Metabonomic study of the biochemical profiles of heterozygous myostatin knockout swine

Abstract

Myostatin is a transforming growth factor-β family member that normally acts to limit skeletal muscle growth. Myostatin gene (MSTN) knockout (KO) mice show possible effects for the prevention or treatment of metabolic disorders such as obesity and type 2 diabetes. We applied chromatography and mass spectrometry based metabonomics to assess system-wide metabolic response of heterozygous MSTN KO (MSTN^+/-) swine. Most of the metabolic data for MSTN^+/- swine were similar to the data for wild type (WT) control swine. There were, however, metabolic changes related to fatty acid metabolism, glucose utilization, lipid metabolism, as well as BCAA catabolism caused by monoallelic MSTN depletion.The statistical analyses suggested that: (1) most metabolic changes were not significant in MSTN^+/- swine compared to WT swine; (2) only a few metabolic properties were significantly different between KO and WT swine, especially for lipid metabolism. Significantly, these minor changes were most evident in female KO swine and suggested differences in gender sensitivity to myostatin.

Keywords

myostatin / transforming growth factor-β family / skeletal muscle / metabolic disorders / chromatography / mass spectrometry / metabonomics

1 Introduction

Metabonomics is a systems approach for studying in vivo metabolic profiles, which promises to provide information on drug toxicity, disease processes and gene function at several stages in the discovery and development process. This approach can also be readily adapted to investigate the functional consequences of genetic variation and transgenesis^[ 1^]. Metabolic changes are directly linked to phenotypic changes, whereas gene expression changes are not; they merely indicate the potential for an end-point change. As such, metabonomics provides a useful connection between the ‘omics’ platforms and actual tissue histology and physiology.

Nuclear magnetic resonance, liquid chromatography- gas chromatography (LC-MS) and gas chromatography- mass spectrometry (GC-MS) are now routinely applied for determination of the changes in metabolite profiles associated with toxicity^[ 2^– 8^], human disease^[ 9^– 11^], pharmaceutical drugs^[ 12^– 15^], genetic modification^[ 1^, 16^– 20^], and influencing factors such as age, strain, gender, dietary difference and diurnal variations in rodents^[ 21^– 30^].

Myostatin, also called growth differentiation facor-8 (GDF-8) previously, is a transforming growth factor-β (TGF-β) family member essential for the regulation of muscle development or function^[ 31^, 32^]. The predicted myostatin protein includes an N-terminal signal sequence, a dibasic proteolytic processing site, and a C-terminal domain following the processing site. The N-terminal fragment following proteolytic processing has been most commonly referred to as the propeptide, which is presumed to be essential for the proper folding of the C-terminal domain into a cystine knot structure. In the C-terminal region, the myostatin sequence shows significant homology to other family members, and is also capable of binding the activin type II receptors, ActRIIA and ActRIIB in vitro^[ 32^– 34^]. Loss of myostatin activity has been studied in mice^[ 35^– 39^], cattle^[ 40^– 43^], sheep^[ 44^], dogs^[ 45^], humans^[ 46^], and swine (unpublished data). The results indicate that the function of myostatin appears to have been conserved across these species. Significantly, myostatin gene (MSTN) knockout (KO) mice have beneficial effects on fat and glucose metabolism and this approach may possibly be effective for the prevention or treatment of obesity and type 2 diabetes^[ 33^, 38^]. This study aimed to research the metabolic effects in heterozygous MSTN KO (MSTN^+/-) swine by LC-MS and GC-MS.

2 Materials and methods

2.1 Animals

All experiments were conducted in accordance with the guidelines of China on the Use and Care of Laboratory Animals. Swine were housed in the same conditions and fed food and water ad libitum. MSTN^+/- Large White pigs were produced by homologous recombination and somatic cell nuclear transfer according to our previously published methods^[ 47^, 48^]

2.2 Sample collection

Blood of 26 F₁MSTN^+/- pigs () were obtained in the morning following an overnight fast to minimize metabolic effects of inter-individual variations of food intake. Blood was centrifuged and plasma was taken and divided into two parts, one part (about 1 mL) was sent to the veterinary hospital of China Agricultural University, Beijing, China for clinical laboratory analysis. Another part (about 500 μL) was stored at –80°C until metabonomics analyses.

