Expression of calpastatin isoforms in muscle and functionality of multiple calpastatin promoters

Archives of Biochemistry and Biophysics 427 (2004) 8–15

ABBwww.elsevier.com/locate/yabbi

Expression of calpastatin isoforms in muscle and functionalityof multiple calpastatin promoters

Tim Parr,* Kirsty K. Jewell, Paul L. Sensky, John M. Brameld,Ronald G. Bardsley, and Peter J. Buttery

Division of Nutritional Biochemistry, School of Biosciences, University of Nottingham, Sutton Bonington Campus, Loughborough,

Leicestershire LE12 5RD, UK

Received 13 January 2004, and in revised form 6 April 2004

Available online 6 May 2004

Abstract

Calpastatin is the specific endogenous inhibitor of calpain proteinase that is encoded by a single gene. Transient transfection

assays in both a non-fusing skeletal muscle and non-muscle cell-line demonstrated that the putative porcine calpastatin promoter

regions 50 to exons 1xa, 1xb, and 1u were functional. Both real-time quantitative and semi-quantitative RT-PCR on porcine skeletal

muscle total RNA indicated that steady-state expression of Type I and III mRNAs containing exons 1xa and 1u, respectively, was at

equivalent levels whilst the expression of Type II mRNA containing exon 1xb was significantly less (p < 0:001). Immunoprobing of

Western blotted muscle extracts with an antibody raised against a peptide sequence encoded by exon 1xa indicated that Type I

protein was expressed and that there was significantly more Type I protein in cardiac than skeletal muscle (p < 0:001). The resultssuggest that the expression of the single calpastatin gene was differentially controlled at several levels.

� 2004 Elsevier Inc. All rights reserved.

Keywords: Calpain; Calpastatin; Promoter analysis; Skeletal muscle; Cardiac muscle

Calpains were first characterised as intracellular cal-cium-dependent cysteine proteinases that are present in

all mammalian cells [1,2]. The conventional, or ubiqui-

tous, l- and m-calpain isoforms appear mainly to cat-

alyse the limited proteolysis of cytoskeletal and

membrane proteins and are regulated both by calcium

ion concentration and the specific protein inhibitor

calpastatin. In striated muscle, the calpain/calpastatin

system has been implicated in the regulation of proteinturnover and growth [3], myoblast fusion [4], cardio-

myopathies [5], and in meat texture development [6].

Calpastatin has four homologous C-terminal inhibi-

tory domains (I–IV) downstream of a non-inhibitory

leader domain (L)1 of largely unknown function [7].

Recently, an additional N-terminal peptide sequence

* Corresponding author. Fax: +44-115-951-6122.

E-mail address: [email protected] (T. Parr).1 Abbreviations used: L, leader domain; ECL, enhanced chemilu-

minescence; DMEM, Dulbecco’s modified Eagle’s medium; FCS,

foetal calf serum.

0003-9861/$ - see front matter � 2004 Elsevier Inc. All rights reserved.

doi:10.1016/j.abb.2004.04.001

(XL) was identified in bovine [8]. Different XL and Lleader sequences have subsequently been shown to

generate a series of tissue-specific variants by alternative

splicing of a single gene to yield calpastatins Type I–IV

[9]. The murine calpastatin gene comprises 32 exons

distributed over �60 kb DNA [9], although the first 6

exons of the porcine gene already extend over �60 kb

DNA [10].

A putative bovine calpastatin promoter was isolatedby Cong et al. [8] and shown to be functional in cell

transfection studies and inducible by dibutyryl cAMP.

Subsequent work based primarily on analysis of mRNA

transcripts predicted four potential promoter regions in

the murine gene [9], three of which have been identified

in the partial porcine gene structure [10]. In the present

report, a series of promoter constructs of the porcine

sequences 50 to exons 1xa, 1xb, and 1u, which are be-lieved to be the transcription initiation sites of Type I,

II, and III calpastatin mRNAs [9,10], respectively, were

generated to characterise these putative promoter re-

gions. Analysis of mRNA transcripts originating from

mail to: [email protected]

T. Parr et al. / Archives of Biochemistry and Biophysics 427 (2004) 8–15 9

these three promoters was carried out by ‘‘real-time’’quantitative RT-PCR along with a partial characteri-

sation of calpastatin protein expression through the use

of an isoform-specific antibody.

