The role of Ca2+-dependent proteases (calpains) in post mortem proteolysis and meat tenderness

229 Fig 1. Electron micrographs showing the effects of ~t- or m-calpain on bovine skeletal muscle. Glyerinated bovine skeletal muscle strips were rinsed with phosphate-buffered saline, and one strip each was then immersed in one of three different solutions: 1) 50 llg ~t-calpain/ml, 100 mM KCi, 50 mM Tris-HCl, pH 7.5, 1.0 mM CaCI2, 0.1% MCE; 2) 50 lig m-calpain/ml in the same solution described for I.tcalpain; and 3) a control solution similar to that described for li-calpain but containing 1.0 mM EDTA instead of ! m M CaCI: and no calpain. After incubation for 1.5 h at 25°C, the strips were transferred to a second set of incubation tubes containing the same solution but having fresh calpain (the calpains autolyze and lose their activity after approximately 1.5 h at 25°C). The strips were incubated in the second solution for 1.5 h at 25°C and were then removed, washed with phosphate-buffered saline, and processed for electron microscopy. A. Micrograph of a strip incubated in a control solution containing EDTA and no calpain; Z-disks are prominent and faint N, lines (arrow) are evident. B. Micrograph of a strip incubated with m-calpain and Ca2+; the N_, lines are gone and gaps or breaks in the continuity of the Z-disk (arrow) can be seen; this section was taken from the interior of the strip, and the calpain had just reached this area after 3 h of incubation. C. Micrograph of a strip incubated with g-calpain and Ca"+; density of the Z-disks is gone but filaments still remain in the Z-disk area; this section was taken from a distance intermediate between the surface and the interior of the strip. D. Micrograph of a strip incubated with ~t-calpain and Ca'-+; the Z-disks are completely gone in this section that was taken from the surface of the strip and only an open gap remains where the Zdisks once were. The bar in A represents 1 lam; the bars in B, C, and D represent 0.25 ~tm. soluble in 2.5% trichloroacetic acid. These results emphasize that, if the calpains are involved in muscle protein turnover, their action is directed exclusively at the myofibrillar or cytoskeletal proteins where their effects result in disassembly of the myofibril and in release of large polypeptide fragments and not in liberation of free amino acids.

Possible mechanism for initiation of myofibrillar protein turnover The arguments summarized in the two preceding sections suggest that myofibrillar protein turnover is initiated by disassembly of intact myofibrils to filaments or polypeptide fragments that are subsequently degraded to free amino acids. The schematic diagram in figure 3 summarizes how myofibrillar protein turnover may occur. A portion of a longitudinal section of a myofibril containing four thick filaments and five pairs of thin filaments per sarcomere is shown at the top of the diagram. Myofibrils in mammalian skeletal muscle are cylindrical structures between 1 and 3 ~tm in diameter and extending from one end of the muscle

cell to the other (-- 1--40 mm). Consequently, a myofibril is much larger than lysosomes or ihe multicatalytic protease (macropain) complex, which is approximately 11 nm high and 16 nm in diameter [28]. If the outer layer of filaments in a myofibril were released, the remaining myofibril would have a smaller diameter (two thick filaments and four pairs of thin filaments in fig 3) but would remain functionally intact (although its strength would eventually be diminished). The released filaments could either reassociate with its parent or another myofibril (arrows in both directions) or could be degraded to free amino acids by one or more cytosolic proteases or by lysosomal cathepsins. The intact, released filaments would still be large compared with muscle lysosomes (mammalian skeletal muscle thin filaments are 1000 nm long and 4 - 6 nm in diameter, and mammalian skeletal muscle thick filaments are 1500 nm long and 1416 nm in diameter), and it seems unlikely that the released filaments could be taken up by lysosomes unless they were at least partly degraded by some cytosolic protease. It has recently been shown that skeletal muscle cells contain the multicatalytic protease or macropain system [29, 30], and this proteolytic system would be a good candidate for degradation of filaments released from myofibrils to amino acids. Release of filaments from the surface of myofibrils as shown in figure 3 would require that the Z-disk, which anchors the thin filaments to the myofibril, and titin and nebulin, which anchor both the thin filaments and the thick filaments to the myofibril [31], be cleaved. Dissociation of thin filaments to actin monomers and of thick filaments to myosin monomers would be facilitated if troponin and tropomyosin (thin filaments) and C-protein (thick filaments) were also degradated (fig 3). Dissociation of the thick and thin filaments into monomers would likely enhance their susceptibility to cytosolic proteases or to being taken up by lysosomes. Consequently, a protease that degrades Z-disks, titin, nebulin, tropomyosin, troponin, and C-protein but that leaves the myosin and actin molecules intact (thereby allowing the possibility for these major myofibrillar proteins to reassemble into functional structures and conserve metabolic energy) would be the ideal enzyme for initiating disassembly of myofibrils.

Evidence indicating that the calpain system initiates turnover of myofibrillar protein A number of lines of evidence (summarized in fig 4) indicate that the calpain system has a role in initiating myofibrillar disassembly. As shown earlier in figures 1 and 2, the calpains selectively remove Z-disks and

230

p.-Colpoin

200kOo-

42 kOo-

II

eilgll

--

~ i

m-Colpoin

I

QIIH

~'~

m...., m i l i p

alI

I

I

qim

25kOo-

Fig 2. SDS-PAGE of rabbit skeletal muscle myofibrils after incubation with g- or m-calpain. Conditions: 2.0 mg of rabbit skeletal muwle myofibrils/ml in 71 mM KCI, 50 mM Tris-HCI, pH 7.5, 0.1% MCE were incubated with 2.0 mM CaCI2 and 0.04 mg ~t- or m-calpain/ml (I calpain: 50 myofibrils, w/w) at 25°C for the times indicated. The digestion was stopped by adding EDTA to a final concentration of 10 mM; the EDTA was added before the Ca2+ for the zero-time samples. After incubation, the myofibrils were centrifuged at 25°C and 1200 g (max) for 10 min to remove those proteins and peptide fragments that had been released by the calpains (t~-actinin released from myofibrils by the calpains rebinds if the myofibrils are incubated or centrifuged at 0-2 °C). The minigels had 10 l.tg protein (the left sides, lanes 1-4 for the I.t-calpain and m-calpain gels) or 2.5 ktg protein (the right sides, lanes 5-8 for the g-calpain and m-calpain gels) loaded on each lane.

