| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Human Health and Nutritional Sciences, University of Guelph, Guelph, Ontario, Canada
Submitted 9 November 2006 ; accepted in final form 16 February 2007
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
O2 max) to volitional EXH (117.8 ± 4.2 min) on two separate occasions. Subjects were administered carbohydrate (CHO; 1 g·kg–1·h–1 Gatorlode) or water [placebo (PL)] during 5 h of recovery. Muscle biopsies were taken at rest, at EXH, and following 30, 60, 120, and 300 min of recovery. At EXH, total glycogen concentration was reduced (P < 0.05). However, GN-1 protein and mRNA content did not change. By 5 h of recovery, glycogen was resynthesized to 
60% of rest in the CHO trial and remained unchanged in the PL trial. GN-1 protein and mRNA level did not increase during recovery in either trial. We observed modest amounts of apo-GN-1 at EXH, suggesting complete degradation of some granules. These data suggest that GN-1 is conserved, possibly as very small, or nascent, granules when glycogen concentration is low. This would provide the ability to rapidly restore glycogen during early recovery. skeletal muscle; carbohydrate; exhaustion; resynthesis; recovery
Although it is appreciated that GN-1 is essential for the formation of a glycogen granule, its regulation has rarely been studied in human tissue, and its fate is unknown in circumstances where glycogen stores decrease. Work with cultured myotubes (10) suggests that some glycogen granules may be completely catabolized when glycogen stores are decreased. Under these circumstances, it is unknown whether the protein is degraded or is liberated as apoprotein (apo)-GN-1 (i.e., unglucosylated) and/or is translocated. Presumably, GN-1 would have to be available rapidly when glycogen stores are restored. Examinations of GN-1 in human muscle have focused predominantly on "activity" and mRNA during exercise and recovery. There are reports that GN-1 mRNA increases approximately twofold during prolonged exercise and/or the early part of recovery (14, 27, 28). Surprisingly, the effects of a carbohydrate (CHO) diet intervention for 48 h following glycogen-lowering exercise indicated that GN-1 mRNA was nearly 50% lower in the low-CHO intervention (6). GN-1 activity has been proposed (8, 17) as a reflection of concentration, and we observed a decrease of 40% in GN-1 activity upon exhaustion (EXH) following prolonged submaximal exercise (27). Additionally, we (28) revealed a more than twofold increase in GN-1 activity during recovery from a prolonged bout of cycling partnered with a 70% increase in GN-1 protein relative to that at EXH. Together, these two studies (27, 28) suggest that, during prolonged exercise, glycogen granules are degraded, and GN-1 is degraded and then rapidly resynthesized during recovery.
Our previous studies (27, 28) were informative but limited, as GN-1 activity was used to represent protein level in the first (27) investigation. In the second study (28), we were able to measure the protein but had to use EXH rather than rest as our reference point. In both cases, only the supernatant extract was measured.
Glycogen synthase, a protein normally associated with GN-1 and the granule, has been shown to translocate when glycogen levels are reduced (21, 24). Similar examinations of GN-1 do not exist. When glycogen concentration is not compromised, GN-1 is bound to glycogen in muscle (30) and localizes in the cytosol (5). Previous studies of GN-1 in skeletal muscle have examined the protein only in the supernatant fraction of the muscle homogenate (11, 27, 28). The possibility that GN-1 may also translocate when glycogen concentration is reduced and granules are catabolized has not been addressed.
Another aspect that has received minimal attention is whether GN-1 exists as apo-GN-1 when glycogen stores are degraded. Under resting conditions, all GN-1 is bound covalently to glycogen and apo-GN-1 is virtually nonexistent. Modest amounts of the apoprotein have been detected when glycogen was reduced by at least 60% by either electrical stimulation of rabbit hindlimb muscles or administration of a pharmacological level of epinephrine (30), suggesting that granules were completely degraded and at least some of the GN-1 "freed". apo-GN-1 was demonstrated qualitatively in human muscle in one study of two subjects. A thin band corresponding to apoprotein has been detected in the quadriceps femoris following a prolonged bout of cycling exercise (13), but we (28) were not able to confirm this observation.
