Abstract

Methodological issues relevant to studies using microarrays and reverse transcription–polymerase chain reaction (RT–PCR) in human aging have rarely been evaluated. Because aging may accentuate biological differences between muscles, we compared transcriptome expression patterns, targeted messenger RNA (mRNA) abundance, strength, and muscle fiber type in the right and left legs of older adults. Muscle biopsies were taken from each Vastus lateralis in eight older (71 ± 2 years) men, and isometric strength was determined. Samples were analyzed using an Affymetrix gene array, ATPase histochemistry, and RT–PCR for mRNA species involved in metabolism, apoptosis, vascular growth, and antioxidant status. Microarray analysis found that 31 of 5499 genes (0.6%) were significantly different between legs (negative log of the p value [NLOGP] ≥ 2.0, but fold < 1.5), with only one gene, jumonji domain containing 1C (JMJD1C), being significantly different by ≥ 1.50-fold. None of the mRNA species, or muscle fiber type, size, or strength, was different between legs. These findings are important for the design and analysis of studies using muscle data in older adults.

MICROARRAY technology has become increasingly common in aging research (1–3), and is a powerful tool to study the response of tissues to physiological and pathological stressors (4–6). Despite literature regarding technical interpretation of array data (1,7–9), there is little information regarding study design issues relevant to muscle physiology research using microarrays. With human-based research, the genetic backgrounds of the participants in a between-group study are inherently different, and variance in small nucleotide polymorphisms is considered to be a major source of variability in gene array studies (10). Consequently, a larger sample size is often required to detect significant treatment effects where they occur.

The muscle biopsy method has been widely used in neuromuscular, exercise, and muscle physiology research for > 40 years (11–16). Recently, there has been some concern when using the multiple biopsy method in the same leg in that the biopsy measurement may induce alterations in the outcome variable of interest (i.e., inflammation, gene expression) due to the trauma from the biopsy inducing a “field effect” on adjacent tissue (17,18). By separating two biopsy sites by a larger (4–5 cm) physical distance to avoid field effects, preliminary results suggest that one biopsy does not appear to alter messenger RNA (mRNA) content in the subsequent biopsy (19). A powerful extension of the within-group study design is the unilateral leg model, where one leg is exposed to an intervention such as exercise, and the contralateral leg serves as an “internal control” for nuclear and mitochondrial DNA heteroplasmic between-group differences (20,21). This design also overcomes any possible biopsy-induced field effects as biopsies are taken from different legs. An assumption of the unilateral leg model is that both legs will be similar with respect to the outcome of interest prior to the intervention. Most studies have looked at the variability of measurements within the same leg but from multiple sites (22,23). Studies have examined the between-leg variability for muscle damage markers (24) and muscle fiber characteristics (25) in young men. One study found that there was more variance in muscle fiber cross-sectional area (CSA) in older than in younger adults (26). Given the accumulation of muscle damage, potential asymmetrical neurogenic changes, and orthopedic issues such as osteoarthritis, it is probable that leg to leg variance in gene expression would become more divergent with aging.

Consequently, the purpose of the current study was to compare the transcriptome expression pattern and muscle fiber characteristics in the Vastus lateralis between the right and left leg of the same eight older adults, taken simultaneously. The current report showed similar mRNA abundance, strength, and muscle fiber characteristics between legs; these results will allow for future study designs such as the use of one leg as a baseline with comparison to samples taken from a contralateral leg after exposure to an intervention such as immobilization or exercise in older adults.

Methods

Participants

A total of eight older males (> 65 years) volunteered for the study after being informed of the risks and benefits of participation and after signing a consent form. The study was approved by the McMaster University Medical Center Ethics Committee. All participants were relatively sedentary, did not participate in any formal exercise training, and had not performed resistance exercise for at least 10 years prior to the start of the study. All participants were right handed and did not perform any sports or activities that required unilateral leg dominance.

Design

Participants had their maximal isometric knee extension strength and body anthropometry, as well as fat-free mass by bioelectrical impedance (BIA 101A; RJL Systems, Clinton Twp, MI), determined. Maximal knee extension force was recorded with participants seated in an isokinetic dynamometer (Biodex System 3; Biodex Medical Systems, Shirley, NY). Participants were strapped into the dynamometer such that their upper body movement was restricted and they were only allowed to hold onto handles under their seat while generating force. The knee was positioned such that the rotation of the fulcrum of the Biodex was aligned with the center of the knee joint while the participant's thigh was strapped to minimize movement. Participants had their knees positioned at 120° of extension (where 180° is full extension), as this is the optimal joint angle for force generation. Three attempts (per leg), each lasting 5 seconds, interspersed with 2-minute rest periods, were made to generate a maximal effort with the highest peak torque being recorded and judged to be maximal. At least 2 weeks after strength testing, participants had a muscle biopsy taken from the right and left Vastus lateralis muscle, ∼15 cm proximal to the lateral joint line, under local anesthesia (2% lidocaine) and 4 hours postabsorptive. Care was taken to advance the needle ∼1.5–2.0 cm beyond the outer fascia to minimize variance induced by the depth of the muscle biopsy sample (26). The samples were taken in the afternoon, and participants did not perform any formal physical activity for > 72 hours before the procedure. A sample (∼75 mg) was immediately cut into two pieces (∼50 mg) for gene array analysis and 25 mg for reverse transcription–polymerase chain reaction (RT–PCR), and placed into individual RNase-free polyethylene tubes and immersed in liquid nitrogen. Another sample was mounted in optimal cutting temperature (OCT) medium for subsequent histochemical analysis. All samples were stored at −86°C prior to analysis.