Materials and methods

Experimental animals

Animal studies were conducted according to the

provisions of the UK Home Office Animals (Scientific

Procedures) act of 1986. All the pigs used were from the

same group of animals that were matched for breed and

age. The animals were slaughtered by electrical stunning

and severance of the carotid arteries. Samples of ven-

tricular and longissimus dorsi muscle were removedfrom the carcass within 5min of slaughter and frozen in

liquid nitrogen for subsequent storage at )70 �C.

Reverse transcriptase-PCR of calpastatin transcripts

Total RNA was prepared from porcine longissimus

dorsi (skeletal muscle) using the RNeasy Midi Kit

(Qiagen). Real-time quantitative RT-PCR was used forthe analysis of Type I, II, and III calpastatin mRNAs

(GenBank Accession Nos. AJ583407, AJ583408, and

AJ583409). Primers and dual-labelled fluorescent oli-

gonucleotide probes (50FAM, 30TAMRA) were pur-

chased from Sigma-Genosys and are listed in Table 1.

All primer and probe combinations, except that for

Type III transcripts, amplified across an exon boundary.

A pool of first strand cDNA was made from 0.5 lg ofskeletal muscle total RNA using random hexamers

(Promega) as primers, essentially as described previously

[10]. For real-time PCR using fluorescent probes, reac-

tion mixtures contained 12.5 ll of 2� Universal Master

Mix (Applied Biosystems), 5 ll cDNA template (1:40

dilution of reverse transcriptase generated cDNA pool

[10]) in a final volume of 25 ll containing primers

Table 1

Primer and probe combinations used for real-time quantitative RT-PCR

Target

gene

Target

region

Amplicon

length

Primer/

Probe

Calpastatin Exon 1xa 91 Forward

Reverse

Probe

Calpastatin Exon 1xb 104 Forward

Reverse

Probe

Calpastatin Exon 1u 60 Forward

Reverse

Probe

Actin Actin 100 Forward

Reverse

(300 nM) and probe (200 nM). Actin expression wasmonitored as an internal standard using real-time

quantitative RT-PCR and SYBR Green Fluorescence

on the cDNA template as described above (Applied

Biosystems). All real-time PCRs were carried out in

triplicate on an ABI PRISM 7700 Sequence Detection

System (Applied Biosystems) using standard default

thermal cycling conditions. A pool of porcine skeletal

muscle cDNA was used to create a standard curve forquantification of the transcripts using a relative stan-

dard curve method as described by Applied Biosystems

[11]. From this the Ct value of a particular variant could

be converted to ng total RNA equivalent used for first

strand synthesis. Semi-quantitative end-point RT-PCR

analysis was carried out as described previously [10] on

calpastatin transcript variants with novel 50termini and

alternative usage of exon 3.

Western blot analysis of muscle extracts with anti-

calpastatin antisera

Porcine skeletal and cardiac muscle samples were

pulverised in liquid nitrogen then homogenised as de-

scribed by Geesink et al. [12]. Homogenates were spun

at 13,000g for 15min at 4 �C after which the supernatantwas heated in a boiling water bath for 5min and then

recentrifuged at 13,000g for 15min at 4 �C. To concen-

trate the heated extract, the solution was dialysed

overnight against distilled water. Ice-cold trichloroacetic

acid was added to 13.3% (w/v) and the precipitate was

centrifuged at 13,000g for 15min at 4 �C. The pellet waswashed five times with ice-cold acetone then dried.

Muscle extracts and concentrated pellets were analy-sed by 8% SDS–PAGE and Western blotted [10].

