Meat Tenderisation – The Role of Calpains

P. L. Sensky, T. Parr, R. G. Bardsley & P. J. Buttery

Division of Nutritional Biochemistry, School of Biosciences, University of Nottingham, Sutton Bonington
Campus,
Loughborough, LE12 5RD, UK
Introduction In recent years there has been a shift in emphasis in livestock production away from increased
muscle
growth towards improved meat quality. The final eating quality of meat depends on a number of organoleptic
properties
including appearance, colour, fat content, taste, texture and tenderness. Whilst colour and fat content are
important in
influencing meat purchase, consumer studies indicate that it is the degree to which muscle tenderises after
slaughter that
is the most important factor contributing to overall meat quality (Warkup et al, 1995). Despite efforts to
standardise
breeding, husbandry, nutrition, transport, lairage and slaughter regimes, ensuring a consistently tender product
still
remains difficult to control or predict. The problem is international, with beefsteak toughness a major concern in
the
USA and pork toughness difficult to eradicate in the UK. The tenderisation process involves complex changes
in
muscle metabolism in the immediate post slaughter period and is dependent on genetic makeup, protein
complement,
metabolic status and environmental factors such as physiological stress. In the early postmortem period,
glycogen
depletion, lactic acid accumulation, pH decline and rate of entry and exit into rigor can all influence the ultimate
tenderness of the meat some 8 - 20 days later following a period of conditioning (Goll et al, 1995). However,
the main
determinant of ultimate tenderness appears to be the extent of proteolysis of key target proteins within muscle
fibres
(Taylor et al, 1995). Research in all major livestock species has pointed to the calpain proteolytic enzyme family
being
a major factor responsible for key peptide bond cleavage (Koohmaraie, 1996). Whilst opinion is divided as to
which
isoform of calpain is the most important under specified conditions, most workers agree that the major factor is
the level
at slaughter of the specific calpain inhibitor calpastatin. The evidence for this is reviewed here, highlighting
potential
means of regulating the system in order to assure a consistently high quality tender product.
The calpain system The calpain system comprises two ubiquitous - and m- isoforms, active in vitro at low and
high
calcium ion concentration respectively, a growing number of tissue-specific variants, all the products of
different genes,
and a specific endogenous inhibitor, calpastatin (Suzuki et al, 1995; Sorimachi et al, 1997). The system is
highly
sensitive to fluctuating levels of calcium ion, pH and temperature, all of which change rapidly in the immediate
postmortem period. Micro- and m-calpain consist of a large 80 kDa subunit and a small 30 kDa subunit, both of
which
can be readily truncated at their N termini, thereby modifying their membrane-binding properties and
calciumrequirement.
The extent to which this autolytic modification is physiologically significant has been difficult to assess
and reports suggest that separation of the subunits is all that is necessary to render both -and m-large subunits
fully
active at low physiological (<1 M) concentrations of calcium (Suzuki et al, 1995). Of the ‘novel’ calpains,
nCL1 or
p94 is of particular interest, given that it is expressed almost exclusively in skeletal muscle and that the lack of
expression of a functional p94 gene in human populations is responsible for limb girdle muscular dystrophy type
IIA
(Richard et al, 1995). The p94 polypeptide closely resembles the large subunits of - and m- isoforms, albeit
with three
additional inserted sequences, one of which confers instability by exposing the polypeptide to an intramolecular
autolytic cleavage event. This instability has so far prevented the isolation of p94 by conventional
chromatography so
that its enzymological properties remain unestablished. However, it is known that p94 is tightly bound to the
giant
elastic protein titin in a region that corresponds to the sarcomeric N2 line (Labeit and Kolmerer, 1995). This
region is
known to be susceptible to postmortem cleavage. Consequently, p94, either alone or in conjunction with - and
mcalpain,
may have an important function in meat undergoing conditioning.