Microarray Analysis

Frozen human Vastus lateralis muscle was homogenized in TRIzol (Life Technologies, Rockville, MD) using tungsten carbide beads by shaking in a mixer mill as recommended by the manufacturer (Qiagen, Life Technologies, Rockville, MD). RNA samples were prepared according to Affymetrix (Santa Clara, CA) recommendations. After the TRIzol extraction, the RNA was purified with an RNeasy Mini Kit (Qiagen, Chatsworth, CA). Reverse transcription was performed on 10 μg of total RNA with the use of SuperScript II Reverse Transcriptase and a T7-(dT)24 primer followed by second-strand DNA synthesis using T4 DNA polymerase as recommended by the manufacturer (Life Technologies). Contaminants were removed from the double-stranded complementary DNA (cDNA) by phenol/chloroform/isoamyl alcohol extraction, and then cDNA was recovered by ethanol precipitation. An RNA Transcript Labeling Kit (Enzo Diagnostics, Farmingdale, NY) was used for production of biotin-labeled complementary RNA (cRNA) targets by in vitro transcription from T7 RNA polymerase promoters, all as recommended by the kit manufacturer. The cDNA prepared from total RNA was used as a template in the presence of a mixture of unlabeled ATP, GTP, CTP, and UTP and biotinylated CTP and UTP. In vitro transcription products were purified with an RNeasy Mini Kit (Qiagen, Life Technologies) to remove unincorporated NTPs and fragmented to approximately 35–200 bases by incubation at 94°C for 35 minutes in fragmentation buffer containing Tris-acetate, potassium acetate, and magnesium acetate. Fragmented cRNA was stored at −20°C until the hybridization was performed.

Biotinylated and fragmented cRNA was hybridized for 16 hours at 45°C to a set of human HG_U95A arrays (Affymetrix) in a GeneChip Hybridization Oven 640 (Affymetrix). A series of stringency washes and staining with streptavidin-conjugated phycoerythrin was then performed in a GeneChip Fluidic Station 400 according to the protocol recommended by Affymetrix. Probe arrays were then scanned with a GeneArray Scanner (Agilent, Palo Alta, CA). The images were analyzed with GeneChip Analysis software (Affymetrix). A total of 16 arrays were used in the current study with each leg (8 × 2) being hybridized to a single chip.

Gene Expression Analysis

Statistical analysis was based on the Affymetrix signal (MAS 5.0 algorithm). Data quality and outliers were checked using exploratory statistical tools, such as summary statistics, pair plots, and principal components analysis. The data from only one Affy chip was questionable based on background and noise comparison with the other chips. The pair plots of that array showed slight nonlinear deviation from the 45° line (the line of agreement with other replicates). Quantile normalization was used after MAS 5.0 calculations to account for nonlinear deviation, and the data from this chip were used in the final analysis (http://biosun01.biostat.jhsph.edu/∼ririzarr/papers/index.html).

Gene filtering was performed based on the minimum number of Affy Absent Calls. Genes were removed from further analysis if an Absent Call was obtained for both legs. After Absent Calls filter, there were 5499 genes (of 12,558 genes on an HG_U95A chip) left for further statistical analysis. Log (base 2) of the Affy signal was used as a gene expression measure in the statistical analysis of the 5499 genes that are not filtered. The negative log of the p value (NLOGP) was calculated from −log10 (NLOGP 2 = p value threshold of.01), where the p value is a summary measure of the statistical significance for the corresponding comparison (analysis of variance [ANOVA] model and t test).

A variety of public resources as well as proprietary tools was used to annotate the expressed genes, including Netaffx (http://www.affymetrix.com/index.affx) and GeneCards (Weizmann Institute Crown Human Genome Center, http://bioinfo.weizmann.ac.il). Gene Ontology (http://www.geneontology.org) was used to cluster genes into functional categories. For those uncharacterized transcriptional probes or expressed sequence tags (ESTs), basic local alignment search tool (BLASTN and BLASTX) searches of nonredundant databases and BLASTN searches of the data base of EST at the National Center for Biotechnology Information were performed.

RNA Preparation for RT–PCR

Total RNA was extracted from ∼25 mg of frozen muscle and treated with DNase I as previously described (6,27). Briefly, ∼25 mg of muscle was removed from the freezer and immediately immersed in 1 mL of TRIzol reagent (catalog no.15596; Invitrogen, Carlsbad, CA) and homogenized for 30 seconds on ice using a glass homogenizer. The homogenate sat at room temperature for 5–10 minutes, 0.2 mL of chloroform was added, and samples were vortexed for 15 seconds. After a 5-minute incubation at room temperature, the sample was centrifuged at 14,500 g at 4°C for 15 minutes. The aqueous phase was transferred to a fresh polyethylene tube, mixed with 500 μL of isopropanol, left for 10 minutes at room temperature to precipitate the RNA, and centrifuged at 14,500 g at 4°C for 10 minutes. The RNA pellet was washed twice with 75% ethanol at room temperature, air dried and dissolved in 14 μL of diethyl pyrocarbonate-treated ddH2O (RNase-free), aliquoted, and stored at −80°C.

The concentration and purity of the RNA was determined in duplicate using ultraviolet (UV) spectrophotometry (GeneQuant pro; GE Healthcare Bio-Sciences Corp, Piscataway, NJ) by measuring the absorbance at 260 (optical density [OD]260) and 280 (OD280) nm with an average coefficient of variation of < 10%. The average purity (OD260/OD280) of the samples was > 1.6, and RNA integrity was further assessed in a randomly chosen subset of samples using agarose gel electrophoresis, with OD ratios of 28S to 18S ribosomal RNA (rRNA) being consistently > 1 for each sample checked, indicating high quality RNA.

