Group Title: BMC Genomics
Title: Changes in skeletal muscle gene expression following clenbuterol administration
CITATION PDF VIEWER THUMBNAILS PAGE IMAGE ZOOMABLE
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00100015/00001
 Material Information
Title: Changes in skeletal muscle gene expression following clenbuterol administration
Physical Description: Book
Language: English
Creator: Spurlock, Diane
McDaneld, Tara
McIntyre, Lauren
Publisher: BMC Genomics
Publication Date: 2006
 Notes
Abstract: BACKGROUND:Beta-adrenergic receptor agonists (BA) induce skeletal muscle hypertrophy, yet specific mechanisms that lead to this effect are not well understood. The objective of this research was to identify novel genes and physiological pathways that potentially facilitate BA induced skeletal muscle growth. The Affymetrix platform was utilized to identify gene expression changes in mouse skeletal muscle 24 hours and 10 days after administration of the BA clenbuterol.RESULTS:Administration of clenbuterol stimulated anabolic activity, as indicated by decreased blood urea nitrogen (BUN; P < 0.01) and increased body weight gain (P < 0.05) 24 hours or 10 days, respectively, after initiation of clenbuterol treatment. A total of 22,605 probesets were evaluated with 52 probesets defined as differentially expressed based on a false discovery rate of 10%. Differential mRNA abundance of four of these genes was validated in an independent experiment by quantitative PCR. Functional characterization of differentially expressed genes revealed several categories that participate in biological processes important to skeletal muscle growth, including regulators of transcription and translation, mediators of cell-signalling pathways, and genes involved in polyamine metabolism.CONCLUSION:Global evaluation of gene expression after administration of clenbuterol identified changes in gene expression and overrepresented functional categories of genes that may regulate BA-induced muscle hypertrophy. Changes in mRNA abundance of multiple genes associated with myogenic differentiation may indicate an important effect of BA on proliferation, differentiation, and/or recruitment of satellite cells into muscle fibers to promote muscle hypertrophy. Increased mRNA abundance of genes involved in the initiation of translation suggests that increased levels of protein synthesis often associated with BA administration may result from a general up-regulation of translational initiators. Additionally, numerous other genes and physiological pathways were identified that will be important targets for further investigations of the hypertrophic effect of BA on skeletal muscle.
General Note: Periodical Abbreviation:BMC Genomics
General Note: Start page 320
General Note: M3: 10.1186/1471-2164-7-320
 Record Information
Bibliographic ID: UF00100015
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: Open Access: http://www.biomedcentral.com/info/about/openaccess/
Resource Identifier: issn - 1471-2164
http://www.biomedcentral.com/1471-2164/7/320

Downloads

This item has the following downloads:

PDF ( PDF )


Full Text



BMC Genomics io Central



Research article

Changes in skeletal muscle gene expression following clenbuterol
administration
Diane M Spurlock* 1, Tara G McDaneldI and Lauren M Mclntyre2


Address: 'Department of Animal Sciences, Iowa State University, Ames, IA, USA and 2Department of Molecular Genetics and Microbiology,
University of Florida, Gainesville, FL, USA
Email: Diane M Spurlock* moodyd@iastate.edu; Tara G McDaneld tmcd@iastate.edu; Lauren M McIntyre mcintyre@ufl.edu
* Corresponding author



Published: 20 December 2006 Received: 19 May 2006
BMC Genomics 2006, 7:320 doi:10.1186/1471-2164-7-320 Accepted: 20 December 2006
This article is available from: http://www.biomedcentral.com/1471-2164/7/320
2006 Spurlock et al; licensee BioMed Central Ltd.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0),
which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.



Abstract
Background: Beta-adrenergic receptor agonists (BA) induce skeletal muscle hypertrophy, yet
specific mechanisms that lead to this effect are not well understood. The objective of this
research was to identify novel genes and physiological pathways that potentially facilitate BA
induced skeletal muscle growth. The Affymetrix platform was utilized to identify gene expression
changes in mouse skeletal muscle 24 hours and 10 days after administration of the BA
clenbuterol.
Results: Administration of clenbuterol stimulated anabolic activity, as indicated by decreased
blood urea nitrogen (BUN; P < 0.01) and increased body weight gain (P < 0.05) 24 hours or 10
days, respectively, after initiation of clenbuterol treatment. A total of 22,605 probesets were
evaluated with 52 probesets defined as differentially expressed based on a false discovery rate
of 10%. Differential mRNA abundance of four of these genes was validated in an independent
experiment by quantitative PCR. Functional characterization of differentially expressed genes
revealed several categories that participate in biological processes important to skeletal muscle
growth, including regulators of transcription and translation, mediators of cell-signalling
pathways, and genes involved in polyamine metabolism.
Conclusion: Global evaluation of gene expression after administration of clenbuterol identified
changes in gene expression and overrepresented functional categories of genes that may regulate
BA-induced muscle hypertrophy. Changes in mRNA abundance of multiple genes associated with
myogenic differentiation may indicate an important effect of BA on proliferation, differentiation,
and/or recruitment of satellite cells into muscle fibers to promote muscle hypertrophy.
Increased mRNA abundance of genes involved in the initiation of translation suggests that
increased levels of protein synthesis often associated with BA administration may result from a
general up-regulation of translational initiators. Additionally, numerous other genes and
physiological pathways were identified that will be important targets for further investigations of
the hypertrophic effect of BA on skeletal muscle.







Page 1 of 15
(page number not for citation purposes)







http://www.biomedcentral.com/1471-2164/7/320


Background
Anabolic effects of P-adrenergic receptor agonists (BA)
have been widely studied for potential applications in the
prevention of muscle atrophy [1,2] and improvement of
the efficiency of muscle growth in production livestock [3-
5]. Clenbuterol is a p2-adrenergic receptor agonist that
has been shown to have a significant effect on muscle
metabolism in a variety of muscle atrophy models,
including hind-limb suspension atrophy [6,7], starvation
induced atrophy [8], and denervation induced atrophy
[9,10]. Additionally, clenbuterol is known to induce a sig-
nificant repartitioning effect by increasing the growth of
skeletal muscle at the expense of fat tissues in most live-
stock species [3,5]. Although it is known that clenbuterol
initiates these effects via activation of the p2-adrenergic
receptor [3,5], the downstream mechanisms by which
activation of these receptors results in increased muscle
growth or decreased muscle atrophy are not clear. To date,
the expression and activity of specific genes have been
investigated in a variety of models in order to implicate
specific pathways with the skeletal muscle response to BA.
For example, increased abundance of myofibrillar and
structural proteins has been demonstrated and appears to
result from increases in both transcription and translation
of these genes [11-13]. Additionally, endogenous protein-
ases including genes of the ubiquitin-proteasome path-
way and calcium-dependent proteolytic enzymes have
been reported to mediate protein turnover in skeletal
muscle after administration of BA [14]. Finally, changes in
skeletal muscle expression of IGF1 and IGF2 mRNA have
been observed shortly after the administration of clen-
buterol to rodents [15], suggesting the regulation of these
growth factors may be important in the initial response of
skeletal muscle to BA.

Although investigations of candidate genes have been
informative in terms of implicating individual pathways
in the skeletal muscle response to BA, they have not pro-
vided a global view of changes occurring in the tissue, and
have been limited to the investigation of genes with
known functions. Thus, the objective of the current
research was to define changes in the global gene expres-
sion profile of skeletal muscle in response to administra-
tion of the BA clenbuterol. Two time points relative to
clenbuterol administration were investigated in order to
compare gene expression profiles following short- (24
hour; 24 h) and long- (10 day; 10 D) term clenbuterol
administration in mice. The Affymetrix platform was cho-
sen for the analysis of gene expression in order to investi-
gate the most comprehensive collection of genes
available.


Results
Clenbuterol stimulated an anabolic response
A significant effect of clenbuterol on blood urea nitrogen
(BUN) was observed due to decreased BUN in 24 h com-
pared to control (C) treatment groups (P < 0.01; Figure 1).
Although average BUN of 10 D treated mice was less than
that of C mice, this difference was not statistically signifi-
cant (P > 0.05). Body weight gain tended to differ among
the three groups of mice (P = 0.06), with a significant
increase in body weight gain observed following 10 D
clenbuterol administration compared to the C group (P <
0.05; Figure 2).

Clenbuterol stimulated changes in mRNA abundance
A total of 22,605 probesets were deemed detected across
the MOE430A and MOE430B chips and were included in
all subsequent analyses. All data were deposited in the
GEO data base (Accession number GSE4490). A total of
137, 56 and 4 probesets were differentially expressed
based on false discovery rates (FDR) of 20, 10, and 5%,
respectively, which correspond to P-values less than
0.0012, 0.00025, and 0.0000082, respectively. We consid-
ered genes that were significant at the 10% FDR threshold
significantly differentially expressed. While individual
nominal significance levels are often set at 5 %, when mul-
tiple tests are performed the impact of both the type I
(false positive) and type II (false negative) error rate
should be carefully considered. The FDR provides a way of
controlling the expected number of false positives in the
list of tests rejected [16,17]. With 56 rejections a 10% FDR
says that the expectation is less than six false positives. In
this case, the substantial increase in the number of genes

I1C II1D EI10D
9 t .--------------------


20 +


05 ------ - -


Figure I
Blood urea nitrogen (BUN) levels verify ananabolic
response. Blood urea nitrogen levels of mice following ten
days of injections, including 24 hours (24 h) or 10 (10 D) days
of clenbuterol administration, or a vehicle control (C). The
overall effect of treatment on BUN was significant (P < 0.01),
and significance values from contrasts comparing each clen-
buterol treatment to the control group are shown.


