Adenovirus Expressing the Myostatin-Somatostatin-fusing Gene Increases Mouse Growth Rate

Background Myostatin (MSTN), a type of transforming growth factor, can negatively regulate skeletal muscle growth. Immunisation of mice with recombinant yeast expressing MSTN increased bodyweight and muscle composition of them. Somatostatin (SST) is an inhibitory effector of growth hormone (GH), and active immunisation against SST with a DNA vaccine improves the growth performance of piglets via an inuence on GH secretion. Here, a recombinant adenovirus was constructed and used to deliver MSTN-SST to mice to achieve the goal of regulation the growth performance of the mice. Results A recombinant adenovirus, rAd-MSTN-SST, expressing the MSTN-SST fusing protein was successfully rescued in HEK293A cells. The expressions of MSTN-SST were conrmed by western blotting and indirect immunouorescence. Mice immunised with rAd-MSTN-SST were successfully induced to have an immune response against MSTN-SST, which increased their growth rate and muscle mass. In addition, a booster immunisation was benecial in terms of higher antibody response and mouse growth rate. Conclusions


Background
Myostatin (MSTN), a member of the transforming growth factor (TGF)-ß superfamily, is also known as growth and differentiation factor 8 (GDF8) [1]. MSTN has been found to be capable of modulating bodyweight and muscle composition in laboratory and farm animals [2]. Oral feeding mice with recombinant yeast Saccharomyces cerevisiae expressing mammalian MSTN from a plasmid or chromosomal integration gene elicits antigen-speci c immune responses, resulting in increased bodyweight and muscle composition [3,4]. Downregulation of MSTN expression by siRNA or gene knockout also increased muscle mass [2,5]. Accordingly, MSTN is an ideal target for regulating animal meat products or human muscle wastage. Somatostatin (SST) is known to inhibit the release of growth hormone (GH) from the anterior pituitary [6]. Reduction in the concentration of SST in the blood accelerates growth in animals. Immunisation of animals to SST and the resulting induction of related antibodies (anabolic factors) is a means of removing SST's normal inhibitory effects. However, SST is a short peptide hormone with only 14 amino acids and its half-life in the bloodstream is only several minutes. Thus, SST conjugates with various proteins are used for immunisation [7]. Here, MSTN was fused to SST to induce the related immune response, accelerate the growth of animals, and enhance meat products.
Adenovirus vectors are the most commonly employed viral vector for gene therapy and delivery of vaccine antigens. They have been used for the delivery of rabbit haemorrhagic disease virus antigen VP60 [8] and many other virus antigens such as the antigens of porcine reproductive and respiratory syndrome virus [9] and foot-and-mouth disease virus [10]. Immunisation with recombinant adenovirus induces a robust immune response and protects against infection. Here, we used adenovirus to deliver MSTN and SST, which may induce a strong immune response that regulates the growth rate of immunised animals.

Construction of recombinant viruses
Shuttle vectors containing the MSTN-SST gene under the control of the CMV early promoter were constructed and veri ed by sequencing. By recombination with adenovirus backbone vector pAdEasy-1 in BJ5183 cells, recombinant adenoviral plasmids, pAd-MSTN-SST, were obtained. Then, the recombinant plasmids were linearised with endonuclease Pac I and transfected into HEK293A to generate recombinant adenoviruses, rAd-MSTN-SST. After about 10 days' incubation, those recombinant adenoviruses were successfully packaged with a characteristic cytopathic effect in transfected cells. Meanwhile, mocktransfected cells (control samples) retained their singularity.
The obtained recombinant adenoviruses, rAd-MSTN-SST, were puri ed three times by the plaque-puri ed method and titred in HEK293A cells. The titre of the recombinant adenoviruses was 3.6 × 10 8 vp mL −1 . The expressions of the MSTN-SST proteins were con rmed by indirect immuno uorescent assay (IFA) and western blotting (WB).
Green uorescence, indicating the expression of MSTN-SST, was observed in rAd-MSTN-SST-infected cells, but not in wtAdinfected cells (Fig. 1A). Correspondingly, the immunoreactive protein bands of MSTN-SST were observed in Lines 1, 2 and 3 The antibody against MSTN-SST was detected as early as 7 days post-inoculation. And reached a peak at day 21 by a single immunisation (Fig. 2). Antibody titres were still increasing at the end of the experiments in group C, which was immunised twice. The immunised groups gained signi cantly high levels of antibody titre since 7 days post-inoculation (P ≤ 0.05). Boost immunisation to group C induced signi cant higher level of humoral response after 14 days (day 28 , P ≤ 0.05; Fig. 2) comparing with that from group B.

