The expression of hypoxia-inducible factor (HIF) is upregulated under hypoxic
conditions[1], [2]. Upon the expression of HIF, several microRNAs
(miRNAs) are upregulated and miR-210 is one of the major hypoxia-inducible miRNAs[1]. miRNA-210 is reported to be highly expressed
under hypoxic conditions, such as in the cardiovascular system[3], brain[4],
skin[5], [6], and tumor tissues[1], and increased miRNA-210 expression was observed in a peripheral
ischemic mouse skeletal muscular injury model[7]. In our previous study, miRNA-210 expression was observed in myoblasts and
macrophages 3 days after ischemic muscular injury, indicating its association with muscular
regeneration[7]. Zaccagnini et
al. have reported that miRNA-210 inhibition increased ischemic tissue damage, and
miRNA-210 overexpression attenuated tissue damage and the inflammatory response with a
reduced number of infiltrative macrophages after hindlimb ischemia[8], [9].
Interestingly, miRNA-210 expression increases during myogenic differentiation under normoxia
with an HIF-1α dependent mechanism in in vitro conditions[10]. In addition, miRNA-210 elevation has been
reported in the regeneration phase after cardiotoxin-induced muscular injury in
mice[10]. Therefore, the elucidation of
HIF-1α and miRNA-210 expression profiles in non-ischemic muscular injury is important to
further understand their biological significance.Several skeletal muscular injury models have been developed to study skeletal muscle repair
and regeneration[11], [12], [13], [14], [15].
However, volumetric muscle loss and mechanical trauma models require validation by a
surgical procedure or establishment of a drop-tower device for trauma under
anesthesia[11], [12], [13], [14]. Cardiotoxin-induced muscular injury can be simply reproduced by
intramuscular injection[15]. However,
protein kinase C (PKC) inhibition is known to be one of the pharmacological actions of the
cardiotoxin[16], and PKC zeta plays an
important role in HIF-1α activation[17]. To
avoid the possible pharmacological effect of the cardiotoxin on HIF-1α, we chose acetic acid
in this study as an alternative chemical to induce non-ischemic muscular injury
model[18], [19], which can be reproduced with intramuscular
injection.In the present study, we investigated the detailed histopathological time course in a mouse
model of acetic acid-induced muscular injury by profiling HIF-1α and miRNA-210 expression.
In addition, we investigated the expression of growth factors, such as hepatocyte growth
factor (HGF), insulin-like growth factor (IGF)-1, and vascular endothelial growth factor
(VEGF), since these growth factors are considered to be related to muscle homeostasis and
regeneration[20], [21], [22] and HGF and VEGF can be upregulated by HIF-1α[23], [24].
Materials and Methods
Six-week-old male C57BL/6J mice purchased from Charles River Laboratories Japan, Inc.
(Tokyo, Japan) were used in this study. Twenty animals were maintained on a laboratory chow
diet (CE-2, CLEA Japan, Inc., Tokyo, Japan) and allowed free access to water and food before
and during the experiments. The care and use of the animals as well as the experimental
protocols used in this study were approved by the Institutional Animal Care and Use
Committee of Takeda Pharmaceutical Co., Ltd. (approval code: 00020216).
Animal model of muscular injury with acetic acid
We used acetic acid, a well-known irritative agent[18], for inducing the skeletal muscle injury model in this study. The
animals were administered 3% (v/v) acetic acid (Wako, Osaka, Japan) diluted with saline
intramuscularly at 50 µL/site into the left side of the gastrocnemius. Four animals per
time point were euthanized at 3 h and 1, 3, 7, and 14 d after acetic acid administration.