Page 2 of 15
(page number not for citation purposes)


BMC Genomics 2006, 7:320







http://www. biomedcentral.com/1471-2164/7/320


DC E1D I10D


Figure 2
Body weight gain measurements verify an ana-
bolicresponse. Body weight gain of mice following ten days
of injections, including 24 hours (24 h) or 10 (10 D) days of
clenbuterol administration, or a vehicle control (C). The
overall effect of treatment on body weight gain showed a
trend for significance (P = 0.06), and significance values from
contrasts comparing each clenbuterol treatment to the con-
trol group are shown.



identified is much larger than the anticipated increase in
type I error and so we opted to place more weight on
reducing the type II error and increasing power. Among
these 56 probesets, four displayed evidence for non-nor-
mal distributions of residuals. In order to be conservative
we do not consider these further, leaving 52 probesets
deemed differentially expressed (Table 1). The mRNA
abundance of 63% of these probesets differed at both 24
h and 10 D, with 24 transcripts having consistently more
and 8 consistently less mRNA relative to C at both time
points. One gene, A530047J11Rik, changed in opposite
directions relative to C in the 24 h and 10 D treatment
groups. A total of 20 probesets showed altered mRNA
abundance in the 24 h but not 10 D treatment, with
approximately equal numbers increasing (11) and
decreasing (8) relative to the C group.

A recent paper from Yi and Xu 2006 [18] describes a
method for grouping or clustering genes based upon a
time series that places genes with similar expression pat-
terns, regardless of statistical significance, into clusters.
We applied this technique to our data and found that the
22,605 detected genes were grouped into a total of 10
clusters describing gene expression profiles over time (Fig-
ure 3). The number of probesets included in each cluster
ranged from eight to 21,738 (Table 3), with the largest
group representing genes with a 'flat' profile, or no evi-
dence of changes in gene expression. However, this
method identified 867 genes whose mean gene expres-


sion changed over time. Expression changes represented
by each cluster are shown in Figure 3, and the cluster to
which each probeset belongs is given in Supplementary
Table 1.

The discrepancy between the number of genes whose
mean behaviour appears different, and the number which
are statistically significant after correction for multiple
tests, may be indicative of weaker signals in these data
than in many array experiments where the treatment con-
ditions produce a dramatic direct response. If clenbuterol
indirectly affects a gene, by perhaps targeting an upstream
regulator, the magnitude of the effect seen in the down-
stream targets would be lower. In order to include genes
potentially indirectly regulated by clenbuterol administra-
tion in our exploratory analysis of the functional groups
underlying response to clenbuterol, a nominal threshold
of P < 0.01 was used. This group of 575 genes was exam-
ined further to help identify biological processes poten-
tially influenced by clenbuterol treatment. A total of 309
of these genes were categorized into 24 Gene Ontology
biological process categories (Level 2) with two or more
members, while 242 genes remained unclassified (Table
4). Categories with the greatest number of genes are con-
sistent with the model of skeletal muscle hypertrophy.
Analysis results for all genes, including the overall and
contrast P-values, cluster, and mean expression for each
treatment (for log transformed and raw data) are provided
in Supplementary Table 1.

Clenbuterol stimulated changes in families and categories
of genes
Five hundred and seventy five genes (P < 0.01) were fur-
ther evaluated using EASE score [19] to identify categories
of over-represented genes, based on the Gene Ontology
biological process annotation. Briefly, the EASE score
compares the proportion of differentially expressed genes
found within a category to the proportion of genes in that
category detected from the GeneChip. The EASE score test
statistic accounts for issues related to multiple testing
across many biological process categories in a semi-con-
servative manner [ 19 1. A total of 19 biological process cat-
egories were over-represented in our set of genes of
interest, based on an EASE score < 0.05. After accounting
for redundant categories, the differentially expressed
genes were classified into 10 over-represented groups
(Table 5). Categories of particular interest to the model of
clenbuterol stimulated muscle hypertrophy include: intra-
cellular signaling cascade, amino acid and derivative
metabolism, translation, and transcription from the Pol II
promoter. Specific genes from these categories are
described in detail in the Discussion and highlighted in
Table 1 and Table 2. The list of genes of interest was also
searched for members that had previously been associated
with BA induced muscle hypertrophy, including structural


Page 3 of 15
(page number not for citation purposes)


BMC Genomics 2006, 7:320








http://www. biomedcentral.com/1471-2164/7/320


Table 1: Significantly differentially expressed genes (10% FDR threshold).


Gene Name


AffylD


1424638_
1454971_
1417962_
1417109_
1439836


1418571_at Tumor necrosis factor receptor superfamily,
member 12a
1416505_at Nuclear receptor subfamily 4, group A,
member I
1416328_aat ATPase, H+ transporting, VO subunit E
1434303_at Ortholog of human Ras association and
pleckstrin homolog domains I
1438510_aat Histidyl-tRNA synthetase
1418326_at Solute carrier family 7 cationicc amino acid
transporter, y+ system), member 5


Homebox only domain
Protein kinase, cGMP-dependent, type I
Pannexin I


Ranka P-valueb Change 24 hc Change 10 Dd Change 10 D/24 he


Cyclin-dependent kinase inhibitor IA (P21)
TSC22 domain family, member I
Growth hormone receptor
Tubulointerstitial nephritis antigen-like
Ankyrin repeat and SOCS box-containing
protein 15
Emerin
Myeloid/lymphoid or mixed-lineage
leukaemia
Vinculin
Eukaryotic translation initiation factor 4AI
Chemokine ligand 12
Actin-binding Rho activating protein
Microfibrilar-associated protein 3-like
Down syndrome critical region gene I-like I
YYI transcription factor
Thimet oligopeptidase I
RIKEN cDNA 4631423 F02 gene
ADP-ribosyltransferase I
Actin related protein 2/3 complex, subunit 3
Tubulin, beta 5
Unknown
Unknown
Fructose bisphosphatase 2
Interleukin 6 signal transducer
Enhancer of yellow 2 homolog
Serine (or cysteine) preptidase inhibitor,
clade A, member I b
Contactin associated protein-like 2
Kelch repeat and BTB (POZ) domain
containing 5
Myosin, light polypeptide kinase 2, skeletal
muscle
Angiotensinogen


28 0.000094


0.0000957
0.000104156


31 0.000105021


0.00010965
0.000 112542


0.000118695
0.000121684


0.000122982
0.000125102
0.000125378


Page 4 of 15
(page number not for citation purposes)


3.62E-06
4.49E-06
6.20E-06
8.15E-06
0.0000111


0.0000145
0.0000179

0.0000234
0.0000273
0.0000281
0.0000285
0.0000346
0.0000361
0.0000363
0.0000407
0.0000411
0.0000451
0.0000453
0.0000563
0.0000619
0.0000684
0.0000703
0.0000733
0.0000756
0.0000774


0.0000822
0.0000929


1417357 at
1416313 at


1416156_
1434985_
1417574_
1458455_
1428804_
1421425_
1435824_
1448907
1452707
1451372_
1448279_
1416256_
1440838_
1439616_
1449088_
1452843_
1460392_
1455218


1422798_at
1431043_at


1427556_at


1423396_at


23.11
-1.52
-2.02
2.36
-5.38


5.56
-2.69

2.04
1.94
2.77
4.17
n/s
-1.99
1.40
2.23
4.34
-2.11
2.43
2.33
-2.01
3.57
3.95
1.6 1
2.74
2.23


-2.84
2.36


-2.26


7.84
35.40


-1.65


1.76
2.60


1.78
-2.20


2.37


-2.62
-15.73


1.43


n/s
-1.63


1428662_
1444232_
1416379


BMC Genomics 2006, 7:320








http://www. biomedcentral.com/1471-2164/7/320


Table I: Significantly differentially expressed genes (10% FDR threshold). (Continued)


1449036_at
1416431_at
1448484_at
1447160 at


Ring finger protein 128
Tubulin, beta 6
S-adenosylmethionine decarboxylase I
Unknown


1453851 _aat Growth arrest and DNA-damage-inducible
45 gamma
1422557_sat Metallothionein I
1440884_s_at Ribonucleotide reductase M2B 9TP53
inducible


Bromodomain containing 4
Insulin-like growth factor I
Dehydrogenase/reductase (SDR family)
member 7


1424058_at RIKEN cDNA I 190002C06 gene
1455643_sat TSR I, 20S rRNA accumulation, homolog


1451751_at
1421466_at

1416831 at


DNA-damage-inducible transcript 4-like
Ankyrin repeat and SOCS box-containing
protein 10
Neuraminidase I


1451924_aat Endothelin I
1427400_at Lady bird-like homeobox I homolog
(Drosophila)
1425099_aat Aryl hydrocarbon receptor nuclear
translocator-like