Fig 3. Schematic diagram showing how the calpains could initiate turnover of myofibrillar proteins. The angled lines on the thick filaments represent bands of C-protein that occur at regular intervals along the length of the thick filament. The crossbridges and C-protein bands on the thick filaments are not drawn to scale and represent only a few of the total cross-bridges and C-protein bands on a thick filament. The lines on the surface of the thin filaments represent troponin molecules, which are distributed at 38.5-nm intervals along the surface of the thin filament (again, not all the troponin molecules that actually exist on thin filaments are shown). Both It- and m-caipain degrade Z-disks, troponin T, and C-protein, so the thick and thin filaments released by the calpains no longer have C-protein or troponin. After incubation with the calpains, the resulting myofibril is narrower by two thick filaments and two thin filaments but is otherwise unchanged. The thick and thin filaments released by the calpain may either reassociate with the myofibril or be degraded by cytosolic proteases, possibly the multicatalytic protease (macropain).

231 !. Comparison of the rates of hydrolysis of casein and rabbit skeletal muscle sarcoplasmic and myofibrillar protein fractions by m-calpaina

Table

Protein substrate

surface of myofibrils, and the calpains are unique among the known proteases in that they do not cleave actin and myosin. No other proteolytic enzyme has yet been detected that has these unique properties, tha't is present inside muscle cells, and that is active at the physiological conditions of pH and ionic strength that exist inside living striated muscle cells. Consequently, if, as the available evidence indicates, myofibrillar proteins must be disassembled from myofibrils before they can be turned over metabolically, the calpains are the only known proteolytic enzymes that could perform this task. Quantitative immunoelectron localization studies indicate that both It- and m-calpain are located exclusively inside skeletal muscle cells; that they tend to be associated with subcellular structures such as mitochondria or myofibrils rather than free in the cell cytoplasm, and that their relative concentrations are two times greater at the Z-disk - the structure they remove - than elsewhere on tee myofibril [32]. The immunogold miclographs in figure 5 also indicate that the calpains are more prevalent at the Z-disk than else-

0D278 units solubilized mg m-calpain

Casein Myofibrillar protein Sarcoplasmic protein Sarcoplasmic protein + casein

81.94 __+ 4.90 (6) 45.02 + 1.51 (8)

1.33 + 0.44 (8) 73.80 _+ 5.65 (4)

aAssay conditions: 100 mM KCI, 100 mM Tris-acetate, pH 7.5, 10 mM MCE, 1 mM NAN,, 5.0 mg of the indicated protein/ml except for those assays having casein and sareoplasmic protein where 2.5 mg casein/ml and 2.5 mg sarcoplasmic protein/ml were used, 25°C for 30 min. The data were taken from Tan et al [26].

degrade titin, nebulin, C-protein, tropomyosin, and troponin (see also [15]) without cleaving myosin and actin. These cleavages are exactly the cleavages needed for a protease to remove filaments from the

I

'

i~-

I bond I

I

:,-I I

One Sorcomere I I I

Abond

j

~I~" I

!

'

H zone I~I

I

t

I band-I i

1

!

'

t9 t') ,q

_

q~

.~.~

edd=== Ii I I !

I

Z line

j

I

PseudoH zone

IV|

I|ql |~

I I I I

I t I I

Z line

It O D t')_/~'O

~ - ~_~-~ a

,,,

III

Ill

III Udv-vu I. Loss of C-protein, tropomyosin ond troponin favors dissociotion to monomers. 2. Monomers ond filoments (?) ore degroded to omino acids by lysosomol cothepsins ond/or cytoplosmic proteoses, such as fhe multicatolytic proteose (mocropoin).

232 where on the myofibril. Immunogold particles in figure 5 seem more numerous along the edges of the myofibrils (which would be on the surface of the cylindrical myofibrils in living cells) than they are in the interior of the thick and thin filament lattice. Evidently, the Z-disk lattice is large enough to permit penetration of the calpain molecules to the center of the myofibril, but the lattice in the remainder of the myofibril (which contains titin and nebulin filaments in addition to thick and thin filaments) does not permit entry of calpain molecules. Consequently, immunolocalization studies show that the calpains are located in Z-disks and on the ~urface of the remainder of the myofibril and are positionecl perfectly to initiate myofibrillar disassembly as depicted in figure 3. Several studies have shown that striated muscle cells contain approximately 5-10% of their total myofibrillar protein in the |brm of easily releasable myofilaments [34, 35]. These easily releasable myofilaments are removed from the remainder of the myofibril by gentle agitation in the presence of ATP, suggesting that they are loosely associated with the surface of myofibrils. Pulse-labeling studies show that

specific activities of the myosin heavy chain in easily releasable filaments are 3-6 times higher than in the myosin heavy chain in the remainder of the myofibril [35]. These properties indicate that easily releasable flaments are intermediates in the turnover of myofibrillar proteins [34, 35]. Incubation of muscles with leupeptin, a protease inhibitor that inhibits the calpains (and several other proteases), or passive stretch, a treatment that reduces the rate of muscle protein turnover, decreases the amount of the easily releasable filaments [35]. Incubation of muscles with a calcium ionophore, A23187, which would activate the calpains, or fasting the animals or treating them with corticosterone, both of which would increase the rate of myofibriilar protein turnover, increases the size of the easily releasable filament pool. The increase due to fasting could be reduced by treatment of the animals with E-64, a cysteine protease inhibitor that inhibits the calpains. Treatment of isolated myofibrils with a crude calpain preparation releases filaments from the myofibrils that are identical to the easily releasable filaments (they lack ¢t-actinin and highmolecular weight proteins). Studies done 20 years ago

A The available evidence indicates that myofibrils grow and turnover by adding and removing filaments from their surfaces; approximately 5-10% of total myofibriilar protein is in the |brrn of easily released myofilaments that are rapidly labeled and seem to be intermediates in metabolic turnover of myofibrils (Dahlmann et al [34]; van der Westhuyzen et al [35]; and growing myofibrils add radiolabeled amino acids to their surfaces (Morkin et al [361). 1 The unique specificity of calpain proteolysis would result in release of filaments from myofibrils: the proportion of easily released myofibrilr~increases in the presence of excess Ca -'+ and decreases in the presence of compounds that inhibit the calpains. B One of the most consistent structural features of rapidly atrophying muscle is degradation of Z-disks and various Z-disk aherations such as 'streaming'. 1 The calpains are unique among the known proteases in their specific degradation of Z-disks. C SDS-PAGE of myofibrils isolated from rapidly atrophying muscle shows that many of the myofibrillar proteins including myosin and actin are intact in these myofibrils. ! The calpains are unique among the known pmtea~es in that they do not degrade undenatured myosin and actin. D Immunolocalization studies indicate that the calpains are located inside muscle cells on myofibrils with their highest concentration at the Z-disk, the structure they degrade. 1 The calpains are optimally active at intracellular pH and ionic strength (but not intraceilular Ca2+ concentration). E The calpains do not degrade myosin or actin and do not cause bulk degradation of sarcoplasmic proteins. I The calpains do not degrade muscle proteins to amino acids. F Studies that measure muscle protein degradation by the release of free amino acids agree that elevated Ca2~-concentrations increase the rate of unuscle protein degradation (to free amino acids), but do not agree as to whether calpain inhibitors affect this Ca2÷-induced increase in rate of muscle protein degradation. I These discrepancies may arise because the calpains are involved in only myofibrillar protein turnover and in only the first step of myofibrillar protein turnover: other proteases, some of them possibly activated by Ca2÷, are required to degrade the released myofilaments to amino acids,

Fig 4. Summary of the evidence suggesting that the calpain system has a role in muscle protein turnover by initiating myofibrillar disassembly but does not degrade myofibrillar or sarcoplasmic proteins to amino acids.