TaqMan Real-Time RT–PCR

Prior to TaqMan real-time RT–PCR, RNA samples were treated with DNase I for 25 minutes at 37°C to remove any contaminating DNA (DNA-free, catalog no. 1906; Ambion Inc, Austin, TX) according to the manufacturer's instructions. DNase I was inactivated by a commercially available inactivation solution (DNase inactivation solution, catalog no.1906; Ambion) according to the manufacturer's instructions. Specific primers and probe to each target gene were designed based on the cDNA sequence in GenBank (http://www.ncbi.nlm.nih.gov/entrez/query.fcgi) with primer 3 designer (http://frodo.wi.mit.edu/cgi-bin/primer3/primer3_www.cgi). All probes were dual-labeled with fluorophores, with a fluorescent reporter dye at its 5′ end (FAM, 6-carboxyfluorescein) and a corresponding quencher dye at its 3′ end (BHQ-1). A list of the mRNA species quantified in the current study and the primer/probe sequences is found in Table 1. In order to determine whether any significantly different gene array results were indicative of false-positive calls, we measured mRNA content using RT–PCR and also determined whether there were any differences in the mRNA content of several mRNA species that are involved in a variety of cellular functions implicated in aspects of aging including apoptosis (B cell CCL/lymphoma 2 [bcl-2]), anti-oxidant enzyme (catalase), mitochondrial biogenesis (peroxisome proliferator-activated receptor-gamma, coactivator 1-alpha [PGC1-α]), fatty acid synthesis (long-chain acyl-coenzyme A synthase [LCAS]), uncoupling proteins (UCP-2, UCP-3), and blood vessel growth (vascular endothelial growth factor-1 [VEGF]). We also chose to look at the aforementioned mRNA species for most are acutely sensitive to an acute exercise stimulus (6), and thus could be altered if one leg was preferentially used in daily activities.

One-tube (i.e., RT and PCR step occur in the same tube) real-time RT–PCR was performed using a real-time PCR system (iCycler; Bio-Rad Laboratories, Hercules, CA) in the one-step Taqman RT-PCR master mixture reagents (part no. 4309169; Applied Biosystems, Foster City, CA) according to the manufacture's instruction. For testing the target gene mRNA expression, duplex RT–PCR was performed with both target gene primers and probe and internal standard gene primers and probe in the same reaction. Given our experience with human β2-microglobulin as a stable reference gene (27), we chose to use this as our internal standard. We optimized the PCR conditions for each gene so that the amplification efficiency for both target gene and internal reference was close to 1 (Table 2). The difference between the slopes of the regression curves of each target gene and the corresponding internal standard gene (Ct vs amount of RNA) was < 10% in all cases. All samples were run in duplicate. Fluorescence emission was detected through a filter corresponding to the reporter dye at the 5′ end of each probe, and Ct was automatically calculated. These transcripts were extensively optimized, run simultaneously with RNA- and RT-negative controls, and agarose gel electrophoresis was used to confirm the specificity of the priming. The RT–PCR conditions for the evaluation of β2-microglobulin were the same as we have previously reported (27).

Histochemical Analysis

The OCT-embedded blocks were sectioned using a cryotome (Microm, Heidelberg, Germany) at 10-μm thickness and stained for myosin ATPase activity at a preincubation pH of 4.6 as previously described by our group (28). After being stained, pictures of the muscle were collected under ×200 magnification with the use of a microscope (Olympus BX60; Melville, NY) and camera with software (SPOT; Diagnostic Instruments Inc., Houston, TX). Images were analyzed for fiber number, size, and area by using the Image-Pro Plus (version 4; Media Cybernetics, Silver Spring, MD) computer program. Muscle fibers were classified as type I, IIa, IIax, or IIx based on staining intensity. To aid in the categorization of fibers, visual as well as OD ‘bins’ were defined based on a 255 grayscale shaded image of each slide based on the darkest fibers' (type I) staining intensity, the lightest fibers' (type IIa) staining intensity, and the intermediate fibers' (IIx or IIax) staining intensity. What could be classified as ‘pure’ type IIax fibers were not always readily distinguishable, so we distributed the fibers classified as type IIax fibers in the analysis to the other fiber types IIa (50%) and IIx (50%) pools, as has been described previously (29).

A minimum of 50 fibers were counted for each fiber type. Total fiber counts per image at each time point to determine fiber type and fiber CSA were 185 ± 3.2, which was comprised of a mean of 57 ± 9 type I, 60 ± 17 type IIa, 17 ± 9 IIax, and 30 ± 7 type IIx fibers. The total CSA of each fiber type was summed using the SPOT software, and the relative fiber percent for each fiber type (I, IIa, IIx) was then calculated as the sum of that fiber's CSA divided by the sum of the total fiber CSA (I + IIa + IIx).

Statistical Analysis (Other Than Gene Array)

The analysis in the current study represents a repeated-measures analysis within the same human (i.e., identical genetic background), and the only variable was the right and left leg Vastus lateralis muscle. For the RT–PCR confirmation and exploratory data analysis, as well as for leg fiber type and strength data, we performed a paired t test using a statistical analysis package (Excel; Microsoft, Redmond, WA). Statistical analyses regarding mRNA expression were performed using linear data and 2−Ct for evaluation of internal standards and 2−ΔCt for target gene normalized with internal reference (30). All data were analyzed using paired t test, and a p value of.05 or less was considered significant. All data in tables and figures are presented as mean ± standard error of the mean (SEM). For jumonji domain containing 1C (JMJD1C) mRNA confirmation (see below), we used a one-tailed test given that the direction was predicted a priori by the gene array, and used a two-tailed test for the other mRNA species to be more conservative and given that no a priori directional change was specified.

Results

Participant Characteristics

The mean age of the participants was 71 ± 2 years, height = 175 ± 2 cm, weight = 88.8 ± 3.1 kg, body mass index = 28.9 ± 0.9 kg/m2, and fat-free mass = 66.9 kg. Maximal isometric knee extension torque was similar between legs (left leg = 136 ± 12 Nm; right leg = 127 ± 16 Nm; p = nonsignificant).

Muscle Fiber Type Differences Between Legs

There were no between-leg differences in muscle fiber type proportion (% type I, IIA, or IIX), and the mean fiber area of the different types was not statistically different between legs (Table 3).

Microarray Analysis

Of the 5499 mRNA species detected in skeletal muscle of the participants, only 31 (0.6%) were considered to be significantly (i.e., NLOGP ≥2.0) different between the legs. Traditionally, a fold change of at least 2.0 has been used to indicate “significance,” with some studies using a cutoff of at least a 5-fold change (31). Even with a 2-fold threshold, none of the mRNA species in the current study would be considered to be “significant” by some nonstatistical, but traditional, criteria. Thirty of the mRNA species were < 1.5-fold different between legs, with one gene (0.02% of the total number detected on the array) showing just over a ≥1.5-fold difference (JMJD1C = + 1.52-fold, Table 4). Many of the differentially expressed genes were of unknown function, and there were no consistent patterns or groups of functionally related genes. A complete list of the microarray data conforming to minimal information about microarray experiments (MIAME) criteria can be found at http://www.ebi.ac.uk/arrayexpress/ with the accession number, E-MEXP-740.

