actin in pregnant rat myometrium during
gestation. In the gravid uterine horn, localization of g-actin is diffuse within the cytoplasm at early and midgestation. By Day 18 (A) g-actin staining
of some SMCs in the gravid horn myometrium formed a continuous ring
of staining that localized to the area under the cell membrane. This localization was pronounced by Days 21 (B) and 23 (C) but by Day 1 PP
(D) g-actin again resumed the diffuse cytoplasmic staining that was present during early gestation to midgestation. Original magnification 3400.
Bar 5 25 mm.
nancy and did not show any gestational changes (Fig. 7).
As with g-actin, a-SM-actin immunostaining was present
exclusively in the cytoplasm of uterine myocytes throughout pregnancy.
DISCUSSION
FIG. 5. Localization of g-actin within the longitudinal SMCs of the nongravid myometrium from unilaterally pregnant rats during gestation. Uterine tissue was collected from empty horns of unilaterally pregnant rats on
different gestational days, including Day 6 (A), Day 12 (B), Day 15 (C),
Day 19 (D), Day 21 (E), and Day 1 PP (F). Tissues were labelled with antig-actin antibody and light microscopy images of cross-sections were collected. Localization of g-actin was observed mainly within the cytoplasm
of longitudinal muscle SMCs of the empty uterine horn and was much
less abundant in the circular muscle layer. No gestational changes in
localization were observed. The lack of staining (Day 21 empty horn myometrium) after incubation of myometrial tissue with nonspecific sheep
IgGs (G) or with anti-sheep secondary antibodies in the absence of primary antibody is shown in H. Original magnification 3400. Bar 5
25 mm.
enhancement of contractile activity in the circular muscle
of the gravid horn during late pregnancy. The spatial distribution of g-actin immunoreactivity within myometrial
SMCs was similar in both uterine muscle layers. Positive
signals for g-actin were detected throughout the cytoplasm
and in the regions closely associated with plasma membrane of myometrial SMCs (Figs. 4 and 6). At early pregnancy and midpregnancy, the distribution of g-actin was
rather diffuse. In contrast, during late pregnancy, significant
amount of g-actin was found to concentrate close to the
cell membrane making a continuous ring of actin staining
localized around the SMC periphery (Fig. 6).
Expression of a-SM-actin was detected in both (longitudinal and circular) rat myometrial layers throughout preg-
Our data demonstrated differential expression of two
major actin isoforms in the rat myometrium during pregnancy. We found that g-actin (the major isoform of visceral
muscle) is markedly upregulated during mid and late pregnancy, but that expression of a-actin (the major actin isoform in vascular smooth muscle) is relatively unchanged
during pregnancy. In addition, the increase in g-actin during
late pregnancy was limited to the circular muscle layer of
the gravid horn, whereas this isoform was highly expressed
in longitudinal muscle throughout pregnancy in both gravid
and nongravid horns. Together these data indicate a complex spatial and temporal regulation of actin isoforms within the myometrium and suggest that the elevated expression
of g-actin in the circular SMCs contributes to the ability of
this layer to generate optimal contractile activity during labor.
The suggestion that g-actin is the major myometrial actin
isoform during pregnancy is supported by the report of Cavaille [16], who found that total g-actin protein was increased in human myometrial samples at term compared to
samples from nonpregnant uterus. Two types of g-actin,
smooth muscle and cytoplasmic, have been reported based
on amino acid [2] and cDNA sequencing [17]. Whereas
Cavaille did not distinguish between these isoforms, our
mRNA analysis (which used primers specific for the
smooth muscle form) suggests that it is the smooth muscle
isoform that is predominantly increased in the late pregnant
rat myometrium. This may in part explain the limited
changes in protein expression found by this study, in that
778
SHYNLOVA ET AL.
FIG. 7. Immunolocalization of a-SM-actin in pregnant rat myometrium
throughout gestation. Immunohistochemical examination was performed
on sections of uterus from these phases of gestation: nonpregnant (A),
early gestation (B, Day 6), late gestation (C, Day 18), and labor (D, Day
23). A high expression level of a-SM-actin in both (longitudinal and circular) myometrial layers was detected during rat pregnancy. The lack of
staining (Day 6 of pregnancy) after incubation of myometrial tissue with
nonspecific sheep IgGs (E) or with anti-sheep secondary antibodies in the
absence of primary antibody (F). Original magnification 3200. Bar 5
50 mm.
the antibody used does not differentiate between the two
isoforms of g-actin.
This observation suggests that the mRNA and protein
data may not be directly comparable. Whereas the primers
for PCR recognize only smooth-muscle-specific g-actin the
antibodies for Western blotting recognized both smoothmuscle and cytoplasmic isoforms. We would suggest that
the cytoplasmic g-actin, that is a component of stable myometrial cytoskeletal stress fibrils, may represent a part of
total g-actin protein and, therefore, may account for the
maintained high protein expression in postpartum tissues.