39 0.000144025 4.70


40 0.000146038
41 0.00015137
42 0.00015393
43 0.000154877


32.86
-3.31
1.56


44 0.000164138 1 1.1 1
45 0.000169965 -1.73


46 0.000174896
47f 0.000176747
48 0.000182143


0.000188845
0.000191388
0.000210837
0.000224062


53 0.000239547 1.65
54 0.000244042 2.76
55 0.000244505 -1.82

56 0.000246238 3.07


aRank of P-value based on analysis of variance of all expressed genes (n = 22,605)
bP-value resulting from analysis of variance. Model included treatment (control, 24 h, and 10 D clenbuterol administration) as a fixed effect. False
discovery rates of 5, 10, and 20% correspond to P-values of 0.0000082, 0.00025, and 0.0012, respectively.
cFold change in mRNA abundance for 24 h clenbuterol treatment relative to control. A positive fold change indicates increased mRNA abundance in 24
h treatment relative to control, a negative fold change indicates decreased mRNA abundance in 24 h treatment relative to control, and n/s indicates no
significant difference between 24 h treatment and control (contrast P-value > 0.05).
dFold change in mRNA abundance for 10 D clenbuterol treatment relative to control. A positive fold change indicates increased mRNA abundance in
10 D treatment relative to control, a negative fold change indicates decreased mRNA abundance in 10 D treatment relative to control, and n/s
indicates no significant difference between 10 D treatment and control (contrast P-value > 0.05).
eFold change in mRNA abundance for 24 h clenbuterol treatment relative to 10 D clenbuterol treatment. A positive fold change indicates increased
mRNA abundance in 10 D treatment relative to 24 h treatment, a negative fold change indicates decreased mRNA abundance in 10 D treatment
relative to 24 h treatment, and n/s indicates no significant difference between 10 D treatment and 24 h treatment (contrast P-value > 0.05).
fResiduals of four genes meeting the 10% FDR threshold showed evidence of a non-normal distribution.


genes, IGF1 and its receptors and binding proteins, and
genes involved in ubiquitination and proteasomal degra-
dation (Table 2).

Validation of differential gene expression
Differential expression of four genes was validated in an
independent experiment using a different set of mice from
the microarray experiment. Genes were selected from the
list of statistically significantly differentially expressed
genes to represent different biological processes impor-
tant to muscle physiology, including genes that interact
with structural proteins (Arpc3 and vinculin), a regulator
of transcription (Hod), and a growth factor implicated in
multiple facets of muscle growth and differentiation
(IGF1). In concordance with the microarray data, mRNA


abundance of each of these genes increased 24 h after
administration of clenbuterol treatment relative to C ani-
mals (P < 0.05), while GAPDH represented a housekeep-
ing gene whose mRNA abundance did not differ between
treatments (Table 6). Thus, these data provide strong bio-
logical validation of results from the GeneChip experi-
ment.

Discussion
Although it has been known for decades that BA stimulate
muscle hypertrophy [5], the data described herein are the
first to provide an overview of global changes in gene
expression associated with this biological model. These
results also contribute to a growing body of literature
describing gene expression changes associated with vari-



Page 5 of 15
(page number not for citation purposes)


14 18 111_
1437401
1426440_


-2.34
-12.59
2.22
-1.30


1.20


-1.86
-2.31
-1.54
2.80

- 1.5 1
-2.02
1.40

-3.13


BMC Genomics 2006, 7:320








http://www. biomedcentral.com/1471-2164/7/320


Cluster 4


Cluster 2







time

Cluster 5







time

Cluster 8

E -



i


Figure 3
Clustering of all expressed genes according to themethod of orthogonal polynomials. Ten clusters representing
different patterns of mean expression changes for response to clenbuterol. The number of genes in each cluster is given in
Table 2. Cluster I is not shown because minimal change in gene expression was apparent. Specific genes included in each clus-
ter are given in Additional File I.


ous models of skeletal muscle hypertrophy and atrophy
[20-22]. Together, these data facilitate the opportunity to
define genes across multiple models of muscle physiol-
ogy, as well as to identify genes specific to the effects of
BA. One unique aspect of our experiment is the investiga-
tion of gene expression at two time points relative to the
administration of clenbuterol. Previous studies have
focused on altered gene and/or protein expression after
significant changes in muscle mass have occurred
[12,13,23]. However, results of our study indicate that
changes in gene expression are equally abundant at an
early time point relative to the initiation of clenbuterol
administration. These initial changes may represent
important alterations of physiological pathways that cul-
minate in altered protein turnover and/or recruitment of
satellite cells to support muscle growth.


Our analysis showed that four of the 56 differentially
expressed genes (10% FDR) met an even more stringent
5% FDR threshold: cyclin dependent kinase inhibitor 1A
(p21) (Cdknla), TSC22 domain family, member 1
(Tsc22dl), growth hormone receptor (GHR), and tubu-
lointerstitial nephritis antigen-like (Tinagl). Two of these
genes, Cdknla and GHR, are of particular interest to this
experimental model because of their well-established
roles in muscle growth and development [24-28]. Cdknl a
inhibits cyclin dependent kinase 2 activity, contributing
to the irreversible withdrawal from the cell cycle and ter-
minal differentiation of myocytes [24]. Although the pres-
ence of Cdknla is not required for normal development
of mice [25], the absence of this gene product prevents
normal regeneration of muscle through satellite cells [26].
Therefore, the significant upregulation of Cdknla mRNA
abundance following both short and long-term clen-


Page 6 of 15
(page number not for citation purposes)


time

Cluster 7







time


Cluster 10


E


time

Cluster 6


time


BMC Genomics 2006, 7:320








BMC Genomics 2006, 7:320




Table 2: Selecteda genes of interest.


http://www. biomedcentral.com/1471-2164/7/320


Rankb P-valuec Change 24 hd Change 10 De Change 10 D/24 hf


143771 lxat Ornithine decarboxylase, structural

1452114_sat Insulin-like growth factor binding protein 5

1422648_at Solute carrier family 7 cationicc amino acid
transporter, y+ system), member 2

1424112_at Insulin-like growth factor 2 receptor

1448166_aat Proteasome (prosome, macropain) subunit,
beta type I

1420971_at Ubiquitin protein ligase E3 component n-
recognin I

1449547_at Ankyrin repeat and SOCS box-containing
protein 14

1449940_aat Eukaryotic translation initiation factor 2B,
subunit 4 delta

1421027_a at Myocyte enhancer factor 2C

1451422_at Myosin XVIIla

1448171 at Seven in absentia 2

1451982_at Mitogen activated protein kinase kinase 4

1426850_aat Mitogen activated protein kinase kinase 6

1448990_aat Myosin I B

1423449_a_at Actinin alpha 4

1454664_aat Eukaryotic translation initiation factor 5

1435787_at Protein phosphatase I (formerly 2C)-like

1451136_aat Eukaryotic translation initiation factor 2B,
subunit 2 beta

1419391 _at Myogenin

1424268_at Spermine oxidase

1426833_at Eukaryotic translation initiation factor 4
gamma, 3


70 0.000372006 1.68

83 0.000505398 -4.01

84 0.000505714 -1.81


95 0.000653571 -1.58

96 0.000658819 1.39


169 0.001652967


218 0.002452054 -2.64


268 0.003245646


294 0.003678395


301 0.003765249 1.68

342 0.004599517 4.39

363 0.004917906 1.44

393 0.00551809 -2.64

405 0.005781979 -1.52


422 0.006130447


450 0.006788818 1.63

454 0.006890512 -1.79

455 0.006894746 1.46


460 0.007033703 4.07

473 0.007450143 -2.54

569 0.009710166 -1.41


aSelected genes of interest that did not meet the 10% FDR criteria, but showed evidence of differential expression (P < 0.01) are presented and
discussed in the text.
bRank of P-value based on analysis of variance of all expressed genes (n = 22,605)
cP-value resulting from analysis of variance. Model included treatment (control, 24 h, and 10 D clenbuterol administration) as a fixed effect. False
discovery rates of 5, 10, and 20% correspond to P-values of 0.0000082, 0.00025, and 0.0012, respectively.
dFold change in mRNA abundance for 24 h clenbuterol treatment relative to control. A positive fold change indicates increased mRNA abundance
in 24 h treatment relative to control, a negative fold change indicates decreased mRNA abundance in 24 h treatment relative to control, and n/s
indicates no significant difference between 24 h treatment and control (contrast P-value > 0.05)
eFold change in mRNA abundance for 10 D clenbuterol treatment relative to control. A positive fold change indicates increased mRNA abundance
in 10 D treatment relative to control, a negative fold change indicates decreased mRNA abundance in 10 D treatment relative to control, and n/s
indicates no significant difference between 10 D treatment and control (contrast P-value > 0.05)
'Fold change in mRNA abundance for 24 h clenbuterol treatment relative to 10 D clenbuterol treatment. A positive fold change indicates increased
mRNA abundance in 10 D treatment relative to 24 h treatment, a negative fold change indicates decreased mRNA abundance in 10 D treatment
relative to 24 h treatment, and n/s indicates no significant difference between 10 D treatment and 24 h treatment (contrast P-value > 0.05)









Page 7 of 15
(page number not for citation purposes)


AffylD


Gene Name


-1.28

2.63

2.23


n/s

-1.13


2.12


2.08


-1.65


1.58

-1.22

-1.71

-1.23

2.51

1.63

-1.69

-1.79

1.49

-1.54


-3.61








http://www. biomedcentral.com/1471-2164/7/320


Table 3: Summary of results from cluster analysis.