RT–PCR

There were no significant differences between right and left legs in β2-microglobulin mRNA expression; consequently, we used this as our internal control for RT–PCR expression. In contrast with the array results, the JMJD1C mRNA content was not statistically different between legs using the RT–PCR method (p = nonsignificant, Figure 1), and we consider the JMJD1C array result to be a false-positive call. We also did not find any statistically significant between-leg differences in mRNA content for any of the other species of interest: PGC-1α, UCP2, Bcl-2, and VEGF (Figure 1A) or catalase, UCP3, and LCAS (Figure1B).

Discussion

Our data show that there are very few statistically significant between-leg differences in the transcriptome expression pattern in older men. Furthermore, there is little between-leg variance in isometric strength, fiber type distribution, and muscle fiber area. These results indicate that a pre- to postinterventional comparison within the same participant (with biopsies taken from different legs) would show little between-leg variance and provide valid results regarding muscle mRNA abundance, muscle fiber characteristics, and isometric strength.

One study has reported that the transcriptome expression pattern was similar in the right, as compared with the left, Vastus lateralis in four young (∼22 years) men whose biopsies were taken 4 weeks apart (32). Only two genes (phosphoinositide-3-kinase regulatory subunit [+ 1.87-fold] and thyroid hormone receptor–associated protein [+ 1.92-fold]) were differentially expressed using a fold ≥1.5 and statistical analysis in that study (32). One strength of the current study was that our biopsies were taken at the same time [< 10 minutes apart vs 4 weeks (32)] and from twice the number of participants (32), which reduces variability and the risk of a type II statistical error. However, our data do support the previous conclusion (32) that biopsies taken from the right and left leg in the same person show similar transcriptome expression patterns. Importantly, with the interest in the effects of aging on skeletal muscle structure and function and gene expression (33), it is important that our data now extend to this older age group.

The extremely small number of between-leg genes that were found to be marginally significantly different (0.6%) was well below the typical cutoff for false discovery rates (FDR) of 5% commonly used in the analysis of microarray data. We found that only a single mRNA species (JMJD1C) was considered to be significantly differentially expressed between right and left legs using Affymetrix gene chip analysis. Given that this gene was not confirmed (using a targeted RT–PCR approach) to be significantly different between legs, we consider the array result to be a false positive. Given that this was chosen to be the most likely of all the genes shown to be significant by gene array analysis and all of the other mRNA species on the array were all < 1.5-fold induced, it is likely that they too would not be confirmed by RT–PCR analysis and be deemed to be a false discovery. Furthermore, the fact that none of the other specifically targeted mRNA species that we measured by RT–PCR were different between legs suggests that the transcriptome expression pattern was truly similar between legs. It is important to note that the current study was completed in the Vastus lateralis muscle, and it is possible that more between-limb variance would be apparent in a muscle such as the Biceps brachii, where upper limb dominance, due to handedness, may have a significant effect on muscle activity and alter the stimulus for gene expression due to physical activity (6). Our data also support the fact that the mRNA steady-state content was similar between legs using statistical analysis of RT–PCR data for all eight of the targeted genes studied. We did find that one gene (VEGF) was > 1.5-fold different. This finding further emphasizes that statistical analysis must be applied to both gene arrays as well as to RT–PCR analysis to avoid type I errors. For example, due to high values in two individuals in one leg for the VEGF RT–PCR analysis, the fold change was > 1.5, yet it was not statistically significant. This result reveals the pitfall of relying solely on fold changes in mRNA abundance, which can lead to false-positive results. In general, a low sample size and a variety of noncontrollable factors add to variance in human experimentation and can lead to fold changes that appear significant when not considering statistical analysis of the data.

From a practical perspective, the current study demonstrates that the transcriptome pattern from one leg can be used as a baseline for a postinterventional biopsy obtained in the contralateral leg. One value of such a strategy is that the contralateral leg would not require a preintervention biopsy, which would minimize the potential influence on the expression pattern of inflammation and/or damage from any previous biopsy. For example, one study found that repeated anatomically adjacent biopsies resulted in elevations in mRNA species that were considered to be exercise-responsive genes (18). Another study has implied that much of the inflammation reported after exercise was due to local inflammation around the previous biopsy site (17). The current data indicate that, at least with mRNA and fiber type, a basal biopsy could be taken in one leg, exercise could be performed in the contralateral leg, and a postintervention biopsy could be taken from that leg with no antecedent biopsy to influence the results. Furthermore, in combination with the data showing stability in the transcriptome pattern between legs after a 4-week interval (32), our data imply that a biopsy taken in one leg could be used as a valid “basal” sample for within-group comparison to samples taken up to 4 weeks later in both the ipsilateral leg (with 4–5 cm of separation) and in the contralateral leg after unilateral intervention. This latter strategy would reduce the total number of muscle biopsies, which may be an issue for participant ethics and recruitment issues, particularly in older adults and patient populations.

Finally, we have shown that strength, fiber type proportion, and fiber area were similar between legs in healthy older adults. Lexell and Taylor (25) found that the mean muscle fiber size was < 25%, and mean fiber type proportion was < 5%, different between legs for young men. Our data showed that the variance in fiber type proportion was marginally higher at 10.5%, yet fiber size was more homogenous (2.7%) in older adults. Furthermore, strength was within 6.6% between legs in the current study for older adults. The latter finding is important, for our data are not likely to be relevant to older adult populations where one leg is disproportionately affected by a disease process such as unilateral osteoarthritis. Consequently, it may be prudent to use a noninvasive measurement such as knee extension strength to ensure similarity between legs before assuming relative homogeneity in mRNA abundance or muscle fiber characteristics.