Calpastatin is a five-domain inhibitory protein of predicted molecular weight 77 kDa (Takano et al, 1988; Lee
et al,
1992). It is expressed in all tissues expressing calpains and in skeletal muscle at a higher level of activity than
calpains
themselves. Of the five domains, the N terminal leader (L) domain does not appear to have any inhibitory
activity, but
may be involved in targetting or intracellular localisation, whilst the other domains (I – IV) are highly
homologous and
are each capable of inhibiting calpain. Although there are at least eight calpain genes (Braun et al, 1999), it is
believed
that there is only a single calpastatin gene. It is predicted that the gene exceeds 100 kb and that considerable
tissuespecific
microheterogeneity is thought to occur as a result of alternative splicing (Takano et al, 1999; Takano et al,
2000). The microheterogeneity of calpastatin in different cells and tissues (Arnold et al, 1995) may determine its
intracellular localisation and its physiological role.
Calpains and postmortem proteolysis The apparent requirement of the calpains for supra-physiological
concentrations of calcium and the natural excess of calpastatin in most muscle types has, in the past, led to the
dismissal
of a role for calpains in postmortem tenderisation. Other proposed candidate enzymes have included the
lysosomal
cathepsins, whose acid pH optima were presumed to suit the low pH conditions that occur postmortem due to
glycolysis
and lactic acid accumulation, and the cytosolic multicatalytic ATP dependent protease or proteasome. However,
neither
of these enzyme groups have been shown to cleave the spectrum of protein substrates known to be degraded in
situ
during the conditioning process, based on SDS PAGE. This contrasts with the calpains, whose substrates are
believed to
include a number of myofibrillar, Z line and costamere proteins, but significantly exclude the major myofibrillar
proteins actin and myosin and the major Z line protein actinin, all of which remain intact during the normal
tenderisation process (Goll et al, 1991, 1992; Koohmaraie, 1994; Taylor et al, 1995). The calpain system is
therefore
likely to be functional postmortem and have a role in determining the proteolytic events associated with meat
tenderisation. However, the time frame within which the most significant proteolysis occurs has proved difficult
to
ascertain. Both μ-calpain and calpastatin decline relatively rapidly postmortem while tenderisation continues for
up to 4
weeks, depending on the species. Nevertheless, there is now considerable evidence linking the calpains to
tenderisation
in beef, lamb and pork. Correlations have shown that the different tenderisation rates between species (beef <
lamb <
240
pork) relate inversely to the ratio of calpastatin:calpain (beef >lamb > pork) (Koohmaraie et al, 1991). Infusion
of CaCl2 in
beef carcasses increases the rate at which the meat tenderises, whilst infusion of the calpain inhibitor, ZnCl 2,
reduces the
rate of tenderisation in both beef and lamb (Geesink et al, 1994; Koohmaraie, 1990).
Although arising from growth rather than meat quality studies, the observation that livestock treated with -
adrenergic
adrenergic agonists developed pronounced hypertrophy of the predominantly fast fibre muscle types has added
to the
evidence for calpain involvement in muscle protein turnover. These effects were accompanied by significant
changes in
the extractable activity of calpain and calpastatin. Decreased -calpain and elevated calpastatin in treated
animals
suggested that myofibrillar degradation could be reduced by a receptor-mediated mechanism (Higgins et al,
1988;
Wang and Beermann, 1988; Bardsley et al, 1992). The -agonist effect was also evident at the level of mRNA
expression, suggesting that the activity measurements were in fact reflecting changes in gene expression and not
merely
some other altered property of the hypertrophying muscle (Parr et al, 1992; Killefer and Koohmaraie, 1994).
Whilst -
agonists were clearly shown to induce muscle hypertrophy, they did so at the expense of producing a tougher
meat
product than in untreated animals (Kretchmar et al, 1990; Wheeler and Koohmaraie, 1992). Bardsley et al.