Tab.1 Experimental animals

Group	Sample	Bodyweight (mean±SD)/kg	Generation	Description
WT	Male (n = 11)	114.36±4.02	F1	Wild type adult swine (MSTN +/+)
Female (n = 3)	113.92±5.40
KO	Male (n = 7)	113.74±5.35	F1	Heterozygous MSTN KO adult swine (MSTN+/-)
Female (n = 5)	114.43±5.24

2.3 Sample preparation

Plasma samples (100 mL) were extracted using an automated MicroLab STAR^® system (Hamilton Company, UT, USA) and maintained at 4°C throughout the extraction process. Recovery standards were added prior to the first step in the extraction process for quality control (QC) purposes. The samples were extracted using a proprietary series of organic and aqueous extractions according to the methods of Lawton and colleagues^[ 49^]. The resulting extract was divided into two equal aliquots, one for LC/MS and one for GC/MS. Aliquots were placed on a TurboVap^® (Zymark, Runcorn, UK) to remove the organic solvent, frozen and dried under vacuum overnight. Samples were then prepared for the appropriate instrument, either LC/MS or GC/MS.

2.4 LC/MS and GC/MS analysis

LC/MS was conducted using Waters Acquity UPLC (ultra performance liquid chromatography) equipment (Waters, Milford, MA) coupled to an LTQ mass spectrometer (Thermo Fisher Scientific, Waltham, MA) equipped with an electrospray ionization (ESI) source and linear ion-trap (LIT) mass analyzer. Two separate UPLC/MS injections were performed on each sample which contained 11 or more injection standards at fixed concentrations. One aliquot was analyzed using acidic positive ion optimized conditions and the other using basic negative ion optimized conditions in two independent injections using separate dedicated columns. Extracts reconstituted in acidic conditions were gradient-eluted using water and methanol both containing 0.1% formic acid, while the basic extracts, which also used water/methanol, contained 6.5 mmol·L^–1 ammonium bicarbonate. The MS analysis alternated between MS and data-dependent MS² scans using dynamic exclusion.

The derivatized samples destined for GC/MS were analyzed on a Thermo-Finnigan Trace DSQ fast-scanning single-quadrupole MS (ThermoElectron Corporation) using electron impact ionization. The GC column was 5% phenyl, the initial oven temperature was 40°C and then ramped to 300°C for a 16 min. period. The samples were re-dried under vacuum desiccation for a minimum of 24 h before being derivatized under dried nitrogen using bistrimethy-siyl-triflouroacetamide (BSTFA). GC/MS was tuned and calibrated daily for mass resolution and mass accuracy^[ 50^].

2.5 Compound identification

Compounds were identified by automated comparison to Metabolon’s reference library entries using Metabolon’s proprietary software (http://www.metabolon.com) developed for creating library entries from known chemical entities and then automatically fitting those spectra to experimentally derived spectra. As of writing, more than 1000 commercially available purified standard compounds had been acquired and registered in the laboratory information management system for distribution to both the LC and GC platforms for determination of their analytical characteristics. The combination of chromatographic properties and mass spectra give an indication of a match to the specific compound or an isobaric entity. Additional entities could be identified by virtue of their recurrent nature (both chromatographic and mass spectral). These compounds have the potential to be identified by future acquisition of a matching purified standard or by classical structural analysis.

2.6 Data normalization

For studies spanning multiple days, a data normalization step was performed to correct variation resulting from instrument inter-day tuning differences. Essentially, each compound was corrected in run-day blocks by adjusting medians to 1.0 and normalizing each data point proportionately (i.e., block correction, ). For studies that did not require more than one day of analysis, no normalization was necessary, other than for purposes of data visualization.

Fig.1 Visualization of data normalization

Full size|PPT slide

2.7 Statistical analysis

Welch’s t-tests and random forest analyses were used for significance tests and classification analysis, respectively. Welch’s t-test is an adaptation of Student’s t-test intended for use with two samples having possibly unequal variances^[ 51^]. Random forest is a supervised classification technique based on an ensemble of decision trees^[ 52^, 53^]. In contrast to a t-test, the random forest method tests whether the unknown means for two populations are different or not. An ANOVA was also performed with sex factor in this study. Statistical analyses are performed with the program R (http://cran.r-project.org/).

3 Results

3.1 Clinical laboratory analysis

As shown in , most clinical laboratory data revealed no significant differences between the WT and KO swine, indicating that the two groups of swine had a similar health status. The plasma level of alanine transaminase and lactic dehydrogenase (LDH) was significantly lower in KO male swine (P<0.05), and the plasma LDH level was very significantly lower in KO female swine (P<0.01). These minor differences in clinical data suggest that there may have been minor metabolic differences between the two group of swine.