The Western blots were probed with a polyclonal anti-

serum (1:1000) raised against a peptide sequence

(MSQPGQKPAASPRP) which was encoded by the

open reading frame in 1xa exon of porcine calpastatin

[10]. Bands were visualised using CDP-Star Western-

blotting reagents (Amersham Biotech). Western blots

Primer

name

Nucleotide sequence (50–30)

1xaF CCGGCGAGCTGCTGC

1xabR TGATCCTTTCTGTTCTCCTGACTTG

1xaP CCCGCCACACCCAGGAGCA

1xbF GCAAATTGGTTGTCGCCAT

1xabR TGATCCTTTCTGTTCTCCTGACTTG

1xbP CGTCTTGTACCACCAGCTTGCGAATG

1uF CCGAGCCCAACCAGGAAT

1uR CTTCCCGGCCGAGAGACT

1uP CAAACATCCCCAAACGCCGCTG

actF GTGGCCCTGGACTTCGAG

actR TTGCCGATGGTGATGACCT

10 T. Parr et al. / Archives of Biochemistry and Biophysics 427 (2004) 8–15

were reprobed with a generic anti-porcine calpastatinantiserum [10,13] at 1:1000 dilution and visualised using

enhanced chemiluminescence (ECL) reagents (Amer-

sham Biotech) as described previously [10].

Generation of promoter constructs and transfections

A porcine BAC clone PigE BAC 045d24 (synthesised

by S. Anderson and A. Archibald, Roslin Institute,Edinburgh, UK) containing a 50-terminal region of the

porcine calpastatin gene [10] (GenBank Accession Nos.

AJ583410 and AJ583411) was used to generate the pu-

tative promoter constructs by PCR using the primers

listed in Table 2 and Expand High Fidelity PCR System

(Roche). Amplicons were ligated into pGL3 Basic vector

(Promega), sequenced to confirm identity with the ori-

ginal BAC clone, and then used in transfections.Transfections were carried out using L6G8 myoblasts

(ECACC No. 92102117), a muscle cell-line originally

derived from rat skeletal muscle, and HEK293 cells, a

cell-line originally derived from human embryonic kid-

ney (kindly donated by Dr. K. Mellits, University of

Nottingham). Frozen stocks of cells were thawed and

cultured in 75 cm2 flasks in Dulbecco’s modified Eagle’s

medium (DMEM) supplemented with 8% foetal calfserum (FCS), 4mM glutamine, 100U penicillin/ml and

100 lg streptomycin/ml (growth medium). Cells were

harvested (trypsin/EDTA), resuspended in growth me-

dium, plated onto 48-well culture plates (Falcon), and

transfected when 50–70% confluent. Four hours prior to

transfection, the medium was replaced with either 200 llprewarmed growth medium (L6G8) or with 200 ll se-rum-free DMEM containing 0.5% BSA and antibiotics(HEK293). For each construct, an aliquot of serum-free

DMEM (no BSA or antibiotics added) was thoroughly

mixed with the required quantity of GeneJuice Trans-

fection Reagent (Novagen) at a GeneJuice:DNA ratio of

6:1 (ll/lg). After 5min incubation at room temperature,

Table 2

Primer combinations used to generate promoter constructs

Primer Nucleotide sequence (50–30)

ApromB AAAACGCGTGTAATTGGGTTCATGGCTTGG

BpromB AAAACGCGTTCGTAATGTATAATAGTGCCG

CpromB AAAACGCGTGCGTTGTTCAGTCAGCCAGTC

EpromB AAAACGCGTACCGCCCCGCCGTAACTCTG

AncaspromB AAAAGATCTTCTGCTTCTGGCTTCCCGGCC

Prom1xaa AAAACGCGTGGGGAGAGGGGGCAACACTG

Prom1xab AAAACGCGTGTATAGGGGGTTAAGTTTGGG

Prom1xac AAAACGCGTATTCCCTTCTGCCAAGTTCCAC

Anc1xa AAAAGATCTCGGGAGGGCACAGCGGAGAC

Prom1xba AAAACGCGTCGTGGCTGCCTGGGGAAATA

Prom1xbb AAAACGCGTGCACTCCTGCCAGAGCCAGAG

Anc1xb AAAAGATCTGGCGACAACCAATTTGCTCTG

Forward primers have MluI sites (ACG CGT) at 50 end. Reverse (Anc)

0.02 lg pRL-SV40 DNA (transfection efficiency control)was added to each tube followed immediately by 0.2 lg ofeither promoter reporter construct DNA, pGL3-SV40

Control (positive control) or pGL3 Basic DNA (negative

control). After 15min incubation at room temperature,

8 ll GeneJuice/DNA mixture was added to cells. Four

hours after transfection, themedium on theHEK293 cells

was replaced with 500 ll DMEM including 8% FCS and

the incubationwas allowed toproceed. Firefly andRenillaluciferase activities were determined 24 h (L6G8) and 48 h

(HEK293) after transfection using Dual-Luciferase Re-

porter Assay System (Promega). The transfection effi-

ciency was corrected for by the ratio of firefly luciferase

activity to Renilla luciferase activity and the level of

promoter activity was calculated relative to the pGL3-

SV40 control DNA which was taken as being 100%.