233

Properties and regulation of the calpain system

Fig 5. Electron micrographs showing immunogold localization of the calpains in bovine skeletal muscle. Samples were prepared foe immunoelectron localization by using the procedures described by Yoshimura et al [33]. Sections were incubated with a monoclonai antibody that reacts with the 80-kDa subunits of both It- and m-calpain followed by incubation with protein A conjugated to lO-nm gold particles. The calpains are located principally at the Z-disks (arrow) and on the fibers themselves. Bar represents 0.5 Itm.

indicated that newly synthesized myofibrillar proteins (containing radiolabeled amino acids) were added to the surface of existing myofibrils in skeletal muscle and did not assemble to form new myofibrils de n o v o [36]. Together, these findings strongly suggest that metabolic turnover of myofibrillar proteins involves at least several different processes and that it is initiated by disassembly of myofilaments from the surface of myofibrils as shown in figure 3. Consequently, the mechanism for metabolic turnover of myofibrillar proteins differs substantially from that for sarcoplasmic or cytosolic proteins, which do not require disassembly before being degraded to amino acids, and regulation of myofibrillar protein turnover is likely to differ substantially from regulation of cytosolic protein turnover. A numbe:- of studies have indicated that myofibri!!ar and non.myofibrillar proteins are degraded by different pathway~ [2, 4]°

Because the calpain system likely has an important role in initiating metabolic turnover of myofibrillar proteins, understanding the properties of the calpain system and how its activity is regulated in living cells is important in understanding the regulation of myofibrillar protein turnover. A great deal of information on the properties of the calpain system has been obtained since m-calpain was first purified in 1976 [13-16]. Some general properties of the calpain system are summarized in figure 6. The cDNAs for I.tcalpain, m-calpain, and calpastatin from several different species have been cloned and sequenced (see [37, 38] for reviews), and a cDNA for a third calpain found only in skeletal muscle tissue has been sequenced [39]. SDS-PAGE of the purified la- and mcalpains had shown earlier that these molecules each contained two polypeptide chains, one of 80 kDa and one of 28 kDa [14, 18, 191. The cDNA-derived sequences demonstrated that the 28-kDa subunit of ~t- and m-calpain was identical and that the 80-kDa subunits of the two calpains were highly related with 50% sequence homology [37]. The cDNA-derived sequences also showed that the C-terminal domain of the 80-kDa subunits of la- and m-calpain had an amino acid sequence homologous to calmodulin with four sequences similar to the E-F hand Ca2+-binding sequences. The amino acid sequence of the C-terminal domain of the 28-kDa subunit common to It- and mcalpain also is homologous to calmodulin and contains four sets of sequences analogous to the E-F hand Ca2÷-binding sequences. Consequently, the amino acid sequences of the la- and m-calpain molecules predict eight potential Ca2+-binding sites on each molecule. The number of Ca2+ atoms actually bound by the calpains has not yet been established although it is probably five or more. Some of the physical properties of the calpain molecules are summarized in table II. Calpain activity in living cells is almost certainly regulated by Ca2+ and by calpastatin, the protein inhibitor specific for the calpains. The nature of this regulation, however, is still unclear. The Ca2÷ concentration required for calpastatin to bind to and inactivate the calpains is approximately the same (~tcalpain) or significantly less (m-caipain) than the Ca2+ concentration required for proteolytic activity of the calpains themselves [41]. Consequently, if the calpains and calpastatin are located in the same places inside skeletal muscle cells, as immunolocalization studies show they are [32], rising Ca 2÷ concentrations would induce calpastatin binding before activating proteolytic activity of the calpains. Recent results in our laboratory indicate that, in the absence of Ca 2÷, the active, catalytic site in the calpain molecules is blocked

234

A It has been detected in all vertebrate cells that have been examined for its presence and in Drosophila but has not been found in plants. B It consists of at least four proteins. 1 It-Calpain - proteinase requiring 5-70 gM Ca2÷ for activity. 2 m-Calpain - proteinase requmng 100--2000 ItM Ca2+ for activity. 3 A third calpain - identified only in muscle cells; sequence homology to It- and m-calpain- a n-calpain or high- Ca2+-cal pain. 4 Calpastatin - an inhibitor specific for the calpains. C All three proteinases require Ca2÷ for activity, and Ca2+ is required for calpastatin to bind to the proteinases. D The physiological function of the system remains unclear; it has been shown to degrade three classes of substrate proteins in

vitro. 1 Cytoskeletal proteins 2 Kinases and phosphatases 3 Hormone receptors a In each instance, calpain degradation seems specific and leaves large peptide fragments, sometimes with altered physiological properties. Fig 6. Summary of some of the general properties of the Ca2+-dependent proteinase (calpain) system. and is unavailable to small (sterically) inhibitors such as E-64 [42]. Addition o f Ca 2+ concentrations high enough to induce proteolytic activity unblocks these active sites and makes them available to E-64 (and presumably also to protein substrates). Hence, Ca 2÷ concentrations high enough to induce proteolytic ac-

tivity evidently cause a significant conformational change in the calpain molecule, making its active site available to substrate. We recently found that Ca2+dependent binding o f calpastatin occurs at two places on the calpain molecule; one on the calmodulin-like domain o f the 28-kDa subunit and one on the cal-

Table !I. Comparison of some properties of the autolyzed and unautolyzed forms of It- and m-calpain from bovine skeletal muscle "~.

Propert3'

Autoly',ed

Autolyzed

It-calpain

It-calpain

m-calpain

m-calpain

Stokes radius b (A)

41.0 + !.0

39.4 + 1.0

41.5 + !.0

39.8 4- 1.0

Molecular dimensions c

18 × 84

21 x 73

18 x 90

20 × 76

% ¢t-helixa

29.9 + 1.5

29.8 + 1.6

27.2 + 1.6

25.8 + 2.6

% 13-sheeta

5.0 + 2.8

2.7 + 1.7

7.6 4. 3.3

11.3 + 5.4

(20% hydration) (A)

[Ca2+l for half-

0,60

7.1

180

1000

7.6

7.5

7.4

7.6

-maximal activity (itM) pH optimum

aData were taken from Edmunds et ai [40]: bobtained from gel permeation chromatography; ccalculated from Stokes' radius and molecular weight obtained from the amino acid composition; aobtained from far ultraviolet circular dichroism spectra.