Tab.2 Clinical laboratory analysis of KO (MSTN+/-) and WT (MSTN swine

Item	KO (MSTN+/-)	WT (MSTN +/+)	P
Male	Female	Male	Female	Male	Female
WBC/( × 109·L-1)	20.8 (4.8)	23.5 (1.0)	26.1 (6.8)	25.4 (7.4)	0.0923	0.6984
RBC/( × 1012·L-1)	6.3 (0.8)	6.9 (0.5)	6.9 (0.5)	6.0 (0.7)	0.1446	0.1387
HGB/(g·L-1)	106 (12.8)	120.2 (7.8)	117.8 (3.8)	107 0.0 (15.9)	0.0742	0.2838
HCT/%	34.7 (4.4)	39.1 (2.9)	38.1 (2.0)	34.3 (4.1)	0.1221	0.1666
MCV/fL	54.9 (1.2)	56.4 (2.5)	55.2 (2.4)	56.7 (1.0)	0.7692	0.8638
MCH/pg	16.8 (1.0)	17.4 (1.1)	17.1 (1.2)	17.6 (0.7)	0.5846	0.7050
MCHC/(g·L-1)	307 (22.9)	308 (10.3)	310.3 (15.2)	312.0 (15.0)	0.7272	0.7094
PLT/( × 109·L-1)	228 (119.7)	321 (103.7)	329.1 (103.2)	351.7 (92.4)	0.1933	0.6870
TBIL/(mmol·L-1)	3.57 (2.44)	4.80 (3.11)	3.10 (1.91)	1.00 (0)	0.6773	0.0525
DBIL/(mmol·L-1)	12.14 (1.46)	12.80 (2.17)	12.60 (1.17)	11.33 (0.58)	0.5067	0.2135
ALT/(U·L-1)	56.43 (11.31)a	69.60 (9.91)	71.70 (9.15)a	58.00 (7.21)	0.0128	0.1091
AST/(U·L-1)	113.00 (83.66)	83.20 (7.76)	121.40 (63.12)	71.33 (34.02)	0.8265	0.6091
TP/(g·L-1)	63.03 (8.08)	69.82 (2.19)	73.24 (5.75)	71.23 (9.38)	0.0164	0.8200
ALB/GLO	1.51 (0.32)	1.60 (0.19)	1.43 (0.17)	1.27 (0.24)	0.5586	0.1297
GGT/(U·L-1)	37.14 (9.42)	110.80 (40.54)	40.90 (16.35)	40.33 (19.09)	0.5587	0.0164
ALP/(U·L-1)	19.43 (17.92)	10.00 (7.91)	11.40 (7.03)	5.33 (3.21)	0.2956	0.2895
BUN/(mmol·L-1)	4.06 (1.45)	4.61 (0.45)	4.55 (0.90)	4.85 (1.25)	0.4515	0.7780
CRE/(mmol·L-1)	156 (17.72)	152 (10.92)	145 (19.46)	152 (28.36)	0.2551	0.9972
GLU/(mmol·L-1)	4.27 (1.81)	3.43 (0.67)	4.62 (1.32)	3.99 (0.43)	0.6700	0.2018
TG/(mmol·L-1)	0.56 (0.24)	0.52 (0.13)	0.40 (0.09)	0.36 (0.12)	0.1367	0.1372
CHO/(mmol·L-1)	1.94 (0.33)	2.33 (0.15)	2.08 (0.19)	2.10 (0.15)	0.3294	0.0899
CK/(U·L-1)	2215 (750.5)	2014.00 (1121.3)	1726 (769.7)	758 (321.9)	0.2122	0.0659
LDH/(U·L-1)	731 (221.3)a	894 (78.3)b	1170 (531.3)a	475 (42.0)b	0.0362	0.0001
AMY/(U·L-1)	2442 (962.5)	3160 (1126.1)	2179 (1028.3)	1857 (976.0)	0.5989	0.1463

Note: Data are mean (SD). WBC, white blood cell; RBC, red blood cell; HGB, hemoglobin; HCT, hematocrit; MCV, mean corpuscular volume; MCH, mean corpuscular hemoglobin; MCHC, mean corpuscular hemoglobin concentration; PLT, platelets; TBIL, total bilirubin; DBIL, direct bilirubin; ALT, alanine transaminase; AST, aspartate transaminase; TP, total protein; ALB, albumin; GLO, globulin; GGT, gamma-glutamyl transpeptidase; ALP, alkaline phosphatase; BUN, blood urea nitrogen; CRE, creatinine; GLU, glucose; TG, triglyceride; CHO, cholesterol; CK, creatine kinase; LDH, lactic dehydrogenase; AMY, amylase. a, significantly different between two groups (P<0.05); b, very significantly different between two groups (P<0.01).