Our understanding of GN-1 in human muscle is in its infancy. The data available for GN-1 under conditions of changing glycogen concentration are not comprehensive enough to determine whether the protein is degraded during exercise and then newly synthesized during recovery or the protein is conserved and exists as apo-GN-1 and/or is translocated. In the present study, we employed prolonged submaximal exercise to drastically reduce the muscle glycogen and then high CHO or water [i.e., placebo (PL)] ingestion to alter the rate of glycogen resynthesis during recovery. We tested for changes in GN-1 protein and mRNA levels under these conditions. We also addressed the following research questions 1) does GN-1 exist in the myofibrillar pellet, 2) is apo-GN-1 created when glycogen is reduced, and 3) does GN-1 translocate between the soluble and pellet fractions under these circumstances?
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
Ten, healthy, recreationally active males [age 22.6 ± 0.8 yr, height 183.6 ± 2 cm, body mass 80.3 ± 3 kg, maximal O2 uptake (O2 max) 54.5 ± 2.1 ml·kg–1·min–1] volunteered for the study. The experimental protocol received approval from the University of Guelph Human Ethics Committee. Subjects were informed, both orally and in writing, of any potential risks involved with the procedure, and consent forms were signed prior to study commencement.
Preliminary Testing
Subjects performed a O2 max protocol 1–2 wk prior to the experiment. O2 max was determined with a cycle ergometer using a graded workload protocol that consisted of increasing resistance by 50 W every 2 min to 300 W followed by increases of 25 W every minute until voluntary EXH. Breath-by-breath expired volume of oxygen (O2) and volume of carbon dioxide (CO2) were recorded.
Experimental Protocol
Approximately 36–40 h prior to each of two trials, subjects performed a 1-h bout of cycling at 65% of O2 max at a predetermined, constant cadence to standardize the amount of physical activity pretrial. Subjects were then provided with prepackaged meals for the 36 h leading up to their arrival in the laboratory for testing. Subjects were advised to eat only the food provided for them and to refrain from caffeine-containing products, alcohol, and physical activity for 3 days prior to each experiment. This combination of a standard pretrial exercise and diet was designed to ensure that the subjects began each trial under similar metabolic conditions and endurance potential. The muscle glycogen concentrations at rest for test days 1 and 2 (426 ± 28 and 484 ± 26 mmol/kg, respectively) and exercise endurance (119.8 ± 3.9 and 115.8 ± 4.5 min, respectively), which were not significantly different, suggests that the standardization was successful.
On the morning of each trial, subjects consumed a standardized breakfast (55% CHO, 30% fat, 15% protein) 2 h prior to cycling. A catheter was inserted into the antecubital vein for mixed venous blood sampling. Following a resting muscle biopsy and blood sampling, subjects exercised at 65% of O2 max on a cycle ergometer until volitional EXH. Approximately 2 wk later, when subjects repeated the exercise protocol, they were encouraged to cycle for the same duration of time in the second trial as in the first trial. The duration of cycling was not significantly different between PL and CHO trials (PL: 117.1 ± 4.9 min; CHO: 118.5 ± 3.5 min). Subjects consumed water ad libitum throughout each exercise bout. O2 measurements were taken at 20-min intervals to ensure that the subject was exercising at the desired intensity. Blood samples were drawn every 30 min during exercise, at EXH, and at 30-min intervals postexercise for the first 2 h and then at 3, 4, and 5 h. Muscle biopsies were taken at rest, at EXH, and at 30, 60, 120, and 300 min postexercise. Muscle biopsies were taken from the vastus lateralis muscle by percutaneous needle biopsy technique under local anesthesia (12). Two biopsies were taken from each incision in directions opposite to one another. In the first trial, the first four biopsies were taken from one leg and the last two from the other; the order of sampling from the legs was switched for the second trial.