After necropsy, both sides of the gastrocnemius were removed, trimmed, and fixed in 10%
(v/v) neutral buffered formalin (NBF). Portions of the gastrocnemius muscles were immersed
in RNAlater (Ambion, Inc., Austin, TX, USA) for 24 h in a refrigerator and preserved at
−20°C until quantitative polymerase chain reaction (qPCR) analysis. The
acetic-acid-related effects were determined through a comparison with the non-treated
gastrocnemius muscles (right side) as a control. After fixation in NBF, the gastrocnemius
muscles were embedded in paraffin, sectioned transversely, and stained with hematoxylin
and eosin (H&E). Using sequential sections, immunohistochemistry (IHC) for F4/80,
iNOS, and galectin-3 (Gal-3) as pan-, M1, and M2 macrophage markers,
respectively[25],
[26], were carried out for all
samples showing cellular infiltration in H&E sections, for macrophage
characterization. For the interpretation of the qPCR results, IHC for HIF-1α and HGF and
in situ hybridization (ISH) for miRNA-210 were performed with
representative samples obtained at 3 d after the administration of acetic acid. The IHC
staining conditions for each primary antibody are summarized in Table 1. ISH was carried out according to the instruction manual v3.0 of the miRCURY
LNATM microRNA ISH Optimization Kit (FFPE) developed by Exiqon (Vedbaek,
Denmark) with minor modifications. We used digoxigenin (DIG)-labeled locked nucleic acid
(LNA)-modified probes for miRNA-210 (hsa-miR-210, Product. No. 18103-15), positive control
(U6 snRNA, product. No. 90010), and negative control (scrambled microRNA, Product. No.
90010) from Exiqon. After deparaffinization, proteinase-K (Exiqon: Product No. 90010,
miRCURY LNA; microRNA Detection, control probes, buffer, and ProtK) at 10 μg/mL was
incubated at 37 °C for 13 min. After washing with PBS, the LNA-probes, namely miRNA-210
(400 nM), scramble-miRNA (400 nM), and U6 (1 nM), were hybridized at 60 °C for 60 min.
After immersion in 5 × SSC buffer at room temperature, the specimens were washed in
pre-heated SSC buffer (55 °C), 1 × 5 min in 5 × SSC, 2 × 5 min in 1 × SSC, 2 × 5 min in
0.2 × SSC, and finally 1 × 5 min in 0.2 × SSC at room temperature and then immersed in
PBS. Sections were blocked against non-specific binding in blocking solution composed of
2% (v/v) sheep serum and 1% (v/v) bovine serum albumin in PBS-T for 15 min at room
temperature (RT). Alkaline phosphatase (AP)-conjugated anti-DIG (Product No. 11093274910,
Roche, Mannheim, Germany) (1:200) was incubated for 60 min at RT for immunological
detection. After washing with PBS-T, the substrate enzymatic reaction was carried out with
NBT/BCIP (Product No. 1697471001, Roche) at 30 °C for 2 h. The reaction was stopped with a
2 × 5 min wash in KTBT buffer (50 mM Tris–HCl, 150 mM NaCl, and 10 mM KCl). Sections were
counterstained with Nuclear Fast Red at room temperature for 3 min and then rinsed with
tap water. After dehydration by increasing the gradient of the ethanol solutions, the
specimens were mounted with Permount (Fisher Scientific Inc., Pittsburgh, PA, USA).
Table 1.
Primary Antibodies and Reaction Conditions for immunohistochemistry
mRNA and miRNA quantification
For relative quantification of Hif1α, Hgf,
Igf1, Vegf, and miRNA-210 using the comparative Ct
method (2-ΔΔCt), qPCR was conducted with all animal samples after extraction of
total RNA and reverse transcription to complementary DNA (cDNA). Total RNA was extracted
from cryopreserved muscle samples using the miRNeasy Mini Kit (Product No. 217004, QIAGEN,
Hilden, Germany). Reverse transcription of cDNA for mRNA and miRNA quantification was
carried out using the SuperScript VILO cDNA Synthesis Kit (Product No. 11754-250, Thermo
Fisher Scientific, Waltham, MA, USA) and Universal cDNA Synthesis Kit II (Product No.
203301, Exiqon), respectively. Subsequently, the relative mRNA levels of Hif1a,
Hgf, Igf1, and Vegf (probe product
Nos.: Mm00468869_m1, Mm01135184_m1, Mm00439560_m1, and Mm00437306_m1, respectively; Thermo
Fisher Scientific) were analyzed using TaqMan Fast Advanced Master Mix (Product No.
