J Appl Physiol 110: 820–825, 2011.
First published November 18, 2010; doi:10.1152/japplphysiol.01082.2010.
Innovative Methodology
Enhanced technique to measure proteins in single segments of human skeletal
muscle fibers: fiber-type dependence of AMPK-␣1 and -␤1
Robyn M. Murphy
Department of Zoology, La Trobe University, Melbourne, Victoria, Australia
Submitted 10 September 2010; accepted in final form 13 November 2010
muscle protein; fiber type; single fibers; Western blotting; adenosine
5=-monophosphate-activated protein kinase; SERCA; tropomyosin;
rodent
PHYSIOLOGICAL STUDIES INVOLVING biochemical analyses of human skeletal muscle samples are often confounded by the
heterogenous nature of the sample. A typical sample from the
vastus lateralis muscle is composed of a mixed-fiber population, broadly termed type I and type II fibers, or slow-twitch
and fast-twitch fibers, respectively. In a given individual, the
proportion of slow-twitch and fast-twitch fibers is reliant on
activity status, genetic composition, health status, and age.
Data obtained from analyses of whole muscle biopsies can be
limiting when there is a differential fiber-type expression of the
proteins being examined. The technologies developed and
presented here enable a single human muscle fiber segment,
dissected from a freeze-dried skeletal muscle biopsy, to be
quantitatively analyzed for a number of different proteins.
Importantly, the advances presented demonstrate the identification of proteins at both the level of the contractile apparatus,
namely the myosin heavy chain (MHC) and the tropomyosin
isoforms present, as well as the presence or absence of the
fast-twitch isoform of the sarcoplasmic reticulum (SR) sarco(endo)plasmic reticulum Ca2⫹-ATPase type 1 (SERCA1). The
technique adopts previously described methods to separate and
obtain individual fiber segments (e.g., Refs. 3, 13); however, it
Address for reprint requests and other correspondence: Dept. of Zoology, La
Trobe Univ., Melbourne, Victoria 3086, Australia (e-mail: r.murphy@latrobe.
edu.au).
820
eliminates the need to pool fibers following fiber typing,
previously necessary to obtain sufficient material to examine
the proteins of interest. The technical steps, normalization
procedure, fiber-type assignment, and data analyses are presented. This advance will allow definitive fiber-type expression
of proteins to be elucidated following certain interventions
(e.g.. exercise) or in certain muscle diseases.
METHODS
Materials and antibodies. All chemicals used were from Sigma
(Sydney, Australia), unless otherwise stated. Antibodies used and
their relevant dilutions or final concentrations used in 1% bovine
serum albumin in phosphate-buffered saline with 0.025% Tween were
as follows: AMP-activated protein kinase (AMPK)-␣1 and AMPK-␤1
(both 1 in 1,000, rabbit polyclonal) (2), MHC type I [MHC I, 0.19
␮g/ml, mouse monoclonal IgM, clone A4.840, Developmental Studies Hybridoma Bank (DSHB), University of Iowa] and MHC II (0.15
␮g/ml, mouse monoclonal IgG, clone A4.74, DSHB), SERCA1 (0.06
␮g/ml, mouse monoclonal, clone CaF2–5D2, DSHB), actin (4 ␮g/ml,
rabbit polyclonal, A-2066, Sigma), tropomyosin (0.04 ␮g/ml, mouse
monoclonal, CH1, DSHB), and ryanodine receptor 1 (RyR1, 0.15
␮g/ml, mouse monoclonal, 34C clone, DSHB). Human recombinant
AMPK heterotrimer consisting of ␣1/␤1/␥1-isoforms was expressed
in, and purified from, a recombinant baculovirus expression system, as
described previously (5), and kindly provided by Dr. Jonathan Oakhill
(St Vincent’s Institute, Melbourne, Australia).
Human muscle biopsies. Human muscle samples were collected
from the vastus lateralis muscle from three young, healthy, active, but
not trained, male volunteers using the percutaneous needle biopsy
technique modified for suction. Samples were rapidly frozen in liquid
nitrogen and stored at ⫺80°C until analyzed. The human muscle
samples analyzed in the present study are spare tissue originally
collected for other completed research projects approved by The
Human Research Ethics Committee at The University of Melbourne.