Cluster ID Number of Genes


21,738
14
93
152
23
36
88
69
410


Percentage of Genes

96.05
0.06
0.41
0.67
0.10
0.16
0.39
0.30
1.81
0.04


buterol administration in our experiment may suggest an
increased potential for terminal differentiation and
recruitment of myogenic precursor cells in support of
muscle hypertrophy. GHR is the transmembrane receptor
for growth hormone (GH), a hormone essential for nor-
mal growth [27]. Although GH effects are largely medi-
ated through the stimulation of synthesis and secretion of
insulin-like growth factor 1 (IGF1), it has been demon-
strated that GH has important roles in the postnatal regu-
lation of skeletal muscle growth that are independent of
IGF1 [28]. This work demonstrates GH signalling from
GHR influences muscle hypertrophy by facilitating the
fusion of myoblasts with myotubes [28]. Our data reveal
a significant down-regulation of GHR in skeletal muscle
following one and ten days of clenbuterol administration.

Table 4: Differentially expressed genes of interest (P < 0.01)
categorized according to Gene Ontology biological process
terms (level 2).

Gene Ontology Biological Process Categorya Number of Genes


Cellular physiological process
Metabolism
Cell communication
Regulation of physiological process
Localization
Morphogenesis
Organ development
Response to stimulus
Organismal physiological process
Regulation of cellular process
Cell differentiation
Death
Homeostasis
Regulation of development
Reproduction
Other categories


aA total of 309 unique gene identifiers were placed in categories
with a minimum of 2 members, while 242 remained unclassified.
Categories containing a minimum of 7 members, and the number of
genes belonging to the category are shown.
bOther categories include: embryonic development, regulation of
enzyme activity, growth, mesoderm development, pattern
specification, coagulation, extracellular structure organization and
biogenesis, pathogenesis, rhythmic process.


This change in GHR expression is consistent with previous
reports of decreased GHR mRNA in a compensatory over-
load model [29], and fiber-type specific increased GHR
mRNA in a hindlimb suspension model of muscle atro-
phy [30]. Together, these reports clearly implicate GHR as
a potential regulator of skeletal muscle growth and atro-
phy across multiple experimental models.

The EASE classification of potential genes of interest also
identified overrepresented functional categories known to
be important in muscle physiology and growth. Closer
evaluation of genes within these categories reveals multi-
ple pathways that potentially contribute to BA induced
muscle hypertrophy. Each of the genes discussed below is
included in Table 1 or 2, which include the level of signif-
icance, fold and direction of change for each gene. Addi-
tionally, all genes grouped in the categories of interest
described below are identified in Supplementary Table 1.

Amino acid and derivative metabolism
The group of 16 genes categorized as being involved in
'amino acid and derivative metabolism' includes three
genes critical to polyamine metabolism (S-adenosylme-
thionine decarboxylase 1 [Amdl; P = 0.00015], ornithine
decarboxylase, structural 1 [Odcl; P= 0.00037], and sper-
mine oxidase [Smox; P = 0.00745]). Polyamines, includ-
ing spermine, spermidine, and putrescine, are
polycationic compounds found in both prokaryotic and
eukaryotic cells that are known to be crucial to growth and
proliferation of mammalian cells [31-341. The three genes
differentially expressed in our experiment represent criti-
cal steps in the biosynthesis of polyamines. Polyamines
have previously been associated with cardiac hypertrophy
stimulated by BA [35-37]. For example, the Odcl inhibi-
tor DFMO successfully blocked cardiac hypertrophy nor-
mally associated with clenbuterol [35,36], and
overexpression of Odcl in a transgenic model resulted in
a significant increase in BA stimulated cardiac hypertro-
phy relative to non-transgenic mice [37]. These experi-
ments demonstrate a clear interaction between
polyamines and beta adrenergic receptor stimulated car-
diac hypertrophy. However, the relationship between
polyamine metabolism and beta adrenergic receptors
remains to be defined for skeletal muscle. Our data sug-
gest that clenbuterol administration increases mRNA
abundance of Odcl while decreasing the abundance of
Amdl and Smox. This differential regulation is difficult to
interpret since each of these genes participate in the syn-
thesis of polyamines. However, increased Odcl expres-
sion is consistent with cardiac models that demonstrate
an interaction between Odcl activity and BA. It is also of
interest to note that mRNA abundance was altered for two
genes that function in the transport of Arginine solutee
carrier family 7 cationicc amino acid transporter, y+ sys-
tem], members 2 and 5 [Slc7a2; P = 0.00051 and Slc7a5;


Page 8 of 15
(page number not for citation purposes)


BMC Genomics 2006, 7:320








http://www. biomedcentral.com/1471-2164/7/320


Table 5: Evaluation of differentially expressed genes of interest (P < 0.01) to identify over-represented families of genes.


Gene Ontology Biological Process Categoryb

Homeostasis (Cell homeostasis, Cation homeostasis, Cell ion homeostasis, Ion homeostasis)
Amino acid and derivative metabolism (Amine metabolism, Amino acid metabolism)
Intracellular signalling cascade
Neurogenesis
Transcription from Pol II promoter (Positive regulation of transcription)
Translation (Translational initiation)
Physiological processes
Morphogenesis
Lymphocyte activation
Protein-mitochondrial targeting


List Hitsc Population Hitsd EASE Scoree


63
196
519
209
229
144
6030
617
42
8


aData were evaluated using EASE software [66] and Gene Ontology biological process categories.
bCategories with evidence for over-representation (EASE score < 0.05) are shown, with redundant categories listed in parentheses.
cThe number of genes from the list of differentially expressed genes within each over-represented category. A total of 331 differentially expressed
genes were assigned to a category.
dThe number of genes included on the MOE430A and B chips within each over-represented category. A total of 7,464 genes were assigned to a
category.
eThe EASE score is the upper bound of the distribution of Jackknife Fisher exact probabilities.


P = 0.00012]), the precursor of omithine. However,
mRNA abundance for these transporters changed in
opposite directions following clenbuterol administration.
The identification of differential expression of multiple
genes involved in polyamine metabolism, combined with
previous data linking polyamines to BA induced cardiac
hypertrophy define polyamine metabolism as a critical
target for further investigation of BA induced skeletal mus-
cle growth.

Translation
This gene ontology category was represented by 13 genes,
including four eukaryotic translation initiation factors.
Eukaryotic initiation factor 2 (Eif2) is involved in the first
step of translation through the formation of a ternary
complex between the initiator tRNA and GTP [38,39].
This complex then binds to the 40S ribosomal subunit to
initiate translation. Following start codon recognition,
GTP is hydrolyzed and the resulting GDP-Eif2 complex is
released. A new cycle of initiation of translation requires
Eif2b to catalyze the exchange of Eif2-bound GDP for


GTP, which is an important step in the regulation of trans-
lation initiation. Because Eif2b is present in low amounts,
it is an important factor controlling the global rate of pro-
tein synthesis [38,39]. Our data show that mRNA abun-
dance of two subunits of Eif2b, Eif2b4 (P = 0.00325) and
Eif2b2 (P = 0.00690), are upregulated in response to clen-
buterol administration. Increased abundance of these
subunits may contribute to increased activity of Eif2b and
the global upregulation of protein synthesis in skeletal
muscle previously associated with clenbuterol administra-
tion [40,41]. An association between Eif2b and adrenergic
receptors has also been described as Eif2b 1 directly inter-
acts with the beta-2 adrenergic receptor [42]. Overexpres-
sion of Eif2b 1 in 293 cells caused a small but significant
increase in beta adrenergic receptor signalling activity.
Although expression of Eif2b 1 was not altered in our
experiment, it is not known if the effect of Eif2b 1 on beta
receptor activity depends on association with other pro-
teins, such as Eif2b2 and Eif2b4 that may be present in
limiting quantities. Therefore, the observed upregulation
of Eif2b2 and Eif2b4 in our study may represent a mech-


Table 6: Confirmation of differential gene expression 24 hours after clenbuterol administration by quantitative PCR.

Log of starting copy number


Clenbuterol

5.12
7.19
3.50
5.58
3.89


Standard Error

0.10
0.06
0.07
0.06
0.07


P-value

0.0001
0.15
0.0001
0.0001
0.0001


aArpc3: actin related protein 2/3 complex subunit; GAPDH: glyceraldehyde 3-phosphate dehydrogenase; Hod: homeobox only domain; IGF I: insulin
like growth factor I



Page 9 of 15
(page number not for citation purposes)


Genea

Arpc3
GAPDH
Hod
IGFI
Vinculin


Control


BMC Genomics 2006, 7:320







http://www. biomedcentral.com/1471-2164/7/320


anism by which clenbuterol administration enhances sig-
nalling activity from beta adrenergic receptors. In addition
to these Eif2b subunits, elongation initiation factor 5
(Eif5; P = 0.00679) and elongation initiation factor 4g3
(Eif4g3; P = 0.00971) also had increased mRNA abun-
dance 24 hours after clenbuterol administration.
Together, these results are consistent with a general upreg-
ulation of translational machinery occurring shortly after
the administration of clenbuterol.