Data analysis

Band intensities were quantified using the FluorS-

Max Imaging system and Quantity-One software (Bio-

Rad). Real-time PCR data were analysed using SDS

software (Applied Biosystems). Data were compared

using a one-way ANOVA using Genstat for Windows (v

6, IACR-Rothamsted).

Results

Expression of variant calpastatin transcripts in skeletal

muscle

Skeletal muscle calpastatin mRNA variants with al-ternative 50termini containing 1xa, 1xb, and 1u exons

were monitored using real-time quantitative RT-PCR.

Although the absolute level of expression could not be

determined, there was no significant difference between

the relative expression of 1xa (Type I) and 1u (Type III)

Forward/reverse

primer to exon

Construct synthesised

by forward primer

GAA Forward 1u Aprom1u

TAG Forward 1u Bprom1u

Forward 1u Cprom1u

Forward 1u Eprom1u

Reverse 1u —

G Forward 1xa Aprom1xa

AATG Forward 1xa Bprom1xa

AT Forward 1xa Cprom1xa

T Reverse 1xa —

Forward 1xb Aprom1xb

Forward 1xb Bprom1xb

Reverse 1xb —

primers have BglII sites (AGA TCT) at 50termini.

Fig. 1. (A) Schematic representation of the predicted products generated from porcine calpastatin mRNAs with alternative 50 sequences (templates

indicated by box) using RT-PCR. The position of the 50 end exon unique to each calpastatin variant and exon3 is indicated in the template. The real-

time RT-PCR primers are indicated by the arrows below the template and the circle indicates the position of the probe. The semi-quantitative RT-

PCR primers are indicated by the arrows above the template. The predicted size of products generated by semi-quantitative RT-PCR with or without

exon 3 is listed. (B) Real-time RT-PCR analysis of Type I, II, and III calpastatin mRNA expression in porcine skeletal muscle (n ¼ 9). Values are

expressed as the ratio of ng total RNA equivalent used for first strand synthesis calpastatin mRNA to actin mRNA (internal standard). Data are

expressed as means+SEM. ***p < 0:001 Type II compared with Type I and Type III (C) Semi-quantitative RT-PCR products amplified from

skeletal muscle total RNA samples analysed on a 1.5% agarose gel. The products amplified from Type I, II, and III transcripts are indicated above

each lane for individual samples and marker sizes are shown.


containing transcripts (Fig. 1B). The detection of cal-

pastatin transcripts containing exon 1xb (Type II) was

at the limits of sensitivity and was significantly lower

(p < 0:001) than both Type I and Type III (Fig. 1B). The

position of the probes and primers used in the real-time

PCR (Fig. 1A) gave no information on whether these

variants contained exon 3 which encodes for a regionwithin L-domain that is alternatively spliced [12]. Semi-

quantitative end-point RT-PCR analysis was carried out

on porcine skeletal muscle total RNA using a series of

variant-specific primers which spanned exon 3 (Fig. 1C).

Whilst both Type I and III transcripts were detected

there was no Type II. Based on predicted amplicon size,

only Type I transcripts contained both exon 3 plus and

exon 3 minus variants whilst the Type III transcript wasexon 3 minus. There was no significant difference in the

total band intensity for Type I and Type III mRNAs as

well as between the exon 3 plus and minus bands of

Type I transcripts (p > 0:2, n ¼ 5, data not shown).

Type I calpastatin in skeletal muscle

Immunoblots probed with the generic anti-calpasta-tin antibody detected a 135 kDa immunoreactive band

in skeletal muscle and a doublet of 145 and 135 kDa in

cardiac muscle as previously reported [12,14] (Fig. 2A).