3.2 Metabolomic analysis

Using LC/MS and GC/MS analysis, 300 compounds matched a named structure in Metabolon’s reference library; these compounds came from the following super metabolic pathways: amino acid, peptide, energy, lipid, nucleotide, xenobiotics, cofactors and vitamins. Of these, 43 biochemicals were different significantly (P≤0.05) between the experimental groups. The statistical changes were calculated by the ratio of their group means and are listed in . These data suggest that loss of myostation has minimal effects on the biochemical profile of MSTN^+/- swine. Significantly, these minor changes were most evident in female MSTN KO plasma (female to male ratio of 35€:€10, ) and may reflect gender specific differences in sensitivity to myostatin. Furthermore, the total number of significant changes observed in male MSTN KO plasma (10) was below the number expected by random chance (15) and therefore suggests the metabolic profile of MSTN KO male and wild-type porcine is similar. This overlap between WT and MSTN KO male biochemical profiles is further revealed by principal component analysis (). Nominal changes observed between WT and MSTN KO female plasma are highlighted below.

Tab.3 Significantly altered biochemicals between the experimental groups

Biochemicals	KO/WT	KO-M/WT-M	KO-F/WT-F
Total (P≤0.05)	43	10	35
Fold of change≥1.00	29	2	25
Fold of change<1.00	14	8	10

Note: KO, (MSTN+/-); WT, (MSTN ; M, male; F, female.

Fig.2 Principle component analysis (PCA) for the two experimental groups

Full size|PPT slide

3.2.1 Fatty acid metabolism

As shown in Appendix A ( S1), an accumulation of essential, dicarboxylate and long-chain fatty acids was observed in KO swine plasma compared to WT. Importantly, these differences were predominantly observed in myostatin deficient females, but not males. Elevated free fatty acid levels in the blood are often suggestive of an increase in lipolysis. Lipolysis is the process by which triglycerides are hydrolyzed into free fatty acids that can then be further degraded through beta oxidation. Similarly, free carnitine levels were decreased in MSTN KO plasma, while carnitine conjugated lipids such as palmitoylcarnitine were modestly elevated ().This suggests that the conjugation of long-chain free fatty acids to carnitine prepared it for transport into the mitochondria and subsequent oxidation. In agreement with markers of lipid oxidation, modest elevations in glycerol and 3-hydroxybutyrate (BHBA) was also observed in MSTN KO female plasma (). While glycerol can be utilized for the synthesis of triacylglycerides, it is also a catabolic product that is generated during their degradation. 3-hydroxybutyrate is a ketone body that is generated from excess acetyl-CoA that often results from enhanced fatty acid oxidation. Together, these results suggest the absence of myostatin in female swine may selectively result in increased fatty acid oxidation.

Fig.3 Lipid metabolism. (a) Box plots of representative biochemicals between groups, median scaled values are presented on the y-axis; (b) lipid metabolic pathway.

Full size|PPT slide

3.2.2 Glucose utilization

Both the levels of glucose and glycolytic intermediates, such as glucose 6-phosphate, 3-phosphoglycerate and lactate, failed to differ in comparison to WT plasma (). Similarly, significant changes were not observed in the sorbitol pathway, which can increase in the presence of excess glucose, or pentose phosphate pathway.

Fig.4 Glucose metabolism. (a) Glucose metabolic pathway; (b) box plots of representative glucose and glycolytic intermediates between groups.

Full size|PPT slide

3.2.3 Branched chain amino acid catabolism

In addition to protein synthesis, branched chain amino acids (BCAA) are important in replenishing the tricarboxylic acid cycle (TCA) cycle through the generation of acetyl-CoA and succinyl-CoA. In KO female plasma, the BCAA catabolic products 4-methyl-2-oxopentanoate, 3-methyl-2-oxobutyrate, and 3-hydroxyisobutyrate were elevated compared to WT (). Considering that glucose and lipid metabolism should be sufficient to fuel the TCA cycle, the accumulation of BCAA intermediates may be indicative of muscle turnover.

Fig.5 BCAA catabolism. (a) Box plots of the BCAA catabolic key products, there were significant differences (P<0.05) only between females; (b) BCAA catabolic pathway.

Full size|PPT slide

3.2.4 Other metabolic pathway

Besides the above pathways, this work also compared metabolites of other pathways between KO and WT swine, such as amino acid, peptide, carbohydrate, energy, nucleotide, xenobiotics, cofactors and vitamins. Although minor differences in metabolites were present, by and large, there were no significant metabolic differences between the two groups (data not shown).

4 Discussion

4.1 Influencing factors in metabolite profile

Bollard et al. found by metabonomics that time of sampling (day and night) markedly affected the metabolites profiles^[ 26^]. In the case of C57BL10J mice, urine samples collected in the morning were found to contain higher concentrations of creatin, hippurate, trimethylamine, succinate, citrate and a-ketoglutarate, and decreased amounts of taurine, trimethylamine-N-oxide, spermine and 3-hydroxy-iso-valerate compare with samples collected in the afternoon^[ 27^, 54^]. Except for the diurnal variation, there are many other factors, such as age, strain, diet and gender, that also affect metabolites profiles^[ 23^, 24^, 30^, 55^]. This study attempted to minimize the impact of such factors on the metabolic results, by sampling in the morning (8:00–10:00), after overnight fasting, feeding the same forage, selecting age and gender matched controls and half-sibling animals.