Bodyweight gain
All mice were about the same weight with no statistical difference ( Fig. 3) at beginning of the experiment. At the end of the experiment, however, they weighed from 34.0 to 40.3 g and the average weights varied among groups. The bodyweight of each animal was recorded weekly during the 4 experimental weeks. And then the mice were sacri ced.
The average increase in bodyweight of the mice was 14.18 ± 0.36 g in Group A, whereas it was 16.70 ± 0.25 g in Group B, suggesting that vaccination against MSTN-SST enhanced bodyweight increase. In addition, the increases in bodyweight in Groups C (19.20 ± 0.63 g) were signi cantly higher than those of Group B, suggesting that immunisation boosting was effective, which further con rms the role of immunisation MSTN-SST on growth rate regulation in mice (Fig. 3).

Muscle morphology observation
To show the effect of the immunisation on mouse muscle, post-mortem examinations were performed. The muscles of immunised mice were more de ned and the increased sizes could be recognised by the naked eye (Fig. 4). The weights of the longissimus dorsi (LD) and biceps femoris (BF) were measured and shown in Fig. 5. The average LD and BF sizes in the control group were 0.358 ± 0.002 g and 0.239 ± 0.005 g, respectively. Immunisation increased these to 0.463 ± 0.004 g and 0.314 ± 0.002 g in group B and 0.494 ± 0.003 g and 0.367 ± 0.005g in group C, respectively.

Histological examination
Histological examination was used to show the micro-level differences of the mice muscle bres among the groups. The BF muscle bres of each group were observed under an optical microscope (Fig. 6). Thickened muscle bres in groups B and C were observed compared with those from group A. In addition, the densities of muscle bres increased signi cantly as the numbers of bres increased signi cantly. According to the statistical results of ten views with the graphical analysis software Image J, the average proportions of muscle bres increased from 63.43% ± 3.28% to 72.18% ± 4.28% and 83.72 %± 1.67%, respectively. The differences between each group were signi cant (P ≤ 0.05 ).

Discussion
Adenovirus vectors are the most commonly used viral vectors in gene therapy and the delivery of vaccine antigens. Adenovirus vectors used as vaccines are mostly replication-defective, with certain essential viral genes deleted and replaced by a foreign gene expression cassette [11,12]. Here, the MSTN-SST fusion open reading frame was inserted into the Ad5 genome, which was used to express the protein MSTN-SST. Immunisation with the recombinant adenovirus induced humoral response against the recombinant protein MSTN-SST. The activated immune response against MSTN-SST resulted acceleration of the bodyweight and muscle mass gains in mice. In previous research, MSTN was delivered though a heat-killed whole recombinant yeast S. cerevisiae expressing mammalian MSTN from a plasmid, which elicited antigen-speci c cell and humoral responses in mice [4]. Immunisation increased bodyweight and muscle composition in mice. Soon after, another report demonstrated that heat-inactivated MSTN-recombinant yeast can promote growth in rabbits and signi cantly increase muscle development [13]. Then, S. cerevisiae harbouring MSTN in the genome was constructed and used to deliver the MSTN gene to mice by oral immunisation [3], which produced similar results. In addition, monoclonal anti-MSTN antibody injected into the yolk of chicken embryos signi cantly increased the bodyweight (4.2%) and muscle mass (5.5%) of the chicks [14]. Though oral immunisation with S. cerevisiae or direct administration of a monoclonal antibody can induce an immune response against MSTN and accelerate growth in animals, such immune operations are di cult to conduct in the eld. Immunisation by muscle injection is mostly used in clinical production because it is easy to standardise. Here, adenovirus expression MSTN-SST was given by intramuscular injection and the immunisation induced similar antibody levels.
Animals immunised with SST had increased mean daily weight gain of 10-20%, appetite reduction of 9%, and an 11% increase in food utilisation e ciency [7]. Improved absorption of food components and slower passage of food through the gastrointestinal tract due to sluggish peristalsis were also observed. Animals immunised with SST and their offspring had correct proportions, and the weight distribution of muscle, bone and fat was the same as in controls [7,15]. Another study found that immunisation of pregnant goats increased the weight of new-borns by 10% and increased their milk yield [7]. Here, according to our results, immunisation with MSTN-SST improved the daily weight gain of mice as well as their gross weight.
However, we did not measure their food consumption and the mice were allowed to feed freely.