4444557, Thermo Fisher Scientific) with TATA-binding protein as a reference gene (Probe
product No. Mm01277042_m1; Thermo Fisher Scientific), while the relative miRNA-210
expression levels (hsa-miR-210, Probe product No. 204333; Exiqon) were analyzed using
ExiLENT SYBR Green master mix (Product No. 203402, Exiqon) with miRNA-191-5p as a
reference miRNA (has-miR-191-5p, Probe product No. 204306, Exiqon). All PCR procedures
were conducted using a 7900 HT Fast Real-Time PCR System (Applied Biosystems, Foster City,
CA, USA).The qPCR data for miRNANA and mRNA were tested by the F-test for homogeneity of variance
between the non-treatment and treatment sides. When the variances were homogeneous,
Student’s t-test was used, and when the variances were heterogeneous, Aspin–Welch t-test
was performed to compare the mean of the control group with that of the dosage
group[27]. The F-test was conducted at
a significance of 0.20, and the other tests were conducted at two-tailed significance
levels of 0.05, 0.01, and 0.001. Statistical analyses were performed using SAS version 9.3
(SAS Institute Inc., Cary, NC, USA).
Results
Histopathology and IHC for macrophage markers (F4/80, iNOS, and Gal-3)
After a single intramuscular dose of 3% (v/v) acetic acid, focal necrosis/degeneration of
muscle fibers was noted at the injection site after 3 h, which progressed to coagulation
necrosis thereafter. Its severity was prominent at 1 and 3 d and decreased at 7 and 14 d
after administration. The other findings mentioned below were located adjacent to the
necrotic area. The regeneration of the muscle fiber was observed at 3 d after the
administration of acetic acid with the appearance of basophilic myoblasts and at 7 and 14
d with the central nucleus fibers. Minimal neutrophil infiltration (only in one animal)
and minimal to mild mononuclear cell infiltration were observed in the interstitial tissue
around the necrotic area from 3 h; its severity and/or incidence increased over time and
were most prominent at 3 d when minimal neutrophil and mild mononuclear cell infiltration
around necrosis were noted in all animals. The main component of the infiltrated cells at
7 and 14 d after the administration of acetic acid was mononuclear cells. Most
infiltrative mononuclear cells were positive for F4/80 (a pan-macrophage marker) and Gal-3
(M2 macrophage marker) throughout the experimental period, and their distributions were
almost comparable. On the other hand, no positive reaction for iNOS (M1 macrophage marker)
was noted from 3 h to 1 d. Only some mononuclear cells were positive for iNOS from 3 d to
14 d, but they were a minor component in comparison to the F4/80- and Gal-3-positive
cells. These immunohistochemical results suggested M2 dominant macrophage infiltration
throughout the study period. A few animals showed minimal hemorrhage in and around the
necrotic area focally at 1 and 7 d, but no other findings indicating circulatory
impairment, such as vascular injury, were noted in the intact area of the treated muscle.
Interstitial fibrosis was noted at 3 d and thereafter, and the incidence was high at 7 and
14 d (Figs. 1 and 2; Tables 2 and 3).
Fig. 1.
Low magnification images (labeled as low) show the distribution of moderate focal
coagulation necrosis of muscle fibers in acetic acid-induced muscular injury model
at 1 and 3 d. Bar=400 μm. Images in the middle and lower panels show the time course
of the histological features from control and 3 h to 14 d after the treatment at
high magnification (labeled as high). Bar=100 μm.
Fig. 2.
Immunohistochemistry of F4/80, iNOS, and Gal-3 at 1, 3, and 14 d after the
treatment, respectively. Arrowheads show iNOS-positive cells at 14 d. Bar=90 μm.
Table 2.
Incidence and Severity of the Histopathological Findings in the Muscle Tissue
after a Single Dose of Acetic Acid
Table 3.