Collection and preparation of single-muscle fibers. Portions
(30 – 60 mg) of muscle biopsies were freeze-dried for 24 h. Samples
were brought to room temperature in a dessicator for ⬎20 min.
Segments of individual fibers were dissected from the freeze-dried
portions (3) using a dissecting microscope and fine jeweller’s forceps.
The microscope was connected to a TV screen, and the average size
of some of the segments collected was estimated by measuring the
length of the fibers (1.2 ⫾ 0.3 mm, mean ⫾ SD, n ⫽ 75). About 50
fibers were collected from each biopsy. With the use of forceps, fibers
were placed into 5 ␮l of 3 ⫻ SDS solubilizing buffer (0.125 M
Tris·Cl, pH 6.8, 4% SDS, 10% glycerol, 4 M urea, 10% mercaptoethanol, 0.001% bromophenol blue) diluted 2:1 (vol/vol) with 1 ⫻
Tris·Cl, pH 6.8, and stored at ⫺20°C until Western blotting. Samples
were not heated before Western blotting, since it was found that there
was no difference in the Western blots of samples that were either
heated (95°C for 3 min) or not heated (Supplemental Fig. S1; the
online version of this article contains supplemental data). Visual
observation attempted to confirm that the fiber was no longer attached
to the forceps; however, despite this, sometimes no fiber was found to
have been in the tube.
8750-7587/11 Copyright © 2011 the American Physiological Society
http://www.jap.org
Downloaded from http://jap.physiology.org/ by 10.220.33.1 on June 17, 2017
Murphy RM. Enhanced technique to measure proteins in single
segments of human skeletal muscle fibers: fiber-type dependence of
AMPK-␣1 and -␤1. J Appl Physiol 110: 820 – 825, 2011. First published November 18, 2010; doi:10.1152/japplphysiol.01082.2010.—
Human physiological studies typically use skeletal muscle biopsies
from the heterogeneous vastus lateralis muscle comprised of both
fast-twitch and slow-twitch fiber types. It is likely that potential
changes of physiological importance are overlooked because fibertype specific responses may not be apparent in the whole muscle
preparation. A technological advance in Western blotting is presented
where proteins are analyzed in just one small segment (⬍2 mm) of
individual fibers dissected from freeze-dried muscle samples using
standard laboratory equipment. A significant advance is being able to
classify every fiber at the level of both contractile (myosin heavy
chain and tropomyosin) and sarcoplasmic reticulum [sarco(endo)plasmic reticulum Ca2⫹-ATPase type 1] properties and then being able to
measure specific proteins in the very same segments. This removes the
need to fiber type segments before further analyses and, as such,
dramatically reduces the time required for sample collection. Compared with slow-twitch fibers, there was less AMP-activated protein
kinase (AMPK)-␣1 (⬃25%) and AMPK-␤1 (⬃60%) in fast-twitch
fibers from human skeletal muscle biopsies.
Innovative Methodology
PROTEIN ANALYSIS IN INDIVIDUAL HUMAN SKELETAL MUSCLE FIBERS
RESULTS
In the present study, the relative amounts of the proteins
AMPK-␣1 and AMPK-␤1 were being compared between fasttwitch and slow-twitch human skeletal muscle fibers. These
proteins were successfully detected in most of the skeletal
muscle fibers examined. When there was no detectable signal,
as determined by the absence or very weak staining of MHC on
J Appl Physiol • VOL
the Coomassie-stained gel and the actin immunoblot, it was
typically due to either the fiber being too small or the fiber not
being successfully placed into the solubilizing buffer (lane 1,
Fig. 1 and lane 3, Fig. 2). When human single fibers were
examined in their entirety (i.e., no sample fractionation undertaken) using both AMPK-␣1 and AMPK-␤1 antibodies, multiple bands were detected (Fig. 1). For AMPK-␣1, the nonspecific band appeared at a very high molecular mass (⬎250 kDa,
Fig. 1A, top), and the band migrating at ⬃63 kDa had a similar
mobility (slightly faster) to the AMPK-␣1 in the pure AMPK
protein, and this band was thus identified as AMPK-␣1. The
AMPK-␤1 antibody identified a number of bands in the region
25– 45 kDa. The band identified as the ␤1-isoform of AMPK
was the band migrating slightly faster than the pure AMPK
protein, similar to the slightly faster migration of AMPK-␣1
(Fig. 1). This band (lanes 6 –13, Fig. 1) was also confirmed as
being absent in skeletal muscle from AMPK-␤1 knockout mice
(Supplemental Fig. S2). It is not known if the other bands
correspond to variations of the specific AMPK protein, for
example, glycosylation modifications, which would affect mobility in SDS-PAGE gels, or if they are detecting nonspecific
proteins.