Each subject completed two trials consisting of the identical protocol prior to and during a bout of prolonged exercise. On one occasion, subjects ingested 1 g·kg body mass–1·h–1 of CHO in the form of Gatorlode starting at EXH and every hour thereafter for 5 h (401.7 ± 13.8 g of total CHO) to facilitate glycogen repletion. In the other experiment, subjects ingested water (PL) equal in volume to the quantity of fluid calculated to provide 1 g·kg body mass–1·h–1 in the CHO trial during the 5-h recovery. The same pattern of fluid ingestion was used during both trials. The CHO and PL trials were separated by a period of at least 2 wk, and the order of the trials was randomized.
Muscle Biopsy Analyses
Muscle biopsies were quickly dissected free of visible connective tissue and fat, frozen in liquid nitrogen, and stored at –80°C for subsequent analyses. Due to a limited availability of muscle tissue, analyses performed were prioritized: 1) glycogen (n = 10 subjects; 120 biopsies), 2) immunoblotting of GN-1 protein in the soluble fraction (n = 10 subjects; 120 biopsies), 3) GN-1 mRNA level (n = 10 subjects; 120 biopsies), and 4) GN-1 activity assay (n = 7 subjects; 84 biopsies). In addition, immunoblotting methods were also employed to 1) test for GN-1 protein in pellet fractions (n = 5 subjects; 60 biopsies), 2) compare the proportion of total GN-1 detected in the soluble vs. myofibrillar pellet fractions (15 biopsies; 15 matching pairs of soluble and pellet fractions), and 3) analyze for ‘free’ apo-GN-1 in soluble and pellet fractions [n = 8 trials (PL and CHO); 48 biopsies]. The selection of samples for these latter comparisons was done based on amount of tissue available.
Glycogen
A 15- to 20-mg sample of wet muscle tissue from each biopsy was freeze-dried, dissected free of blood, connective tissue, and fat, and finely powdered. Proglycogen (PG), macroglycogen (MG), and total glycogen (PG + MG) concentrations were measured as described previously (1, 27, 28) and reported in millimoles glucosyl units per kilogram dry weight (mmol glucosyl units/kg dry wt).
Protein
Immunoblot. GN-1 protein levels were measured semiquantitatively as previously described (11, 28). Additionally, the "muscle extract," or soluble fraction, was separated from the myofibrillar pellet fraction (11, 28, 30). Pellets were washed three times with homogenizing buffer to eliminate any unbound protein and resuspended in 5 volumes of homogenizing buffer. Total protein for each soluble and pellet fraction was measured (Coomassie Plus Protein Assay Reagent Kit, Pierce).
As described above, GN-1 protein level in the soluble fraction was measured for all subjects (120 biopsies), whereas pellet fractions for five subjects (60 biopsies) were analyzed. Additionally, the soluble and pellet fractions from 15 biopsies [3 different subjects; 1 PL trial (Rest to 300 min), 2 CHO trials (Rest to 300 min for one and Rest to 30 min for the other)], were used to compare the proportion of total GN-1 detected in each of the two fractions. Soluble fractions (10 µg total protein) and pellet fractions (20 µg total protein) were incubated with type IX-A human salivary 
-amylase (Sigma, final conc. 10 µg/ml, 1 h, 37°C) to digest CHO prior to SDS-PAGE. To quantitatively compare the proportion of total GN-1 in the soluble and pellet fractions, the densitometry value for each pellet fraction was halved to account for the differences in the total quantity of protein loaded per fraction. As described previously, soluble fractions were also used to detect for "free" apo-GN-1. Samples (48 biopsies) from five subjects were not treated with -amylase prior to SDS-PAGE. Fifty micrograms of protein were loaded per sample in anticipation of low quantities of free protein.