Moreover, studies of aortic smooth muscle documented a
switch of actin isoform from a-smooth muscle actin in normal quiescent state to nonmuscle actins (including cytoplasmic g-actin) during injury and tissue repair. These led
us to believe that cytoplasmic g-actin protein might be expressed higher in the postpartum myometrium when there
is tremendous tissue involution and repair (reviewed in
[18]).
Although g-actin expression was already elevated in early pregnancy, the predominant increase was observed in
mid- to late pregnancy, a time when the myometrial cells
have been shown to exhibit a hypertrophic phenotype. A
similar positive relationship between hypertrophy and increased expression of g-actin has been reported in bladder
(visceral) [8] and in hepatic portal vein (vascular) muscle
[19]. Importantly, as in our analysis of the myometrium,
these authors found no increase in expression of a-SMactin in association with muscle hypertrophy. This linkage
of actin isoform expression with the differentiation state of
smooth muscle has been reported widely in other muscles,
with the differentiated state being associated with increased
g-actin and decreased a-actin and the dedifferentiated state
associated with a reversal of this pattern of actin expression
[6]. Moreover, when mice deficient in cardiac-specific aactin were rescued by the ectopic expression of visceral SM
g-actin, their hearts were enlarged and hypertrophied [20].
Together these results strongly suggest that increase in gactin contributes to myometrial smooth muscle hypertrophy
during pregnancy.
Our previous in vivo data in unilaterally pregnant rats
suggest that myometrial hypertrophy (increase in protein:
DNA ratio) during mid- to late pregnancy is controlled by
mechanical (because of stretch of the uterine wall by the
growing fetus) and hormonal (progesterone) signals [21].
Stretch has also been reported to increase hypertrophy of
the nonpregnant [22] and to increase protein synthesis in
the pregnant uterus [23]. Our finding that the increase in gactin occurred predominantly in the gravid uterine horn
during midpregnancy suggested that stretch played a more
important role in controlling the expression of this actin
isoform than did elevated levels of progesterone, because
there was no significant change in g-actin levels of empty
horns. Interestingly, mechanical stretch has been reported
to decrease the expression of a-SM actin mRNA and protein [24] and to induce degradation of the a-SM actin filaments in vascular smooth muscle cells [25]. This is consistent with our observation that expression of this isoform
does not increase (and may even decrease) when labor approaches. Other studies [26, 27] have reported that estrogen
can induce increased expression of cytoplasmic actin isoforms (b- and g-actin) in cultured uterine smooth muscle
cells. It is possible that some of the increase in g-actin
mRNA (but not protein) expression observed in early pregnancy may be caused by the elevated levels of estrogen
present at this time.
The myometrium is the muscular part of the uterus that
is responsible for generating the force to expel the fetus at
term. It is composed of three well-defined layers. The circumferentially oriented SMCs of the circular muscle of the
myometrium surround the endometrium. The SMCs of the
outer longitudinal muscle layer of the myometrium align
along the long axis of the uterus. Separating these two muscle layers is the vascular plexus, which is an extensive network of blood vessels that nourish the muscle layers. In
order to investigate the distribution of g-actin in the myometrial tissue throughout gestation, we have analyzed all
three layers of myometrium. The differential expression of
g-actin in circular and longitudinal uterine smooth muscle
is intriguing. During pregnancy the uterus initially grows
longitudinally, with circumferential growth occurring from
midpregnancy onwards in association with fetal growth. It
would be expected that tension on the uterine wall at this
time would be predominantly on the circular smooth muscle layer. Such data provide additional support for a role of
mechanical signals in the expression of g-actin. These muscle layers exhibit quite different phenotypes during pregnancy. Our previous studies have shown expression of the
gap junction protein, connexin43 (also known as gap junction membrane channel protein alpha 1), is limited to uterine circular muscle [28], whereas others have reported differential production of prostaglandins [29], as well as dif-
ACTIN ISOFORM PATTERN IN PREGNANT MYOMETRIUM
ferent responses to stretch, noradrenaline and estrogen stimulation in circular versus longitudinal muscle [28, 30]. Such
data imply specific roles for these layers during late pregnancy and labor. The circular muscle would provide the
primary contractile response, whereas the role of the longitudinal layer is to shorten the uterus upon expulsion of
each fetus.
The smooth muscle cells of the uterine vasculature express g-actin. Our visual observation of g-actin staining in
vasculature of nonpregnant and early and late pregnant
myometrium indicated no marked changes in the intensity
of g-actin immunostaining, whereas there was a dramatic
change in the intensity of uterine muscle g-actin staining.