235

II

INACTIVE

III

Co 2+

ft.,"

II

i ""

III

I

•

I

INACTIVE

? I / c . ,+ I I

It

/ I

NI

ACTIVE

~

Ill

V

INACTIVE

Fig 7. Schematic diagram showing interaction of calpain with Ca2+ and calpastatin. The calpain molecules are shown with an axial ratio of 3.5-3.7 as suggested by hydrodynamic measurements. The active site (the SH group) is shown as being sterically blocked by the N-terminal regions of the 28- and 80-kDa subunits. Roman numerals l-IV represent the four domains of the 80-kDa subunit and Roman numerals V and VI represent the two domains of the 28-kDa subunit. Only a small part of the elongated calpastatin molecule (dashed line) is included because one calpastatin molecule can inhibit four calpain molecules.

modulin-like domain of the 80-kDa subunit of the molecule [43]. Calpastatin is a competitive inhibitor of the calpains [44, 45], so it must also bind to a third site, the active site, on the calpains. Because the Ca2+ concentration required for calpastatin to bind to the calpains is lower than that required for proteolytic activity (ie for the conformational change that opens the active site), calpastatin initially probably binds just to the two calmodulin-like domains at the Ctermini of the two polypeptide chains of the inactive calpain molecule (upper half of fig 7). Rising Ca2+ concentrations would then induce the conformational change that opens the active site, and the calpastatin molecule already bound to the calmodulin-like domains would immediately bind to the active site and inactivate the enzyme (lower right molecule in fig 7). If calpastatin was not located in the same place

in the cell as the calpain molecule, rising Ca2* concentrations could induce the conformational change and open the active site on the calpain molecule creating an active protease before it bound calpastatin (lower left hand molecule in fig 7). If calpastatin relocated to the same place as the calpain, it would then bind to the calpain and inactivate it (the Ca 2÷ concentrations would already be high enough to induce calpastatin binding, so no additional Ca e+ would be required lower half of fig 7). No evidence yet exists, however, to indicate that calpastatin undergoes such relocations in living cells. The Ca2~- concentrations required to induce proteolytic activity (or the conformational change in the caipain molecule) are much higher than the free Ca2* concentrations that normally exist inside living cells (2-20 ~tM free Ca 2÷ or 200-700 ~tM free Ca 2÷

236 requiled for half-maximal activity o f IX- and m - c a l pain, respectively, c o m p a r e d with 0.2-0.8 laM free Ca -'+ inside cells). Consequently, cells must contain some m e c h a n i s m to lower the C a 2+ concentration required for proteolytic activity o f the calpains. T h e nature o f this m e c h a n i s m is unknown, but it too probably requires Ca 2+. This m e c h a n i s m would also nullify the inhibition by calpastatin because the Ca2+ concentration required for calpastatin to bind to the calpains is also higher than the free C a 2÷ concentrations inside cells. Hence, this ' a c t i v a t o r ' , if it exists, would have a crucial role in regulating calpain activity and therefore, in regulating the rate o f calpaininduced myofibrillar protein turnover. Obviously, a great deal remains to be learned about the function o f the calpain system and how its activity is regulated in living cells, and additional studies are needed in this area.

10 11

12 13

14

15

Acknowledgments

We thank Janet Christner for her excellent work in preparing this manuscript in the very short time allowed her. The research reported in this paper that originated from the authors' laboratory was supported by USDA Competitive Grant 87-CRCR-I2283, by the Muscular Dystrophy Association, by the National Livestock and Meat Board, and by the Arizona Agriculture Experiment Station. Project 28, a contributing project to USDA Regional Project NC- 131. References

1 Goldberg AL (1980) The regulation of protein turnover by endocrine and nutritional factors. In: Plasticity of Muscle (Pette D, edl Walter de Gruyter and Co, Berlin, 469--492 2 Sugden PH. Fuller SJ (1991) Regulation of protein turnover in skeletal and cardiac muscle. Biochem J 273, 21-37 3 Tischler ME ( 1981 ) Hormonal regulation of protein degradation in skeletal and cardiac muscle. Life Sci 28, 25692576 4 Kettelhut IC. Wing SS, Goldberg AL (1988) Endocrine regulation of protein breakdown in skeletal muscle. Diabetes/Metabolism Rex, 4, 751-772 5 Furuno K, Goodman MN, Goldberg AL (1990) Role of the different proteolytic systems in the degradation of muscle proteins during denervation atrophy. J Bioi Chem 265, 8550-8557 6 Reeds PJ (1989) Regulation of protein turnover. In: Animal GtY~wth Regulation (Campion DR, Hausman G J, Martin RJ, eds) Plenum Press, New York, 183-210 7 Lowell BB, Ruderman NB, Goodman MN (1986) Evidence that iysosomes are not involved in the degradation of myofibrillar proteins in rat skeletal muscle. Biochem J 234, 237-240 8 Smith DM, Sugden PH (1986) Contrasting response of protein degradation to starvation and insulin as measured by release of N-methylhistidine or phenylalanine from the perfused heart. Biochem J 237, 391-395 9 Odedra BR, Bates PC, Millward DJ (1983) Time course of the effect of catabolic doses of corticosterone on protein