(1992)
suggested that the agonist effect on calpains could be a reflection of the natural stress hormone response
involving
adrenaline oversecretion, which could happen in livestock during production, or in the transport and lairage
events
leading up to slaughter. Data from porcine LD has shown that elevated plasma adrenaline increases calpastatin
activity
and expression, implying that the link between stress and meat toughness may indeed be mediated via the
calpain
system (Sensky et al, 1996; Parr et al, 2000).
The sequence of proteolytic events leading to conditioning have led to a general consensus that calpain activity
is the
most likely causative agent for specific peptide bond cleavage postmortem. Since both μ- and m-calpain appear
to be
capable of cleaving the same target proteins it is difficult to determine unequivocally which is likely to be most
important during the postmortem conditioning period. In general, the m- isoform persists longer than the less
stable - in
ageing muscle from all species studied, including pig (Sensky et al., 1996). Despite the potential for p94 to play
an
important role in postmortem tenderisation, unequivocal evidence for this is lacking at present. In porcine LD,
p94
expression does not differ between tough and tender animals and shows no correlation with 8 d shear force
values (Parr
et al, 1999b), whereas recent reports in ovine and bovine muscle shows a positive correlation between p94 and
tenderness (Ilian et al., 2001). In general, the most important component of the calpain system with respect to
meat
tenderisation is currently considered to be calpastatin.
Relationship between calpastatin and meat toughness Studies on a random selection of commercially
slaughtered
pigs have shown that a high level of calpastatin in the first few hours after slaughter is associated with an
increased
incidence of toughness (Sensky et al, 1998; Parr et al, 1999a). This is consistent with relationship between
calpastatin in
the muscle of ruminant species 24 h after slaughter and the degree of tenderization achieved after conditioning,
with
differences in calpastatin accounting for 40% of the variation in tenderness (Shackelford et al, 1991;
Koohmaraie et al,
1995). By monitoring calpastatin at these times it should therefore be possible to predict whether or not any
given
carcass will tenderise to an acceptable degree. This provides an opportunity to grade carcasses at a much earlier
time
postmortem than is currently possible. Measurement of
calpastatin using specific antibodies has shown that a
number of bands appear on SDS PAGE, suggesting that
proteolytic processing may occur, a fact that may be of
considerable importance when considering how
calpastatin may be regulated. In porcine LD, the
principle bands run at molecular weights of 172 and
135 kDa (whole muscle) and 135 and 80 kDa (muscle
extracts). Of these bands, the intensity of the extracted
135 kDa isoform correlates most closely with
toughness, identifying over 70% of ultimately tough
carcasses (Figure 1; Parr et al, 1999a). It has also been
shown that short term -agonist treatment in pigs leads
to a differential increase in the 135 kDa immunoreactive
band, suggesting that expression of 135 kDa can be
modified and may be stress sensitive (Parr et al, 1999b).
Furthermore, isoelectric focussing of the 135 kDa
protein reveals multiple bands, which may be minor
splicing or phosphorylated variants, possibly truncated by proteolytic editing. The identification of these
isoforms and the
mRNA species from which they originate may lead to an even closer predictor of toughness/tenderness.
Calpastatin is also known to vary between different breeds within a species in a manner consistent with the
prevalence
of meat toughness associated with each breed. For example, the callipyge sheep, characterised by increased
hypertrophy
of specific muscles, notably in the hind quarters, produces extremely tough meat and has a high level of
calpastatin in
the muscle at slaughter (Koohmaraie et al, 1995). -agonist treatment does not induce further hypertrophy nor
does it
raise calpastatin any higher in these animals, suggesting that the muscle growth rates in the callipyge sheep are
maximal
leaving no room for further increases in response to -agonist treatment (Koohmaraie et al, 1996). The
mechanism by
which the callipyge gene causes muscle hypertrophy is therefore likely to be similar to that by which -agonists
act.