Moreover, we found equally high immunostaining in myometrial vessels from empty and gravid myometrial horns of
unilaterally pregnant rats. The presence of intense g-actin
immunostaining in nonpregnant myometrium as well as in
both the empty and gravid horns of pregnant rats suggested
that g-actin expression in myometrial vasculature is not a
function of pregnancy. However, we visually detected an
increase in actual size of myometrial blood vessels throughout pregnancy, although we did not find a difference in a
number of blood vessels in individual fields of view. Because in our study we used whole myometrium to generate
proteins, our Western blot results would include a component of vascular g-actin. Therefore, we cannot exclude
some contribution of vascular g-actin to the Western blot
data.
Our immunohistologic analysis revealed a change in the
cellular distribution of g-actin from a diffuse, cytoplasmic
localization during early pregnancy to midpregnancy to a
more membrane-associated distribution during late pregnancy. Because antibody used in this study does not discriminate between smooth muscle and cytoplasmic g-actin
isoforms it is not possible to conclude which isoform is
involved in this redistribution. Myometrial cells are very
rich in actin fibers, which generally appear as parallel bundles along the longest axis of the cells [31], and their presence is consistent with the contractile function of these
cells. The cytoskeleton is known to act as a cell volume
sensor and a structure primarily involved in linking protein
filaments to each other and to anchoring sites in the adhesion plaques (reviewed in [31]). We have previously reported a marked increase in FAK activity in the myometrium during late pregnancy, together with increased expression of proteins within the focal adhesion complex adjacent to the cell membrane [12]. We speculate that the
increased expression of g-actin and its translocation to the
membrane may facilitate smooth muscle contraction by 1)
increasing the contractile protein content in a tissue and 2)
supporting the interaction of myocytes with the surrounding
collagen matrix, which is especially important at late gestation and during coordinated labor contractions.
In conclusion, our data demonstrated an intricate regulation of smooth muscle actin isoforms within the pregnant
myometrium. The a and g isoforms exhibit differential spatial and temporal expression within the myometrium and
our data imply complex interactions in the control of gactin (but not a-actin) by hormonal and mechanical signals.
We speculate that the increase in the expression and cellular
distribution of g-actin protein in circular smooth muscle
cells of gravid horns during the latter half of gestation supports myometrial hypertrophy and, as labor approaches,
contributes to the restoration of a contractile phenotype of
myometrial smooth muscle cells.
779
REFERENCES
1. Small JV. Structure-function relationships in smooth muscle: the missing links. Bioessays 1995; 17:785–792.
2. Vandekerckhove J, Weber K. At least six different actins are expressed
in a higher mammal: an analysis based on the amino acid sequence
of the amino-terminal tryptic peptide. J Mol Biol 1978; 126:783–802.
3. Schoenenberger CA, Steinmetz MO, Stoffler D, Mandinova A, Aebi
U. Structure, assembly, and dynamics of actin filaments in situ and in
vitro. Microsc Res Tech 1999; 47:38–50.
4. North AJ, Gimona M, Lando Z, Small JV. Actin isoform compartments in chicken gizzard smooth muscle cells. J Cell Sci 1994; 107
(pt 3):445–455.
5. Owens GK. Regulation of differentiation of vascular smooth muscle
cells. Physiol Rev 1995; 75:487–517.
6. Blank RS, McQuinn TC, Yin KC, Thompson MM, Takeyasu K,
Schwartz RJ, Owens GK. Elements of the smooth muscle alpha-actin
promoter required in cis for transcriptional activation in smooth muscle. Evidence for cell type-specific regulation. J Biol Chem 1992; 267:
984–989.
7. Saga H, Kimura K, Hayashi K, Gotow T, Uchiyama Y, Momiyama T,
Tadokoro S, Kawashima N, Jimbou A, Sobue K. Phenotype-dependent expression of alpha-smooth muscle actin in visceral smooth muscle cells. Exp Cell Res 1999; 247:279–292.
8. Kim YS, Wang Z, Levin RM, Chacko S. Alterations in the expression
of the beta-cytoplasmic and the gamma-smooth muscle actins in hypertrophied urinary bladder smooth muscle. Mol Cell Biochem 1994;
131:115–124.
9. Skalli O, Vandekerckhove J, Gabbiani G. Actin-isoform pattern as a
marker of normal or pathological smooth-muscle and fibroblastic tissues. Differentiation 1987; 33:232–238.
10. Shynlova O, Mitchell JA, Tsampalieros A, Langille BL, Lye SJ. Progesterone and gravidity differentially regulate expression of extracellular matrix components in the pregnant rat myometrium. Biol Reprod
2004; 70:986–992.