Conclusion
We constructed an adenovirus which expressed the fusing proteins, MSTN-SST. Mice immunised with this adenovirus had enhanced daily weight gain and muscle mass. This recombinant adenovirus can be used to increase animal meat production and may also be able to treat muscular atrophy disease of human. However, experiments on pigs are needed and serious ethical issues should be considered before such vaccines go on the market.

Virus and cells
The human type 5 adenovirus expression system (replication-defective) was purchased from TaKaRa (Dalian, China).
All the cells were cultured in Dulbecco's modi ed Eagle's medium (DMEM) and supplemented with 10% foetal bovine serum, 2 mM L-glutamine, 100 U/mL penicillin, and 100 μg/mL streptomycin at 37 °C in a humidi ed atmosphere of 5% CO 2 .

Construction of recombinant adenoviruses (rAd-MSTN-SST)
The open reading frames of porcine MSTN (GenBank No. AY208121) were ampli ed by PCR using the primers (MSTN-SSTs and MSTN-SSTa, Table 1) with a plasmid harbouring arti cially synthesized MSTN gene (General boil, Co.,Ltd, Anhui ) as templates. SST gene was fused into MSTN by a second round PCR reaction with the primers of MSTN-SSTs and MSTN-SSTa2.
The PCR products were cloned into the transfer vector, pAd-shuttle-CMV. The recombinant adenoviral vectors were generated by homologous recombination of linearised transfer vectors with pAdEasy-1 in Escherichia coli BJ5183 strain and con rmed by restriction enzyme digestion (New England Biolabs). The recombinant adenovirus was generated by transfection of 1 μg plasmid (Pac I linearised) using 3 μL of Trans Fast TM Transfection Reagent (Promega, Madison, USA). When 90% of the cells showed cytopathic effect, adenoviruses were released by three cycles of rapid freezing and thawing and stored at −80 ℃ after addition of 10% glycerol.  β-actin was used as a reference and the primary antibody against β-actin was purchased from Boster Co., Ltd. (Wuhan, BM0627).

Immunisation
Thirty 4-week-old male Chinese Kunming mice were purchased from Chengdu Dashuo Experimental Animal Co. Ltd. and randomly divided into three groups of ten. The mice had very similar weights. The rst group received wtAd (at day 0) and served as a control group (Group A). The second group was given rAd-MSTN-SST at day 0 by intramuscular injection (Group B). The third group was given rAd-MSTN-SST at day 0 and boosted at day 14 (Group C). Mice were kept in cages with an automatic ventilation system. Five mice were kept in each cage and were allowed to feed freely.
The mice were weighed every week and blood was sampled on days 14 and 28. All the mice euthanised by CO2 inhalation. CO2 was infused in mice home cage at 10%, 30%, or 100% volume per minute displacement rates. When mice were in deep narcosis, they were sacri ced by cervical dislocation. To observe the muscles, the skins of the mice were removed and the muscle bres (biceps femoris muscle) were observed by making tissue slices.

Histopathology tests
The collected biceps femoris muscles were xed and embedded in para n wax and cut into 4-5 μm slices. For microstructural observation, the slices were stained with haematoxylin and eosin. The densities of muscle bres in ten views were analysed using Image J software and the average densities of muscle bre were used to express the increase in muscle quantity.
Enzyme-linked immunosorbent assay (ELISA) ELISA plates were coated with the recombinant protein MSTN-SST (0.2 µg per well). The recombinant protein was expressed in E. coli. It was stored at −80 °C after puri cation until assayed. The sera of the mice collected at different times were detected with the ELISA plates as reported [12]. Commercial peroxidase-conjugated rabbit anti-mice IgG (Sigma, Shanghai, China) was used as the secondary antibody. After reaction, the absorbance values of the reaction system at 450 nm (OD450) were used to represent the antibody levels of the mice. Each sample was assayed in triplicate.

Statistical analysis
The bodyweights of mice are expressed as means ± standard deviation (SD) and evaluated with ANOVA. Differences between control and immunised groups were analysed by two-tailed independent Student's t-tests. Differences  The authors declare no con ict of interest. Authors' contributions GX and NZ conducted the experiment. SL and TW made the gures and revised the manuscript. XC and DY did the data analysis and XW prepared the manuscript. All authors have read and approved the manuscript.