Duroc and Large White pigs provide another instance where calpastatin has been shown to differ between
breeds within
a species. Immunoblotting techniques reveal that the 80 kDa calpastatin band is more abundant in Duroc
genotypes than
0
2
4
6
8
10
12
14
16
tough tender
abs/50mg protein
0
1
2
3
4
5
6
7
8
9
10
0 5 10 15 20 25
Calpastatin immunoreactivity
Shear Force
************
a)
b) c)
Figure 1. a) Immunoblot of tough (*) and tender pork samples taken 2 h
postmortem after probing for calpastatin. The 135 kDa band is indicated ( ).
b) Quantification of 135 kDa calpastatin band intensity in tough and tender
samples. c) Relationship between 135 kDa calpastatin immunoreactivity and 8
d shear force values. Dashed/dotted lines represent mean values.
241
it is in Large White breeds (Sensky et al, 1999), a fact that may be of significance given the reduced incidence
of
toughness seen in Duroc pigs.
Extensive intraspecies studies in cattle of diverse genotype have demonstrated the heritability of the link
between
calpastatin level and beef toughness (Shackelford et al, 1994). Given the potential this has for marker-assisted
breeding
programmes, a number of groups have now identified calpastatin gene polymorphisms and have attempted to
link these
to growth and meat quality (Lonergan et al, 1995). A major problem in this approach is the fact that the size of
the
calpastatin gene inevitably means that most polymorphisms will tend to be intronic. Furthermore, the
measurement of
calpastatin inhibitory activity in muscle extracts is problematic because the protein tends to be unstable
postmortem,
and quantitative measurements will differ with different methods of extraction and assay. Additionally, the
purified
protein exhibits anomalous molecular weights on SDS PAGE, compounded by the fact that calpastatin can exist
in a
number of alternatively-spliced isoforms or proteolytically-processed fragments (Takano et al, 1999). The
capacity of
any one of these potential variants to inhibit one or both calpains under conditions likely to prevail in
postmortem
muscle has not yet been established. Whilst polymorphisms in the calpastatin gene have been detected in cattle
and
sheep, the significance of this with respect to toughness remains to be clarified (Killefer and Koohmaraie, 1994;
Palmer
et al, 2000). A QTL for beef tenderness has been identified (Keele et al, 1999). However, the QTL is located on
chromosome 15, not on chromosome 7, where the calpastatin locus is part of a linkage group. This illustrates
that care
should be exercised when ascribing the control of meat tenderness solely to calpastatin.
Thus there is evidence that variations in calpastatin that relate to meat toughness/tenderness are subject to
environmental and genetic regulation. Understanding the mechanism by which such regulation can control
calpastatin
function is therefore of paramount importance to further improvement of meat quality.
Regulation of calpastatin On the basis of the data linking -adrenergic stimulation to increased calpastatin and
increased toughness, a physiological mechanism for increased calpastatin expression is likely to relate, at least
in part,
to the -receptor/cAMP/protein kinase A axis. Given that this pathway is activated as part of the natural stress
response,
this mechanism would go a long way towards explaining the molecular and biochemical link between stress and
poor
meat quality. Support for this has arisen recently with the publication of a partial calpastatin gene structure in
bovine
(Cong et al, 1998). A potential promoter was identified and shown to contain a cAMP responsive element
(CRE)
indicating sensitivity to -receptor linked signalling pathways
(Figure 2). Transcription from this promoter generated a novel
mRNA transcript that included sequence from three additional N
terminal exons. These transcripts indicated that the Leader (L)
domain of the corresponding protein was longer than previously
believed, containing an extended Leader sequence (XL)
incorporating the peptide encoded by the additional exons. The
XL-sequence was shown to contain four potential
phosphorylation sites based on the amino acid sequence deduced
from bovine calpastatin cDNA, rendering it susceptible to
posttranslational modification. This means of regulating
calpastatin is also likely to be of significance in attempts to
influence postmortem tenderisation, given that phosphorylation
by cAMP-dependent kinase in vitro can alter the specificity of the
inhibitor for the calpain isoforms (Salamino et al, 1994).