Transcription from Pol II promoter
A total of 18 genes were found in this category, including
four genes with known functions in muscle growth and
development: homeobox only domain (Hod; P =
0.00012), YY1 transcription factor (Yyl; P = 0.00004),
myocyte enhancer factor 2C (Mef2c; P = 0.00368), and
myogenin (Myog; P = 0.00703). The YY1 transcription
factor has been associated with myogenic differentiation
[431. In skeletal muscle, Yyl inhibits transcription of the
alpha actin gene, and down-regulation of Yyl is necessary
for alpha actin expression and myogenic differentiation to
proceed [43]. Our data show a significant increase in Yyl
mRNA abundance following 24 hours and 10 days of
clenbuterol administration, potentially contributing to
decreased differentiation of myogenic cells. However,
alterations in Yyl protein abundance have been shown to
occur through proteolytic regulation, independent of
changes in mRNA [43]. Thus, future investigation of the
regulation of Yyl in response to BA stimulation will
require careful consideration of changes at both the
mRNA and protein level. Another transcription factor that
was differentially expressed in our experiment was Hod,
an unusual homeodomain protein that modulates cardiac
growth and development [44,45]. Hod functions by inter-
acting with serum response factor (SRF) to inhibit activa-
tion of SRF-dependent transcription [46]. Inactivation of
Hod in transgenic mouse models confirm that an absence
of Hod results in an imbalance between proliferation and
differentiation of cardiomyocytes, culminating in
impaired cardiac development [44,45]. In contrast, trans-
genic mice that overexpress Hod develop severe cardiac
hypertrophy that is mediated by inhibition of SRF-
dependent transcriptional activity. Our data show a signif-
icant increase in Hod mRNA 24 hours after clenbuterol
administration. Although the effect of Hod on skeletal
muscle growth and differentiation has not been described,
its effect on cardiac hypertrophy makes it an intriguing
candidate as a potential mediator of the growth promot-
ing effects of clenbuterol. Additionally, the regulation of
Hod by BA may have important implications regarding
cardiac hypertrophy commonly observed in response to
these compounds [45]. Myogenin and Mef2c are well
characterized transcription factors known to be essential
for myoblast differentiation [47,48]. Although significant
alteration of satellite cell differentiation and recruitment


into muscle fibers in response to BA has not been
described, the observed increase in mRNA of transcription
factors that contribute to muscle cell differentiation sug-
gest the recruitment and differentiation of pre-myogenic
cells may be involved in the physiological response of
skeletal muscle to clenbuterol administration.

Intracellular signalling cascade
Of the 34 genes of interest categorized as intracellular sig-
nalling molecules, four belong to the mitogen activated
protein (MAP) kinase signalling pathway (growth arrest
and DNA-damage-inducible 45 gamma [Gadd45g; P =
0.000151, mitogen activated protein kinase kinase 4
[Map2k4; P = 0.00492], mitogen activated protein kinase
kinase 6 [Map2k6; P = 0.005521, and protein phosphatase
1 (formerly 2C)-like [Ppmll; P = 0.006891). The MAP
kinase signal transduction pathway is of particular interest
to this skeletal muscle model because it is a key mediator
of the cellular response to insulin-like growth factor 1
(IGF1), a known regulator of myogenic cell proliferation,
differentiation, and protein turnover [49,50]. In our
experiment, mRNA abundance of IGF1 increased 24
hours after clenbuterol administration (P = 0.00018), but
was not different from C after 10 days. Inconsistent results
describing the effect of BA on IGF1 have been reported,
but these likely reflect differences in the timing and sam-
pling (local IGF1 production versus circulating levels)
across experiments. Additionally, IGF2 receptor (P =
0.00065) and IGF binding protein 5 (P = 0.00051) had
decreased mRNA abundance at both 24 h and 10 D time
points. Although our experiment was not designed to clar-
ify the relationship between IGF1 and BA stimulated mus-
cle growth, it does provide data suggesting an interaction
between these pathways. One potential explanation is
that cross-talk between these pathways facilitates complex
regulation of the recruitment of myogenic precursor cells
or protein turnover within muscle fibers.

The intracellular signalling cascade category also includes
three members of the ankyrin repeat and SOCS-box (Asb)
containing gene family: AsblO0 (P = 0.00022), Asb14 (P =
0.00245) andAsb15 (P = 0.00001). Members of this gene
family have been shown to act as an E3 ligase to target spe-
cific proteins for degradation through the ubiquitin-pro-
teasome degradation pathway [51-53]. Although specific
proteins targeted for degradation through interactions
with Asb10, Asb14 and Asb15 have not been identified,
we have previously reported that Asb 15 mRNA is down-
regulated in response to BA in other animal models [54].
Additionally, we have demonstrated that localized over-
expression of Asb15 stimulates muscle hypertrophy in
vivo, and that its over-expression causes a delay in differ-
entiation and increase in protein synthesis in C2C12 cells
[55]. The current experiment provides additional support
for the hypothesis that Asb family members participate in


Page 10 of 15
(page number not for citation purposes)


BMC Genomics 2006, 7:320







http://www. biomedcentral.com/1471-2164/7/320


the regulation of intracellular signalling via beta adrener-
gic receptors by showing that multiple members of the
Asb family are coordinately regulated at the mRNA level.
Although much remains to be learned about the Asb gene
family and its role in targeting proteins for ubiquitination,
the down-regulation of three members of the Asb family
supports the idea of changes in the turnover of specific,
targeted proteins in response to BA.

The ubiquitin-proteasomal pathway is known to be a
major contributor to protein degradation in skeletal mus-
cle, and decreases in components of the ubiquitin-protea-
some pathway have been associated with decreased
protein degradation leading to muscle hypertrophy [56].
Additionally, administration of clenbuterol has been
reported to alter expression of ubiquitin-proteasome con-
jugates in mice following hindlimb suspension [14].
Although the EASE analysis of our data did not identify
ubiquitin related genes to be significantly over-repre-
sented, three genes involved in the ubiquitin-proteasomal
pathway, including ubiquitin protein ligase E3 compo-
nent n-recognin 1 [Ubrl; P = 0.00165], seven in abstentia
2 [Siah2; P = 0.00460], and proteasome subunit, beta type
1 [Psmbl; P = 0.00066], were differentially expressed. It
should also be noted that the Asb gene family members
previously discussed were not associated with gene ontol-
ogy terms related to the ubiquitin pathway at the time of
analysis. The mRNA abundance of the three ubiquitin
related genes was upregulated following clenbuterol
administration in our experiment. This result was unex-
pected because an increase in components of the ubiqui-
tin-proteasomal pathway is associated with increased
protein degradation [57,58], while clenbuterol is known
to decrease protein degradation [3]. One potential expla-
nation for this finding is that skeletal muscle undergoes
significant fiber-type transition in response to clenbuterol
administration, and the ubiquitin-proteasome system has
been implicated in this process [59]. Therefore, our results
may reflect changes in skeletal muscle specifically associ-
ated with fiber-type transition, rather than a general
increase in protein degradation.

Structural proteins
Few examples of structural proteins typically associated
with skeletal muscle hypertrophy were found in the list of
differentially expressed genes, although myosin IB
[Myolb; P = 0.00578], actinin alpha 4 [Actn4; P =
0.00613], and myosin XVIIIa [Myol8a; P = 0.00377] had
altered mRNA abundance associated with clenbuterol
administration. Additionally, genes encoding proteins
that interact with structural proteins were regulated. For
example, actin related protein 2/3 complex subunit
[Arpc3; P = 0.00005], and vinculin [Vcl; P = 0.00002]
increased in mRNA abundance after administration of
clenbuterol. Vinculin is an integrin-associated protein


located at junctions between actin and the plasma mem-
brane, while Arpc3 is a subunit of the Arp2/3 complex
which functions in actin remodeling [60-62]. Vinculin is
thought to facilitate actin organization by recruiting the
Arp2/3 complex which binds to actin [60-62]. Both Arpc3
and Vcl had increased mRNA abundance following clen-
buterol administration. Thus, it is reasonable to hypothe-
size that Arpc3 and Vcl are upregulated in the present
study in order to support an increase in structural proteins
associated with muscle hypertrophy.

Conclusion
Global evaluation of gene expression after administration
of clenbuterol identified changes in gene expression and
overrepresented functional categories that may regulate
BA-induced muscle hypertrophy. Changes in the mRNA
abundance of multiple genes associated with myogenic
differentiation may indicate an important effect of BA on
the proliferation, differentiation, and/or recruitment of
satellite cells into muscle fibers to promote muscle hyper-
trophy. Additionally, increased mRNA abundance of
genes involved in the initiation of translation suggests
that increased levels of protein synthesis often associated
with BA administration may result from a general up-reg-
ulation of translational initiators, rather than a sustained
up-regulation of gene expression at the transcriptional
level. Finally, numerous other genes and physiological
pathways were identified that will be important targets for
further investigations of the hypertrophic effect of BA on
skeletal muscle.