4.2 Metabolomic analysis

Published studies have clearly indicated that adipogenesis decreased in MSTN KO mice^[ 33^, 38^]. This study showed that fatty acid oxidation increased in MSTN^+/- female swine, it also indicated that adipose may be decreased in MSTN^+/- female swine. Together with reduced adipose and increased fatty acid oxidation, the lipid profile in MSTN null animals was improved. Adult male homozygous MSTN KO (MSTN^-/-) mice had significantly lower serum cholesterol and triglyceride levels compared to WT mice^[ 38^, 56^, 57^]. This study has shown higher long chain fatty acids and glycerol levels in KO swine, and that KO swine had lower triglyceride levels compared with WT swine. However, no significant difference in cholesterol between KO and WT swine was found.

As found in fatty acid oxidation, this study also observed difference in the MSTN KO female in BCAA catabolism. Similar to Lee’s findings^[ 39^], this study also indicated a maternal effect on metabolic reaction results, most likely from prenatal transfer of myostatin or some other mediator from mother to fetus.

Circulating glucose levels often serve as an indicator of systemic metabolic status. However, no significant difference in glucose metabolism was observed in KO swine in this study. Together with lipid metabolism, the results suggests that glucose utilization was unaltered in MSTN^+/- swine, which differs from published reports of increased glycolysis and decreased lipid oxidation in myostatin deficient skeletal muscle in rodents^[ 37^– 39^].

The published studies, mentioned above, have also demonstrated: (1) loss of myostatin activity can have beneficial metabolic effects in two genetic models of obesity and type 2 diabetes, and (2) myostatin inhibitors may be useful agents for the prevention or treatment of metabolic disorders such as obesity and type 2 diabetes^[ 33^, 38^]. While the current study did not find any strong evidence for the prevention or treatment of obesity and type 2 diabetes in swine by lipid and glycose metabolism, there were minor changes in lipid and BACC metabolism. We conclude that the effect may not be obvious in MSTN^+/- animals, and we will confirm and improve these results when enough MSTN ^-/- swine are produced in future.

5 Conclusions

This study demonstrated a few significant differences between WT and MSTN^+/- swine plasma. In the absence of myostatin, lipid oxidation, but not glucose utilization, was selectively elevated in female plasma. In contrast, the metabolic profile of WT and MSTN KO males were indistinguishable. Ultimately, these findings have important implications for livestock development, suggesting MSTN^+/- swine may be metabolically similar to WT swine.

References

[1] Nicholson J K, Connelly J, Lindon J C, Holmes E. Metabonomics: a platform for studying drug toxicity and gene function. Nature Reviews Drug Discovery, 2002, 1(2): 153–161

[2] Nicholson J K, Lindon J C, Holmes E. ‘Metabonomics’: understanding the metabolic responses of living systems to pathophysiological stimuli via multivariate statistical analysis of biological NMR spectroscopic data. Xenobiotica, 1999, 29(11): 1181–1189

[3] Holmes E, Nicholls A W, Lindon J C, Ramos S, Spraul M, Neidig P, Connor S C, Connelly J, Damment S J, Haselden J, Nicholson J K. Development of a model for classification of toxin-induced lesions using ¹H NMR spectroscopy of urine combined with pattern recognition. NMR in Biomedicine, 1998, 11(4–5): 235–244

[4] Beckwith-Hall B M, Nicholson J K, Nicholls A W, Foxall P J, Lindon J C, Connor S C, Abdi M, Connelly J, Holmes E. Nuclear magnetic resonance spectroscopic and principal components analysis investigations into biochemical effects of three model hepatotoxins. Chemical Research in Toxicology, 1998, 11(4): 260–272

[5] Robertson D G, Reily M D, Sigler R E, Wells D F, Paterson D A, Braden T K. Metabonomics: evaluation of nuclear magnetic resonance (NMR) and pattern recognition technology for rapid in vivo screening of liver and kidney toxicants. Toxicological Sciences, 2000, 57(2): 326–337

[6] Williams R E, Cottrell L, Jacobsen M, Bandara L R, Kelly M D, Kennedy S, Lock E A. 1H-Nuclear magnetic resonance pattern recognition studies with N-phenylanthranilic acid in the rat: time- and dose-related metabolic effects. Biomarkers, 2003, 8(6): 472–490

[7] Idborg-Björkman H, Edlund P O, Kvalheim O M, Schuppe-Koistinen I, Jacobsson S P. Screening of biomarkers in rat urine using LC/electrospray ionization-MS and two-way data analysis. Analytical Chemistry, 2003, 75(18): 4784–4792