Samples were transferred to polyvinylidene difluoride membranes following SDS-PAGE (1-mm 10% gels) (15). Blots were stained with Ponceau S, and densitometry was used to normalize for any loading differences between lanes within each blot. Membranes were completely destained in 1% Tris-buffered saline-Tween solution (TBS-T) and blocked at 4°C overnight in TBS-T containing 5% BSA (Sigma) and 2% skim milk (Nestle) with gentle agitation. Blocked membranes were incubated with guinea pig anti-human GN-1 monoclonal antibody (a generous donation by Dr. P. Roach, Indiana University, Bloomington, IN) diluted 1:5,000 in TBS-T for 1 h at room temperature. The membranes were washed with TBS-T for 30 min (3 10-min intervals) and were subsequently incubated for 1 h at room temperature with goat anti-guinea pig IgG HRP-conjugated antibody (Chemicon) diluted 1:3,000 with blocking solution. GN-1 protein (37 kDa) was visualized by enhanced chemiluminescence (Amersham) according to the manufacturer's instructions, and protein was detected on film (Kodak BioMax XAR, Fisher) developed using X-omat automated developer system (Kodak). Relative densities were quantified using the ChemiGenius 2 imaging system (Syngene), and all data were expressed in arbitrary units. Histidine (His)-tagged recombinant human GN-1 (a generous donation by Dr. P. Roach, Indiana University) was used as a positive control.
Activity.
Activity should not be mistaken for the classical experiments performed to assay the autoglucosylation of GN (16, 22, 31). Importantly, this assay does not imply a more or less active state of the protein. This activity assay was used as an additional, but indirect, method to measure GN-1 protein level. Originally described by Manzella et al. (17), this assay quantified the radiolabeled product generated when GN-1 was incubated in the presence of a donor substrate, UDP-glucose, and an exogenous acceptor, dodecyl-
- D-maltoside (DBM). Manzella et al. (17) reported that the transglycosylating activity of purified GN-1 was directly proportional to GN-1 concentration.
Essentially, apo-GN-1 was generated in vitro by amylolysis; therefore, changes in GN-1 activity should reflect changes in GN-1 concentration in the muscle. Although the assay may seem redundant, considering we had directly measured GN-1 protein by immunoblot analysis, it provided a direct comparison between protein levels as measured by activity vs. Western blot. The assay proceeded as described by us (27, 28) and by Hansen et al. (11).
mRNA Transcription
RNA Isolation. Total RNA was extracted from 20–30 mg wet tissue using a modified Chomczynski and Sacchi method (9) as previously described (27, 28).
cDNA synthesis. Reverse transcription was performed using SuperScript II Reverse Transcriptase (Invitrogen). Briefly, 1 µl (150 ng) of random primers (Invitrogen) and 1 µl of dNTP Mix (Invitrogen) were combined with 1 µg of total RNA and sterile, distilled water (to 12 µl). RNA was denatured by heating for 5 min at 65°C (Techgene thermocycler, Cambridge, MA) before addition of 4 µl of 5x First-Strand Buffer and 2 µl of 0.1 M dithiothreitol. Samples were incubated at 25°C for 2 min before addition of 1 µl of Superscript II RT and diethyl pyrocarbonate-treated water (to 20 µl). Samples were incubated at 25°C for 10 min followed by 50 min at 42°C and finally at 70°C for 15 min. cDNA was stored at –20°C until ready to proceed with real-time RT-PCR.
RT-PCR
Primers were designed using the Premier Biosoft primer program (Netprimer). GenBank sequences were used to design primers for GN-1 (accession no. AH007714) and 18S rRNA (accession no. M10098.1). GN-1 forward and reverse primers were 5'-CGG AGT CTT CGT TTA TCA GCC T-3' and 5'-GTC CCC ACC ATC AAA ACT ACC TT-3', respectively, with an amplicon size of 88 bp. 18S rRNA forward and reverse primers were 5'-GAC TCA ACA CGG GAA ACC TCA C-3' and 5'-ATC GCT CCA CCA ACT AAG AAC G-3', respectively, with an amplicon size of 114 bp (Lab Services, University of Guelph). 18s was selected as the endogenous control gene because it did not change throughout the experiment.