Methods
Animal management and tissue collection
A total of 36 C57BL/6J male mice (Jackson Labs, Bar Har-
bor, ME), 3 to 5 wk of age, were utilized in the experiment.
Mice were housed three per cage and maintained at 25 C
with a 12:12-h light-dark cycle and ad libitum access to
water and feed (Rodent Laboratory Chow, Ralston Purina,
St. Louis, MO) throughout the experiment. All animals
were handled in accordance with the protocol approved
by the Purdue Animal Care and Use Committee. Four
cages (three mice per cage) were randomly assigned to one
of three experimental treatments (n = 12 mice per treat-
ment): vehicle control (C), 24 hour clenbuterol treatment
(24 h), or 10 day clenbuterol treatment (10 D). Clen-
buterol (Sigma, St. Louis, MO) was prepared by dissolving
2.5 mg in 1 mL of ethyl alcohol, diluting to 0.25 mg/ml
with 1:1 PEG200:phosphate buffered saline, and steri-
lized by filtration. The vehicle control treatment was pre-
pared in the same manner, except that no clenbuterol was
added. Following a seven day acclimation period, all mice
were administered daily intraperitoneal injections of con-
trol or clenbuterol treatments for ten days. The C group
received the control treatment all ten days, the 24 h group
received the control treatment for nine days and den-


Page 11 of 15
(page number not for citation purposes)


BMC Genomics 2006, 7:320







http://www. biomedcentral.com/1471-2164/7/320


buterol on day 10, and the 10 D group received clen-
buterol for all ten days. The clenbuterol treatment was
administered at a dosage of 1 mg/kg of body weight, and
an equivalent volume of the control treatment was
injected. All mice were euthanized by CO2 asphyxiation
24 hours following the tenth injection. The gracillus mus-
cle was collected from each animal, immediately frozen in
liquid nitrogen, and stored at -80 C pending RNA extrac-
tion. The gracillus muscle was selected for analysis since it
contains a mixed fiber type of both Type I and Type II fib-
ers.

Body weight change and blood urea nitrogen level (BUN)
were measured as indicators of a physiological response to
the clenbuterol treatment. Individual body weights were
recorded daily at the time of treatment administration
(approximately 10 am), and body weight gain (WG) was
calculated as the difference in body weight between days
ten and one. Truncal blood was collected at the time of tis-
sue collection, immediately following euthanasia. Serum
was isolated and stored at -20 C pending analysis. Blood
urea nitrogen was measured via methods of Kerscher and
Ziegenhorn [63]. WG and BUN were evaluated by analysis
of variance using the GLM procedure of SAS [64] to test
the main effect of clenbuterol treatment. Means of the 24
h and 10 D groups were compared to the C group using
contrasts to determine if differences were significant.

RNA isolation and hybridization
A total of 27 mice (9 per treatment) were selected in order
to maximize the difference in BUN and WG between the
24 h and 10 D treatments relative to C. Total RNA was
extracted from the gracillus muscle of each mouse using
the Qiagen RNeasy Mini kit following the manufacturer's
recommended protocol, including an additional step of
protein kinase digestion for muscle tissue (Qiagen, Valen-
cia, CA). Quality and quantity of individual RNA samples
were examined using a 2100 Bioanalyzer (Agilent, Palo
Alto, CA). Three pools of RNA representing each treat-
ment were made. Mice were randomly assigned to pools
within treatment (three mice per pool) such that each
pool contained an equal quantity of RNA from each
mouse, for a total of greater than 50 gg of RNA at a con-
centration of at least 2.5 gg/gl.

Messenger RNA, cRNA synthesis and labelling reactions
were performed independently for each replicate follow-
ing the recommendations of the Gene Chip Expression
Analysis technical manual (Affymetrix, Santa Clara, CA).
The MOE430A and B chips were hybridized to the frag-
mented cRNA, stained and washed according to the rec-
ommendations of the Gene Chip Expression Analysis
technical manual (Affymetrix, Santa Clara, CA) in the Pur-
due University Genomics Core facility. Image data were


quantified using GeneChip Analysis Suite/Microarray
Suite 5.0 (MAS 5.0).

Statistical analysis
If all replicates for a particular probeset were deemed
'absent', that probeset was removed from further consid-
eration (n = 22,478) and the remaining probesets were
analyzed (n = 22,605). Transcript levels were normalized
to the chip median and log transformed. For each
probeset, which represents the combined expression data
from all relevant probe pairs on the chip, the generalized
linear model Y, = p + B1Ti + eij was fit. In each ANOVA, Yj
is a the log normalized transcript level for the ith treatment
and the jth replicate, p is the overall mean expression for
the feature and T, represents the ith treatment (C, 24 h, and
10 D). An F test of the effect of treatment for each probeset
was conducted as the ratio of the mean squares for treat-
ment over the mean squares for error, and the P-value for
the test of the null hypothesis t2 = tl = to (i.e., mean expres-
sion not different among the three treatments) was calcu-
lated. We examined the model for conformation to the
assumption of normality of the residuals testing the null
hypothesis that the residuals for each gene were normally
distributed using the Shapiro-Wilkes Test. All analyses
were performed in SAS (SAS Institute, Cary NC [64]). An
FDR [17] level of 10% was used for declaring findings sig-
nificant, and a stringent rate of 5% was also examined. A
nominal threshold of P < 0.01 was defined to identify
genes of potential interest for the EASE analysis. If the test
of the null hypothesis of difference across treatments was
rejected, and we had no evidence for departure from nor-
mality of the residuals, we declared the gene differentially
expressed across treatments and qualitatively assessed
additional contrasts comparing the effect of the treat-
ments (C versus 24 h and C versus 10 D).

We used the orthogonal polynomials method recently
described in Yi and Xu 2006 [18] to group the 22,605
probesets into clusters. Briefly, genes are clustered based
on non-linear association between gene expression and
time using orthogonal polynomials under the Gaussian
mixture model. The whole data set is assumed to be a mix-
ture of K clusters. The gene expression for individual
probesets is the sum of a cluster mean (fixed effect), a nor-
mally distributed random effect that specifies deviation
from the cluster mean, and a random measurement error.
All genes are centered so that average expression is zero
before clustering. In this way, clustering is based solely on
the pattern of association of their expression with time
and not the magnitude of gene expression. Bayesian Infor-
mation Criterion (BIC) is used to choose the optimal
number of the clusters.






Page 12 of 15
(page number not for citation purposes)


BMC Genomics 2006, 7:320







http://www. biomedcentral.com/1471-2164/7/320


Table 7: Sequences of PCR primers used for quantitative PCR.


Forward Primer (5'-3')

CACCAAGCTCATCGGTAACA
GAACATCATCCCTGCATCCA
AGCAGACGCAGAAATGGTTT
GGCATTGTGGATGAGTGTTG
AGCTCGGAAATGGTCTAGCA


Reverse Primer (5'-3')

ATGTCCTGTCCGCTTCATTC
CCAGTGAGCTTCCCGTTCA
GTAAGCCGAGGGAAGGAAGA
GTCTTGGGCAGTTCAGTGTG
GAATAAGTGCCCGCTTGGTA


aArpc3: actin related protein 2/3 complex subunit; GAPDH: glyceraldehyde 3-phosphate dehydrogenase; Hod: homeobox only domain; IGF I: insulin
like growth factor I


Functional characterization of differentially expressed
genes
Gene Ontology biological process categories were defined
for genes of interest using DAVID [65] with the Entrez
Gene ID as the primary identifier. Over-represented func-
tional categories were identified using the EASE score
[ 16 1, as determined by the Expression Analysis Systematic
Explorer (EASE) software [66], with the Entrez Gene ID as
the primary identifier. An EASE score <0.05 was used to
define over-represented categories [16].

Validation of GeneChip results
Ten additional independently reared C57BL/6J male mice
(Jackson Labs, Bar Harbor, ME), 3 to 5 wk of age, were uti-
lized in the experiment. Mice were housed two to three
per cage and maintained at 25 C with a 12:12-h light-
dark cycle and ad libitum access to water and feed (Rodent
Laboratory Chow, Ralston Purina, St. Louis, MO). All ani-
mals were handled in accordance with the protocol
approved by the Purdue Animal Care and Use Committee.
Mice were randomly assigned to one of two experimental
treatments: vehicle control (C) or 24 hours (24 h) of clen-
buterol treatment. Clenbuterol (Sigma, St. Louis, MO)
was prepared as described previously. All mice were
administered intraperitoneal injections of C or 24 h treat-
ments at approximately 10 am. The clenbuterol treatment
was administered at a dosage of 1 mg/kg of body weight,
and an equivalent volume of the control treatment was
injected. All mice were euthanized by CO2 asphyxiation
24 hours following the injection. The gracillus muscle was
collected from each animal, immediately frozen in liquid
nitrogen, and stored at -80C pending RNA extraction by
TRIzol (Invitrogen, Carlsbad, California).

Primers specific to mouse sequence for Arpc3, Hod, IGF1,
vinculin, and GAPDH were used in the qPCR experiment
(Table 7). All qPCR assays were standardized to starting
quantity of RNA. One jig of total RNA was reverse tran-
scribed using the iScript cDNA Synthesis Kit (Biorad, Her-
cules, CA), and one il of the resulting cDNA was used for


qPCR. The qPCR assay was carried out in the BioRad iCy-
cler (Biorad, Hercules, CA) in a 25 tiL final reaction vol-
ume. Quantification of PCR products was achieved using
SYBR Green (Biorad, Hercules, CA) reagents following the
manufacturer's recommended protocol with the follow-
ing thermal cycling conditions: 95 0C, 10 min (1 cycle);
950C, 15 sec, 580C, 15 sec, 720C, 15 sec (35 cycles); 40C.
The PCR products were visualized on an agarose gel to
ensure there was no non-specific PCR amplification. All
assays were done in triplicate in a 96-well plate format.
Standard controls were prepared by cloning the PCR prod-
uct into a vector and making serial dilutions of known
starting copy number. Control samples included on each
96-well plate were used to establish a standard curve for
determining the log starting copy number (LSCN) of each
experimental cDNA sample. The LSCN were analyzed by
analysis of variance using mixed model procedures of JMP
SAS (JMP 5.1, 2004). The model included treatment as a
fixed effect and cDNA sample within treatment as a ran-
dom effect. GAPDH was evaluated as a housekeeping gene
and consistent expression was observed across experimen-
tal treatments. Therefore, no further normalization of the
qPCR data was done.