[8] Lenz E M, Bright J, Knight R, Wilson I D, Major H. A metabonomic investigation of the biochemical effects of mercuric chloride in the rat using ¹H NMR and HPLC-TOF/MS: time dependent changes in the urinary profile of endogenous metabolites as a result of nephrotoxicity. Analyst, 2004, 129(6): 535–541

[9] Lamers R J, van Nesselrooij J H, Kraus V B, Jordan J M, Renner J B, Dragomir A D, Luta G, van der Greef J, DeGroot J. Identification of an urinary metabolite profile associated with osteoarthritis. Osteoarthritis and Cartilage, 2005, 13(9): 762–768

[10] Brindle J T, Antti H, Holmes E, Tranter G, Nicholson J K, Bethell H W, Clarke S, Schofield P M, McKilligin E, Mosedale D E, Grainger D J. Rapid and noninvasive diagnosis of the presence and severity of coronary heart disease using ¹H-NMR-based metabonomics. Nature Medicine, 2002, 8(12): 1439–1444

[11] Yang J, Xu G, Hong Q, Liebich H M, Lutz K, Schmülling R M, Wahl H G. Discrimination of type 2 diabetic patients from healthy controls by using metabonomics method based on their serum fatty acid profiles. Journal of Chromatography B, 2004, 813(1–2): 53– 58

[12] Robertson D G, Reily M D, Baker J D. Metabonomics in pharmaceutical discovery and development. Journal of Proteome Research, 2007, 6(2): 526–539

[13] Louden D, Handley A, Taylor S, Lenz E, Miller S, Wilson I D, Sage A. Reversed-phase high-performance liquid chromatography combined with on-line UV diode array, FT infrared, and ¹H nuclear magnetic resonance spectroscopy and time-of-flight mass spectrometry: application to a mixture of nonsteroidal antiinflammatory drugs. Analytical Chemistry, 2000, 72(16): 3922–3926

[14] Plumb R S, Stumpf C L, Gorenstein M V, Castro-Perez J M, Dear G J, Anthony M, Sweatman B C, Connor S C, Haselden J N. Metabonomics: the use of electrospray mass spectrometry coupled to reversed-phase liquid chromatography shows potential for the screening of rat urine in drug development. Rapid Communications in Mass Spectrometry, 2002, 16(20): 1991–1996

[15] Lenz E M, Bright J, Knight R, Wilson I D, Major H. Cyclosporin A-induced changes in endogenous metabolites in rat urine: a metabonomic investigation using high field ¹H NMR spectroscopy, HPLC-TOF/MS and chemometrics. Journal of Pharmaceutical and Biomedical Analysis, 2004, 35(3): 599–608

[16] Piccioni F, Capitani D, Zolla L, Mannina L. NMR metabolic profiling of transgenic maize with the Cry1Ab gene. Journal of Agricultural and Food Chemistry, 2009, 57(14): 6041–6049

[17] Chassy B M. Can-omics inform a food safety assessment? Regulatory Toxicology and Pharmacology, 2010, 58(3 Suppl): S62–S70

[18] Raamsdonk L M, Teusink B, Broadhurst D, Zhang N, Hayes A, Walsh M C, Berden J A, Brindle K M, Kell D B, Rowland J J, Westerhoff H V, van Dam K, Oliver S G. A functional genomics strategy that uses metabolome data to reveal the phenotype of silent mutations. Nature Biotechnology, 2001, 19(1): 45–50

[19] Delneri D, Brancia F L, Oliver S G, Brancia F L. Towards a truly integrative biology through the functional genomics of yeast. Current Opinion in Biotechnology, 2001, 12(1): 87–91

[20] Fiehn O, Kopka J, Dörmann P, Altmann T, Trethewey R N, Willmitzer L. Metabolite profiling for plant functional genomics. Nature Biotechnology, 2000, 18(11): 1157–1161

[21] Gavaghan C L, Holmes E, Lenz E, Wilson I D, Nicholson J K. An NMR-based metabonomic approach to investigate the biochemical consequences of genetic strain differences: application to the C57BL10J and Alpk:ApfCD mouse. FEBS Letters, 2000, 484(3): 169–174

[22] Lenz E M, Bright J, Wilson I D, Hughes A, Morrisson J, Lindberg H, Lockton A. Metabonomics, dietary influences and cultural differences: a ¹H NMR-based study of urine samples obtained from healthy British and Swedish subjects. Journal of Pharmaceutical and Biomedical Analysis, 2004, 36(4): 841–849

[23] Williams R E, Lenz E M, Rantalainen M, Wilson I D. The comparative metabonomics of age-related changes in the urinary composition of male Wistar-derived and Zucker (fa/fa) obese rats. Molecular BioSystems, 2006, 2(3–4): 193–202