16

17 18

19

20 21 22

23 24

25

turnover in rat skeletal muscle and liver. Biochem J 214, 617-627 Kayali AG, Young VR, Goodman MN (1987) Sensitivity of myofibrillar proteins to glucocorticoid induced muscle proteolysis. Am J Physio1252, E621-626 Reeves JP, Decker RS, Crie JS, Wildenthal K (1981) Intracellular disruption of rat heart lysosomes by leucine methyl ester: effects on protein degradation. Proc Natl Acad Sci USA 78, 4426--4429 Busch WA, Strainer MH, Gall DE, Suzuki A (1972) Ca 2+specific removal of Z-lines from rabbit skeletal muscle. J Cell Bio152, 367-381 Gall DE, Kleese WC, Okitani A, Kumamoto T, Cong J, Kapprell HP (1991) Historical background and current status of the Ca2+-dependent proteinase system. In: lntracellular Calcium-Dependent Proteolysis (Mellgren RL, Murachi T, eds) CRC Press, Boca Raton, Florida, 1-24 Dayton WR, Gall DE, Zeece MG, Robson RM, Reville WJ (! 976) A Ca2+ activated protease possibly involved in myofibrillar protein turnover. Purification from porcine muscle. Biochemistry 15, 2150-2158 Dayton WR, Gall DE, Strainer MH, Reville WJ, Zeece MG, Robson RM (1975) Some properties of a Ca2+-actirated protease that may be involved in myofibrillar protein turnover. In: Cold Spring Harbor Conferences on Cell Proliferation Vol 2, Proteases and Biological Control (Reich E, Rifkin DB, Shaw E, eds) Cold Spring Harbor Laboratory, Cold Spring Harbor, New York, 551-577 Dayton WR, Reville WJ, Gall DE, Strainer MH (1976) A Ca 2+ -activated protease possibly involved in myofibrillar protein turnover. Partial characterization of the purified enzyme. Biochemistry i 5, 2159--2 ! 67 Dayton WR, Scholimeyer JV (1980) Isolation from porcine cardiac muscle of a Ca2+-activated protease that partially degrades myofibrils. J Mol Cell Card 12, 533-55 i Szpacenko A, Kay J, Gall DE. Otsuka Y (1981) A different form of the Cae+-dependent proteinase activated by micromolar levels of Ca2÷. In: Proc Syrup Proteinases and Their Inhibitors: Structure. Function, and Applied Aspects (Turk V, Vitale LJ, eds) Pergamon Press, Oxford, 151-161 Dayton WR, Schollmeyer JV, Lepley RA, Cortes LR (1981) A calcium-activated protease possibly involved in myofibrillar protein turnover. Isolation of a low-calcium requiring form of the protease. Biochim Biophys Acta 659, 48-61 Gall DE, Dayton WR, Singh i, Robson RM (1991) Studies of the a-actinin/actin interaction by using calpain. J Biol Chem 266, 8501-8510 Lazarides E, Burridge K (1975) ot-Actinin: immunofluorescent localization of a muscle structural protein in nonmuscle ceils. Cell 6, 289-298 Endo I", Masaki T (1982) Molecular properties and functions in vitro of chicken smooth-muscle ¢x-actinin in comparison with those of striated muscle ct-actinin, J Biochem 92, 1457-1468 Lane BP, Elias J, Drummond E (1977) lmmunoelectronmicroscopic localization of ~t-actinin in skeletal muscle cells. J Histochem Cytochem 25, 69-72 Olson DG, Parrish FC Jr, Dayton WR, Gall DE (1977) Effect of post mortem storage and calcium activated factor on the myofibrillar proteins of bovine skeletal muscle. J Food Sci 42, 117-124 Wolfe FH, Sathe SK, Gall DE, Kleese WC, Edmunds T, Duperret SM (1989) Chicken skeletal muscle has three Ca2+-dependent proteinases. Biochim Biophys Acta 998, 236-250

237 26 Tan FC, Goll DE, Otsuka Y (1988) Some properties of the millimolar Ca2+-dependent proteinase from bovine cardiac muscle. J Mol Cell Card 20, 983-997 27 Waxman L (1981) Calcium-activated proteases in mammalian tissues. Meth Enzymoi 80, 664-680 28 Tanaka K, Yoshimura T, Ichihara A, Ikai A, Nishigai M, Morimoto Y, Sato M, Tanaka N, Katsube Y, Kameyama K, Toshio T (1988) Molecular organization of a high molecular weight multi-protease complex from rat liver. J Mol Biol 203, 985-996 29 Fagan JM, Waxman L, Goldberg AL (1987) Skeletal muscle and liver contain a soluble ATP+ubiquitin-dependent proteolytic system. Biochem J 243, 335-343 30 Haas A, Riley DA (1988) The dynamics of ubiquitin pools within skeletal muscle, h~: The Ubiquitin System (Schlesinger M, Hershko A, eds) Cold Spring Harbor Laboratory, Cold Spring Harbor, New York, 178-185 31 Wang K, Wright J (1988) Architecture of the sarcomere matrix of skeletal muscle: immunoelectron microscopic evidence that suggests a set of parallel extensible nebulin filaments anchored a! the Z line. J Cell Biol 107, 2199-2212 32 Kumamoto T, Kleese WC, Cong J, Goll DE, Pierce PR, Allen RE (1991) Localization of the Ca2÷-dependent proteinases and their inhibitor in normal, fasting, and denervated rat skeletal muscle. Anat Rec 232, 60-77 33 Yoshimura N, Murachi T, Heath R, Kay J, Jasani B, Newman G (1986) lmmunogold electron-microscopic localization of calpain ! in skeletal muscle of rats. Cell Tissue Res 244, 265-270 34 Dahlmann B, Rutschmann M, Reinauer H (1986) Effect of starvation or treatment with corticosterone on the amount of easily releasable myofilaments in rat skeletal muscles. Biochem J 234, 659-664 35 Van der Westhuyzen DR, Matsumoto K, Etlinger JD (1981) Easily releasable myofilaments from skeletal and cardiac muscles maintained in vitro. Role in myofibrillar assembly and turnover. J Biol Chem 256, 11791-11797

36 Morkin E (197')) Postnatal muscle fiber assembly: localization of newly synthesized myofibrillar proteins. Science 167, 1499-1501 37 Suzuki K (1990) The structure of calpains and the calpain gene. In: lntracellular Calcium-Dependent Proteolysis (Mellgren RL, Murachi T, eds) CRC Press, Boca Raton, Florida, 25-35 38 Maki M, Hatanaka M, Takano E, Murachi T (1990) Structure-function relationships of calpastatins. In: Intracellular Calcium-Dependent Proteolysis (Mellgren RL Murachi T, eds) CRC Press, Boca Raton, Florida, 3754 39 Sorimachi H, Imajoh-Ohmi S, Emori Y, Kawasaki H, Ohno S, Minami Y, Suzuki K (1989) Molecular cloning of a novel mammalian calcium-dependent pro!ease distinct from both m- and It-types. Specific expression of the mRNA in skeletal muscle. J Biol Chem 264, 20106-2011 i 40 Edmunds T, Nagainis PA, Sathe SK, Thompson VE Goll DE (1991) Comparison of the autolyzed and unautolyzed forms of It- and m-calpain from bovine skeletal muscle. Biochim Biophys Acta 1077, 197-208 41 Kapprell HE Goll DE (1989) Effect of Ca2÷ on binding of the calpains to calpastatin. J Biol Chem 264, 17888-17896 42 Thompson VE Goil DE, Kleese WC (1990) Effects of autolysis on the catalytic properties of the calpains. Biol Chem Hoppe-Seyler 371, Suppl 177-185 43 Nishimura T, Goll DE (1991) Binding of calpain fragments to calpastatin. J Biol Chem 266, 11842- I 1850 44 Maki M, Takano E, Osawa T, Ooi T, Murachi T, Hatanaka M (1988) Analysis of structure-function relationships of pig calpastatin by expression of mutated cDNAs in Escherichia coll. J Biol Chem 263, 10254-10261 45 Kawasaki H, Emori Y, Imajoh-Ohmi S, Minami Y, Suzuki K (1989) Identification and characterization of inhibitor sequences in four repeating domains of the endogenous inhibitor for calcium-dependent pro!ease. J Biochem 106, 274-281