The relative amounts of the AMPK isoforms in the fasttwitch EDL muscle and predominantly slow-twitch Sol muscles shown in Fig. 1A were plotted (Fig. 1, B and C). From the
slopes of the curves, it can be seen that there was about twice
as much AMPK-␣1 and about three times as much AMPK-␤1
in Sol muscle compared with EDL muscle. Figure 1, D and E,
shows the pooled data from a number of different samples,
which were in line with the data from the individual gel shown.
There was no difference in the amount of MHC present in
samples from EDL or Sol muscle homogenates (Supplemental
Fig. S3).
Following detection of the proteins of interest, i.e.,
AMPK-␣1 and AMPK-␤1, the membranes were consecutively
probed for a number of proteins to identify the individual fibers
as type I (slow twitch) or type II (fast twitch). It was important
that MHC I and MHC II could be probed for independently on
the same fiber without interference of the previous probe, and
this was made possible because the antibodies were IgM and
IgG, respectively. Other probes (SERCA1, actin, tropomyosin,
or RyR1) were in unique regions of the membrane and,
therefore, were free of interference from each other.
On each gel, standard curves of rat EDL and Sol muscle
homogenates were included. It was important to include this,
first, because it allowed the detection range to be accurately
considered, and, second, it confirmed the expected fiber specificity of the antibodies used.
Given the very small amount of tissue in the samples used in
this study, it was not possible to assess the muscle weight or
total protein in absolute terms. To be able to determine the
relative quantification of the AMPK isoforms between fibers,
for each fiber the densities of the AMPK isoforms were
normalized to the MHC band seen on the Coomassie-stained
gel. The linearity of the signal density to the amount of muscle
loaded was confirmed using the density of the homogenate
samples, as described previously (8) and shown here (Supplemental Fig. S3). To determine the relative amounts of the
AMPK isoforms in the human single fibers, two normalization
procedures were used. The first analyses, referred to as raw
110 • MARCH 2011 •
www.jap.org
Downloaded from http://jap.physiology.org/ by 10.220.33.1 on June 17, 2017
Collection and preparation of rat skeletal muscle homogenates.
With approval of the La Trobe University Animal Ethics Committee,
male Long-Evans hooded rats (⬃6 – 8 mo old) were killed by overdose with fluothane (2% vol/vol) in a restricted air space. Extensor
digitorum longus (EDL) and soleus (Sol) muscles were rapidly excised and homogenized at room temperature using a Polytron homogenizer, 3 ⫻ 8 s (Kinematica, Lucerne, Switzerland) in 10 volumes of
homogenizing solution made of the following (in mM): 165 Na⫹, 1
free Mg2⫹ (10.3 total Mg2⫹), 90 HEPES, 50 EGTA, 8 ATP, 10
creatine phosphate, pH 7.10, with the addition of protease inhibitor
cocktail (PIC, COMplete, Roche Diagnostics, Sydney, Australia), and
strongly buffered to a very low free Ca2⫹ concentration (⬍1 nM) to
avoid activating any Ca2⫹-dependent proteases. Whole muscle homogenate was then diluted 1 in 40 in homogenizing solution and then
further diluted to a concentration of 2.5 ␮g wet wt muscle/␮l with 3 ⫻
SDS solubilizing buffer and stored at ⫺20°C for Western blotting.