Overall, the current data suggest that the transcriptome expression pattern, muscle fiber characteristics, and strength are very similar between legs in older adults. These data will allow for study designs in which one leg can be used as a control for the contralateral leg (i.e., biopsy of one leg exposed to exercise with the other as the simultaneous resting control). Such considerations will avoid any potential confounding influence of a biopsy in one leg influencing the results of a subsequent biopsy.

Decision Editor: Huber R. Warner, PhD

Figure 1.

Messenger RNA (mRNA) content for each of eight selected genes expressed relative to β2-microglobulin (2−ΔCt, arbitrary units). The description of each mRNA species is given in the Methods section. There were no statistically significant differences in mRNA content for the selected genes between the right (open bars) and left (closed bars) legs

Figure 1.

Messenger RNA (mRNA) content for each of eight selected genes expressed relative to β2-microglobulin (2−ΔCt, arbitrary units). The description of each mRNA species is given in the Methods section. There were no statistically significant differences in mRNA content for the selected genes between the right (open bars) and left (closed bars) legs

Table 1.

Primers and Probe Sets.

Genes GenBank Number* Primers and Probe Set
 
 
JMJD1C NM_032776 Left primer 5′-cacctcttgtgtcccagaataa-3′ 
  Right primer 5′-gcaatgccagcatctgtagac-3′ 
  Probe 5′-caaatcccgtaaggttgagccttgttc-3′ 
PGC-1α NM_013261 Left primer 5′-ttgctaaacgactccgagaa-3′ 
  Right primer 5′-tgcaaagttccctctctgct-3′ 
  Probe 5′-aacagttgggctgtcaacattcaaagc-3′ 
Bcl-2 NM_000633 Left primer 5′-tctttggaaatccgaccacta-3′ 
  Right primer 5′-caacatggaaagcgaatctatg-3′ 
  Probe 5′-ccacctggatgttctgtgcctgtaaa-3′ 
LCAS D10040 Left primer 5′-ggaagcgttcgtgtttgact-3′ 
  Right primer 5′-tctgcaacatgaggtgactgta-3′ 
  Probe 5′-aatgctagaggaaacagaacaccgcct-3′ 
Catalase NM_001752 Left primer 5′-actgaggtccaccctgactac-3′ 
  Right primer 5′-tcgcattcttaggcttctca-3′ 
  Probe 5′-ccaggctcttctggacaagtacaatgc-3′ 
VEGF AF022375 Left primer 5′-ccttgccttgctgctctac-3′ 
  Right primer 5′-cgctgatagacatccatgaact-3′ 
  Probe 5′-cacttcgtgatgattctgccctcct-3′ 
UCP3 AF001787 Left primer 5′-gaaaggaactttgcccaacat-3′ 
  Right primer 5′-ttgtcagtgagcaggtggtagt-3′ 
  Probe 5′-cagcacagttgacgatagcattcctca-3′ 
UCP2 NM_003355 Left primer 5′-tcatcacctttcctctggatac-3′ 
  Right primer 5′-agaatggtgcccatcacac-3′ 
  Probe 5′-cctgactttctccttggatctgtaaccg-3′ 
Genes GenBank Number* Primers and Probe Set
 
 
JMJD1C NM_032776 Left primer 5′-cacctcttgtgtcccagaataa-3′ 
  Right primer 5′-gcaatgccagcatctgtagac-3′ 
  Probe 5′-caaatcccgtaaggttgagccttgttc-3′ 
PGC-1α NM_013261 Left primer 5′-ttgctaaacgactccgagaa-3′ 
  Right primer 5′-tgcaaagttccctctctgct-3′ 
  Probe 5′-aacagttgggctgtcaacattcaaagc-3′ 
Bcl-2 NM_000633 Left primer 5′-tctttggaaatccgaccacta-3′ 
  Right primer 5′-caacatggaaagcgaatctatg-3′ 
  Probe 5′-ccacctggatgttctgtgcctgtaaa-3′ 
LCAS D10040 Left primer 5′-ggaagcgttcgtgtttgact-3′ 
  Right primer 5′-tctgcaacatgaggtgactgta-3′ 
  Probe 5′-aatgctagaggaaacagaacaccgcct-3′ 
Catalase NM_001752 Left primer 5′-actgaggtccaccctgactac-3′ 
  Right primer 5′-tcgcattcttaggcttctca-3′ 
  Probe 5′-ccaggctcttctggacaagtacaatgc-3′ 
VEGF AF022375 Left primer 5′-ccttgccttgctgctctac-3′ 
  Right primer 5′-cgctgatagacatccatgaact-3′ 
  Probe 5′-cacttcgtgatgattctgccctcct-3′ 
UCP3 AF001787 Left primer 5′-gaaaggaactttgcccaacat-3′ 
  Right primer 5′-ttgtcagtgagcaggtggtagt-3′ 
  Probe 5′-cagcacagttgacgatagcattcctca-3′ 
UCP2 NM_003355 Left primer 5′-tcatcacctttcctctggatac-3′ 
  Right primer 5′-agaatggtgcccatcacac-3′ 
  Probe 5′-cctgactttctccttggatctgtaaccg-3′ 

Notes: *GenBank numbers can be accessed at http://www.ncbi.nlm.nih.gov/entrez/query.fcgi

JMJD1C = jumonji domain containing 1C; PGC-1α = peroxisome proliferator-activated receptor gamma, coactivator 1-alpha; bcl-2 = B cell CCL/lymphoma 2; LCAS = long-chain acyl-coenzyme A synthase; VEGF = vascular endothelial growth factor-1; UCP = uncoupling protein.

Table 2.

TaqMan Real-Time RT–PCR Reaction Conditions.