11. Ingber DE. Tensegrity: the architectural basis of cellular mechanotransduction. Annu Rev Physiol 1997; 59:575–599.
12. MacPhee DJ, Lye SJ. Focal adhesion signaling in the rat myometrium
is abruptly terminated with the onset of labor. Endocrinology 2000;
141:274–283.
13. Ou CW, Chen ZQ, Qi S, Lye SJ. Increased expression of the rat
myometrial oxytocin receptor messenger ribonucleic acid during labor
requires both mechanical and hormonal signals. Biol Reprod 1998;
59:1055–1061.
14. Piersanti M, Lye SJ. Increase in messenger ribonucleic acid encoding
the myometrial gap junction protein, connexin-43, requires protein
synthesis and is associated with increased expression of the activator
protein-1, c-fos. Endocrinology 1995; 136:3571–3578.
15. Williams SJ, White BG, MacPhee DJ. Expression of {alpha}5 integrin
is elevated in the rat myometrium during late pregnancy and labor:
implications for development of a mechanical syncytium. Biol Reprod
2005; 72:1114–1124.
16. Cavaille F, Leger JJ. Characterization and comparison of the contractile proteins from human gravid and non-gravid myometrium. Gynecol
Obstet Invest 1983; 16:341–353.
17. Erba HP, Gunning P, Kedes L. Nucleotide sequence of the human
gamma cytoskeletal actin mRNA: anomalous evolution of vertebrate
non-muscle actin genes. Nucleic Acids Res 1986; 14:5275–5294.
18. Herman IM. Actin isoforms. Curr Opin Cell Biol 1993; 5:48–55.
19. Malmqvist U, Arner A. Isoform distribution and tissue contents of
contractile and cytoskeletal proteins in hypertrophied smooth muscle
from rat portal vein. Circ Res 1990; 66:832–845.
20. Kumar A, Crawford K, Close L, Madison M, Lorenz J, Doetschman
T, Pawlowski S, Duffy J, Neumann J, Robbins J, Boivin GP, O’Toole
BA, Lessard JL. Rescue of cardiac alpha-actin-deficient mice by enteric smooth muscle gamma-actin. Proc Natl Acad Sci U S A 1997;
94:4406–4411.
21. Lye SJ, Mitchell J, Nashman N, Oldenhof A, Ou R, Shynlova O,
Langille L. Role of mechanical signals in the onset of term and preterm labor. Front Horm Res 2001; 27:165–178.
22. Csapo A, Erdos T, De Mattos CR, Gramss E, Moscowitz C. Stretchinduced uterine growth, protein synthesis and function. Nature 1965;
207:1378–1379.
23. Goldspink DF, Douglas AJ. Protein turnover in gravid and nongravid
horns of uterus in pregnant rats. Am J Physiol 1988; 254:E549–E554.
24. Wang J, Seth A, McCulloch CA. Force regulates smooth muscle actin
780
SHYNLOVA ET AL.
in cardiac fibroblasts. Am J Physiol Heart Circ Physiol 2000; 279:
H2776–H2785.
25. Goldman J, Zhong L, Liu SQ. Degradation of alpha-actin filaments
in venous smooth muscle cells in response to mechanical stretch. Am
J Physiol Heart Circ Physiol 2003; 284:H1839–H1847.
26. Hsu CY, Frankel FR. Effect of estrogen on the expression of mRNAs
of different actin isoforms in immature rat uterus. Cloning of alphasmooth muscle actin message. J Biol Chem 1987; 262:9594–9600.
27. Cicatiello L, Ambrosino C, Coletta B, Scalona M, Sica V, Bresciani
F, Weisz A. Transcriptional activation of jun and actin genes by estrogen during mitogenic stimulation of rat uterine cells. J Steroid Biochem Mol Biol 1992; 41:523–528.
28. Doualla-Bell F, Lye SJ, Labrie F, Fortier MA. Differential expression
and regulation of connexin-43 and cell-cell coupling in myocytes from
the circular and longitudinal layers of bovine myometrium. Endocrinology 1995; 136:5322–5328.
29. Doualla-Bell F, Bonneau MJ, Labrie F, Fortier MA. Differential effects
of a steroidal antiestrogen, EM-139, on prostaglandin and cyclic adenosine 39,59-monophosphate production in the circular and longitudinal
layers of bovine myometrium. Biol Reprod 1995; 52:1358–1363.
30. Matsumoto J. Changes of the contraction of the circular and longitudinal muscles of pregnant rat myometrium during pregnancy (author’s
transl.). Nippon Sanka Fujinka Gakkai Zasshi 1980; 32:1749–1757.
31. Yu JT, Lopez BA. The cytoskeleton of human myometrial cells. J
Reprod Fertil 1998; 112:185–198.