Biochimie (1992) 74, 239-245

239

© Soci6t6 franqaise de biochimie et biologic mol6culaire / Elsevier, Paris

The role of Ca'÷-dependent proteases (calpains) in post mortem proteolysis and meat tenderness M Koohmaraie USDA, ARS, Roman L Hruska US Meat Animal Research Center; PO B o x / 6 6 , Clay Centep; NE 68933, USA

(Received 12 April 1991; accepted 18 January 1992)

Summary - - This manuscript summarizes research results from our laboratory regarding the role of endogenous proteases in post mortem proteolysis resulting in meat tenderization. Proteolysis of key myofibrillar proteins is the principal reason for ultrastructural changes in skeletal muscle associated with meat tenderization. Proteases should have the following characteristics to be considered as possible candidates for bringing about post mortem changes: i) to be located within skeletal muscle cells; ii) to have access to the substrate ie, myofibrils); and iii) to be able to hydrolyze the same proteins that are degraded during post mortem storage. Of the proteases located within skeletal muscle cells and thus far characterized, only calpains have all of the above characteristics. Numerous experiments conducted in our laboratory have indicated that the calcium-dependent proteolytic system (calpains) is responsible for post mortem proteolysis. Some of this evidence includes: I) inc~bation of muscle slices with buffer containing Ca 2+ accelerates post mortem proteolysis; 2) incubation of muscle slices with Ca 2+ chelators inhibits post mortem proteolysis; 3) infusion or injection of carcasses with a solution of calcium chloride accelerates post mortem proteolysis and the tenderization process such that post mortem storage beyond 24 h to ensure meat tenderness is no longer necessary; 4) infusion of carcasses with zinc chloride, a potent inhibitor of calpains, blocks post mortem proteolysis and the tenderization process; and 5) feeding a [I-adrenergic agonist to lambs results in a reduction of the pmteolytic capacity of the calpain system, which leads to a decreased rate of post mortem proteolysis and produces tough meat. Based on these results, we have concluded that calpains are the main proteolytic system responsible for post mortem proteolysis, and that one of the main regulators of calpains is their endogenous inhibitor, calpastatin. calpains / calpastatin / proteolysis / post mortem / calcium

Introduction It is now well documented that p o s t m o r t e m storage o f carcasses at refrigerated conditions results in a significant improvement in meat tenderness. The first scientific report regarding p o s t m o r t e m tenderization o f meat was that o f L e h m a n [1], in 1907, who reported that there was a 30% increase in meat tenderness during an eight-day p o s t m o r t e m storage period. Although the improvement in meat tenderness is an accepted phenomenon, the mechanism through which these changes are brought about has remained a controversial issue. Tenderness is probably the most important organoleptic characteristic o f red meat. Therefore, it is important that the mechanism(s) o f p o s t m o r t e m tenderization be identified, so that methodology can be developed to manipulate the process advantageot~sly. D e v e l o p m e n t o f such m e t h o d o l o g y will have at least two major effects on the animal industry as a whole and specifically on the meat industry: 1) p o s t m o r t e m

storage of up to 21 days to ensure meat tenderness may no longer be necessary; and 2) perhaps would eliminate the toughness problems that occur when p o s t m o r t e m storage fails to produce tender meat (such as with meat from B o s indicus breeds o f cattle and from animals fed [3-adrenergic agonists). The purpose o f this manuscript is to review and summarize the results of experiments conducted in our laboratory regarding the mechanism o f p o s t m o r t e m tenderization. Whenever possible, I will compare and discuss our results with those reported by other laboratories. Throughout this manuscript p o s t mortern storage is defined as holding o f carcasses at refrigerated temperatures (2-4°C) and should be distinguished from other methods o f storage, such as storage at higher temperatures. Also, research efforts from this laboratory have been directed towards explaining the observed variation in meat tenderness from slaughter-age animals and should not be extrapolated to explain tenderness variation in meat from younger or older animals.

240

Role of proteolysis in meat tenderization During post mortem storage of carcasses, numerous changes occur in skeletal muscle which result in loss of structural integrity of this tissue. This loss of structural integrity is responsible for meat tenderization (table I).

Table !. Summary of key changes that occur in skeletal muscle during post rnortern storage at 2--4°C (adapted from [2-4]).

Table II. Experimental evidence demonstrating the role of proteolysis in post rnortern meat tenderization. I. Incubation of muscle slices with calcium chloride induces proteolysis of myofibrillar proteins and fragmentation of myofibrils. However, incubation of muscle slices with calcium chelators (EDTA and EGTA) prevents both degradation of myofibrillar proteins and myofibrii fragmentation [7l. 2. Infusion of carcasses with calcium chloride accelerates post mortem changes (degradation of myofibrillar proteins, tenderness) in skeletal muscle such that post rnortern storage to ensure meat tenderness is no longer necessary [8-10]. 3. Infusion of carcasses with zinc chloride inhibits all post mortern changes measured (degradation of myofibrillar proteins, myofibril fragmentation, tenderization) [ 111, 4. Muscle from [i-adrenergic agonist (BAA)-fed lambs, which does not undergo post mortem proteolysis (no detectable degradation of myofibrillar proteins and myofibril fragmentation during post mortem storage), is tougher than muscle from untreated lambs [12-14]. However, calcium chloride infusion of carcasses from BAA-fed lambs induces degradation of myofibrillar proteins and eliminates their meat toughness [ 15].

I. Z-disk weakening and/or degradation which leads to fragmentation of myofibrils. 2. Disappearance of troponin-T and simultaneous appearance of polypeptide with molecular weight of 28-32 kda This is perhaps the most publicized post rnortem change in skeletal muscle. However, because of its location in the myofibrils, the exact relationship between meat tendemess and troponin-T is not yet understood. 3. Degradation of desmin which leads to fragmentation of myofibrils, probably through disruption of transverse cross-linking between myofibrils. 4. Degradation of titin. Effects of titin degradation on meat tenderness are not yet understood. 5. Degradation of nebulin. Effects of nebulin degradation on meat tenderness are not yet understood. 6. Appearance of a 95 000 kDa polypeptide, probably from degradation of myofibrillar proteins with molecular weights of greater than 95 000 kDa. Neither its origin or significance to meat tenderness is known. 7. Perhaps the most significant observation is that the major contractile proteins, myosin and actin, are not affected even after 56 days of post rnortern storage.