Incidence and Severity of Immunohistochemistry for F4/80, INOS, and Gal-3 in
Infiltrative Cells
Low magnification images (labeled as low) show the distribution of moderate focal
coagulation necrosis of muscle fibers in acetic acid-induced muscular injury model
at 1 and 3 d. Bar=400 μm. Images in the middle and lower panels show the time course
of the histological features from control and 3 h to 14 d after the treatment at
high magnification (labeled as high). Bar=100 μm.Immunohistochemistry of F4/80, iNOS, and Gal-3 at 1, 3, and 14 d after the
treatment, respectively. Arrowheads show iNOS-positive cells at 14 d. Bar=90 μm.
Expressions of HIF-1α and miRNA-210
The qPCR data showed that the gene expression of Hif1α in the treated
muscles was higher than that in the non-treated muscles, with statistically significant
difference from 3 h to 7 d after the administration of acetic acid, and was prominent from
3 h to 3 d after the treatment. A statistically significant increase in
Hif1α expression was also noted at 14 d but was not considered to
indicate pathophysiological significance since the extent of increase was small (less than
1.5 times higher). A statistically significant increase in the expression of miRNA-210 was
observed only at 3 d after the treatment (Fig.
3).
Fig. 3.
Relative quantification of Hif1α, Hgf,
Igf1, and Vegf and miRNA-210, using the
comparative Ct method (2−ΔΔCt) was conducted by qPCR. Fold changes in the
treated muscle (blue bar) as compared to the control muscle (white bar) are shown.
Data are presented as mean ± SD. *: p<0.05, **: p<0.01, and ***:
p<0.001.
Relative quantification of Hif1α, Hgf,
Igf1, and Vegf and miRNA-210, using the
comparative Ct method (2−ΔΔCt) was conducted by qPCR. Fold changes in the
treated muscle (blue bar) as compared to the control muscle (white bar) are shown.
Data are presented as mean ± SD. *: p<0.05, **: p<0.01, and ***:
p<0.001.IHC and ISH showed comparable positive reactions of HIF-1α and miRNA-210 in the cytoplasm
of macrophages and regenerative muscle fibers, myoblasts, around the necrotic area at 3 d
after the treatment (Fig. 4). Positive reactions were not observed in the intact muscle fibers.
Fig. 4.
Immunohistochemistry (IHC) or in situ hybridization (ISH) for
HIF1α (IHC), miRNA-210 (ISH), and HGF (IHC), control and 3 d after the treatment.
Bar=100 μm.
Immunohistochemistry (IHC) or in situ hybridization (ISH) for
HIF1α (IHC), miRNA-210 (ISH), and HGF (IHC), control and 3 d after the treatment.
Bar=100 μm.
Expressions of growth factors: HGF, IGF-1, and VEGF
The gene expression of Hgf in the treated muscles was significantly
higher than that in the non-treated muscles throughout the experimental period, and was
prominent (more than a 10-fold increase) at 3 d after the treatment. The gene expression
of Vegf decreased from 1 d to 7 d after the treatment, which was
statistically significant. Although a statistically significant increase in
Vegf expression was noted at 3 h, it was not pathophysiologically
significant (approximately 1.2 times higher). No statistically significant changes were
noted in Igf1 expression throughout the experimental period, although it
tended to increase at 7 and 14 d after the treatment (Fig. 3).Immunohistochemical signals of HGF were observed in macrophages and myoblasts at 3 d
after the treatment, which was consistent with the distribution of HIF-1α and miRNA-210
(Fig. 4). Immunohistochemical signals were not
observed in the intact muscle fibers.