Western blotting. Samples were analyzed for AMPK-␣1, AMPK␤1, MHC I, MHC II, SERCA1, actin, tropomyosin, and RyR1 protein
contents by Western blotting using a method similar to that described
previously (8). To determine the relative amount of given proteins in
the human single-muscle fibers, they were always loaded onto a gel
with standard curves of muscle homogenates prepared from EDL and
Sol muscle (1.5, 3, 6, and 12 ␮g muscle from each homogenate), so
that the linear response of each of the proteins analyzed was known
(see Ref. 8). Proteins were separated on 18- or 26-well 4 –12%
gradient Criterion SDS-PAGE gels (BioRad, Hercules, CA). On most
gels, two protein standard ladders were loaded (Hi-mark, Invitrogen,
Sydney, Australia, and Fermentas, UK). In some instances, when rat
muscle homogenates were examined, Criterion Stainfree gels (BioRad) were used, which allowed digital images of the total protein on
the gels to be collected following UV activation using a Stainfree
Imager (BioRad). Following UV activation (Stainfree gels) or immediately following electrophoresis, proteins in gels were wet transferred
to nitrocellulose for 60 min at 100 V in a circulating ice-cooled bath,
using transfer buffer containing 140 mM glycine, 37 mM Tris-base,
20% methanol. Following transfer, the non-Stainfree SDS polyacrylamide gels were stained with BioSafe Coomassie Stain (BioRad) for
detection of MHC. In both this study and as shown previously (9),
MHC is a reliable indicator of the amount of protein present when
small sample sizes are used (Supplemental Fig. S3). Immediately
following transfer, membranes were incubated for 10 min in Pierce
Miser solution (Pierce, Rockford, IL), a reagent that improves binding
of antibodies to antigen. Following five quick washes in doubledistilled water, membranes were blocked in 5% skim milk powder in
Tris-buffered saline-Tween. Membranes were then cut into three
portions (⬎100 kDa; ⬍100 kDa and ⬎55 kDa; ⬍55 kDa), and each
portion was incubated in antibodies at concentrations detailed in
Materials and antibodies above (2-h rocking at room temperature and
overnight rocking at 4°C). Images were collected following exposure
to SuperSignal West Femto (Pierce) using a charge-coupled device
camera attached to a ChemiDoc XRS (BioRad) and using Quantity
One software (BioRad). When required, membranes were reprobed
with an alternate antibody following 30 – 60 min wash in Tris-buffered
saline-Tween (note: membranes were not stripped). Densitometry was
performed using the Quantity One software.
Data analyses. Data are expressed as means ⫾ SE, unless otherwise indicated. Student’s t-tests were performed using GraphPad
Prism4. Significance was set at P ⬍ 0.05.
821
Innovative Methodology
822
PROTEIN ANALYSIS IN INDIVIDUAL HUMAN SKELETAL MUSCLE FIBERS
Downloaded from http://jap.physiology.org/ by 10.220.33.1 on June 17, 2017
Fig. 1. Detection and identification of AMP-activated
protein kinase (AMPK)-␣1 and -␤1 isoforms in human skeletal muscle fibers and rodent skeletal muscle. A: protein separated on 4 –12% SDS-PAGE gels
and transferred to nitrocellulose. Loaded onto the gel
were five individual muscle fibers (fibers A–E, lanes
1–5); 1.5- to 12-␮g muscle from rat extensor digitorum longus (EDL; lanes 6 –9) and soleus (Sol) muscle (lanes 10 –13); 1.3-, 2.5-, and 5-ng pure AMPK
protein (lanes 14 –16); and two different molecular
mass markers (lanes 17 and 18, taken as white light
image immediately before capture of chemiluminescent image without moving the membrane and the
two images superimposed, separated by black line).
The markers were HiMark (HiM) and Fermentas
(Fer; see METHODS). Myosin heavy chain (MHC) is
shown in the posttransferred Coomassie blue-stained
gel, indicating the presence of muscle samples in the
given lanes, and that most protein, other than the
abundant contractile protein, MHC, transferred out
of the gel. Note that lane 1 has no muscle fiber (see
RESULTS). The middle blot shows detection of
AMPK-␣1 at ⬃63 kDa, and the bottom blot shows
detection of AMPK-␤1 at ⬃38 kDa. Both proteins
migrate slightly faster than the pure protein. In the
human muscle fibers, which comprised the entire
tissue sample, bands migrating at different molecular
masses were identified (NS) with both antibodies
used. Plots show density vs. amount of protein for
EDL and Sol muscle, shown in A, for AMPK-␣1 (B)
and AMPK-␤1 (C). Linear regression equations are
indicated for each plot. For both proteins, the slopes
were significantly different (P ⬍ 0.01). Pooled data
are shown in D (AMPK-␣1) and E (AMPK-␤1).
*P ⬍ 0.05, Student’s t-test.