Which proteolytic system causes tenderization

Clearly, the majority of the changes that occur in skeletal muscle (table !) which lead to the disruption of the muscle cell and meat tenderization are the result of proteolysis. In fact, as early as 1917, Hoagland et ai [5] reported that proteolysis is an important factor contributing to post m o r t e m changes in skeletal muscle, including meat tenderness. Penny [6] reported that: ~ There is no doubt that proteolytic enzymes are responsible for the changes during conditioning (post mortem storage) >~. To further substantiate the argument that proteolysis is the principal reason for the observed meat tenderization during post mortem storage, the results of several experiments will be summarized (table II). In addition to those mentioned in table II, we have also found that the major reason for the observed differences in meat tenderness between Bos taurus (tender) and Bos indicus (tough) breeds of cattle is the reduced rate of myofibrillar protein degradation during p o s t mortem storage [16, 17]. Also, differences in the rate o f post mortem tenderization and proteolysis in skeletal muscle from pigs, sheep and cattle were apparently due to the differences in the rate of myofibrillar protein degradation. The results clearly indicate that pro-

Skeletal muscle is composed of three classes of proteins: sarcoplasmic, connective tissue and myofibrillar. Although some proteolytic degradation of sareoplasmic proteins may occur during post m o r t e m storage, their degradation probably does not contribute directly to increased tenden~ess (for review see [2]). Also, proteolytic changes in collagen (the principal protein of the connective tissue fraction) during post mortem storage comparable to those o f myofibrillar proteins have not been observed [18]. Therefore, the principal mechanism of post m o r t e m tenderization is limited to the proteolysis of myofibriUar proteins. It is important to bear in mind that throughout this manuscript and for the most part - all experiments conducted in our laboratory - we have studied the cause of tenderizatio,1 in animals of similar age. Therefore, we agree with the conclusion of Tarrant [18] as long as animal age is kept constant. Indeed, we accept the argument that the connective tissue fraction could play a significant role in meat tenderness from animals of different ages. Proteases should have the following characteristics to be considered as possible candidates for bringing about post mortem changes that result in meat tender-

teolysis of key myofibril!ar proteins is the principal reason for the ultrastructural changes in skeletal muscle resulting in the loss of muscle cell integrity ( ie, tenderization).

241 ization: i) they should be located within the skeletal muscle cell (for details see [21); ii) have access to the substrate (ie, myofibrils); and iii) have the ability to hydrolyze the same proteins in an in vitro system that are degraded during p o s t m o r t e m storage. If a proteolytic system had these characteristics, it would be impossible to exclude its potential involvement in p o s t rnortern proteolysis and the tenderization process. There are many proteases in skeletal muscle; however, thus far only calpains and certain lysosomal enzymes have been shown to degrade myofibrillar proteins. For this reason, over the last decade we have focused our attention on these two proteolytic systems to sort out their involvement in post m o r t e m proteolysis. This by no means excludes the possible direct or indirect involvement of other proteolytic systems (such as the multicatalytic proteolytic system [191) in this process. Based on the observations reported in table III (cathepsins degrade myosin efficiently and p o s t mortern storage has no effect on myosin) and because of the location of iysosomal proteases in the skeletal muscle cell, we have excluded their involvement in p o s t rnortern proteolysis. These enzymes are normally located in lysosomes and presumably have to be released to have access to myofibrils. It has been assumed that during p o s t rnortem storage lysosomes are ruptured and thereby cathepsins are released into the cytosol. However, the only experiment conducted to test the accuracy of this assumption has indicated that even after 28 days of p o s t rnortem storage, lysosomal rupture was not evident [20]. Some of the other reasons that have led us to believe that lysosomal proteases are not involved in this process are explained in detail elsewhere [4, 111. Table IlL Effect of post mortem storage, calpains and cathepsins on myofibrils (adapted from [2-4, 21]). Post mortem storage Calpains Cathepsins

Z-disk degradation Titin degradation Nebulin degradation Myosindegradation a-Actinin degradation l)esmin degradation Actin degradation Troponin-T degradation Appearance of 30 K

+ + + -+ + +

+ + + + + +

+ + + + + + + +

In contrast to lysosomal enzymes, considerable experimental evidence supports the hypothesis that calpains are the primary enzyme system responsible for post mortem proteolysis and tenderization. The properties and regulation of the calpain proteolytic system have been discussed in detail [22]. Therefore, I shall focus on experimental evidence regarding the role of this proteolytic system in post mortem proteolysis and tenderization process. There is considerable experimental evidence indicating that elevated calcium ion concentration is responsible for the weakening of myofibrillar structures that results in tenderization. I shall first review this experimental evidence and then attempt to explain the mechanism of action of calcium in this process. The first report linking calcium ions to post mortem tenderization is perhaps that of Davey and Gilbert [23]. They reported that the weakening and disappearance of Z-disks was inhibited by EDTA. They also speculated that EDTA may exert its effect by chelating calcium ions. Their observations were later supported by others [7, 24]. Busch et a l [24] demonstrated that myofibril fragmentation was inhibited by EDTA, but was induced by calcium ions. Koohmaraie et al [7] demonstrated that the disappearance of the Zdisk and myofibril fragmentation were inhibited by EDTA (a general chelator of divalent cations) and EGTA (specific for calcium in the presence of magnesium) and were accelerated in the presence of calcium chloride. The results of these experiments 17, 23, 24] and others had convinced us that elevation of calcium ions is the cause of the post mortem tenderization process. We, therefore, attempted to reproduce these observations in situ. Immediately after slaughter and after electrical stimulation (to exhaust ATP and prevent supercontraction), the carcasses were infused with a solution of calcium chloride via the vascular system. Results indicated that the post mortem tenderization processes were accelerated such that ultimate tenderness values were obtained within 24 h instead of after 7-14 days of post mortem storage [81. We then focused our attention on attempting to elucidate the mechanism of action of calcium ions. There are at least three possible mechanisms through which calcium ions can exert their effect on post mortem tenderization: i) protein solubilization due to a salting-in action by calcium chloride; ii) non-enzymatic weakening of structural proteins involved in stability of Z-disk proteins; and iii) activation of calpains. To determine whether elevated ion concentration was the mechanism of action of infused-calcium chloride (salting-in action). carcasses were infused with calcium chloride or sodium chloride at the same ionic strength 19]. Infusion of carcasses with sodium chloride did not result in acceleration of post rnortem proteolysis or the