Discussion
In this study, focal muscular necrosis/degeneration was observed a few hours after a single
intramuscular administration of acetic acid in mice, which eventually progressed to
coagulation necrosis. Its severity was prominent at 1 and 3 d after the treatment. Muscular
regeneration with the appearance of basophilic myoblasts was initiated 3 d after the
treatment. Macrophage infiltration was observed in the interstitial tissue from tissue
injury to the recovery process, which was the most prominent 3 d after the treatment. In
general, infiltrative macrophages are categorized into pro-inflammatory M1 and
anti-inflammatory M2 phenotypes[26]. In
skeletal muscle injury, M1 macrophages infiltrate early to promote the clearance of necrotic
debris during the inflammation phase shortly after neutrophil infiltration, whereas M2
macrophages appear later to sustain tissue healing[28] Notably, the pattern of macrophage infiltration among the injury models
seemed to vary. Martins et al. have reported M1 macrophage infiltration
just after traumatic injury that slowly increased over time until 15 d after the treatment,
and M2 macrophage infiltration peaked at 4 d after the treatment[14]. In the cardiotoxin-induced model, Bouredji et
al. showed M1 and M2 macrophage infiltration at 3 d after the treatment, with a
tendency to decrease and increase, respectively at 7 d[29]. However, in the acetic acid-induced muscular injury, neither prominent
M1 macrophage infiltration nor neutrophilic infiltration was observed, and most of the
infiltrative macrophages were of the M2 phenotype.As assessed by qPCR, Hif1α elevation was prominent from 3 h to 3 d after
the treatment and remained until 7 d, which was consistent with the severity of muscular
necrosis. On the other hand, miRNA-210 increased transiently at 3 d after the treatment,
when macrophage infiltration was most prominent and muscular regeneration was initiated, and
the elevated levels returned to the baseline at 7 d. In IHC or ISH staining, positive
reactions of HIF-1α and miRNA-210 were observed in the cytoplasm of infiltrative macrophages
and regenerative muscle fibers around the time of necrosis at 3 d after treatment. The
comparable histological distribution indicated that miRNA-210 elevation was induced in a
HIF-1α-dependent manner. It has been reported that HIF-1α-dependent miRNA-210 induction was
observed in murine macrophages[30] and
myoblasts[10] under normoxic in
vitro conditions. The increased expression of HIF-1α was considered to be induced
by muscular necrosis or possibly by acidic conditions after acetic acid injection as HIF-1α
can be activated when it comes in contact with cell debris derived from myotubes or low pH
conditions independent of hypoxia[31],
[32]. Moreover, in this study, the
existence of muscular necrosis/degeneration and a few cases of minimal hemorrhage indicated
the possibility of focal hypoxia, which could be partially related to the increased
expression of Hif1α. However, its contribution was considered less dominant
than that observed in the hindlimb ischemia model, wherein a severe reduction in blood flow
was noted after arterial ligation[7], whereas
no other histopathological findings indicating circulation impairment were noted in the
intact area of the treated muscles in this study. Under hypoxic conditions, HIF-1α can
continuously act as a transcriptional factor after transfer to the nucleus, and miRNA-210
expression can be upregulated thereafter[1], [33]. In the
hindlimb ischemia model, significantly increased miRNA-210 expression was noted from 3 d to
14 d after femoral artery dissection using qPCR[8]. In contrast, in acetic acid-induced muscular injury, qPCR data showed
that miRNA-210 increased transiently at 3 d, the timing of the initiation of muscular
regeneration. The association between miRNA-210 expression and muscular regeneration was
also demonstrated in the hindlimb ischemia model because increased miRNA-210 expression was
noted in the myoblasts and macrophages during the muscular regenerative phase[7]; however, the transient miRNA-210 expression
seen in the acetic acid-induced non-ischemic muscular injury model may suggest a more
specific role of miRNA-210 expression in the process from muscular injury to regeneration.