[24] Williams R E, Lenz E M, Lowden J S, Rantalainen M, Wilson I D. The metabonomics of aging and development in the rat: an investigation into the effect of age on the profile of endogenous metabolites in the urine of male rats using ¹H NMR and HPLC-TOF MS. Molecular BioSystems, 2005, 1(2): 166–175

[25] Tate A R, Damment S J, Lindon J C. Investigation of the metabolite variation in control rat urine using ¹H NMR spectroscopy. Analytical Biochemistry, 2001, 291(1): 17–26

[26] Bollard M E, Holmes E, Lindon J C, Mitchell S C, Branstetter D, Zhang W, Nicholson J K. Investigations into biochemical changes due to diurnal variation and estrus cycle in female rats using high-resolution ¹H NMR spectroscopy of urine and pattern recognition. Analytical Biochemistry, 2001, 295(2): 194–202

[27] Gavaghan C L, Wilson I D, Nicholson J K. Physiological variation in metabolic phenotyping and functional genomic studies: use of orthogonal signal correction and PLS-DA. FEBS Letters, 2002, 530(1–3): 191–196

[28] Griffin J L, Walker L A, Garrod S, Holmes E, Shore R F, Nicholson J K. NMR spectroscopy based metabonomic studies on the comparative biochemistry of the kidney and urine of the bank vole (Clethrionomys glareolus), wood mouse (Apodemus sylvaticus), white toothed shrew (Crocidura suaveolens) and the laboratory rat. Comparative Biochemistry and Physiology Part B: Biochemistry & Molecular Biology, 2000, 127(3): 357–367

[29] Holmes E, Nicholson J K, Tranter G. Metabonomic characterization of genetic variations in toxicological and metabolic responses using probabilistic neural networks. Chemical Research in Toxicology, 2001, 14(2): 182–191

[30] Gavaghan McKee C L, Wilson I D, Nicholson J K. Metabolic phenotyping of nude and normal (Alpk:ApfCD, C57BL10J) mice. Journal of proteome research, 2006, 5(2): 378–384

[31] McPherron A C, Lawler A M, Lee S J. Regulation of skeletal muscle mass in mice by a new TGF-β superfamily member. Nature, 1997, 387(6628): 83–90

[32] Lee S J. Regulation of muscle mass by myostatin. Annual Review of Cell and Developmental Biology, 2004, 20(1): 61–86

[33] Lin J, Arnold H B, Della-Fera M A, Azain M J, Hartzell D L, Baile C A. Myostatin knockout in mice increases myogenesis and decreases adipogenesis. Biochemical and Biophysical Research Communications, 2002, 291(3): 701–706

[34] Rebbapragada A, Benchabane H, Wrana J L, Celeste A J, Attisano L. Myostatin signals through a transforming growth factor β-like signaling pathway to block adipogenesis. Molecular and Cellular Biology, 2003, 23(20): 7230–7242

[35] Bogdanovich S, Krag T O, Barton E R, Morris L D, Whittemore L A, Ahima R S, Khurana T S. Functional improvement of dystrophic muscle by myostatin blockade. Nature, 2002, 420(6914): 418– 421

[36] Bogdanovich S, Perkins K J, Krag T O, Whittemore L A, Khurana T S. Myostatin propeptide-mediated amelioration of dystrophic pathophysiology. FASEB Journal, 2005, 19(6): 543–549

[37] Wagner K R, McPherron A C, Winik N, Lee S J. Loss of myostatin attenuates severity of muscular dystrophy in mdx mice. Annals of Neurology, 2002, 52(6): 832–836

[38] McPherron A C, Lee S J. Suppression of body fat accumulation in myostatin-deficient mice. Journal of Clinical Investigation, 2002, 109(5): 595–601

[39] Lee S J. Quadrupling muscle mass in mice by targeting TGF-β signaling pathways. PLoS ONE, 2007, 2(8): e789

[40] Grobet L, Martin L J, Poncelet D, Pirottin D, Brouwers B, Riquet J, Schoeberlein A, Dunner S, Ménissier F, Massabanda J, Fries R, Hanset R, Georges M. A deletion in the bovine myostatin gene causes the double-muscled phenotype in cattle. Nature Genetics, 1997, 17(1): 71–74

[41] Kambadur R, Sharma M, Smith T P, Bass J J. Mutations in myostatin (GDF8) in double-muscled Belgian Blue and Piedmontese cattle. Genome Research, 1997, 7(9): 910–916

[42] McPherron A C, Lee S J. Double muscling in cattle due to mutations in the myostatin gene. Proceedings of the National Academy of Sciences of the United States of America, 1997, 94(23): 12457–12461

