Application of Fluorogenic Derivatization-Liquid Chromatography-Tandem Mass Spectrometric Proteome Method to Skeletal Muscle Proteins in Fast Thoroughbred Horses Tomoko Ichibangase and Kazuhiro Imai* Research Institute of Pharmaceutical Sciences, Musashino University, 1-1-20 Shinmachi, Nishitokyo-shi Tokyo, Japan 202-8585 Received November 19, 2008
Abstract: To extend the applicability of the fluorogenic derivatization-high performance liquid chromatographytandem mass spectrometry (FD-LC-MS/MS) method, which consists of fluorogenic derivatization (FD), separation by liquid chromatography (LC), and identification by LCtandem mass spectrometric (MS/MS) proteomic analysis, we applied it to Thoroughbred horse muscle. With the optimization of the protein extraction and separation procedure, reproducible chromatograms were obtained and the changes in protein expressions during exercise were able to be analyzed. To quantify the changed protein expressions, the training-to-detraining (+/-) ratios for proteins were calculated, and the correlation of the ratio with the percentage of maximum oxygen consumptions (VO2max; the indicator of the running speed) was investigated. Sixteen proteins involved in energy supply, especially in anaerobic energy production, increased with an increase in VO2max, suggesting that this method was able to suggest the biochemical events in the fasterrunning horse and would be useful for evaluating the training effect in Thoroughbred horses. Keywords: FD-LC-MS/MS method • DAABD-Cl • skeletal muscle • thoroughbred horse • Equus caballus • anaerobic energy production
Introduction Proteomics studies have provided information on dynamic cellular performance at the protein level and an integrated view of an individual biological reaction, which is extremely valuable for clarifying the progression of biological changes (e.g., disease and aging). Therefore, many proteomics approaches have been developed in recent years. Among them, we have recently developed the fluorogenic derivatization-high performance liquid chromatography-tandem mass spectrometry (FD-LCMS/MS) method.1-6 This method involves fluorogenic derivatization of proteins, followed by HPLC separation of the derivatized proteins, isolation of the subject proteins, enzymatic digestion of the isolated proteins, and identification of the * To whom correspondence should be addressed. Kazuhiro Imai, Ph.D., Research Institute of Pharmaceutical Sciences, Musashino University, 1-120 Shinmachi, Nishitokyo-shi, Tokyo, Japan, 202-8585. Tel.: +81-42-468-9787. Fax: +81-42-468-9787. E-mail:
[email protected]. 10.1021/pr801004s CCC: $40.75
2009 American Chemical Society
proteins utilizing HPLC and tandem MS with a databasesearching algorithm. The method has unique features, differing from other proteome approaches in using a fluorogenic reagent to derivatize proteins and HPLC to separate the derivatized proteins. Although the method initially used a water-soluble and small-mass derivatized reagent, 7-fluoro-2,1,3-benzoxadiazole-4-sulfonate (SBD-F), as a fluorogenic reagent, the SBDpeptides after digestion with enzymes could not be sensitivity detected by MS to our satisfactory levels owing to the negatively charged sulfonyl group in the SBD moiety.1 Therefore, the water-soluble, small-mass and positively charged fluoroganic derivatized reagents for thiols, such as 7-chloro-N-[2-(dimethylamino)ethyl]-2,1,3-benzoxadiazole-4-sulfonamide (DAABDCl), were subsequently developed.2 The developed method using DAABD-Cl enables highly sensitive detection of derivatized proteins at the femtomol level due to the fluorogenic derivatization which utilizes a nonfluorescent reagent to yield highly fluorescent products with protein thiol residues; therefore, background interference from excess or degraded reagents is avoided. In addition, separation by HPLC led to highly reproducible quantification. Furthermore, a protein can be isolated after the protein go thorough a flow cell of the HPLCfluorescence detector, and then the isolated protein itself is digested into peptides and identified without losing any amino acid sequence information, including protein isoforms and post-translational modifications. As an additional benefit, the simple