The elevated levels of miRNA-210 observed in this study may play anti-inflammatory or
cytoprotective roles in the muscular tissue during the regeneration phase, as has also been
demonstrated in the literature; for example, inhibition of NF-κB expression in macrophages
followed by a decrease in TNF-α, IL-6, and iNOS[30], improved angiogenesis and tissue repair[9], and tissue-protective effects during myogenic differentiation
of myoblasts[8], [10].Increased growth factor expression has been observed after muscular injury or during the
regeneration processes, such as an increase in HGF mRNA and protein expression during the
regeneration from ischemic injury or muscular atrophy[34], [35],
elevation in IGF-1 mRNA levels from tissue injury to regeneration phases after muscle
contusion[36], and a transient increase
in VEGF mRNA 12 h after a single freezing injury[37]. In the acetic acid-induced muscular injury, increased
Hgf expression was noted throughout the experimental period, and its
expression was most prominent at 3 d after the treatment. The increased Hgf
level was considered to be a result of its expression in the infiltrative macrophages and
regenerative muscle fibers based on the results of IHC for HGF.HGF is secreted by M2 macrophages as a means of tissue repair[38], and its expression in the muscular tissue also exerts
protective effects in muscular injury; HGF can increase the number of myoblasts in a damaged
muscle tissue, but can also inhibit muscle differentiation[20]. Therefore, increased Hgf expression and the
immunohistochemical results could be indicative of their roles in tissue repair in
infiltrative macrophages and regenerative muscle fibers.The expression of miRNA-210 in infiltrative macrophages and regenerative muscle fibers
could be related to the expression of HGF, since increased expression of HGF was observed in
acute myocardial infarction rat models who received miRNA-210 agonist[39]. In addition, HGF expression can be
upregulated by HIF-1α under hypoxic in vitro condition[23], and our results showed a simultaneous
increase in the expressions of Hif1α and Hgf.Meanwhile, a clear elevation in the level of Igf1 was not induced in this
study, although a slight increase in Igf1 was noted at 7 and 14 d after the
treatment. Thus, there seemed to be some differences in the expression pattern of
Igf1 from that in other muscular injury models, although the underlying
reasons remained unclear. In the muscle contusion or cardiotoxin-induced model, elevation in
Igf1 levels was observed from tissue injury to regeneration phases as a
factor that affected the diameter of muscle fibers[36], [40],
[41]. In particular, in
cardiotoxin-induced muscular injury, macrophage-derived IGF-1 is considered a key factor in
inflammation resolution and macrophage polarization during muscle regeneration[40]. Although we could not compare the degree of
macrophage infiltration between the cardiotoxin-[40] and acetic acid-induced models, the possibility of low intensity of
macrophage infiltration might be related to only a slight increase in Igf1
level observed in this study.It has been reported that Vegf expression can be upregulated by HIF-1α or
miRNA-210 under hypoxic in vitro condition[24], [42]; however, Vegf expression was found to significantly
decrease from 1 d to 7 d in this study. A similar decrease in Vegf
expression in the muscle tissue was noted after 1 to 3 d after a single dose of cardiotoxin
in heme oxiganase-1 knockout and wild type mice[41], despite reports of increased Hif1α or miRNA-210
expression levels in cardiotoxin-induced muscular injury[10], [43]. Ochoa
et al. also showed a decrease in VEGF protein expression after 1 to 3 d
of cardiotoxin-induced muscular injury[44].
Furthermore, they demonstrated that the decrease in VEGF expression was exacerbated and
lasted until 14 d under CCR-dependent impairments of F4/80+ macrophage
infiltration[44].In conclusion, dominant infiltration of M2 macrophages was confirmed in the area adjacent
to the focal coagulation necrosis of muscle fibers, where following muscular regeneration
was observed after a single intramuscular injection of acetic acid in mice. M1 macrophage or
neutrophil infiltration was not prominent. An increase in the Hif1α level
was observed from 3 h to 7 d after the treatment, and the elevations in miRNA-210 and
Hgf levels were clearly noted at 3 d after the treatment at which tissue
regeneration was initiated. miRNA-210, HIF-1α, and HGF were histologically detected in
macrophages and myoblasts. These results indicate the tissue-protective and regenerative
effects of miRNA-210 after muscular injury, independent of ischemia.
Disclosure of Potential Conflicts of Interest
The authors declare that there are no conflicts of interest associated with this
manuscript.
Fig. 1.
Fig. 2.
Fig. 3.
Fig. 4.