J Appl Physiol • VOL
110 • MARCH 2011 •
www.jap.org
Innovative Methodology
PROTEIN ANALYSIS IN INDIVIDUAL HUMAN SKELETAL MUSCLE FIBERS
823
DISCUSSION
This paper describes a significant advance in Western blotting methodology, allowing for detection of specific proteins in
the very small sample size of a segment of a human skeletal
muscle fiber. By using muscle biopsies that have been snapfrozen and then freeze-dried, this application can be used to
detect fiber-type-dependent changes in proteins, as demonstrated previously (13). One overwhelming advance of the
Fig. 2. Fiber-type expression of AMPK-␣1 and -␤1 isoforms along with various
proteins allowing fiber typing of segments of individual human skeletal muscle
fibers. Protein was separated on 4 –12% SDS-PAGE gels and transferred to
nitrocellulose. Eight individual muscle fibers are shown from three subjects.
The fiber in lane 3 was not apparent, likely as a consequence of the difficulty
in sample collection. A and B: detection of AMPK-␣1 at ⬃63 kDa and
AMPK-␤1 at ⬃38 kDa, respectively. Fibers could be classified as fast twitch
or slow twitch based on their sarcoplasmic reticulum properties [sarco(endo)plasmic reticulum Ca2⫹-ATPase type 1 (SERCA1); C] and contractile properties [tropomyosin (D) and MHC (E and F)]. Protein loading was considered
by examining both actin (G) and MHC (H), with the latter being detected in the
posttransferred Coomassie stain of the gel and used for normalization of data
(see METHODS). I: plot showing the relative amounts of AMPK-␣1 and
AMPK-␤1 in fast-twitch and slow-twitch human skeletal muscle fibers. The
density for a given AMPK isoform was normalized to the density of the MHC
and then expressed relative to the average density/MHC for the slow-twitch
fibers on a given gel. The number of fibers analyzed are shown in parentheses.
*P ⬍ 0.01, fast different from slow, Student’s unpaired t-test.
data, determined the amount of the AMPK isoform relative to
the amount of MHC on the Coomassie-stained gel. The second
analysis was a two-step process, referred to as normalized data.
First, the density of a given sample was assigned an arbitrary
value based on the standard curve, and, second, this normalized
value was expressed relative to the normalized value for MHC.
In muscle fibers collected from human muscle biopsies, the
amount of AMPK-␣1 in fast-twitch fibers was ⬃25% lower
J Appl Physiol • VOL
Fig. 3. Detection of ryanodine receptor 1 (RyR1) in human skeletal muscle
fibers. Protein was separated on 4 –12% SDS-PAGE gels and transferred to
nitrocellulose. Two individual muscle fibers (lanes 1 and 2) are shown. The
line indicates the separation of the membrane, which was imaged with white
light to visualize the molecular mass markers (HiM and Fer), and the image
was captured before chemiluminescent detection without the membrane being
moved (see METHODS).
110 • MARCH 2011 •
www.jap.org
Downloaded from http://jap.physiology.org/ by 10.220.33.1 on June 17, 2017
than that seen in slow-twitch fibers (Fig. 2I), and the amount of
AMPK-␤1 in fast-twitch fibers was less than one-half that in
slow-twitch fibers (Fig. 2I). Both raw data and normalized data
gave similar results.
The fiber type of the individual fibers was ascertained based
on the presence or absence of relevant contractile proteins
(MHC and tropomyosin, Fig. 2, H and D) and SR protein
(SERCA1, Fig. 2C). In all cases, segments of fibers that
expressed the slow-twitch isoform of MHC (MHC I, Fig. 2F),
also expressed the slow-twitch tropomyosin (slower migrating
band, i.e., larger molecular mass), and MHC II and SERCA1
were both essentially absent (see Fig. 2). Conversely, fiber
segments that expressed the fast-twitch isoform of MHC
(MHC II) also expressed the fast-twitch tropomyosin (faster
migrating band, i.e., smaller molecular mass), with the fasttwitch SERCA1 protein with MHC I being essentially absent
(see Fig. 2). Hybrid fibers were typically not observed; however, in a small proportion of fibers examined, there was a little
MHC I seen in fast-twitch fibers and MHC II seen in slowtwitch fibers. Nevertheless, this was typically ⬍5% of the
signal of the predominant MHC present, and the fibers were
thus assigned slow-twitch or fast-twitch based on the 95% or
more predominant MHC present.