Authors' contributions
DMS designed and oversaw the research and assisted with
drafting the manuscript. TGM assisted with sample collec-
tion and preparation, validated the microarray results,
and assisted with drafting the manuscript. LMM partici-
pated in design of the experiment and planned and exe-
cuted the microarray data analysis and assisted in writing
the manuscript. All authors read and approved the final
manuscript.


Page 13 of 15
(page number not for citation purposes)


Genea

Arpc3
GAPDH
Hod
IGFI
Vinculin


BMC Genomics 2006, 7:320








BMC Genomics 2006, 7:320




Additional material


Additional File 1
Analysis results for all detected genes. Results from all analyses com-
pleted for all genes are provided in the Excel (.xls) file. 'AffylD', 'Gene
Symbol', and 'Entrez Gene ID' show the annotation used for analysis of
gene ontology categories. 'Cluster' represents the cluster (1 -10, see Figure
3) to which the gene belongs. 'Rank' is the relative order of overall P-value
(lowest= 1, highest= 22,605) for all genes. 'P-value' represents; I,. i ..
of treatment (control, 24 h clenbuterol, 10 D clenbuterol) from analysis
of variance. 'Normality' shows the result of testing the hypothesis that
residuals from the analysis of variance are normally distributed (0 = fail
to reject, 1 = reject the hypothesis, suggesting non-normal distribution of
residuals). 'Contrast 24 h-C' and 'Contrast 10 D-C' and Contrast 24 h-
10 D' are the P-values resulting from the contrast comparing the 24 h
clenbuterol to control, 10 D clenbuterol to control, and 24 h clenbuterol
to 10 D clenbuterol treatments, respectively. 'Mean C (log)', 'Mean 24 h
(log)', and 'Mean 10 D (log)' show the mean normalized expression of
the log transformed data for each treatment. These represent the values
used for all statistical analyses. 'Mean C', 'Mean 24 h', and 'Mean 10 D'
show the mean normalized expression of the raw data for each treatment.
These values were used to estimate fold-change it ...... between treat-
ments. 'GO Category' represents the Gene Ontology category to which the
gene belonged, if the category was found to be over-represented by the
EASE analysis. IS ii,,i -- . iti, cascade, TL translation, TC
transcription, AA = amino acid and derivative metabolism.
Click here for file
[http://www.biomedcentral.com/content/supplementary/1471 -
2164-7-320-Si.xls]




Acknowledgements
This work was supported by the Purdue Agricultural Research Program.
Martin Gonzalo assisted with the cluster analyses.

References
1. Pellegrino MA, D'Antona G, Bortolotto S, Boschi F, Pastoris 0, Bot-
tinelli R, Polla B, Reggiani C: Clenbuterol antagonizes glucocor-
ticoid-induced atrophy and fiber type transformation in
mice. Exp Physiol2004, 89:89-100.
2. Hinkle RT, Hodge KM, Cody DB, Sheldon RJ, Kobilka BK, Isfort RJ:
Skeletal muscle hypertrophy and anti-atrophy effects of
clenbuterol are mediated by the beta2-adrenergic receptor.
Muscle Nerve 2002, 25:729-734.
3. Anderson DB, Veenhuizen EL, Schroeder AL, Jones DJ, Hancock DL:
The use of phenethanolamines to reduce fat and increase
leanness in meat animals. In Proceedings of Symposium on Fat and
Cholesterol Reduced; Advances in Applied Biotechnology Series Edited by:
Haberstroh C, Morris CE. New Orleans: Portfolio Publishing Com-
pany; 1991:43-73.
4. Mersmann HJ: Overview of the effects of b-adrenergic recep-
tor agonists on animal growth including mechanisms of
action. J Anim Sci 1998, 76:160-172.
5. Moody DE, Hancock DL, Anderson DB: Phenethanolamine
Repartitioning Agents. In Farm Animal Metabolism and Nutrition
Edited by: d'Mello JPF. New York, NY: CABI Publishing; 2000:65-96.
6. Dodd SL, Koesterer TJ: Clenbuterol attenuates muscle atrophy
and dysfunction in hindlimb-suspended rats. Aviat Space Environ
Med 2002, 73:635-639.
7. Herrera NM Jr, Zimmerman AN, Dykstra DD, Thompson LV: Clen-
buterol in the prevention of muscle atrophy: a study of hind-
limb-unweighted rats. Arch Phys Med Rehabil 2001, 82:930-934.
8. Higgins JA, Lasslett YV, Bardsley RG, Buttery PJ: The relation
between dietary restriction or Clenbuterol (a selective beta
2 agonist) treatment on muscle growth and calpain protein-


http://www.biomedcentral.com/1471-2164/7/320


ase (EC 3.4.22.17) and calpastatin activities in lambs. BrJ of
Nutr 1988, 60:645-652.
9. Cockman MD, Jones MB, Prenger MC, Sheldon RJ: Magnetic reso-
nance imaging of denervation-induced muscle atrophy:
effects of clenbuterol in the rat. Muscle Nerve 2001,
24:1647-1658.
10. Fitton AR, Berry MS, McGregor AD: Preservation of denervated
muscle form and function by clenbuterol in a rat model of
peripheral nerve injury. J Hand Surg 2001, 26:335-346.
I I. Elsasser TH, Rumsey TS, Kahl S, Czerwinski SM, Moseley WM, Ono
Y, Solomon MB, Harris F, Fagan JM: Effects of Synovex-S and
recombinant bovine growth hormone (Somavubove) on
growth responses of steers: III. Muscle growth and protein
responses. J Anim Sci 1998, 76:2346-2353.
12. Grant AL, Skjaerlund DM, Gelferich WG, Bergen WG, Merkel RA:
Skeletal muscle growth and expression of skeletal muscle a-
actin mRNA and insulin-like growth factor I mRNA in pigs
during feeding and withdrawal of ractopamine. J Anim Sci
1993, 71:33 19-3326.
13. Helferich WG, Jump DB, Anderson DB, Skjaerlund DM, Merkel RA,
Bergen WG: Skeletal muscle alpha-actin synthesis is increased
pretranslationally in pigs fed the phenethanolamine rac-
topamine. Endocrinology 1990, 126:3096-3100.
14. Yimlamai T, Dodd SL, Borst SE, Park S: Clenbuterol induces mus-
cle-specific attenuation of atrophy through effects on the
ubiquitin-proteasome pathway. J Appi Physiol 2005, 99:71-80.
15. Lo HC, Ney DM: GH and IGF-I differentially increase protein
synthesis in skeletal muscle and jejunum of parenterally fed
rats. Am J Physiol 1996, 271 :E872-E878.
16. Benjamini Y, Hochberg Y: Controlling the flase discovery rate -
a practical and powerful approach to multiple testing. J R Stat
Soc 1997, B 57:289-300.
17. Verhoeven KJF, Simonsen KL, McIntyre LM: Implementing flase
discovery rate control: increasing your power. OIKAS 2005,
108:643-647.
18. Yi Xu : Quantitative trait association microarray gene
expression data analysis. Molecular Biology and Evolution in press.
19. Hosack DA, Dennis G Jr, Sherman BT, Lane HC, Lempicki RA: Iden-
tifying biological themes within lists of genes with EASE.
Genome Biology 2003, 4:P4.
20. Carson JA, Nettleton D, Reecy JM: Differential gene expression
in the rat soleus muscle during early work overload-induced
hypertrophy. FASEBJ 2002, 16:207-209.
21. Komamura K, Shirotani-lkejima H, Tatsumi R, Tsujita-Kuroda Y,
Kitakaze M, Miyatake K, Sunagawa K, Miyata T: Differential gene
expression in the rat skeletal and heart muscle in glucocor-
ticoid-induced myopathy: analysis by microarray. Cardiovasc
Drugs Ther 2003, 17:303-310.
22. Batt J, Bain J, Goncalves J, Michalski B, Plant P, Fahnestock M, Wood-
gettJ: Differential gene expression profiling of short and long
term denervated muscle. FASEBJ 2006, 20:115-117.
23. Smith SB, Garcia DK, Davis SK, Anderson DB: Elevation of a spe-
cific mRNA in longissimus muscle of steers fed ractopamine.
J Anim Sci 1989, 67:3495-3502.
24. Guo K, Wang J, Andres V, Smith RC, Walsh K: MyoD-induced
expression of p21 inhibits cyclin-dependent kinase activity
upon myocyte terminal differentiation. Mol Cell Biol 1995,
15:3823-3829.
25. Deng C, Zhang P, Harper JW, Elledge SJ, Leder P: Mice lacking
p21CIPI/WAFI undergo normal development, but are
defective in GI checkpoint control. Cell 1995, 82:675-84.
26. Hawke TJ, Meeson AP, Jiang N, Graham S, Hutcheson K, DiMaio JM,
Garry DJ: p21 is essential for normal myogenic progenitor cell
function in regenerating skeletal muscle. AmJ Physiol Cell Physiol
2003, 285:C1019-C1027.
27. Mukherjee A, Murray RD, Shalet SM: Impact of growth hormone
status on body composition and the skeleton. Horm Res 2004,
62:35-41.
28. Sotiropoulos A, Ohanna M, Kedzia C, Menon RK, Kopchick JJ, Kelly
PA, Pende M: Growth hormone promotes skeletal muscle cell
fusion independent of insulin-like growth factor I up-regula-
tion. Proc Natl Acad Sci 2006, 103:73 15-7320.
29. Yamaguchi A, Ikeda Y, Hirai T, Fujikawa T, Morita I: Local changes
of IGF-I mRNA, GH receptor mRNA, and fiber size in rat
plantaris muscle following compensatory overload. JpnJ Phys-
iol 2003, 53:53-60.