Quantification of mRNA was performed by real-time RT-PCR using the ABI PRISM 5700 sequence detection system (Applied Biosystems). PCR reactions were carried out in triplicate using SYBR Green qPCR SuperMix UDG Kit (Invitrogen). For each reaction, 0.5 µl of cDNA was combined with 12.5 µl of Supermix, 0.25 µl of ROX Passive Reference Dye, 0.5 µl (10 pM) of forward and reverse primers (for 18s or GN-1), and 6.75 µl of sterile, distilled Milli-Q to a final volume of 25 µl.
cDNAs were amplified in triplicate using the following conditions: 50°C for 2 min and 95°C for 10 min followed by 40 cycles of 95°C for 15 s and 60°C for 1 min. For each sample, a critical threshold (CT) was detected that identified the cycle number that gene amplification was first detected. A delta (
)CT value was obtained by subtracting the CT for 18s from that of GN-1. The resting CT was subtracted from exercise and recovery samples to determine a – CT value. The efficiency was assessed to be 100%; therefore, the expression of GN-1 was assessed based on 2
. A separate dissociation step (95°C for 15 s, 65°C for 60 s, and 95°C for 15 s) confirmed a single product for both the endogenous control and the gene of interest. No amplified product was detected from the "no template" control.
Blood Glucose and Insulin
Blood samples collected in heparinized tubes were immediately stored on ice and analyzed for glucose using a YSI 2300 Stat Plus Analyzer. An additional blood sample was collected into a nonheparinized tube, clotted at room temperature and centrifuged (10 min, 2,500 rpm, 4°C) to collect serum (stored at –20°C) for insulin analysis (Coat-a-Count Insulin, Intermedico Diagnostic Products). All samples were assayed in duplicate, and the mean value was taken.
Statistical Analysis
Data are expressed as means ± SE. Data were analyzed statistically using two-way, repeated-measures ANOVA (SigmaStat 3.1). Significance was established at P 
0.05. Where significant interactions were identified, Tukey's post hoc test was used. Linear regression analysis was used to test for a relationship between GN-1 activity and protein level measured by immunoblotting.
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
Blood glucose concentrations at rest, during exercise, and at EXH did not differ between CHO and PL trials. At all time points during recovery, blood glucose concentrations in the CHO trial were significantly elevated above PL trial concentrations. In the CHO trial, blood glucose reached its peak concentration (7.9 ± 0.3 mM) by 90 min of recovery. Similarly, insulin concentrations at rest, during exercise, and at EXH did not differ between trials. Likewise, at all time points during recovery, serum insulin levels in the CHO trial were significantly elevated above PL concentrations (P < 0.05) and peaked at a concentration of 36.5 ± 6.4 µU/ml (data not shown).
Muscle Glycogen Concentration
Total glycogen concentration (Fig. 1) at rest did not differ between trials (PL: 460 ± 28 mmol glucosyl units/kg dry wt; CHO: 450 ± 29 mmol glucosyl units/kg dry wt). At EXH, total glycogen was significantly (P < 0.05) reduced to 16% (73 ± 13 mmol glucosyl units/kg dry wt) and 24% (108 ± 19 mmol glucosyl units/kg dry wt) of resting levels in the PL and CHO trials, respectively. Total glycogen concentration and the time spent cycling at the point of volitional EXH (PL: 117.1 ± 4.9 min; CHO: 118.5 ± 3.5 min) did not differ between trials. Following 2 h of recovery, total glycogen in the CHO trial had doubled and was significantly greater than the corresponding glycogen concentration in the PL trial. Following 5 h of recovery, total glycogen was resynthesized to nearly 60% of resting levels in the CHO trial, whereas no changes were observed in the PL trial throughout recovery.
|
GN-1 Activity
GN-1 activity is depicted in Fig. 2. An effect of time on activity was observed at EXH for both PL and CHO trials (P < 0.05). Activity decreased (15–20%) from 153.9 ± 9.3 and 151.4 ± 8.8 mU·mg protein–1·min–1 at rest to 133.6 ± 7.5 and 119.7 ± 10.7 mU·mg protein–1·min–1 at EXH in the PL and CHO trials, respectively. Within 30 min of recovery, the activity in both trials was not different from levels observed at rest.