It has recently become evident that the region in the calpastatin that encodes the XL sequence of the calpastatin
protein is subject to alternative splicing mechanisms (Takano et al, 2000). Furthermore it is likely that there are
a
number of alternative promoters in the calpastatin gene and that transcription from these could vary between
different
tissue or muscle types (Takano et al, 2000; Bardsley et al., 2001). Another potential promoter has been
identified in the
porcine calpastatin gene that contains several potential cAMP responsive elements (CRE), which would
reinforce the
idea that calpastatin expression is sensitive to adrenergic stimuli (Parr et al, unpublished). The potential
physiological
consequences of calpastatin alternative exon usage include effects on mRNA stability and rate of translation,
posttranslational
modification and the targetting of nascent proteins to subcellular loci where contact between inhibitor and
calpain is restricted. In the postmortem period, the alternatively-spliced isoforms could be localised in different
regions of
muscle fibres or be degraded at different rates.
As well as regulation by -adrenergic pathways there is evidence that calpastatin is responsive to changes in the
IGF/GH
axis. Recent studies have shown that somatotropin-induced growth in pigs actually resulted in a selective
downregulation
of calpastatin mRNA transcripts (Ji et al, 1998). This implies that elevated somatotropin would tend to produce
meat of
low shear force. The mechanism by which this impacts on calpastatin is likely to involve the calcineurin
signalling
pathway and is the subject of continued research.
It is known that under certain conditions, calpain itself (Doumit and Koohmaraie, 1999; Sensky et al., 2001) can
degrade
calpastatin to functional peptides which still retain inhibitory activity. Recent in vitro experiments in ovine
skeletal
muscle extracts have shown that calpain can induce the disappearance of 135 kDa calpastatin and that this
correlates
with loss of inhibitory potency (Doumit and Koohmaraie, 1999). In pigs, analysis of the 135 kDa and 80 kDa
calpastatin
bands in muscle extracts has shown that the intensity of the 80 kDa band is proportional to increased calpain
activity
(Sensky et al., 2001). Whilst this doesn’t alter overall calpastatin activity, the fact that the 80 kDa band intensity
also
correlates negatively to 8 d shear force suggests that cleavage of calpastatin permits calpain mediated
tenderisation of
skeletal muscle to proceed more effectively (Sensky et al., 2001).
XL L 1 2 3 4
CRE
cAMP
PKA 2 receptor
2 agonist
P
CREB
CREB
cytoplasm
AAAA
nucleus
Calpastatin gene (blocks = exons)
Calpastatin mRNA
Figure 2. Hypothetical mechanism of 2-agonist induced changes
in skeletal muscle calpastatin expression via phosphorylation of
the CRE binding protein (CREB).
242
Conclusions Until recently the meat industry did not know if cathepsins, proteasome or calpains were involved
in the
proteolytic phase of the tenderisation process. It had long been suspected that stress played a role in determining
meat
toughness, but the underlying mechanism by which this could be achieved biochemically was unknown. The
close
relationship between elevated calpastatin expression and increased toughness in livestock species has led to
extensive
research on the role this protein has to play in the tenderisation process. The current knowledge of the structure
of the
calpastatin gene and its potential regulation by -agonist/cAMP pathway mechanisms has shown that the
presumed link
between stress and toughness can be explained at a molecular level. Therefore, any efforts that can be
undertaken to
minimise stress arising from production and slaughter procedures can only be of benefit to the livestock
industry. In
addition, by monitoring calpastatin in the early posmortem period it is theoretically possible to predict whether
or not a
carcass will fail to tenderise adequately and this has the potential to be incorporated into meat quality assurance
schemes. It is clear that calpastatin may be regulated in a number of ways and attention is therefore currently
focussed
on determining how calpastatin can be regulated in a more controllable fashion by investigating the pathways
that
impinge on calpastatin expression. Such research could lead to improved nutritional or husbandry regimes, more
accurate testing methods and to the possibility of identifying one or several DNA markers associated with
toughness
that could be incorporated into breeding programmes aimed at reducing the incidence of toughness in British
meat.
References
Arnold et al (1995) Biochemical Society Transactions 23: S454.