A replicate blot probed with the anti-1xa peptide anti-

body detected a 145 kDa band only in cardiac tissue

(Fig. 2A). By concentrating the skeletal muscle extracts

by trichloroacetic acid precipitation an immunopositive

band could be detected at approximately 145 kDa, along

with a minor band at approximately 65 kDa (Fig. 2B).The 65 kDa band could either be a non-specific cross-

reaction of the antibody or a proteolytic fragment of

calpastatin, which is very susceptible to proteolysis

[1,12]. Any proteolytic cleavage must be taking place

towards the C terminus to produce this fragment with

the N-terminal epitope intact, allowing the anti-1xa

peptide antibody to detect this large peptide fragment.

The use of this anti-1xa peptide antibody has indicatedfor the first time that the peptide sequence encoded by

exon 1xa can be expressed in cardiac and skeletal mus-

cle. It is noteworthy that the 1xa-specific antibody does

not detect the band of 135 kDa in either cardiac or

skeletal muscle, unlike the generic anti-calpastatin an-

tibody; accordingly the 135 kDa band is therefore un-

likely to represent a Type I protein lacking exon 3.

Comparisons between bands on immunoblots of pre-cipitated skeletal muscle extracts and cardiac extracts

indicated that the 1xa-peptide immunopositive band was

Fig. 2. (A) Replicate Western blots of porcine cardiac (hrt) and skeletal

(sk) muscle extracts (equivalent to 3.75mg wet weight of tissue ex-

tracted per lane) probed with generic anti-calpastatin antibody (anti-

calstn) or anti-exon 1xa peptide antibody (anti-1xa) as indicated. (B) A

Western blot of porcine sk and hrt extracts (sk and hrt equivalent to

420 and 2.5mg and wet weight of tissue extracted per lane, respec-

tively). The Western blot was probed with anti-exon 1xa peptide an-

tibody (anti-1xa) then reprobed with generic anti-calpastatin antibody

(anti-calstn) as indicated. The positions of molecular weight markers

and major immunopositive bands are indicated.


significantly higher in cardiac muscle per unit wet weight

of tissue extracted (p < 0:001, n ¼ 5, data not shown).

Transfections

In the rat L6G8 myoblasts, transcription/translationfrom three of the four 1u constructs approached that of

the pGL3 SV40 control, indicating very efficient pro-

moter activity (Fig. 3B). The most striking and highly

significant feature was that truncation of construct

Cprom 1u to Eprom1u (removing nucleotides )203 to

)53 and encroaching on the 50 region of a predicted

CpG island [10]) led to almost complete loss of func-

tionality. These data suggest the importance of CpGislands in promoter usage [15] and/or the existence of a

key cis acting enhancer in the )203 to )53 region. In the

L6G8 myoblasts, transcription from the 1xa and 1xb

promoters was generally less strong than from 1u, but

still indicated full functionality (Fig. 4B). In the 1xa

construct loss of nucleotides )1010 to )670 and )669 to

)269 led to a significantly decreased and increased effi-

ciency, respectively, whereas 50 truncation of 1xb did not

have a significant effect. A similar pattern of expressionis seen in the HEK 293 cells indicating that all three

promoters are functional, and generally the constructs

appeared to be more efficient in the HEK293 cells than

in L6G8 myoblasts Figs. 3C and 4C. As with the L6G8

myoblasts, 1xa and 1xb promoters were less effective

than 1u, but still indicated full functionality. Unlike the

myoblasts there was no significant effect of the loss of

nucleotides )1010 to )670, although the deletion ofnucleotides )669 to )269 did lead to significantly in-

creased efficiency. 50 truncation of 1xb had no significant

effect. A possible explanation for the differential re-

sponse of the cell types would be the presence or absence

of muscle-specific suppressors in the L6G8 myoblasts

and HEK293 cells, respectively.

Discussion

Recent studies have indicated that the 50 end of the

calpastatin gene has several putative promoters up-

stream of exons 1xa, 1xb, and 1u that are believed to

initiate transcription of three mRNAs encoding for

Type I, II, and III calpastatin, respectively [9,10].