|
GN-1 mRNA expressed relative to resting levels is depicted in Fig. 3. GN-1 mRNA level did not change in response to a prolonged bout of cycling exercise in either trial. However, a moderate 15–20% overall decrease from rest was observed at 60 and 300 min of recovery (P < 0.05). Although it did not reach significance, the PL trial was lower than the CHO trial.
|
The Western blot data, expressed in arbitrary units, for the soluble and myofibrillar pellet fractions are shown in Fig. 4, A and B. GN-1 protein level in both the soluble and pellet fractions remained unchanged in response to a prolonged bout of cycling. Similarly, net protein expression measured in the soluble fraction was not different from resting levels during recovery in response to either the PL or CHO intervention. However, the level of protein detected in the myofibrillar pellet was less in the PL compared with the CHO trial (P < 0.05) at all time points during recovery. Notably, although glycogen doubled during the recovery period in the CHO treatment, it did not change in the PL trial during recovery, and GN-1 protein level in the pellet fraction decreased in the PL condition to 75% of that of the CHO trial after 5 h.
|
Notably, apo-GN-1 was detectable in several of the soluble fractions tested (samples tested were loaded with five times more total protein than normally used). In 24 of 48 samples tested (8 trials), a band corresponding to free apo-GN-1 was visible. These bands appeared at EXH and at time points throughout recovery, but not at rest. Band intensity varied from trial to trial; in five trials the bands were barely visible, whereas in the remaining three trials the bands were much more intense, with the most intense bands appearing at EXH. Data for the PL trial for an individual subject are shown in Fig. 5. Interestingly, the quantity of free protein detected appears to decrease during recovery despite there being no change in glycogen content during recovery. Furthermore, the glycogen concentration at EXH in the CHO trial for this particular subject was higher than the concentration measured at EXH in the PL trial (141.8 and 92.7 mmol glucosyl units/kg dry wt, respectively) and yielded very little detectable free protein. apo-GN-1 was not detected in the pellet fractions (data not shown), likely because the concentration of GN-1 in the pellet is a small fraction of the total GN-1 in the muscle.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
Although the exercise bout successfully reduced the glycogen by 75–85%, neither did GN-1 protein decrease, nor did its mRNA increase. Neither, was there an indication of an increase in protein nor mRNA level during recovery in either trial, despite a doubling of PG concentration by 2 h of recovery in the CHO trial (P < 0.05). Within 5 h of recovery, total glycogen and PG levels in the CHO trial were resynthesized to 60 and 80% of resting levels, respectively. The initial phase of recovery is characterized primarily by an increase in PG followed by an increase in MG; this may reflect an initial increase in the number of small granules, followed by their enlargement and development into larger, mature granules (2, 28). Elsner et al. (10) demonstrated an "ordered" pattern of granule synthesis using cultured rodent myotubes, in that existing granules appear to accept CHO initially, but the cell quickly begins to generate new granules even though the existing ones were only intermediate in size. However, it needs to be noted that the glycogen stores of the myotubes were only moderately compromised prior to the synthesis period. A rapid increase in PG during the initial phase of recovery may originate from 1) condensation of UDP-glucose to small, preexisting granules that were catabolized during exercise; 2) the genesis of new granules; or 3) a combination of the two. Although we cannot positively ascertain how granule resynthesis occurs, our results argue that the majority of PG was synthesized from small, preexisting granules rather than from the formation of new granules. New granule formation would require the availability of apo-GN-1 to initiate the process; thus, either protein would be synthesized, or existing inactive apo-GN-1 would be activated. In the CHO trial of this study, 80% of PG was resynthesized within 5 h. Therefore, this is the most likely time for new granule formation to occur, but we did not observe a change in protein levels during this period (Fig. 4A). Similarly, the large decrease in muscle glycogen from rest to EXH in both the PL and CHO trials could imply that some granules were completely degraded, freeing up apo-GN-1. However, the GN-1 concentration did not decrease, implying that it was freed or that granules were not completely degraded.