To demonstrate that the method described is able to be used
for proteins with a lower abundance than some of the proteins
shown, the ability to detect the RyR1 in two fiber segments is
shown (Fig. 3).
Innovative Methodology
824
PROTEIN ANALYSIS IN INDIVIDUAL HUMAN SKELETAL MUSCLE FIBERS
J Appl Physiol • VOL
that methodology. Nevertheless, in the findings from human
single fibers, where the fibers were classified as purely type I or
type II, according to fiber-type-specific isoforms of MHC,
tropomyosin, and SERCA1, there was 25% less AMPK-␣1
protein present in fast-twitch compared with slow-twitch fibers
(Fig. 2).
The use of a standard curve in the present study provides a
rigorous test of the detection limits of the system. It also
allowed examination of the relative amount of the AMPK
isoforms in rodent muscle. Twice as much AMPK-␣1 was
found in Sol compared with EDL muscles (Fig. 1). Previously,
approximately twofold greater expression of AMPK-␣1 was
reported in predominantly fast-twitch flexor digitorum brevis
muscles compared with Sol muscles from rats (1), whereas
another study reported ⬃1.8 greater expression of AMPK-␣1
in Sol muscle compared with fast-twitch white gastrocnemius
muscle (11). The difference between those two studies, as
suggested by Putman et al. (11), was in the identification of
predominant MHC isoforms in the mixed-muscle samples.
In both human single fibers and rat skeletal muscle, there
was approximately three times more AMPK-␤1 in slow-twitch
compared with fast-twitch muscle. The relative AMPK-␤1
protein levels in whole muscle preparations have not previously been described, although early work reported a similar
level of AMPK-␤1 immunoprecipitation with AMPK-␣ isoforms from both Sol and EDL muscle (2). Despite AMPK-␤1
being as active as the ␤2-isoform, once extracted from rodent
muscle (14), recent work has identified that the majority
(⬎90%) of the AMPK trimer complexes present in EDL and
Sol muscle from mice contain the AMPK-␤2 isoform, and that
the activity of ␤1-containing trimers constitute ⬍5% and
⬍18% of the total AMPK trimer complex/activity in EDL and
Sol muscle, respectively (15). It is likely that a similar situation
exists in human skeletal muscle, since AMPK-␤1 showed no
detectable immunoprecipitation with any other AMPK isoforms in human skeletal muscle preparations (17).
Using the method described here, it is also possible to detect
many other proteins (not shown). As an example of detection
limitations, RyR1 was examined in two fast-twitch human
skeletal muscle fibers (Fig. 3). RyR1 is a large proteins (⬃550
kDa) present at ⬃200 nM (7) and could be detected using the
same approach as that for the other proteins described. As for
any scientific method, conditions will need to be optimized for
specific proteins; however, it should be reiterated that, for all of
the proteins examined in the present study, information was
obtained for the same segments of individual fibers by reprobing the Western blotting membrane. Therefore, for the array of
proteins measured here, identical sample preparation running
and transferring conditions were used.
In conclusion, the present study presents a refined method
for the determination and relative quantification of various
proteins in very small segments of freeze-dried skeletal muscle
fibers using standard laboratory equipment and commercially
available reagents. For the first time, fibers can be classified as
fast twitch or slow twitch, according to a number of properties,
including contractile and SR, and then further characterized
based on the presence of an array of other proteins.
ACKNOWLEDGMENTS
I thank Glenn McConell (current address Victoria University, Australia) for
providing the human muscle samples and Heidy Latchman for technical
110 • MARCH 2011 •
www.jap.org
Downloaded from http://jap.physiology.org/ by 10.220.33.1 on June 17, 2017
methodology presented is the ability to characterize fiber segments according to their contractile proteins present through
MHC (type I and type II) and tropomyosin properties, as well
as their SR properties (SERCA1, which is exclusively expressed in fast-twitch, type II fibers). Additionally, the method
described avoids the need to use pools of fibers (e.g.. containing ⬎20 –200 fibers) to obtain enough material to reach sensitivity levels for detection of proteins, as well as the prerequisite to fiber-type individual fibers before they can be pooled.
This can extensively save time required for experimental procedures.