From the real-time RT-PCR data it appears that inporcine skeletal muscle Type I and III calpastatin

mRNAs, which contain 1xa and 1u sequences, re-

spectively, are expressed in similar quantities, whilst

Type II transcripts containing 1xb sequence are sig-

nificantly less. This was confirmed by RT-PCR using

an alternative set of primers that detected exon 3 al-

ternative splicing. These data indicated that Type I

transcripts had approximately equal proportions ofmRNAs with or without exon 3. Previously we have

shown that in cardiac tissue there are similar levels of

Type I and III mRNAs expressed [10] and the generic

anti-calpastatin antibody detects a doublet of 145 and

135 kDa on immunoblots of cardiac muscle extracts

[14]. The anti-1xa peptide antibody detected a 145 kDa

protein only in cardiac muscle suggesting that the

nucleotide sequence encoding XL region, includingexon 1xa, from Type I mRNA is translated into

peptide sequence, there being a potential translation

initiation site in exon 1xa [10]. Although there was no

apparent difference in the overall level of expression of

Type I and III mRNAs in skeletal muscle, the generic

antibody detected one band only at 135 kDa in stan-

dard extracts. Only after concentration of the skeletal

muscle extract by trichloroacetic acid precipitation didthe anti-1xa peptide antibody detect a band at

145 kDa. This suggests that in skeletal muscle the full-

length protein encoded by Type I mRNA is not ex-

pressed as efficiently as Type III and the 135 kDa

band is the translation product of Type III mRNA

presumably originating from the AUG codon in exon

2. Alternatively, the translation of Type I mRNA

Fig. 3. (A) Diagrammatic representation of the partial 50porcine gene structure indicating the exon positions and the deletion mutants of porcine 1u

promoter–luciferase constructs. (B,C) Transfection efficiency of different 1u promoter–luciferase constructs in L6G8 myoblasts (B) and HEK 293

cells (C). Values are expressed relative to transfection of pGL3 control DNA and corrected for transfection efficiency relative to transfection of cells

with Renilla (pRL)+SEM. ***p < 0:001 Eprom1u compared with Aprom1u, Bprom1u, and Cprom1u.


could be from the downstream AUG codon in exon 2

to give a 135 kDa protein. A further possibility is thatthe peptide originating from the upstream AUG co-

don in exon 1xa is subsequently cleaved to a truncated

protein in a similar way to that reported for bovine

Type II mRNA [8].

From the analysis of transcript expression, both

previous studies and the present data have suggested

that the 1u and 1xa promoters are used predominantly

more than the 1xb promoter in skeletal and cardiacmuscle [10]. Apart from the bovine promoter (1xb)

study [8], the functionality of these alternative promot-

ers has not yet been described in direct cell transfection

studies. In the present study, results indicated that all

but one of the constructs tested were functional and

remarkably efficient compared to the pGL3 SV40 con-

trol, especially in HEK 293 cells (Figs. 3 and 4). The one

exception was construct Eprom1u, in which only 51 ntof 50 sequence upstream of the transcriptional start site

in exon 1u was retained. The 1u exon encodes 50 un-translated sequence only, becoming spliced to exon 2

which contains an in-frame AUG encoding the N ter-

minus of Type III calpastatin [9,10]. The inclusion of an

additional 151 50 nucleotides in Cprom1u restored full

activity, suggesting the presence of key regulatory fea-ture(s) in this region. The additional 151 nt contains

high-scoring consensus sequences for one CRE and one

MyoD motif. The CRE is unlikely to be essential, since

there is a similar CRE in the Eprom1u construct that

appears almost completely inactive. Similarly the MyoD

motif may not be essential, because the same pattern of

responses was seen in the non-muscle HEK293 cells.

However, truncation of Cprom1u to Eprom1u does re-move 151 nt of a high-scoring CpG island, which is

therefore proposed to be an important point of control

for the 1u promoter.

The 1xa and 1xb promoter constructs are approxi-

mately 50% less effective than 1u in both cell types. For

the 1xa promoter, the shortest construct Cprom1xa

containing 250 nt of 50 sequence is most effective, with

the suggestion that negative enhancer sequences couldbe located further upstream. The serial 50 deletions of

the 1xb promoter carried out in this study did not have a

significant effect on expression.