[43] Grobet L, Poncelet D, Royo L J, Brouwers B, Pirottin D, Michaux C, Ménissier F, Zanotti M, Dunner S, Georges M. Molecular definition of an allelic series of mutations disrupting the myostatin function and causing double-muscling in cattle. Mammalian Genome, 1998, 9(3): 210–213

[44] Clop A, Marcq F, Takeda H, Pirottin D, Tordoir X, Bibé B, Bouix J, Caiment F, Elsen J M, Eychenne F, Larzul C, Laville E, Meish F, Milenkovic D, Tobin J, Charlier C, Georges M. A mutation creating a potential illegitimate microRNA target site in the myostatin gene affects muscularity in sheep. Nature Genetics, 2006, 38(7): 813–818

[45] Mosher D S, Quignon P, Bustamante C D, Sutter N B, Mellersh C S, Parker H G, Ostrander E A. A mutation in the myostatin gene increases muscle mass and enhances racing performance in heterozygote dogs. PLOS Genetics, 2007, 3(5): e79

[46] Schuelke M, Wagner K R, Stolz L E, Hübner C, Riebel T, Kömen W, Braun T, Tobin J F, Lee S J. Myostatin mutation associated with gross muscle hypertrophy in a child. New England Journal of Medicine, 2004, 350(26): 2682–2688

[47] Li H, 2, Feng C, Wang N, Yan J, Ha F, Chen H, Fan B, Pan D. Construction of a myostatin gene-targeting vector and myostatin gene knockout of porcine fetal fibroblasts cells. Letters in Biotechnology, 2010, 21(5): 699–704

[48] Pan D, Zhang L, Zhou Y, Feng C, Long C, Liu X, Wan R, Zhang J, Lin A, Dong E, Wang S, Xu H, Chen H. Efficient production of omega-3 fatty acid desaturase (sFat-1)-transgenic pigs by somatic cell nuclear transfer. Life Sciences, 2010, 53(4): 517–523

[49] Lawton K A, Berger A, Mitchell M, Milgram K E, Evans A M, Guo L, Hanson R W, Kalhan S C, Ryals J A, Milburn M V. Analysis of the adult human plasma metabolome. Pharmacogenomics, 2008, 9(4): 383–397

[50] Evans A M, DeHaven C D, Barrett T, Mitchell M, Milgram E. Integrated, nontargeted ultrahigh performance liquid chromatography/electrospray ionization tandem mass spectrometry platform for the identification and relative quantification of the small-molecule complement of biological systems. Analytical Chemistry, 2009, 81(16): 6656–6667

[51] Welch B L. The generalisation of student’s problems when several different population variances are involved. Biometrika, 1947, 34(1–2): 28–35

[52] Breiman L. Random Forests. Machine Learning, 2001, 45(1): 5–32

[53] Goldstein B A, Hubbard A E, Cutler A, Barcellos L F. An application of Random Forests to a genome-wide association dataset: methodological considerations & new findings. BMC Genetics, 2010, 11(1): 49

[54] Plumb R, Granger J, Stumpf C, Wilson I D, Evans J A, Lenz E M. Metabonomic analysis of mouse urine by liquid-chromatography-time of flight mass spectrometry (LC-TOFMS): detection of strain, diurnal and gender differences. Analyst, 2003, 128(7): 819–823

[55] Zhao B, Wall R J, Yang J. Transgenic expression of myostatin propeptide prevents diet-induced obesity and insulin resistance. Biochemical and Biophysical Research Communications, 2005, 337(1): 248–255

[56] Guo T, Jou W, Chanturiya T, Portas J, Gavrilova O, McPherron A C. Myostatin inhibition in muscle, but not adipose tissue, decreases fat mass and improves insulin sensitivity. PLoS ONE, 2009, 4(3): e4937

[57] McPherron A C. Metabolic Functions of Myostatin and Gdf11. Immunology, Endocrine & Metabolic Agents in Medicinal Chemistry, 2010, 10(4): 217–231

Acknowledgments

This work was funded by the key special projects of breeding new varieties of genetically modified organisms in China (2014ZX08012-002). We thank Metabolon, a diagnostic products and services company, for assistance with the metabolic profile and statistical analysis.

Supplementary materialsƒThe online version of this article at http://dx.doi.org/10.15302/J-FASE-2015045 contains supplementary material (Appendix A).

Compliance with ethics guidelinesƒJianxiang Xu, Dengke Pan, Jie Zhao, Jianwu Wang, Xiaohong He, Yuehui Ma and Ning Li declare that they have no conflict of interest or financial conflicts to disclose.

ƒAll applicable institutional and national guidelines for the care and use of animals were followed.

RIGHTS & PERMISSIONS

Higher Education Press and Springer-Verlag Berlin Heidelberg