Muscle glycogen concentration appears to influence the translocation of regulatory enzymes. Glycogen synthase, which normally is associated with GN-1 and bound to the granule (22), translocates from a glycogen-enriched region to a localized cytosolic distribution when glycogen is reduced by muscle activity (21, 24). The majority of GN-1 is located in the supernatant (5, 30) but it is uncertain whether the protein translocates when glycogen content is reduced. Baque et al. (5) did not address the possible effects of altering glycogen content on GN-1 distribution, and Smythe et al. (30) reported that the sedimentation rate of the protein changed following a reduction in glycogen content. However, they did not address whether GN-1 translocated in vivo.
In the present study, we addressed the possibility that GN-1 translocates between the soluble and the myofibrillar pellet fractions. A relative comparison made between corresponding soluble and pellet fractions from 15 biopsies indicated that the majority of total GN-1 (82%) in the muscle was detected in the supernatant fraction and the remainder sedimented with the myofibrillar pellet. The latter was resistant to repeated washing of the pellet, suggesting that this fraction was protein or membrane bound. We also observed (Fig. 4A) that there were no significant changes in GN-1 concentration in the supernatant over time for either treatment (n = 120), or changes in the pellet GN-1 concentration for the CHO trial (n = 30). This supports the idea that the distribution of GN-1 is normally static and does not translocate between these two fractions. It is noteworthy that the data agree with the observation that 15–20% of glycogen granules in a muscle fiber are located in the intramyofibrillar region in association with the contractile proteins (19). These findings, along with observing no change in the basic distribution of GN-1 over time (Fig. 4A), support earlier reports using other experimental models that the majority of the GN-1 was located in the soluble fraction. However, previous studies have overlooked the GN-1 located in the pellet fraction (20% of total). Due to a limited availability of tissue in the present study, we were unable to conduct tests to examine whether GN-1 translocates between other subcellular fractions.
If GN-1 is not degraded, it may exist as apo-GN-1 when glycogen is reduced. Under resting conditions, GN-1 is bound to CHO (16, 22, 30). Two studies have reported observing apo-GN-1 when glycogen concentration was reduced (13, 30), although the observations in the former study were made in only two subjects and the latter study used pharmacological levels of epinephrine in an animal model to reduce glycogen. In contrast, we were previously unable to detect apo-GN-1 following prolonged exercise in humans (28). The present tests were conducted more systematically and used a larger total protein sample. We confirm that free apo-GN-1 does not exist in resting muscle, although we were able to detect small amounts of free protein in the supernatant fraction at EXH in three of the subjects. The ability to detect apo-GN-1 in this study but not previously (28) may be because we used more total protein for this analysis. Our detection of free protein, together with the observations that GN-1 concentration did not change in the soluble fraction and that there was no evidence of translocation to or from the pellet when glycogen concentration was low, suggests that most granules were not fully catabolized and that GN-1 protein was conserved.
Although no differences in GN-1 protein level in the soluble fraction were observed between the PL and CHO trials, the level of protein detected in the myofibrillar pellet was 15–25% less in the PL than in the CHO trial (P < 0.05) at all time points during recovery. Although the pellet contains the minor (20% of the total GN-1) fraction of the protein, it may represent a specific pool of glycogen granules. We can speculate why the pellets differed in protein level between trials. The results imply that the turnover rate of this pool of GN-1 changed due either to decreased protein stability or to decreased synthesis, potentially in response to less available CHO. If this pool represents that fraction bound to myofibrillar and contractile proteins, it is possible that small changes in protein synthesis would first be apparent in this small fraction of protein most distant from the nucleus. These possibilities require further investigation.