It is concluded that all three promoter sequences

studied in the present report are potentially functional

Fig. 4. (A) Diagrammatic representation of the partial 50porcine gene structure indicating the exon positions and the deletion mutants of porcine 1xa

and 1xb promoter–luciferase constructs. (B,C) Transfection efficiency of different 1xa and 1xb promoter–luciferase constructs in L6G8 myoblasts (B)

and HEK 293 cells (C). Values are expressed relative to transfection of pGL3 control DNA and corrected for transfection efficiency relative to

transfection of cells with Renilla (pRL)+SEM. Deletions which result in significant differences in activity are indicated (*p < 0:05).


with high efficiency and that efficiency can be varied by

the inclusion or removal of nucleotide sequences. It is

important to point out that the rat L6G8 myoblastsused in this study are a non-fusing skeletal muscle cell-

line so that direct comparison between the promoter

analysis carried out in these cells and promoter function

in fully differentiated muscle cells in situ cannot be

made. However, within these sequences are a number of

high-scoring cis acting transcriptional motifs whose

importance will need to be determined by site-directed

mutagenesis in combination with a range of physiolog-ical treatments. In skeletal muscle, the level of tran-

scripts originating from these promoters shows a low

level of Type II calpastatin mRNA, indicating that ei-

ther the 1xb promoter is less active or that Type II

mRNA is rapidly degraded. Although the levels of Type

I and III mRNAs are equal in skeletal muscle, the ex-

pression of Type I calpastatin protein (145 kDa) is less

than Type III (135 kDa). These observations suggest adifferential level of control of the calpastatin gene

products at several levels that may have significance for

calpastatin’s role in inhibiting calpain-mediated prote-

olysis both physiologically and in meat texture devel-opment.

Acknowledgments

We thank Tracey Simpson for her excellent technical

assistance. This work was funded by the Biotechnologyand Biological Sciences Research Council (BBSRC),

UK.

References

[1] D.E. Goll, V.F. Thompson, H. Li, W. Wei, J. Cong, Physiol. Rev.

83 (2003) 731–801.

[2] H. Sorimachi, K. Suzuki, J. Biochem. 129 (2001) 653–664.

[3] D.E. Goll, V.F. Thompson, R.G. Taylor, A. Ouali, Can. J. Anim.

Sci. 78 (1998) 503–512.


[4] S. Barnoy, L. Supino-Rosin, N.S. Kosower, Biochem. J. 351

(2000) 413–420.

[5] Y. Sorimachi, K. Harada, T.C. Saido, T. Ono, S. Kawashima, K.

Yoshida, J. Biochem. 122 (1997) 743–748.

[6] P.L. Sensky, T. Parr, R.G. Bardsley, P.J. Buttery, Proc. Br. Soc.

Anim. Sci. 2001 (2001) 239–243.

[7] E. Takano, M. Maki, H. Mori, M. Hatanaka, M. Marti, K.

Titani, R. Kannagi, T. Ooi, T. Murachi, Biochemistry 27 (1988)

1964–1972.

[8] M. Cong, V.F. Thompson, D.E. Goll, P.B. Antin, J. Biol. Chem.

273 (1998) 660–666.

[9] J. Takano, M. Watanabe, K. Hitomi, M. Maki, J. Biochem. 128

(2000) 83–92.

[10] T. Parr, P.L. Sensky, R.G. Bardsley, P.J. Buttery, Arch. Biochem.

Biophys. 395 (2001) 1–13.

[11] User Bulletin #2:ABI PRISM 7700 Sequence Detection System,

Applied Biosystems, Foster City, CA, USA, 1997.

[12] G.H. Geesink, D. Nonneman, M. Koohmaraie, Arch. Biochem.

Biophys. 356 (1998) 19–24.

[13] P.L. Sensky, T. Parr, A.K. Lockley, R.G. Bardsley, P.J.

Buttery, J.D. Wood, C. Warkup, J. Anim. Sci. 77 (1999) 2956–

2964.

[14] T. Parr, P.L. Sensky, M.K. Arnold, R.G. Bardsley, P.J. Buttery,

Arch. Biochem. Biophys. 374 (2000) 299–305.

[15] H. McQueen, V.H. Clark, A.P. Bird, M. Yerle, A.L. Archibald,

Genome Res. 7 (1997) 924–931.

Expression of calpastatin isoforms in muscle and functionality of multiple calpastatin promoters

Documents

Transcript of Expression of calpastatin isoforms in muscle and functionality of multiple calpastatin promoters