Genes Target Gene Left Primer (μM) Target Gene Right Primer (μM) Target Gene Probe (μM) β2-M Left Primer (μM) β2-M Right Primer (μM) β2-M Probe (μM) RNA Template (ng) Reaction Volume (μL) 
JMJD1C 0.3 0.3 0.1 0.3 0.3 0.05 10.0 25 
PGC-1α 0.9 0.9 0.1 0.3 0.3 0.05 5.0 25 
Bcl-2 0.9 0.9 0.2 0.3 0.3 0.05 12.5 25 
LCAS 0.2 0.2 0.1 1.2 1.2 0.05 7.5 25 
Catalase 0.6 0.6 0.1 0.8 0.8 0.05 10.0 25 
VEGF 0.1 0.1 0.15 0.4 0.4 0.05 10.0 25 
UCP3 0.9 0.3 0.15 0.6 0.6 0.05 5.0 25 
UCP2 0.3 0.3 0.1 0.2 0.2 0.05 7.5 25 
Genes Target Gene Left Primer (μM) Target Gene Right Primer (μM) Target Gene Probe (μM) β2-M Left Primer (μM) β2-M Right Primer (μM) β2-M Probe (μM) RNA Template (ng) Reaction Volume (μL) 
JMJD1C 0.3 0.3 0.1 0.3 0.3 0.05 10.0 25 
PGC-1α 0.9 0.9 0.1 0.3 0.3 0.05 5.0 25 
Bcl-2 0.9 0.9 0.2 0.3 0.3 0.05 12.5 25 
LCAS 0.2 0.2 0.1 1.2 1.2 0.05 7.5 25 
Catalase 0.6 0.6 0.1 0.8 0.8 0.05 10.0 25 
VEGF 0.1 0.1 0.15 0.4 0.4 0.05 10.0 25 
UCP3 0.9 0.3 0.15 0.6 0.6 0.05 5.0 25 
UCP2 0.3 0.3 0.1 0.2 0.2 0.05 7.5 25 

Notes: β2-microglobulin (β2-M) primers and probes were the same as used by Mahoney and colleagues (6). All other primers and probes were designed using Primer3 software. Specificity was checked using BLAST, and thermal dynamics were manipulated by calculating delta G with Analyzer of Oligo (IDT) at http://www.idtdna.com/analyzer/Applications/OligoAnalyzer/Default.aspx

RT–PCR = reverse transcription–polymerase chain reaction; JMJD1C = jumonji domain containing 1C; PGC-1α = peroxisome proliferator-activated receptor gamma, coactivator 1-alpha; bcl-2 = B cell CCL/lymphoma 2; LCAS = long-chain acyl-coenzyme A synthase; VEGF = vascular endothelial growth factor-1; UCP = uncoupling protein.

Table 3.

Muscle Fiber Characteristics.

Characteristic Right Leg Left Leg p Value % Difference 
Fiber proportion     
    Type I (%) 37.4 ± 12.8 43.0 ± 13.7 .51 13.0% 
    Type IIA (%) 40.1 ± 13.7 36.9 ± 16.2 .67 7.9% 
    Type IIX (%) 22.5 ± 9.2 20.1 ± 15.6 .49 10.7% 
Mean fiber area     
    Type I (μm23815 ± 1423 3966 ± 935 .48 3.8% 
    Type IIA (μm24520 ± 1235 4645 ± 1299 .61 2.7% 
    Type IIX (μm23439 ± 1292 3497 ± 1410 .55 1.7% 
Characteristic Right Leg Left Leg p Value % Difference 
Fiber proportion     
    Type I (%) 37.4 ± 12.8 43.0 ± 13.7 .51 13.0% 
    Type IIA (%) 40.1 ± 13.7 36.9 ± 16.2 .67 7.9% 
    Type IIX (%) 22.5 ± 9.2 20.1 ± 15.6 .49 10.7% 
Mean fiber area     
    Type I (μm23815 ± 1423 3966 ± 935 .48 3.8% 
    Type IIA (μm24520 ± 1235 4645 ± 1299 .61 2.7% 
    Type IIX (μm23439 ± 1292 3497 ± 1410 .55 1.7% 
Table 4.

Gene Annotation From Microarray Data.