apparatus, consisting of a pump, a column, and a fluorescence detector, does not require a complex facility for operation. Very recently, to demonstrate the availability of the method utilizing DAABD-Cl, it has been applied to the extracts of Caenorhabditis elegans,3 mice liver,4 breast cancer cell lines,5 and mice brain,6 revealing the proteins related to early stage Parkinson’s disease, hepatocarcinogenesis, metastatic breast cancer, and aging. Because the method is highly sensitive (femtomol level) and reproducible (relative standard deviation: RSD < 10% for between-day protein peak heights), in a study of small mouse brain regions, such as cerebral cortex, hippocampus and brain stem, many kinds of the age-related proteins were demonstrated in each region. The used protein amount for the determination was dozens of times smaller than the amount using other proteomics approach, such as two-dimentional electrophoresis, and the number of the identified proteins is the greatest among the reported age-related proteins in each region of mouse brain.6 Furthermore, in a study of Journal of Proteome Research 2009, 8, 2129–2134 2129 Published on Web 02/17/2009
technical notes
Ichibangase and Imai
Table 1. Values for Body Weight and VO2max (Post-training) of Each Horse Body weight (kg) VO2 max (post-training; mL/(kg · min))
Thoroughbred-1
Thoroughbred-2
Thoroughbred-3
Thoroughbred-4
492 177
477 181
500 193
481 202
breast cancer cell lines, the presumptive mechanisms involved in invasion, metastasis, and proliferation were able to be presented.5 On the basis of this knowledge, we believe that with highly sensitive, reproducible, and quantitative proteomic analysis, the performance of dynamic tissue investigations in animal bodies can be demonstrated. Since horses have extensive interactions with humans in both a wide variety of sports (e.g., horse racing) and in working for transport, many researchers have studied their exercise physiology, histology, and biochemistry. In particular, the Thoroughbred horse is a remarkable animal, with both speed and endurance during running. To train a faster-running horse, the breeding and the training of Thoroughbred horses have been examined for more than 300 years.7-14 In this study, in order to extend the applicability of the method, we applied the FD-LC-MS/MS method to analysis of the changes in protein expressions during exercise in Thoroughbred horse muscles. Since the muscle sample is tougher and more viscous than the internal organs such as liver and brain, the protein extraction and separation procedures were investigated. Consequently, the biochemical events in the faster-running horses were suggested.
Materials and Methods Tissue Samples. Four Thoroughbred horses (males, 3 years old) that weighed 488 ( 10 kg were used in this study. Before the experiments, the horses had been acclimatized to exercise on a treadmill, and had been well-trained at the conventional intensities. Training was performed on a 10% inclined treadmill. The horses exercised for 5.0 min at three different intensities and durations. The exercise intensities were determined by percentage of maximum oxygen consumption (VO2max) (i.e., the maximum rate of oxygen uptake during an increase of workload). In this study, low-level exercise was defined as 60% VO2max (pretraining) for 6 weeks; high-level exercise was defined as 80% VO2max (pretraining) for 3 weeks; and veryhigh-level exercise was defined as 60-100% VO2max (pretraining) for 6 weeks. After the training period, detraining consisted of 3 months of pasture rest to replenish the muscle fiber and recover from physical fatigue. Following the sampling method,7 a percutaneous muscle biopsy needle was used to obtain muscle samples from the middle gluteal muscle at 5.0 cm from the skin surface. To determine protein expressions during exercise, sampling was performed after training and detraining. Muscle biopsy was performed under local anesthetic with 2.0% lidocaine. All the muscle samples were frozen in melting isopentane cooled by liquid nitrogen; they were then stored at -80 °C until analysis. Since the VO2max values generally varied with the training, the values for all the horses were redetermined on the treadmill after the training period. The values of body weight and VO2max (post-training) are summarized in Table 1. All procedures used in this study were approved by the Animal Experiment Committee of the Equine Research Institute, Tochigi, Japan. Preparation of Samples and Determination of Total Proteins. The muscle tissues (20 mg) were homogenized in 100 µL of 10 mM CHAPS (aq) with a Teflon pestle (1.5 mL) on ice. 2130