The technique overcomes issues faced by researchers who
investigate changes in proteins in whole skeletal muscle biopsies collected from vastus lateralis muscles. This muscle is
composed of ⬃50% type I (slow-twitch) and ⬃50% type II
(fast-twitch) fibers in healthy, recreationally active individuals
(12), which is likely similar to the profile of the muscle
biopsies of the individuals in the current study. Typically,
well-trained endurance athletes will have a more slow-twitch
phenotype (e.g., 70 – 80% type I fibers, depending on the
training status), whereas, in sedentary individuals, there would
likely be a more fast-twitch phenotype (12). Given the disparity in fiber types present in different populations, the differential expression of many proteins, the shift in the fiber-type
profile expression in certain disease states (e.g.. Type 2 diabetes) (10), and, importantly, the various diseases that affect
different fiber types differentially, the ability to readily measure proteins in individual fibers is an important step to be
made.
Type I and type IIa fibers, however, make up the major
proportion of fiber types present in vastus lateralis muscle
biopsies (4). Of the fibers examined, all contained proteins that
immunoreacted with antibodies directed toward either the
MHC I or MHC II. The presence of I/II hybrid fibers was not
apparent (see RESULTS); however, it is likely that IIa/IIx hybrid
fibers did exist, as they typically represent ⬃13% of the total
fiber pool (16). It was not possible to measure the presence or
absence of the fast-twitch hybrid fibers because the antibody
used detected type II fibers in general. With the use of type IIaand type IIx-specific antibodies, the presence of such fibers
could be discerned. It is possible that the apparent absence of
hybrid fibers was due to the very small lengths of fiber
segments examined (⬃1.2 mm) and that a single myosin
isoform predominates over that region. In addition to determining the MHC isoform present in a given fiber segment, the
present study demonstrates the ability to also identify a given
fiber segment based on its thin-filament tropomyosin isoform
and SERCA isoform present. Fibers expressed fast-twitch
contractile and SR proteins, or slow-twitch contractile and SR
proteins with ⬎95% homogeneity (Fig. 2).
In the present study, there was ⬃25% more AMPK-␣1 in
human slow-twitch fibers compared with fast-twitch fibers
(Fig. 2, A and I). This is contrary to findings using immunofluorescence in human cross sections, where no fiber-type
difference in AMPK-␣1 was reported in resting muscle (6). In
that study, antibodies used identified the type I and IIa fibers,
with additional information obtained identifying type IIx fibers. It is difficult to make accurate quantitative measurements
across different serial cross sections using the arbitrary units of
fluorescence, and so the relatively small difference seen in the
present study might not have been within the detection limits of
Innovative Methodology
PROTEIN ANALYSIS IN INDIVIDUAL HUMAN SKELETAL MUSCLE FIBERS
assistance. The AMPK antibodies, purified AMPK proteins, and AMPK-␤1
knock-out and wild-type lysates were gifts from Dr. Jonathan Oakhill and
Bruce Kemp, St Vincent’s Institute, Melbourne, Australia. The monoclonal
antibodies directed against adult human MHC isoforms (A4.84 and A4.74)
used in the present study were developed by Dr. Blau, those directed against
SERCA1 were developed by Dr. D. Fambrough, those directed against tropomyosin were developed by Dr. J. Lim, and those directed against RyR1 were
developed by Drs. J. Airey and J. Sutko. All were obtained from the Development Studies Hybridoma Bank, under the auspices of the National Institute
of Child Health and Human Development and maintained by the University of
Iowa, Department of Biological Sciences, Iowa City, IA 52242.
GRANTS
R. Murphy was supported by National Health and Medical Research
Council of Australia Postdoctoral Fellowship (380842).
DISCLOSURES
No conflicts of interest, financial or otherwise, are declared by the author(s).
REFERENCES
J Appl Physiol • VOL
6. Lee-Young RS, Canny BJ, Myers DE, McConell GK. AMPK activation
is fiber type specific in human skeletal muscle: effects of exercise and
short-term exercise training. J Appl Physiol 107: 283–289, 2009.
7. Margreth A, Damiani E, Tobaldin G. Ratio of dihydropyridine to
ryanodine receptors in mammalian and frog twitch muscles in relation to
the mechanical hypothesis of excitation-contraction coupling. Biochem
Biophys Res Commun 197: 1303–1311, 1993.