Page 14 of 15
(page number not for citation purposes)








http://www. biomedcentral.com/1471-2164/7/320


30. Casse AH, Desplanches D, Mayet-Sornay MH, Raccurt M, Jegou S,
Morel G: Growth hormone receptor expression in atrophying
muscle fibers of rats. Endocrinology 2003, 144:3692-3697.
31. Janne J, Alhonen L, Pietila M, Keinanen TA, Uimari A, Hyvonen MT,
Pirinen E, jarvinen A: Genetic manipulation of polyamine catab-
olism in rodents. j Biochem 2006, 139:155-160.
32. Nishimura K, Murozumi K, Shirahata A, Park MH, Kashiwagi K, Igar-
ashi K: Independent roles ofelF5A and polyamines in cell pro-
liferation. BiochemJ 2005, 385(Pt 3):779-785.
33. Igarashi K, Kashiwagi K: Polyamine Modulon in Escherichia coli:
Genes Involved in the Stimulation of Cell Growth by
Polyamines. j Biochem 2006, 139:1 I1-16.
34. Igarashi K, Kashiwagi K: Polyamines: mysterious modulators of
cellular functions. Biochem Biophys Res Commun 2000,
271:559-564.
35. Cubria JC, Reguera R, Balana-Fouce R, Ordonez C, Ordonez D:
Polyamine-mediated heart hypertrophy induced by clen-
buterol in the mouse. j Pharm Pharmacol 1998, 50:91-96.
36. Pegg AE: Effect of alpha-difluoromethylornithine on cardiac
polyamine content and hypertrophy. j Mol Cell Cardiol 1981,
13:881-887.
37. Shantz LM, Feith DJ, Pegg AE: Targeted overexpression of orni-
thine decarboxylase enhances beta-adrenergic agonist-
induced cardiac hypertrophy. Biochem j 2001, 358:25-32.
38. Proud CG: Role of mTOR signalling in the control of transla-
tion initiation and elongation by nutrients. Curr Top Microbiol
Immunol 2004, 279:215-244.
39. Hershey JW: Translational control in mammalian cells. Annu
Rev Biochem 1991, 60:717-755.
40. Bardocz S, Brown DS, Grant G, Pusztai A, Stewart JC, Palmer RM:
Effect of the beta-adrenoceptor agonist clenbuterol and phy-
tohaemagglutinin on growth, protein synthesis and
polyamine metabolism of tissues of the rat. Br j Pharmacol
1992, 106:476-482.
41. Rehfeldt C, Schadereit R, Weikard R, Reichel K: Effect of clen-
buterol on growth, carcass and skeletal muscle characteris-
tics in broiler chickens. Br Poult Sci 1997, 38:366-373.
42. Klein U, Ramirez MT, Kobilka BK, von Zastrow M: A novel interac-
tion between adrenergic receptors and the alpha-subunit of
eukaryotic initiation factor 2B. j Biol Chem 1997,
272:19099-19102.
43. WalowitzJL, Bradley ME, Chen S, Lee T: Proteolytic regulation of
the zinc finger transcription factor YYI, a repressor of mus-
cle-restricted gene expression. j Biol Chem 1998, 273:6656-666 1.
44. Shin CH, Liu ZP, Passier R, Zhang CL, Wang DZ, Harris TM, Yamag-
ishi H, Richardson JA, Childs G, Olson EN: Modulation of cardiac
growth and development by HOP, an unusual homeodomain
protein. Cell 2002, 1I 10:725-735.
45. Chen F, Kook H, Milewski R, Gitler AD, Lu MM, Li J, Nazarian R,
Schnepp R, jen K, Biben C, Runke G, Mackay JP, Novotnyj, Schwartz
RJ, Harvey RP, Mullins MC, Epstein JA: Hop is an unusual homeo-
box gene that modulates cardiac development. Cell 2002,
I 10:713-723.
46. Kook H, Lepore JJ, Gitler AD, Lu MM, Wing-Man Yung W, Mackay J,
Zhou R, Ferrari V, Gruber P, Epstein JA: Cardiac hypertrophy and
histone deacetylase-dependent transcriptional repression
mediated by the atypical homeodomain protein Hop. j Clin
Invest 2003, 112:863-871.
47. Ridgeway AG, Wilton S, Skerjanc IS: Myocyte enhancer factor 2C
and myogenin up-regulate each other's expression and
induce the development of skeletal muscle in PI 9 cells. j Biol
Chem 2000, 275:41-46.
48. Ma K, Chan JK, Zhu G, Wu Z: Myocyte enhancer factor 2
acetylation by p300 enhances its DNA binding activity, tran-
scriptional activity, and myogenic differentiation. Mol Cell Biol
2005, 25:3575-3582.
49. Florini JR, Ewton DZ, Magri K: Hormones, growth factors, and
myogenic differentiation. Annu Rev Physiol 1991, 53:201-216.
50. Florini J, Magri K: Effects of growth factors on myogenic differ-
entiation. Am j Physiol 1989, 256:C701 -C71 I.
51. Chung AS, Guan YJ, Yuan ZL, Albina JE, Chin YE: Ankyrin repeat
and SOCS box 3 (ASB3) mediates ubiquitination and degra-
dation of tumor necrosis factor receptor II. Mol Cell Biol 2005,
25:4716-26.
52. Heuze ML, Guibal FC, Banks CA, ConawayJW, Conaway RC, Cayre
YE, Benecke A, Lutz PG: ASB2 is an Elongin BC-interacting pro-


tein that can assemble with cullin 5 and rbxl to reconstitute
an E3 ubiquitin ligase complex. j Biol Chem 2005, 280:5468-5474.
53. Wilcox A, Katsanakis KD, Bheda F, Phillay TS: Asb6, an adipocyte-
specific ankyrin and SOCS box protein, interacts with APS
to enable recruitment of Elongins B and C to the insulin
receptor signalling complex. j Biol Chem 2004, I39:1794-1800.
54. McDaneld TG, Hancock DL, Moody DE: Altered mRNA abun-
dance ofASBIS and four other genes in skeletal muscle fol-
lowing administration of beta-adrenergic receptor agonists.
Physiol Genomics 2004, 16:275-283.
55. McDaneld TG, Hannon K, Moody DE: Ankyrin repeat andSOCS
box protein (ASB) 15 regulates protein synthesis in skeletal-
muscle. Am Physiol Regul Integr Comp Physiol 2006 in press.
56. Glass DJ: Skeletal muscle hypertrophy and atrophy signalling
pathways. Intj Biochem Cell Biol 2005, 37:1974-1984.
57. Attaix D, Ventadour S, Codran A, Bechet D, Taillandier D, Combaret
L: The ubiquitin-proteasome system and skeletal muscle
wasting. Essays Biochem 2005, 41:173-186.
58. Taillandier D, Combaret L, Pouch MN, Samuels SE, Bechet D, Attaix
D: The role of ubiquitin-proteasome-dependent proteolysis
in the remodelling of skeletal muscle. Proc Nutr Soc 2004,
63:357-361.
59. Ordway GA, Neufer PD, Chin ER, DeMartino GN: Chronic con-
tractile activity upregulates the proteasome system in rabbit
skeletal muscle. j AppI Physiol 2000, 88:1134-1141.
60. DeMali KA, Barlow CA, Burridge K: Recruitment of the Arp2/3
complex to vinculin: coupling membrane protrusion to
matrix adhesion. j Cell Biol 2002, I 59:881-891.
61. Blanchoin L, Pollard TD, Hitchcock-DeGregori SE: Inhibition of the
Arp2/3 complex-nucleated actin polymerization and branch
formation by tropomyosin. Curr Biol 2001, 11:1300-1304.
62. Chopard A, Pons F, Marini JF: Vinculin and meta-vinculin in fast
and slow rat skeletal muscle before and after hindlimb sus-
pension. Pflugers Arch 2002, 444:627-633.
63. Kerscher L, Ziegenhorn J: Urea. In: Methods of Enzymatic Anal-
ysis, 3rd edition, edited by Bergmeyer HU. Deerfield Beach,
FL. Verlag Chemie 1985, 8:444-453.
64. SAS: SAS User's Guide, (Release 6.03). Cary, NC:SAS Inst. Inc;
1998.
65. DAVID: [http://david.abcc.ncifcrf.gov/].
66. EASE: [http://david.abcc.ncifcrf.gov/ease/easel.htm].


Page 15 of 15
(page number not for citation purposes)


Publish with BioMed Central and every
scientist can read your work free of charge
"BioMed Central will be the most significant development for
disseminating the results of biomedical research in our lifetime."
Sir Paul Nurse, Cancer Research UK
Your research papers will be:
available free of charge to the entire biomedical community
peer reviewed and published immediately upon acceptance
cited in PubMed and archived on PubMed Central
yours you keep the copyright

Submit your manuscript here: BioMedcentral
http://www.biomedcentral.com/info/publishing adv.asp


BMC Genomics 2006, 7:320




University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs