Ubiquitous Gasp1 overexpression in mice leads mainly to a hypermuscular phenotype
© Monestier et al.; licensee BioMed Central Ltd. 2012
Received: 14 November 2011
Accepted: 3 October 2012
Published: 10 October 2012
Skip to main content
© Monestier et al.; licensee BioMed Central Ltd. 2012
Received: 14 November 2011
Accepted: 3 October 2012
Published: 10 October 2012
Myostatin, a member of the TGFβ superfamily, is well known as a potent and specific negative regulator of muscle growth. Targeting the myostatin signalling pathway may offer promising therapeutic strategies for the treatment of muscle-wasting disorders. In the last decade, various myostatin-binding proteins have been identified to be able to inhibit myostatin activity. One of these is GASP1 (Growth and Differentiation Factor-Associated Serum Protein-1), a protein containing a follistatin domain as well as multiple domains associated with protease inhibitors. Despite in vitro data, remarkably little is known about in vivo functions of Gasp1. To further address the role of GASP1 during mouse development and in adulthood, we generated a gain-of-function transgenic mouse model that overexpresses Gasp1 under transcriptional control of the human cytomegalovirus immediate-early promoter/enhancer.
Overexpression of Gasp1 led to an increase in muscle mass observed not before day 15 of postnatal life. The surGasp1 transgenic mice did not display any other gross abnormality. Histological and morphometric analysis of surGasp1 rectus femoris muscles revealed an increase in myofiber size without a corresponding increase in myofiber number. Fiber-type distribution was unaltered. Interestingly, we do not detect a change in total fat mass and lean mass. These results differ from those for myostatin knockout mice, transgenic mice overexpressing the myostatin propeptide or follistatin which exhibit both muscle hypertrophy and hyperplasia, and show minimal fat deposition.
Altogether, our data give new insight into the in vivo functions of Gasp1. As an extracellular regulatory factor in the myostatin signalling pathway, additional studies on GASP1 and its homolog GASP2 are required to elucidate the crosstalk between the different intrinsic inhibitors of the myostatin.
Improving muscle mass and function is of a considerable clinical interest in therapeutic strategies for musculoskeletal disorders and has been assessed by several studies [1–3]. In the last decade, among all these approaches, dramatic attention has been focused on the regulation of the myostatin (GDF8) pathway. Indeed, myostatin is a key regulator of skeletal muscle growth and homeostasis. In mice, targeted inactivation of the Gdf8 gene causes a large and widespread increase in skeletal muscle mass, resulting from a combination of muscle cell hyperplasia and hypertrophy. Moreover, postnatal inhibition of myostatin signalling, through the delivery of neutralizing antibodies, myostatin propeptide injection or antisense RNA showed skeletal muscle improvement when administered to mice of different ages [4–12]. The identification of myostatin-binding proteins capable of regulating myostatin activity further expanded the number of potential therapeutic targets . Thus, follistatin (FS) can function as a potent myostatin antagonist, its overexpression in mice is found to enhance muscle growth [13, 14]. The increase in muscle mass observed in transgenic mice overexpressing FS in muscle is even significantly larger than that observed in Gdf8 -/- mice . However, follistatin is not a specific inhibitor for myostatin and binds also to other TGFβ including activin. In addition to follistatin, two other proteins have been identified that are involved in the regulation of the myostatin. Follistatin-related gene is highly similar to follistatin and has also been shown to inhibit activin and multiple bone morphogenic proteins in vitro. Growth and differentiation factor-associated serum protein-1 (GASP1; also called WFIKKN2) is a secreted protein that contains multiple domains associated with protease-inhibitory proteins including a whey acidic protein domain, a Kazal domain, two Kunitz domains, and a netrin domain. It also contains a highly conserved module of cysteine-rich sequence termed the follistatin domain. GASP1 was shown to bind directly but independently to both mature myostatin and the myostatin propeptide and to inhibit myostatin activity but not that of activin or TGFβ1 in vitro. Like its homologous protein GASP2, GASP1 also has a high affinity for GDF11, a secreted factor that regulates anterior/posterior patterning in the axial skeleton . Recent studies showed that both GASP1 and GASP2 bind growth factors TGFβ1, BMP2 and BMP4 but do not inhibit in vitro their signalling activity . GASP1 and 2 show distinct expression patterns both in the developing fetus and the adult. In the developing fetus, GASP1 expression is highest in the brain, skeletal muscle, thymus and kidney while GASP2 is abundant in the lung, skeletal muscle and liver . In the adult, GASP1 is primarily expressed in the ovary, testis, and brain while GASP2 is in the pancreas, liver, and thymus . To date, despite these data, little is known about the precise in vivo functions and protein interactions of these GASP proteins. To highlight the range and extent of Gasp1 roles during mouse development and in adulthood, we have generated and characterized transgenic mouse lines that ubiquitously overexpress Gasp1 under the control of a cytomegalovirus (CMV) promoter. Six transgenic lines have been isolated and two were selected with different levels of overexpression of Gasp1 in muscle, brain, heart, spleen, liver, lung and kidney for analyses. Detailed phenotypic characterization shows muscle abnormalities but no obvious defects in other major organ systems.
Intensive efforts have been made over the last decade to explore the molecular mechanisms underlying the regulation and function of myostatin, a key negative regulator of skeletal muscle growth. The identification of several components of the myostatin-signalling pathway had important implications with respect for testing the therapeutic value of a myostatin antagonist in muscle wasting disorders such as Duchenne muscular dystrophy [22, 23] cachexia and age-related sarcopenia [24–26]. Among the known myostatin-binding proteins, GASP1 has been shown to inhibit myostatin activity in vitro and may maintain myostatin latency but no data was available on the effect of GASP1 when expressed as a transgene in all skeletal muscles of wild type mice. In the present study, we have generated a “gain of function” mouse model to further understand the in vivo roles of Gasp1. These mice carry a transgene containing the Gasp1 coding sequence under the transcriptional regulation of the ubiquitous CMV promoter. As a consequence, Gasp1 mRNA expression is greatly enhanced in several organs including muscle, brain, spleen, liver, heart, lung and kidney. Analysis of the two Gasp1 transgenic lines with the highest transgene expression revealed a skeletal muscle hypertrophic phenotype. This might have been expected based on previous data indicating that GASP1 can function as a myostatin antagonist, similar to follistatin or the follistatin-related gene protein whose overexpression leads to an increase of the muscle mass. Furthermore, viral delivery of a Gasp1 expression cassette into adult muscle has been shown to induce increases in muscle mass and grip strength . The surGasp1 mice have similar average life span compared to standard inbred laboratory mice (http://research.jax.org/faculty/harrison/ger1vi_Lifespan.html) Except the phenotype described in this paper, no other gross abnormalities were noted, even in elder surGasp1 mice (≈ 28 months). Effects of elevated GASP1 on body growth were not observed before day 15 of postnatal life. The enhanced muscle growth occurs in both male and female animals with a more pronounced phenotype in male pectoralis major muscle. The individual weights of the gastrocnemius, rectus femoris and pectoralis major muscles were increased. The masses of other skeletal muscles were also increased. The sizes of other internal organs did not differ from those of control mice despite overexpression of GASP1. Unlike the myostatin-deficient animals or FS overexpressing mice, which exhibit both muscle hypertrophy and hyperplasia, we only observed an increase in myofiber size without a corresponding increase in myofiber number in surGasp1 animals. As the number of myofibers in muscle is largely determined during prenatal development, the overexpression of Gasp1 does not seem to provide prenatal effect. However, we cannot preclude that the lack of hyperplasia is not the result of a too moderate expression of the transgene during embryonic and fetal stages, although Gasp1 is strongly expressed in postnatal 3 days mice (Additional file 2). In the litterature, heterozygous mutations in the myostatin gene or surexpression of its propeptide have been reported to result also in hypertrophy and no hyperplasia . Taken together, these results may be reflective of an incomplete inhibition of the myostatin. Moreover, Zhu et al. have described that mice carrying a dominant negative form of myostatin preventing the release of the mature myostatin from the propeptide exhibited a significant increase in muscle mass that resulted from myofiber hypertrophy and not from myofiber hyperplasia. As Hill et al.  suggested, GASP1 may inhibit the propeptide proteolysis to keep the myostatin in a latent and inactive form. Such possible mechanism may explain the observed phenotype in the surGasp1 mice.
In mammals, myofibers are mainly classified into glycolytic and oxidative fibers based on their metabolic profiles. In mice, fast glycolytic fibers express the type IIB MHC isoform whereas oxidative fibers express type I (slow fibers), the fibers expressing MHC IIA are capable of both oxidative and glycolytic metabolism. We could show that Gasp1 overexpression does not significantly change fiber type distribution while a lack of myostatin results in an alteration in the fiber type composition [33–36]. Glycolytic (phosphofructokinase, lactate dehydrogenase) or oxidative (citrate synthase, isocitrate dehydrogenase, cytochrome c-oxidase) enzyme activities did not show any significant differences between wild-type and surgasp1-20 mice, confirming the above mentioned result (data not shown).
Interestingly, we do not detect a change in fat pad mass in our Gasp1 transgenic mice. This result differs from the results reported for the myostatin knockout mice or transgenic mice overexpressing the myostatin propeptide in which a reduction in adiposity is observed [14, 37, 38]. Recent literature showed that myostatin inhibition in skeletal muscle, but not in adipose tissues, is primarily responsible for a decrease of fat mass . This effect on fat pad could reflect a regulation of myostatin independant from GASP1.
Taken together, our data provide definitive evidence of the role of Gasp1 in vivo, in particular in muscle. In our transgenic mice, estimation of the amount of GDF8 associating with GASP1 will be an important issue, thus providing mechanistic insights into this association. We are currently crossing our transgenic mice overexpressing Gasp1 on the myostatin null background to investigate if GASP1 stimulates muscle growth by additional mechanisms independent of myostatin inhibition like it has been shown in transgenic mice overexpressing follistatin presenting a quadrupling of muscle mass . The related protein, GASP2 has also been shown to be capable of binding myostatin although GASP2 was not detected as one of the proteins bound to endogenous myostatin in serum . GASP2 appears to be similar to GASP1 in its ability to bind GDF8 or GDF11 in vitro but effects of GASP2 on other TGFβ ligands are not yet known . Further research will be required to determine the exact roles of GASP2 play in vivo and its interactions with GASP1.
All mice were bred and housed in the animal facility of Limoges University under controlled specific pathogen free conditions (21°C, 12-h light/12-h dark cycle) with free access to standard mouse chow and tap water. All of the experimental procedures were carried out in accordance with the local ethics commission.
First strand cDNA was synthesized from 1μg of total muscle RNA using the SuperScript® III Reverse Transcriptase (Invitrogen) with oligodT primers. The coding sequence of murine Gasp- 1 was PCR-amplified from muscle cDNA using primers 5′-ATGTGTGCCCCAGGGTATCATCG-3′ located at the position 159-181 bp and 5′-TCATTGCAAGCCCAGGAAGTCCTT-3′ located at the position 1851-1874 bp (transcript sequence ENSEMBL ENSMUST00000061469). The 1716 bp fragment (GenBank accession number: JQ080910) was introduced into the expression vector pcDNATM3.1/V5-His® TOPO (Invitrogen) to generate the psurGasp1 vector. Consequently, Gasp1 is under the human cytomegalovirus immediate-early promoter/enhancer leading to a strong constitutive expression. In our construct, the GASP1 protein is not expressed as a fusion to the V5 epitope and polyhistidine tag since a stop codon is present at the 3′ end of Gasp1 cDNA. A purified 3578 bp Sal1-NsiI fragment was microinjected into the male pronucleus of one-cell fertilized FVB/N embryos. Transgenic founders, or surGasp1 animals were identified by PCR of tail-extracted genomic DNA using primers fwdU2: 5′-AAGTACGCCCCCTATTGACG-3′ and revR1: 5′-CCAGAGGTTGGGGTTCATGT-3′. Founders bearing the transgene were bred to wild-type FVB/N mice to generate F1 offspring. Transgenic F1 mice were bred with other transgenic or wildtype FVB/N mice as necessary to maintain and expand the colony.
Muscle weights were measured following dissection of 12-week-old mice. Individual muscles from both sides of the animal were taken and the average weight was used for each muscle. Body composition analysis to determine fat and muscle contents was performed on conscious mice at 10 weeks of age, using the EchoMRI-500™ whole body composition analyzer (Echo Medical Systems).
Cryosections (14 μm) of rectus femoris muscles were prepared from frozen muscles of wildtype or surGasp1 12-week-old mice. Transverse sections were processed for hematoxylin and eosin staining. For cross sectional area measurement and fiber counting, muscles were stained with a reticulin silver staining kit (04-040801, Bio-Optica Milano S.p.A). Fiber typing was performed using the myosin ATPase method (pH 4.6) allowing differentiating 3 fiber types: I, IIA and IIB. For morphometric analysis, the muscle fiber sizes were measured with Image J software.
Total RNA was isolated from spleen, kidney, liver, heart, lung, brain and skeletal muscles dissected from mice of various ages (8 to 12 weeks, see figures legend) using TRIzol (Invitrogen) according to the manufacturer’s instruction and relative RNA concentration was determined by spectrophotometric analysis. 1.5 μg of total RNA was reverse-transcribed into DNA using the high-capacity cDNA reverse transcription kit (Applied Biosystems). Real-time PCR was performed in triplicate using 50 ng of cDNA for each sample. Relative amounts of transcripts were determined using Taqman probes specific for Gasp1 (Mm00725281_m1), Gapdh (Mm99999915_g1), TfIID (Mm00446973_m1) on an ABI PRISM© 7900 system (Applied Biosystems). Relative mRNA expression values were calculated by the ΔΔCt method with normalization of each sample to the average change in cycle threshold value of the controls. TLDA (Taqman low density array, Applied Biosystems) assays were performed based on the same above conditions, except 100 ng cDNA was used for each fill reservoir. The selected target genes analysed are listed in the Additional file 3: Table 1.