Title Acronym Fold Δ NLOGP Genbank ID 
Matrix-remodeling associated 5 MXRA5 −1.4 2.1 AL0499 
Glutathione transferase zeta 1 GSTZ1 −1.35 2.1 U86529 
Cytoskeleton-associated protein 4 CKAP4 −1.34 2.1 X69910 
Annexin A2 ANXA2 −1.33 2.3 M62895 
Death-associated protein kinase 3 DAPK3 −1.28 2.4 AB0071 
CUG triplet repeat, RNA binding protein 1 CUGPB1 −1.26 2.5 AI0955 
Zinc finger protein 148 (pHZ-52) ZNF148 −1.23 2.1 L04282 
Cysteine-rich protein 1 (intestinal) CRIP1 −1.19 2.0 AI0175 
Component of oligomeric golgi complex 4 COG4 −1.18 2.1 AL0501 
Cytochrome c oxidase subunit IV isoform 1 COX4I1 1.11 2.6 AF0171 
Glutamate receptor interacting protein 2 GRIP2 1.14 3.1 AF0521 
mutL homolog 1, colon cancer, nonpolyposis MLH1 1.18 2.3 U07418 
Testis-enhanced gene transcript (BAX inhibitor 1) TEGT 1.18 2.0 X75861 
Transmembrane protein 4 TMEM4 1.18 2.2 AB0156 
NADH coenzyme Q reductase, 75kDa NDUFS1 1.19 2.4 X61100 
H3 histone, family 3B (H3.3B) H3F3B 1.2 2.6 Z48950 
Ubiquitin specific peptidase 2 USP2 1.2 2.3 AF0795 
Diacylglycerol kinase, zeta 104 kd DGKZ 1.2 2.6 U94905 
Dehydrogenase/reductase (SDR family) member 3 DHRS3 1.2 2.5 AF0617 
Transcription factor 25 (basic helix-loop-helix) TCF25 1.23 2.4 AB0289 
Glyoxylate reductase/hydroxypyruvate reductase GRHPR 1.24 2.4 W28944 
SWI/SNF related, actin dependent regulator of chromatin, subfamily a, member 2 SMARCA 1.26 2.1 D26155 
Guanine nucleotide binding protein (G protein), alpha 11 (Gq class) GNA11 1.26 2.0 N36926 
Aldo-keto reductase family 1, member C3 (3-alpha hydroxysteroid dehydrogenase, type II) AKR1C3 1.27 2.0 D17793 
Aconitase 2, mitochondrial ACO2 1.29 3.6  
Hypothetical protein LOC201229 LOC201229 1.29 2.1 AI6258 
Chromosome 9 open reading frame 97 C9ORF97 1.3 2.2 AI9510 
Clone 23688 mRNA sequence  1.35 2.4 AF0521 
DnaJ (Hsp40) homolog, subfamily B, member 12 DNAJB12 1.37 2.0 W73046 
5,10-methenyltetrahydrofolate synthetase MTHFS 1.37 2.3 L38928 
Nuclear factor I/B NFIB 1.38 2.5 U70862 
Heat shock 70 kd protein 2 HSPA2 1.39 2.0 L26336 
TAF5-like RNA polymerase II, 65 kd TAF5L 1.42 2.0 AJ0097 
Jumonji domain containing 1C JMJD1C 1.52 2.5 L40411 
Title Acronym Fold Δ NLOGP Genbank ID 
Matrix-remodeling associated 5 MXRA5 −1.4 2.1 AL0499 
Glutathione transferase zeta 1 GSTZ1 −1.35 2.1 U86529 
Cytoskeleton-associated protein 4 CKAP4 −1.34 2.1 X69910 
Annexin A2 ANXA2 −1.33 2.3 M62895 
Death-associated protein kinase 3 DAPK3 −1.28 2.4 AB0071 
CUG triplet repeat, RNA binding protein 1 CUGPB1 −1.26 2.5 AI0955 
Zinc finger protein 148 (pHZ-52) ZNF148 −1.23 2.1 L04282 
Cysteine-rich protein 1 (intestinal) CRIP1 −1.19 2.0 AI0175 
Component of oligomeric golgi complex 4 COG4 −1.18 2.1 AL0501 
Cytochrome c oxidase subunit IV isoform 1 COX4I1 1.11 2.6 AF0171 
Glutamate receptor interacting protein 2 GRIP2 1.14 3.1 AF0521 
mutL homolog 1, colon cancer, nonpolyposis MLH1 1.18 2.3 U07418 
Testis-enhanced gene transcript (BAX inhibitor 1) TEGT 1.18 2.0 X75861 
Transmembrane protein 4 TMEM4 1.18 2.2 AB0156 
NADH coenzyme Q reductase, 75kDa NDUFS1 1.19 2.4 X61100 
H3 histone, family 3B (H3.3B) H3F3B 1.2 2.6 Z48950 
Ubiquitin specific peptidase 2 USP2 1.2 2.3 AF0795 
Diacylglycerol kinase, zeta 104 kd DGKZ 1.2 2.6 U94905 
Dehydrogenase/reductase (SDR family) member 3 DHRS3 1.2 2.5 AF0617 
Transcription factor 25 (basic helix-loop-helix) TCF25 1.23 2.4 AB0289 
Glyoxylate reductase/hydroxypyruvate reductase GRHPR 1.24 2.4 W28944 
SWI/SNF related, actin dependent regulator of chromatin, subfamily a, member 2 SMARCA 1.26 2.1 D26155 
Guanine nucleotide binding protein (G protein), alpha 11 (Gq class) GNA11 1.26 2.0 N36926 
Aldo-keto reductase family 1, member C3 (3-alpha hydroxysteroid dehydrogenase, type II) AKR1C3 1.27 2.0 D17793 
Aconitase 2, mitochondrial ACO2 1.29 3.6  
Hypothetical protein LOC201229 LOC201229 1.29 2.1 AI6258 
Chromosome 9 open reading frame 97 C9ORF97 1.3 2.2 AI9510 
Clone 23688 mRNA sequence  1.35 2.4 AF0521 
DnaJ (Hsp40) homolog, subfamily B, member 12 DNAJB12 1.37 2.0 W73046 
5,10-methenyltetrahydrofolate synthetase MTHFS 1.37 2.3 L38928 
Nuclear factor I/B NFIB 1.38 2.5 U70862 
Heat shock 70 kd protein 2 HSPA2 1.39 2.0 L26336 
TAF5-like RNA polymerase II, 65 kd TAF5L 1.42 2.0 AJ0097 
Jumonji domain containing 1C JMJD1C 1.52 2.5 L40411 

Note: NLOGP = negative log of the p value; mRNA = messenger RNA; JMJD1C = jumonji domain containing 1C; PGC-1α = peroxisome proliferator-activated receptor gamma, coactivator 1-alpha; bcl-2 = B cell CCL/lymphoma 2; LCAS = long-chain acyl-coenzyme A synthase; VEGF = vascular endothelial growth factor-1; UCP = uncoupling protein.

This study was funded by CIHR (Institute of Aging). Dr. Tarnopolsky holds an endowed chair in neuromuscular disorders from the Hamilton Hospital's Assessment Center. Some of the equipment used was supported with a grant from the Canadian Foundation for Innovation (CFI).

References

1
Welle S, Brooks AI, Delehanty JM, et al. Skeletal muscle gene expression profiles in 20–29 year old and 65–71 year old women.
Exp Gerontol.
 
2004
;
39
:
369
-377.
2
Welle S, Brooks AI, Delehanty JM, Needler N, Thornton CA. Gene expression profile of aging in human muscle.
Physiol Genomics.
 
2003
;
14
:
149
-159.
3
Zahn JM, Sonu R, Vogel H, et al. Transcriptional profiling of aging in human muscle reveals a common aging signature.
PLoS Genet.
 
2006
;
2
:
e115
.
4
Haslett JN, Sanoudou D, Kho AT, et al. Gene expression profiling of Duchenne muscular dystrophy skeletal muscle.
Neurogenetics.
 
2003
;
4
:
163
-171.
5
Sterrenburg E, Turk R, ‘t Hoen PA, et al. Large-scale gene expression analysis of human skeletal myoblast differentiation.
Neuromuscul Disord.
 
2004
;
14
:
507
-518.
6
Mahoney DJ, Parise G, Melov S, Safdar A, Tarnopolsky MA. Analysis of global mRNA expression in human skeletal muscle during recovery from endurance exercise.
FASEB J.
 