We did not observe an increase in mRNA content during recovery in either the PL or the CHO trial and only modest, albeit insignificant, increases in protein level from EXH in the CHO trial. In comparison, previous studies have reported an increase in mRNA (14, 28), and protein (28) during recovery from an acute bout of exercise. Small differences in experimental design may account for these apparent differences. Comparison of the present results with our previous data (28) is difficult, as a resting biopsy was not taken in the latter investigation. In addition, glycogen was reduced to a lower concentration by EXH than in the current study (59 vs. 108 mmol/kg dry wt, respectively). The previous protocol employed more intense exercise (2 h at 75% O2 max followed by five 30-s bouts of sprinting at 130% O2 max). This protocol may have led to the recruitment of more fast-twitch fibers and a different pattern of GN-1 gene and protein expression than in the present study. A limitation to using any human model is that one does not know how much of the muscle is active, and a biopsy very likely contains samples of fibers that have been inactive to varying degrees. Nevertheless, the exercise protocol of the present study is similar to what we (27) and others (14) have used previously when an increase in GN-1 mRNA was still observed. We are unable to provide any further explanations. However, our present data set is more comprehensive and complete than that of earlier studies, and the protein and mRNA data are internally consistent. Furthermore, the 70% increase in GN-1 protein that we observed previously (28) did not occur until late in recovery (by 5 h), whereas PG synthesis had increased significantly from EXH after only 30 min of recovery. This is consistent with our present conclusion that that the quantity of GN-1 present in the muscle immediately after exercise is sufficient to enable the early, rapid phase of glycogen restoration.
Although the activity assay has been used to indirectly assess protein content (8, 11, 17, 18, 20, 27, 28), our data indicate a discrepancy between activity and Western blot analysis. The significant 15–20% decrease from rest to EXH in activity was not apparent in the protein concentration. The previous observation that GN-1 activity decreased from rest to EXH (27) was interpreted as catabolism of glycogen granules resulting in GN-1 protein being liberated and then degraded, based on the work of others (8, 11, 17, 18, 20). Although we also observed a moderate but transient (15–20%) decrease in GN-1 activity from rest to EXH that returned to resting levels by 30 min of recovery, this was not observed in the direct measurement of protein concentration using Western blotting techniques, which questions the accuracy of activity as a measurement of protein concentration. However, we cannot rule out that larger changes in activity do reflect changes in GN-1 concentration; for example, previously (28) we observed a twofold increase in association with a 70% increase in protein. The activity data suggest that the transglucosylating activity of the protein may be affected by other factors within the muscle extract in response to metabolic perturbations such as exercise and changes in muscle glycogen content. In support of this statement, it is noteworthy that we previously observed a twofold increase in activity within the first 30 min postexercise, which was not accounted for by a corresponding increase in protein (28). Activity, when measured in skeletal muscle at rest (11) or using purified GN-1 (17, 18, 20), appears to be related to protein levels. Thus, the extent of the relationship may be situationally dependent. Clearly, an in-depth examination of other factors present in the muscle extract and their effects on GN-1 activity in vitro is required to fully interpret the data and should be reported with caution in similar experiments in the future.
This was the first study to comprehensively examine GN-1 protein, mRNA, and activity simultaneously in conjunction with PG and MG concentrations in human skeletal muscle in response to changes in glycogen content during prolonged exercise and in response to different rates of glycogen restoration. Our observations support the possibility that few glycogen granules are completely catabolized during prolonged exercise, freeing GN-1. However, the protein neither translocates between the myofibrillar pellet and soluble fraction nor is degraded. Despite significant changes in glycogen content, GN-1 protein appears to remain constant. This is supported further by the finding that there was no increase in mRNA level. The conserved level of protein, presumably in nascent granules, would enable the muscle to quickly restore glycogen immediately postexercise without the need for de novo GN-1 synthesis.
|
GRANTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  S. Roy, S. Khanna, C. Rink, S. Biswas, and C. K. Sen Characterization of the acute temporal changes in excisional murine cutaneous wound inflammation by screening of the wound-edge transcriptome Physiol Genomics, July 1, 2008; 34(2): 162 - 184. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