SurGasp1 F1 animals were genotyped by checking the presence of the CMV sequence using specific primers (fwdU2: AAGTACGCCCCCTATTGACG and revR1: CCAGAGGTTGGGGTTCATGT). These F1 animals are used as calibrator in the next step of the qPCR assay, i.e. their CMV or Gasp1 copy number is arbitrarily defined as 1. Then, SYBR Green–based real-time PCR was carried out by 3 amplifications for each sample (F2 mice) using the ABI PRISM® 7900 system (Applied Biosystems). One reaction that uses Gasp1-Fwd (CAGTCTCAATGGCACAGCTT) and Gasp1-Rev (GAGATTGTGGTGGCAGTGAC) primers was designed to yield a 148 bp product that corresponds to Gasp1 exon 2, the second reaction, by using CMV-Fwd (CCCACTTGGCAGTACATCAA) and CMV-Rev (GCCAAGTAGGAAAGTCCCAT) primers, was designed to yield a 123 bp product that corresponds to the CMV promoter. The last reaction by using CCR5-Fwd (GCACAAAGAGACTTGAGGCA) and CCR5-Rev (GTCATCTCTAGGCCACAGCA) was designed to yield a 81 bp product that corresponds to the CCR5 allele defining our reference gene. All the tests were done in triplicate. Each reaction was carried out in 20 μl of reaction mixture that contained SYBR® Green master mixture, each of the forward and reverse primers at a final concentration of 300 nM, and 5 ng of purified DNA sample. The thermal profile began with incubations at 50°C for 2 min and 95°C for 10 min followed by 40 cycles of amplification alternating between 95°C for 15 sec and 60°C for 1 min. The SYBR® Green fluorescent signal was obtained once per cycle at the end of the extension step. After amplification, melting curve analysis was performed by heating the PCR products to 95°C for 15 sec, then cooling it to 60°C for 15 sec, followed by a linear temperature increase to 95°C at a rate of 0.3°C/sec while continuously monitoring the fluorescent signal. Data were analyzed by the standard software, SDS 2.3, and RQ manager 1.2, included with the real-time PCR system.
Total proteins were extracted from spleen, kidney, liver, heart, pancreas, brain and some skeletal muscles dissected from 12-week-old mice using RIPA buffer (50 mM Tris, pH 8, 150 mM NaCl, 0.1% SDS, 1% NP-40, 0.5% sodium deoxycholate, and protease inhibitors). The Bradford assay was used to quantify protein concentrations at A595nm. The proteins extracted from tissues (50 μg) were mixed with Laemmli loading buffer and heated for 5 min at 95°C. The proteins were separated under denaturating conditions into a 10% polyacrylamide gel and then transferred to a Hybond C-Extra Nitrocellulose membrane (Amersham Biosciences). Unspecific binding was prevented using 5% non fat dry milk (w/v) in TBS-T0.1% buffer (50 mM Tris, 150 mM NaCl, pH 7.4, 0.1% Tween-20) for 1 h at room temperature. First antibodies were diluted in 2% non fat dry milk in TBS-T0.1% buffer [goat anti-GASP1 (AF2070, R&D Systems, 0.2 μg.ml-1); rabbit anti-GASP1 (HPA010953, Sigma, 0.3 μg.ml-1); rabbit anti-GASP1 directed against the peptide DCGEEQTRWFDAQANN (0.9 μg.ml-1); goat anti-GAPDH antibody (R&D Systems, 0.5 μg.ml-1)]. The diluted antibodies were incubated with membrane overnight at 4°C with constant agitation, followed by several washing steps in TBS-T0.1%. Blots were incubated with the secondary antibodies, horseradish peroxidase-coupled anti-goat IgG or anti-rabbit IgG (Dako) at a dilution of 1:2000 in 2% non fat dry milk in TBS-T0.1% buffer for 1 h at RT. After several washing steps in TBS-T0.1%, the immunoblots were processed by chemiluminescence detection (BM Chemiluminescence Western Blotting Substrate (POD), Roche Applied Science) and exposed to a film (Amersham Hyperfilm ECL, Amersham Biosciences).
Unless indicated otherwise, data are presented as mean ± SEM. Data are considered significant with p < 0.05 by two-tailed Student’s t test analysis.
This work was supported by the French National Institute for Agricultural Research and by the Limousin Regional Council.
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.