2005
;
19
:
1498
-1500.
7
Altman N. Replication, variation and normalisation in microarray experiments.
Appl Bioinformatics.
 
2005
;
4
:
33
-44.
8
Short KR, Vittone JL, Bigelow ML, et al. Impact of aerobic exercise training on age-related changes in insulin sensitivity and muscle oxidative capacity.
Diabetes.
 
2003
;
52
:
1888
-1896.
9
Shi L, Tong W, Goodsaid F, et al. QA/QC: challenges and pitfalls facing the microarray community and regulatory agencies.
Expert Rev Mol Diagn.
 
2004
;
4
:
761
-777.
10
Bakay M, Chen YW, Borup R, Zhao P, Nagaraju K, Hoffman EP. Sources of variability and effect of experimental approach on expression profiling data interpretation.
BMC Bioinformatics.
 
2002
;
3
:
4
.
11
Forsberg AM, Nilsson E, Werneman J, Bergstrom J, Hultman E. Muscle composition in relation to age and sex.
Clin Sci (Lond).
 
1991
;
81
:
249
-256.
12
Carter SL, Rennie CD, Hamilton SJ. Tarnopolsky. Changes in skeletal muscle in males and females following endurance training.
Can J Physiol Pharmacol.
 
2001
;
79
:
386
-392.
13
Cartoni R, Leger B, Hock MB, et al. Mitofusins 1/2 and ERRalpha expression are increased in human skeletal muscle after physical exercise.
J Physiol.
 
2005
;
567
:
349
-358.
14
Mihic S, MacDonald JR, McKenzie S, Tarnopolsky MA. Acute creatine loading increases fat-free mass, but does not affect blood pressure, plasma creatinine, or CK activity in men and women.
Med Sci Sports Exerc.
 
2000
;
32
:
291
-296.
15
Burke DG, Chilibeck PD, Parise G, Tarnopolsky MA, Candow DG. Effect of alpha-lipoic acid combined with creatine monohydrate on human skeletal muscle creatine and phosphagen concentration.
Int J Sport Nutr Exerc Metab.
 
2003
;
13
:
294
-302.
16
Harris RC, Soderlund K, Hultman E. Elevation of creatine in resting and exercised muscle of normal subjects by creatine supplementation.
Clin Sci (Lond).
 
1992
;
83
:
367
-374.
17
Malm C. Exercise-induced muscle damage and inflammation: fact or fiction?
Acta Physiol Scand.
 
2001
;
171
:
233
-239.
18
Vissing K, Andersen JL, Schjerling P. Are exercise-induced genes induced by exercise?
FASEB J.
 
2005
;
19
:
94
-96.
19
Lundby C, Nordsborg N, Kusuhara K, Kristensen KM, Neufer PD, Pilegaard H. Gene expression in human skeletal muscle: alternative normalization method and effect of repeated biopsies.
Eur J Appl Physiol.
 
2005
;
95
:
351
-360.
20
Gibala MJ, Interisano SA, Tarnopolsky MA, et al. Myofibrillar disruption following acute concentric and eccentric resistance exercise in strength-trained men.
Can J Physiol Pharmacol.
 
2000
;
78
:
656
-661.
21
Parise G, Phillips SM, Kaczor JJ, Tarnopolsky MA. Antioxidant enzyme activity is up-regulated after unilateral resistance exercise training in older adults.
Free Radic Biol Med.
 
2005
;
39
:
289
-295.
22
Wendling PS, Peters SJ, Heigenhauser GJ, Spriet LL. Variability of triacylglycerol content in human skeletal muscle biopsy samples.
J Appl Physiol.
 
1996
;
81
:
1150
-1155.
23
Simoneau JA, Bouchard C. Human variation in skeletal muscle fiber-type proportion and enzyme activities.
Am J Physiol.
 
1989
;
257
:
E567
-E572.
24
Beaton LJ, Tarnopolsky MA, Phillips SM. Variability in estimating eccentric contraction-induced muscle damage and inflammation in humans.
Can J Appl Physiol.
 
2002
;
27
:
516
-526.
25
Lexell J, Taylor CC. A morphometrical comparison of right and left whole human vastus lateralis muscle: how to reduce sampling errors in biopsy techniques.
Clin Physiol.
 
1991
;
11
:
271
-276.
26
Lexell J, Taylor CC. Variability in muscle fibre areas in whole human quadriceps muscle: effects of increasing age.
J Anat.
 
1991
;
174
:
239
-249.
27
Tarnopolsky MA, Mahoney DJ, Vajsar J, et al. Creatine monohydrate enhances strength and body composition in Duchenne muscular dystrophy.
Neurology.
 
2004
;
62
:
1771
-1777.
28
Beaton LJ, Tarnopolsky MA, Phillips SM. Contraction-induced muscle damage in humans following calcium channel blocker administration.
J Physiol.
 
2002
;
544
:
849
-859.
29
Kim HK, Kim SR, Ahn JY, Cho IJ, Yoon CS, Ha TY. Dietary conjugated linoleic acid reduces lipid peroxidation by increasing oxidative stability in rats.
J Nutr Sci Vitaminol (Tokyo).
 
2005
;
51
:
8
-15.
30
Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-time quantitative PCR and the 2(−Delta Delta C(T)) Method.
Methods.
 
2001
;
25
:
402
-408.
31
Chen YW, Hubal MJ, Hoffman EP, Thompson PD, Clarkson PM. Molecular responses of human muscle to eccentric exercise.
J Appl Physiol.
 
2003
;
95
:
2485
-2494.
32
Hespel P, Op't Eijnde B, Van Leemputte M, et al. Oral creatine supplementation facilitates the rehabilitation of disuse atrophy and alters the expression of muscle myogenic factors in humans.
J Physiol.
 
2001
;
536
:
625
-633.
33
Giresi PG, Stevenson EJ, Theilhaber J, et al. Identification of a molecular signature of sarcopenia.
Physiol Genomics.
 
2005
;
21
:
253
-263.

Author notes

Departments of 1Pediatrics and Medicine and 2Kinesiology, McMaster University, Hamilton, Ontario, Canada.