Journal of Proteome Research • Vol. 8, No. 4, 2009
average ( SD
488 ( 10 188 ( 11
The homogenization was repeated for 3.0 min until the aqueous fluid of low-viscosity was acquired. The extracted proteins were then treated according to the previous report.4 The total muscle protein was also determined according to the previous procedure.4 FD and HPLC Conditions. To determine and quantify the muscle proteins, the FD condition using DAABD-Cl (fluorogenic reagent) and separation for the derivatized proteins using HPLC were reoptimized based on the previous paper.4 Briefly, homogenized tissue was diluted with the CHAPS (aq) to 4.0 mg/mL, and 10 µL of the sample was mixed with 60 µL of a mixture of 0.83 mM Tris (2-carboxyethyl) phosphine hydrochloride (TCEP, Sigma-Aldrich, St. Louis, MO), ethylenediamine-N,N,N′,N′-tetraacetic acid sodium salt (Na2EDTA, Dojindo Laboratories, Kumamoto, Japan), and 16.6 mM CHAPS (Dojindo Laboratories) in the pH 8.7 buffer solution (6.0 M guanidine hydrochloride, Tokyo Chemical Industry, Tokyo, Japan) and 25 µL of the buffer solution and 5.0 µL of 140 mM DAABD-Cl (Tokyo Chemical Industry) in acetonitrile. Then, the reaction mixture was placed in a 40 °C water bath for 10 min; 3.0 µL of 20% trifluoroacetic acid (TFA, Wako Pure Chemicals, Osaka, Japan) was added to stop the derivatization reaction. Twenty microliters of the reaction mixture was injected into the HPLC system. In our procedure, the protein amount for the FD reaction was always constant in all the experiment (4.0 mg/mL, 10 µL), and the amount of the injection to HPLC system was also constant (8.0 µg per injection). The overall system consisted of a Hitachi L-7000 series (Hitachi High Technologies, Tokyo) and a fluorescence detector (Jasco FP2025 plus, Jasco, Tokyo; λex, 395 nm; λem, 505 nm) at a flow rate of 0.55 mL/min. The stationary phase for separation of the derivatized proteins utilized a protein column (Intrada WPRP, 250 × 4.6 mm i.d., Imtakt Co., Kyoto, Japan) with a column temperature of 60 °C. The modified mobile phases consisted of 0.20% trifluoroacetic acid (TFA) in acetonitrile/isopropanol/ water (A) 9/1/90 and 0.15% TFA in (B) 50/20/30. The eluent conditions elaborately combined isocratic with gradient elution as indicated in Figure 1. For the differential analysis, the corresponding peak height obtained from the samples after training (+) was compared with that obtained from the samples after detraining (-) of each horse. The corresponding peak was judged not only from the specific retention time of the derivatives, but also from the confirmation of the protein peak following isolation and identification of the derivatized peaks. Correlations Calculations. To investigate the relationship between the changes in protein expressions during exercise in each horse and each physical ability, the correlation coefficient values between the +/- ratio and VO2max (post-training) were calculated. Among the found proteins which fluctuated in the expressions during exercise in each horse, the fluctuated proteins in all horses were examined. As to the proteins, the correlations were investigated between the each +/- ratio obtained from each horse and VO2max values (post-training) in Table 1. Identification of the Derivatized Proteins. Among the fluctuated proteins in all horses, only the proteins which have correlation (0.200 < r < 1.000) between the +/- ratios
FD-LC-MS/MS Proteomics in Skeletal Muscle of Horse
technical notes
Figure 1. Chromatograms of derivatized proteins with DAABD-Cl (8.0 µg protein per HPLC injection) in horse muscle and the adopted gradient elution program. The chromatograms above (below) were obtained from Thoroughbred-3 after training (detraining). The fluorescence intensity in the retention time range (from 2.5 to 3.0 h and from 5.1 to 5.3 h) indicated the quarter intensity in other ranges.
and VO2max values (post-training) were identified. Each protein peak was isolated and concentrated to 5.0 µL under reduced pressure. To avoid isolation of neighboring protein peaks, only the top peak (peak no. 9 in Figure 1) was isolated. The concentrated residue was diluted with 50 mM ammonium bicarbonate solution (pH 7.8: Sigma-Aldrich) in order to digest it to a peptide mixture. The procedures for the enzymatic digestion reaction and subsequent HPLCtandem MS were described in the previous report.5 Briefly, tryptic peptides were analyzed using a nano-LC-ESI-tandem MS spectrometer (HCT plus, Bruker Daltonics, Bremen, Germany) with an Ultimate/Famos/Switchos suite of instruments (LC Packings, Dionex, IL). The MS/MS data was extracted as mgf files using Data Analysis 3.2 software (Bruker Daltonics) under the following settings: minimum total ion intensity threshold, 20 000; maximum number of compounds 500. Peptides were identified by searching the mgf files against the National Center for Biotechnology Information (NCBI)-C1, which picked the proteins including cysteine residues from the NCBI database updated on Oct. 12, 2008 (170 887 sequences for other mammalian) using the
MASCOT version 2.1.03 search engine. Database searching was performed in accordance with the previous method,5 except that “other mammalian” was selected as the taxonomy parameter (i.e., enzyme, trypsin; peptide tolerance, (0.50 Da for both the MS and the tandem MS ions; maximum allowed missed cleavage, 1; instrument type, ESITRAP). Protein scores were derived from ions scores as a nonprobabilistic basis for ranking protein hits and the protein scores as the sum of peptide scores. The score threshold to achieve p < 0.05 was set using the Mascot algorithm. Under these conditions, multiple proteins shared the found peptide. Among them, a protein with both Equus caballus as the source organism and the highest score among the organism species, except ubiquitous keratin, was determined. Each protein was identified at least twice. Statistical Analysis. The replicate experiments were performed at least three independent derivatization of the same extracted samples, and then the data was subjected to statistical analysis. The significance of the difference in means was determined by a two-tailed Student’s t test. Journal of Proteome Research • Vol. 8, No. 4, 2009 2131
technical notes Results Optimization of the Protein Extraction and Separation Procedure. Our preliminary experiments using mouse liver suggested that the homogenization with a Teflon pestle (1.5 mL) was superior in the extractability and reproducibility to that with sonication (Ultrasonic Disruptor UD-200, Tomy Seiko, Tokyo), homogenizer (Physcotron, Microtec, Chiba, Japan) and BioMasher (Wako Pure Chemicals) (data not shown). Also, the protein peaks decreased with an increase in the time to homogenize with the sonication and homogenizer. In this study, considering the above results, the pestle was utilized. In the previous method, the extraction with the pestle was performed within 1.0 min on ice to avoid the degradation of proteins.4,5 However, due to the properties of the muscle, the tissue required sufficient time to obtain the aqueous fluid of low-viscosity. Consequently, the homogenization was performed within 3.0 min, as mentioned in Materials and Methods. The effect of the isopropanol concentration in the HPLC eluent on the protein separation was also investigated in the range of 0-20%. The low-concentration (lower than 10% in the eluent) suited with the separation at 0-1.0 h and 6.0-8.0 h, while the concentration around the 10% isopropanol in the eluent was suitable to the protein separations in the range of 2.5-6.0 h and 8.0-10 h. Therefore, the concentration in the eluent increased with an increase in the separation time with gradient elution. In the range of 6.0-8.0 h, isocratic elutions were employed to prevent the protein peaks co-eluting at once. Comparison of Protein Expressions between Training and Detraining in Each Thoroughbred. Four Thoroughbred horse muscle tissues were analyzed by optimized FD-HPLCfluorescence detection. Typical chromatograms obtained from the Thoroughbred-3 muscle tissue after training and detraining are depicted in Figure 1. Each peak height shows each protein expression. The total protein amount required for quantification was 8.0 µg per HPLC injection. To identify the small protein peaks (e.g., peak 8), 32 µg protein was required. The accuracy of the method was confirmed based on the reproducibility of the peak heights. The relative standard deviation (RSD) values of peaks 1-28 (Figure 1) were in the range of 0.0-21.6% for between-days (n ) 3). Also, peak 12, which had the highest RSD value for peak height, was employed to calculate the RSD values for retention times on the chromatograms. The betweenday RSD of the retention time (n ) 3) was 0.16%. These values suggested that the method had good reproducibility. To quantify the protein expressions during exercise, the protein expressions obtained from the tissue after training were compared with those from the tissue after detraining in each horse. It was found that the expressions of 36-, 7-, 24-, and 20-proteins significantly fluctuated during exercise in Thoroughbreds 1, 2, 3, and 4. The training-to-detraining ratio (+/ -) of the changed proteins and the chromatograms obtained from each horse are summarized in Supplementary Tables S1-S4 and Figure S1-S4 (Supporting Information). Correlation of Training-to-Detraining Ratio with VO2max. Oxygen uptake (VO2) is related to total aerobic rate, and many investigators have demonstrated a significant correlation between VO2max (post-training) and the running speed of a horse.9-11 Therefore, an increase in the VO2max values (post-training) was used as an indicator of fast running in this study. To investigate the relationship between the changes in protein expressions during exercise in each horse and each physical ability, correlations of the +/- ratios with VO2max 2132
Journal of Proteome Research • Vol. 8, No. 4, 2009
Ichibangase and Imai values (post-training) were calculated. The VO2max values (post-training) of protein peaks 6 and 13 did not correlate with the +/- ratios (r ) 0.151 for peak 6 and 0.196 for peak 13). Moreover, the coefficient values of peaks 2-5, 7-8, 10-11, and 27 were false, since Thoroughbred-3’s value was far from that of the others, so that those values were excluded. The absence and false correlation were shown in Supplementary Table S6 in Supporting Information. Consequently, 17 proteins exhibited a correlation between the VO2max values (post-training) and the +/- ratios, and were isolated and identified. The results of the identification are summarized in Table 2. These proteins involved in energy production, such as aerobic and anaerobic energy supply. The identified proteins were categorized as muscle creatin kinase (M-CK) (peaks 12, 15, 16, 17, 19, and 21), glycolytic enzyme (peaks 9, 14, 18, 20, 22, 23, 24, and 26), and other proteins (peaks 1, 25 and 28).
Discussion This is the first report on proteomic analysis of horse muscle tissues. With the optimization of the protein extraction and separation procedure, reproducible chromatograms were able to be obtained from muscle tissue as well as liver and brain with small protein amounts (8.0 µg per HPLC injection). Although the composition and gradient program of the eluent were thoroughly investigated, separation of a duplicated peak (e.g., peak no. 9 in Figure 1) could not be completely achieved. The subdivided fractions of peak 9 were isolated and subjected to identification, and then all the fractions were given the same protein name, muscle-type aldolase. In this study, when the peak was identified, only the top peak of the duplicated peaks was collected and identified to avoid the isolation of neighboring protein peaks. Additionally, higher peaks, such as peaks 14 and 17, were able to be identified, but not lower peaks, such as peaks 1 and 7. To identify the two latter peaks, the repetitive isolation of each peak and the increase in the injected protein amounts were attempted, but the identification failed. The failure, especially in peak 7, would be derived from not only incomplete genomic analysis for E. caballus but also the interference from the high-abundance proteins. Therefore, in order to achieve baseline separation, the development of a higher-performance column or the improvement of a separation technique such as an introduction of a column switching separation system would be required. To quantify the changed protein expressions in the skeletal muscle during exercise, each chromatogram obtained from the sample after training was compared with that from the same sample after detraining. Because of the high sensitivity and reproducibility of the method, a great number of the proteins were significantly changed during exercise. Furthermore, the number and the kind of the changed protein differed for each horse (Supplementary Tables S1-S4 in Supporting Information). Considering the observed differences above, we speculated that there might be the relationship between the changes in protein expressions during exercise in each horse and physical ability of each horse, such as running speed. The VO2 values are related to endurance and are often used as indicators of running speed in horse studies.9-11 In this study, the VO2max values (post-training) were also used as indicators of running speed, and the correlations between VO2max (post-training) and +/- ratio were calculated. As a result, 16 protein peaks involved in energy production, such as aerobic and anaerobic energy supply, exhibited a good correlation coefficient and were
technical notes
FD-LC-MS/MS Proteomics in Skeletal Muscle of Horse
Table 2. Changed Proteins between Training (+) and Detraining (-) in Each Horse and Correlations of the +/- Ratio with VO2max (Post-training)a +/- ratio peak no.
protein name
similar similar similar similar similar similar
9
similar to Fructose-bisphosphate aldolase A (Muscle-type aldolase) Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) similar to Triosephosphate isomerase 1 similar to Pyruvate kinase 3 similar to Enolase 3, beta muscle isoform similar to Lactate dehydrogenase-A similar to Phosphoglycerate kinase 1 similar to Glucose-6-phosphate isomerase
18 20 22 23 24 26
1 25 28 a
M-CK M-CK M-CK M-CK M-CK M-CK
ND Hemoglobin, beta similar to Malate dehydrogenase, cytoplasmic
Thoroughbred-2
Muscle Creatine Kinase (M-CK) 0.27* 0.55* 0.43* 0.98 0.61* 1.05 0.43* 1.10 0.57* 1.04 0.72* 1.48
12 15 16 17 19 21
14
to to to to to to
Thoroughbred-1
Glycolytic Enzyme 0.48* 0.89
Thoroughbred-3
Thoroughbred-4
correlation
0.77 1.10 1.16 1.13 1.47 1.36
2.14* 1.68* 1.83* 1.51* 1.55* 1.84*
0.917 0.933 0.939 0.867 0.928 0.812
1.08
1.38
0.950
0.64*
1.02
1.08
1.10
0.771
0.43*
0.96
1.07
1.26
0.874
0.33* 0.42*
1.02 0.78
1.27 1.85*
1.59* 1.21
0.914 0.715
0.31* 0.42*
0.87 0.91
2.14* 1.72*
1.85* 1.64*
0.886 0.914
1.50*
0.88
1.22
2.19*
0.665
Other Proteins 0.67* 0.37* 0.63*
1.07 0.87 1.06
1.97* 1.07 1.29
2.05* 1.71* 1.04
0.959 0.953 0.595
Peak numbers correspond to those in Figure 1. Asterisks denote significant differences (two-tailed Student’s t test, p e 0.05).
identified. In horse muscle, all pathways for energy supply are designed to produce adenosine triphosphate (ATP), which is the ultimate substrate.9 The categorized M-CK peaks were detected with various retention times, despite their having the same protein name. In HPLC separation, the retention time depends on the interactions among the stationary phase, the protein molecules and the eluents; thus, those peaks would be different protein molecules, as a result of post-translational modification. To determine whether these proteins were phosphorylated, we further attempted to search the MASCOT database, considering variable modification set not only as DAABD for cysteine residues, but also as phospho- for serine, threonine, and tyrosine residues. However, phosphorylation of these proteins was not detected. Ion trap MS in the present experiment employed collisionally induced dissociation (CID) for fragmentation, which could result in the loss of weakly bonded residues such as phosphorylation and glycosylation.15,16 Hence, further study is required to clarify the same-named proteins with different retention times. However, the present study did demonstrate that the +/- ratios obtained from all M-CK peaks increased with an increase in VO2max (posttraining) values for horses. In contrast, glycolytic enzyme for glycolysis pathways were almost identified in this study and significantly correlated with the VO2max values (post-training). Scheme S1 in the Supporting Information depicts the glycolysis pathway; the detected enzymes are denoted as bold arrows in the scheme. The significant positive correlation of the +/- ratio of the enzyme with VO2max (post-training) strongly suggested that the glycolysis pathway, especially the ATP-producing reaction (phos-
phoglycerate kinase and pyruvate kinase), increased with an increase in the running speed of horses. In this study, the +/- ratios of both M-CK and glycolytic enzymes significantly correlated with the VO2max (post-training) values. However, M-CK and glycolytic enzymes are related to anaerobic energy supply, and the VO2max value indicates aerobic capacity in theory. The product of anaerobic glycolysis, pyruvate, is transported into the mitochondria, and then the metabolized product enters the aerobic TCA cycle; thus, an increase of the anaerobic glycolytic pathway might result in an increase of the aerobic TCA cycle, which produces ATP efficiently. Actually, Thoroughbred-1, which had the lowest +/ratio of the anaerobic glycolytic enzymes and VO2max value, exhibited a significant decrease in the aerobic TCA cycle enzyme, such as malate dehydrogenase (peak 28). Moreover, the expression of oxygen carrier proteins (i.e., hemoglobin (peaks 25)) significantly decreased with Thoroughbred-1. Since the TCA cycle was located in the mitochondrial matrix, the proteins related to the TCA cycle were not found and were undetected in this study. Therefore, the contribution of the aerobic pathway to the exercise was not clearly demonstrated. However, the cooperative increase in the enzymes of the anaerobic pathway, including M-CK and glycolysis, and in the proteins involved in the aerobic pathway, such as hemoglobin in Thoroughbred-4, led to the speculation that these anaerobic pathways effectively activated the aerobic pathway in fasterrunning horse. In general, the anaerobic capacity should participate in energy production during the first few minutes in Thoroughbreds as well as in human athletes, and the aerobic capacity becomes dominant in energy production during an Journal of Proteome Research • Vol. 8, No. 4, 2009 2133
technical notes exercise (e.g., racing). Therefore, it is often reported that aerobic pathways are more important during racing.9 However, measurement of the activity of the individual enzymes in the anaerobic and aerobic pathways has demonstrated that, when the energy demand exceeds the aerobic energy supply during racing, the anaerobic pathway becomes essential for energy production.8,13,14 The results obtained from the present study suggest that anaerobic ATP production and aerobic ATP production are equally important during exercise in fasterrunning horses. The results also confirmed that, in faster horses, the plasma concentration of lactate which is the last product of the anaerobic glycolytic pathway rises and more oxygen is utilized during exercise.11 In conclusion, to extend the applicability of the FD-LC-MS/ MS method, we applied it to Thoroughbred horse muscle. The protein extraction and separation procedure were optimized, enabling to analyze the changes in protein expressions during exercise in each horse. Moreover, the changes in protein expressions were related to running speed of horses, suggesting that the anaerobic pathway effectively activated the aerobic pathway in faster-running horses. Although the improvements of the separation technique and of the identification technique for the post-translational proteins were further required to demonstrate the hypothesis, it was clearly found that the proteomic analysis is very useful in demonstrating the biochemical events in Thoroughbred horses and aided in the evaluation of the training effect in racing horses.
Acknowledgment. We thank Dr. Kurosawa in the Laboratory of Racing Chemistry and Dr. Hiraga in the Equine Research Institute for supplying Thoroughbred horse muscle and for providing valuable advice for the work. A part of this work was supported by MEXT-HAITEKU (2004-2008). Supporting Information Available: . The +/- ratio of the changed proteins and chromatograms obtained from each horse were summarized in Tables S1-S4 and Figures S1-S4. Tables S5 and S6 show the full list of validated MASCOT identification details and of the absence and correlation between the +/- ratio and VO2max, respectively. Scheme S1 depicts the glycolysis pathway and the detected enzymes in this study are denoted as bold arrows in the scheme. This material is available free of charge via the Internet at http:// pubs.acs.org.
References (1) Toriumi, C.; Imai, K. An identification method for altered proteins in tissues utilizing fluorescence derivatization, liquid chromatog-
2134
Journal of Proteome Research • Vol. 8, No. 4, 2009
Ichibangase and Imai
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
(10)
(11)
(12)
(13)
(14) (15)
(16)
raphy, tandem mass spectrometry, and a database-searching algorithm. Anal. Chem. 2003, 75, 3725–3730. Masuda, M.; Toriumi, C.; Santa, T.; Imai, K. Fluorogenic derivatization reagents suitable for isolation and identification of cysteine-containing proteins utilizing high-performance liquid chromatography-tandem mass spectrometry. Anal. Chem. 2004, 76, 728–735. Ichibangase, T.; Saimaru, H.; Takamura, N.; Kuwahara, T.; Koyama, A.; Iwatsubo, T.; Imai, K. Proteomics of Caenorhabditis elegans over-expressing human alpha-synuclein analyzed by fluorogenic derivatization-liquid chromatography/tandem mass spectrometry: identification of Actin and several ribosomal proteins as negative markers at early Parkinson’s disease stages. Biomed. Chromatogr. 2008, 22, 232–234. Ichibangase, T.; Moriya, K.; Koike, K.; Imai, K. A proteomics method revealing disease-related proteins in livers of hepatitisinfected mouse model. J. Proteome Res. 2007, 6, 2841–2849. Imai, K.; Ichibangase, T.; Saitoh, R.; Hoshikawa, Y. A proteomics study on human breast cancer cell lines by fluorogenic derivatization-liquid chromatography/tandem mass spectrometry. Biomed. Chromatogr. 2008, 22, 1304–1314. Asamoto, H.; Ichibangase, T.; Uchikura, K.; Imai, K. Application of an improved proteomics method, fluorogenic derivatizationliquid chromatography-tandem mass spectrometry, to differential analysis of proteins in small regions of mouse brain. J. Chromatogr., A 2008, 1208, 147–155. Yamano, S.; Eto, D.; Hiraga, A.; Miyata, H. Recruitment pattern of muscle fibre type during high intensity exercise (60-100% VO2max) in Thoroughbred horses. Res. Vet. Sci. 2006, 80, 109–115. Eto, D.; Yamano, S.; Mukai, K.; Sugiura, T.; Nasu, T.; Tokuriki, M.; Miyata, H. Effect of high intensity training on anaerobic capacity of middle gluteal muscle in Thoroughbred horses. Res. Vet. Sci. 2004, 76, 139–144. Eaton; D. M. Energetics and Performance. In The Athletic Horse: Prinnciples and Practice of Equine Sports Medicine; Saunders: Philadelphia, PA, 1994; pp 49-61. Eaton, M. D.; Hodgson, D. R.; Evans, D. L.; Rose, R. J. Effects of low- and moderate-intensity training on metabolic responses to exercise in thoroughbreds. Equine Vet. J. Suppl. 1999, 30, 521– 527. Harkins, J. D.; Beadle, R. E.; Kamerling, S. G. The correlation of running ability and physiological variables in thoroughbred racehorses. Equine Vet. J. 1993, 25, 53–60. Yamano, S.; Eto, D.; Sugiura, T.; Kai, M.; Hiraga, A.; Tokuriki, M.; Miyata, H. Effect of growth and training on muscle adaptation in Thoroughbred horses. Am. J. Vet. Res. 2002, 63, 1408–1412. Roneus, N.; Essen-Gustavsson, B. Skeletal muscle characteristics and metabolic response to exercise in young standardbreds. Am. J. Vet. Res. 1997, 58, 167–170. Poso, A. R. Monocarboxylate transporters and lactate metabolism in equine athletes: A review. Acta Vet. Scand. 2002, 43, 63–74. Agnetti, G.; Kane, L. A.; Guarnieri, C.; Caldarera, C. M.; Van Eyk, J. E. Proteomic technologies in the study of kinases: Novel tools for the investigation of PKC in the heart. Pharmacol. Res. 2007, 55, 511–522. Syka, J. E. P.; Coon, J. J.; Schroeder, M. J.; Shabanowitz, J.; Hunt, D. F. Peptide and protein sequence analysis by electron transfer dissociation mass spectrometry. Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 9528–9533.
PR801004S