8. Mollica JP, Oakhill JS, Lamb GD, Murphy RM. Are genuine changes
in protein expression being overlooked? Reassessing Western blotting.
Anal Biochem 386: 270 –275, 2009.
9. Murphy RM, Verburg E, Lamb GD. Ca2⫹ activation of diffusible and
bound pools of mu-calpain in rat skeletal muscle. J Physiol 576: 595–612,
2006.
10. Oberbach A, Bossenz Y, Lehmann S, Niebauer J, Adams V, Paschke
R, Schon MR, Bluher M, Punkt K. Altered fiber distribution and
fiber-specific glycolytic and oxidative enzyme activity in skeletal muscle
of patients with type 2 diabetes. Diabetes Care 29: 895–900, 2006.
11. Putman CT, Martins KJ, Gallo ME, Lopaschuk GD, Pearcey JA,
MacLean IM, Saranchuk RJ, Pette D. Alpha-catalytic subunits of
5=AMP-activated protein kinase display fiber-specific expression and are
upregulated by chronic low-frequency stimulation in rat muscle. Am J
Physiol Regul Integr Comp Physiol 293: R1325–R1334, 2007.
12. Saltin B, Henriksson J, Nygaard E, Andersen P, Jansson E. Fiber types
and metabolic potentials of skeletal muscles in sedentary man and endurance runners. Ann N Y Acad Sci 301: 3–29, 1977.
13. Tannerstedt J, Apro W, Blomstrand E. Maximal lengthening contractions induce different signaling responses in the type I and type II fibers of
human skeletal muscle. J Appl Physiol 106: 1412–1418, 2009.
14. Thornton C, Snowden MA, Carling D. Identification of a novel AMPactivated protein kinase beta subunit isoform that is highly expressed in
skeletal muscle. J Biol Chem 273: 12443–12450, 1998.
15. Treebak JT, Birk JB, Hansen BF, Olsen GS, Wojtaszewski JF.
A-769662 activates AMPK beta1-containing complexes but induces glucose uptake through a PI3-kinase-dependent pathway in mouse skeletal
muscle. Am J Physiol Cell Physiol 297: C1041–C1052, 2009.
16. Tupling AR, Bombardier E, Stewart RD, Vigna C, Aqui AE. Muscle
fiber type-specific response of Hsp70 expression in human quadriceps
following acute isometric exercise. J Appl Physiol 103: 2105–2111, 2007.
17. Wojtaszewski JF, Birk JB, Frosig C, Holten M, Pilegaard H, Dela F.
5=AMP activated protein kinase expression in human skeletal muscle:
effects of strength training and type 2 diabetes. J Physiol 564: 563–573,
2005.
110 • MARCH 2011 •
www.jap.org
Downloaded from http://jap.physiology.org/ by 10.220.33.1 on June 17, 2017
1. Ai H, Ihlemann J, Hellsten Y, Lauritzen HP, Hardie DG, Galbo H,
Ploug T. Effect of fiber type and nutritional state on AICAR- and
contraction-stimulated glucose transport in rat muscle. Am J Physiol
Endocrinol Metab 282: E1291–E1300, 2002.
2. Chen Z, Heierhorst J, Mann RJ, Mitchelhill KI, Michell BJ, Witters
LA, Lynch GS, Kemp BE, Stapleton D. Expression of the AMPactivated protein kinase beta1 and beta2 subunits in skeletal muscle. FEBS
Lett 460: 343–348, 1999.
3. Essen B, Jansson E, Henriksson J, Taylor AW, Saltin B. Metabolic
characteristics of fibre types in human skeletal muscle. Acta Physiol Scand
95: 153–165, 1975.
4. Harridge SDR, Bottinelli R, Canepari M, Pellegrino MA, Reggiani C,
Esbjornsson M, Balsom PD, Saltin B. Sprint training, in vitro and in vivo
muscle function, and myosin heavy chain expression. J Appl Physiol 82:
442–449, 1998.
5. Iseli TJ, Oakhill JS, Bailey MF, Wee S, Walter M, van Denderen BJ,
Castelli LA, Katsis F, Witters LA, Stapleton D, Macaulay SL, Michell
BJ, Kemp BE. AMP-activated protein kinase subunit interactions: beta1:
gamma1 association requires beta1 Thr-263 and Tyr-267. J Biol Chem
283: 4799 –4807, 2008.
825