Bardsley et al (1992) Biochimie 74: 267-273.
Bardsley et al (2001) Journal of Muscle Research and Cell Motility (in press).
Braun et al (1999) Biochemical and Biophysical Research Communications 260: 671-675.
Cong et al (1998) Journal of Biological Chemistry 273: 660-666.
Doumit & Koohmaraie, M. (1999) Journal of Animal Science 77: 1467-1473.
Geesink et al (1994) Sciences Des Aliments 14: 485-502.
Goll et al (1991) Journal of Biological Chemistry 266: 8501-8510.
Goll et al (1992) Biochimie 74: 225-237.
Goll et al (1995) Proceedings 41st Annual International Congress of Meat Science and Technology, San Antonio, Texas
537-544.
Higgins et al (1988) British Journal of Nutrition 60: 645-652.
Ilian et al (2001) Journal of Animal Science 79: 122-132.
Ji et al (1998) Journal of Animal Science 76: 1389-1395.
Keele et al (1999) Journal of Animal Science 77: 1364-1371.
Killefer & Koohmaraie, M. (1994) Journal of Animal Science 72: 606-614.
Koohmaraie (1990) Journal of Animal Science 68: 1476-1483.
Koohmaraie (1994) Meat Science 36: 93-104.
Koohmaraie (1996) Meat Science 43: S193-S201.
Koohmaraie et al (1991) Journal of Animal Science 69: 617-624.
Koohmaraie et al (1995) In:Expression of tissue proteinases and regulation of protein degradation as related to meat quality
(ed.
Ouali et al) pp 395-411. ECCEAMST, Utrecht, The Netherlands.
Koohmaraie et al (1995) Journal of Animal Science 73: 3596-3607.
Koohmaraie et al (1996) Journal of Animal Science 74: 70-79.
Kretchmar et al (1990) Journal of Animal Science 68: 1760-1772.
Labeit & Kolmerer, B. (1995) Science 270: 293-296.
Lee et al (1992) Journal of Biological Chemistry 267: 8437-8442.
Lonergan et al (1995) Journal of Animal Science 73: 3608-3612.
Palmer et al (2000) Animal Biotechnology 11: 63-67.
Parr et al (1992) European Journal of Biochemistry 208: 333-339.
Parr et al (1999a) Journal of Animal Science 77 (suppl.1): 164.
Parr et al (1999b) Journal of Animal Science 77 (suppl. 1): 164-165.
Parr et al (2000) Archives of Biochemistry and Biophysics 374: 299-305.
Richard et al (1995) Cell 81: 27-40.
Salamino et al (1994) Biochemical and Biophysical Research Communications 199: 1326-1332.
Sensky et al (1996) Journal of Animal Science 74: 380-387.
Sensky et al (1998) Proceedings of the British Society of Animal Science 14.
Sensky et al (1999) Proceedings of the British Society of Animal Science 194.
Sensky et al (2001) Proceedings of the British Society of Animal Science (in press).
Shackelford et al (1991) Journal of Food Science 56: 1130.
Shackelford et al (1994) Journal of Animal Science 72: 857-863.
Sorimachi et al (1997) Biochemical Journal 328: 721-732.
Suzuki et al (1995) Biological Chemistry 376: 523-529.
Takano et al (1988) Biochemistry 27: 1964-1972.
Takano et al (1999) Biochemical and Biophysical Research Communications 260: 339-345.
Takano et al (2000) Journal of Biochemistry 128: 83-92.
Taylor et al (1995) Journal of Animal Science 73: 1351-1367.
Wang & Beermann, D. H. (1988) Journal of Animal Science 66: 2245-2550.
Warkup et al (1995) In:Expression of tissue proteinases and regulation of protein degradation as related to meat quality (ed.
Ouali et
al) pp 225-237. ECCEAMST, Utrecht, The Netherlands.
Wheeler & Koohmaraie, M. (1992) Journal of Animal Science 70: 3035-3043.