242 tenderization process. Hence, it was concluded that the observed effects with calcium chloride infusion of carcasses were due to calcium ions and not due to an elevation of ionic strength. Recently, Taylor and Etherington [25] conducted an experiment to determine the mode of action of calcium. Their results indicated that although some solubilization of myofibrillar proteins occurred in the presence of calcium chloride: 'The removal of these proteins would probably not affect the stability of the Z-disk', and therefore should not affect meat tenderness. The second possible mode of action of elevated calcium concentration, from endogenous (due to their release from mitochondria and the sarcoplasmic reticulum) or exogenous (infusion of carcasses with calcium chloride or incubation of muscle slices in calcium chloride solution) sources is a non-enzymatic one [26-28l. Immediately after slaughter, lamb carcasses were infused with a solution of zinc chloride. Guroff [29] reported that zinc chloride was a potent inhibitor of calpains. Therefore, if post mortem tenderization is brought about by non-enzymatic action of calcium ions, the process should not be affected by infusion of carcasses with zinc chloride infusion. However, results indicated that none of the post mortern changes (proteolysis of myofibrillar proteins, myofibril fragmentation or tenderization) occurred in carcasses infused with zinc chloride [11]. We have, therefore, concluded that the action of calcium chloride (from endogenous or exogenous sources) is mediated through the calpain proteolytic system. In support of these findings, Alarcon-Rojo and Dransfield [30l reported that the calcium chloride acceleration of post mortem tenderization was inhibited in the presence of synthetic inhibitor N-Acetyileu-leu-norleucinal), which is a substrate-like inhibitor of calpains. In addition, this inhibitor did not inhibit cathep.sins B and L [30]. There is no doubt that colcram tons induce other changes in the skeletal muscle [25] and that much remains to be learned regarding its mode of action. However, the present experimental evidence suggests that its effects on tenderization are mediated by the calpain proteolytic system.

Regulation of the calpain proteolytic system in post mortem muscle The process of conversion of muscle to meat is complex and involves metabolic, physical and structural changes. In a typical slaughterhouse, the animals are slaughtered by severing the carotid artery and jugular vein. After bleeding and evisceration, the carcasses are stored at about I°C. Due to cessation of blood flow to the tissue, the oxygen supply (ie, source of energy) is cut off. In addition, the products of anaer-

obic metabolism (glycolysis) cannot be removed and accumulate in the tissue, resulting in buildup of lactic acid, which causes a gradual decline in the pH of the tissue from about 7.0 to about 5.6 over a 24-h period. At the same time, the temperature of the carcass falls from about 37°C to about 2°C over a 12-h (in cattle) period [31-33]. The combination of these complex changes results in the generation of a new environment totally different from that of living tissue. These post mortem conditions may change the capacity of different proteolytic systems dramatically. For example, one of the major changes is the elevation in the free calcium concentration due to its release from mitochondria and sarcoplasmic reticulum. While the concentration of free calcium in the resting muscle is less than 1 I~M, in post mortem muscle it could reach 100 I~M. These three dramatic changes alone (gradual fall in pH and temperature, and elevation of calcium concentration) would likely have a dramatic effect on the endogenous proteolytic systems. For example, serine proteases are almost totally inactive at pH values below 6.0. However, the conditions are more favorable for other proteolytic systems, such as catpains and lysosomal cathepsins. Currently, there is a debate on how calpains could possibly function in muscle tissue (for details see [2, 22]). Much of this debate occurs because the calcium requirements for proteolytic activity of the caipains (approximately 10 I~M for ~t-calpain and approximately 200-300 ~tM for m-calpain) are much higher than the free calcium concentrations found in living tissue (< 1 gM). Moreover, the calcium concentration required for binding of calpastatin to calpains is less than that required for proteolytic activity of the calpains themselves [2, 22]. However, these arguments may not be valid in post mortem muscle, because free calcium concentrations are sufficient to activate ~t-calpain. Also, we have recently found that the drop in pH and temperature has a significant effect on the ability of calpastatin to inhibit ~-calpain (at 25°C: pH 7.5 = 87% inhibition; pH 5.7 = 55% inhibition; at 5°C: pH 7.5 = 59% inhibition; pH 5.7 = 6% inhibition) [34]. Previous studies have indicated that under normal post mortem conditions (ie, slaughter and holding at 2°C for up to 14 days) m-calpain is remarkably stable, whereas there is a gradual decline in the activities of g-calpain, and calpastatin loses its activity rapidly [31, 35, 36]. Both ~t- and m-calpain undergo autolysis in the presence of sufficient calcium with the eventual loss of activity [29, 37-45]. However, this loss of enzymatic activity is highly temperature-dependent [9] and greatly reduced in the presence of substrate [42]. We have therefore, [31] suggested that the reason for loss of ~t-calpain p o s t mortem is its autolysis due to the elevated calcium concentration in p o s t mortem muscle. Although this seems to be a plausible

243 hypothesis (elevated calcium concentration activates la-calpain which in turn hydrolyzes the few myofibrillar proteins that it can utilize as substrate, upon depletion of these proteins gt-calpain will undergo autolysis leading to its inactivation), another explanation could be its hydrolysis by another protease. We believe that both of these hypotheses are likely and that the role of a third protease (ie, the one that hydrolyzes ~t-calpain) could be very significant. We are in the process of testing the accuracy of this hypothesis. The third component of the calpain proteolytic system is their specific endogenous inhibitor~ calpastatin. Results of several experiments reported recently seem to indicate that caipastatin is one of the principal regulators of the calpains in post mortem muscle. Firstly, infusion of carcasses with zinc chloride which prevented the p o s t mortem proteolysis and tenderization protess, completely blocked the inactivation of calpastatin [ 11 ]. Secondly, infusion of carcasses with calcium chloride, which results in acceleration of p o s t m o r t e m proteolysis and tenderization process, also accelerates the process of calpastatin inactivation [8, 9]. Thirdly, the rate of inactivation of calpastatin is highly correlated with the rate of p o s t mortem proteolysis and tenderness in meat from B o s indicus breeds of cattle [16, 17] and in meat from animals fed a ~-adrenergic agonist, L644.969[ 13, 14]; and finally, the differences in the rate of post m o r t e m proteolysis and tenderization of meat from different species are negatively correlated with their calpastatin activity [33, 46]. From these results, it becomes apparent that calpastatin is indeed a powerful regulator of the calpains and efforts should be made to elucidate the mechanism of its inactivation. We believe that the inactivation of calpastatin is an enzymatic process and the protease involved is activated with calcium and inhibited by zinc [ 11 ]. Finally, I would like to suggest that our present knowledge of the regulation of calpains in p o s t mortem muscle is by no means complete and much remains to be learned. Although quantification of the components of the calpain proteolytic system may be very useful, the results should be interpreted with caution, because the measured activity and actual proteolytic capacity, in situ, may be very different. One of the best examples to illustrate this point is the activity of the calpains in slow-twitch red vs fast-twitch white muscles. The activity of la-calpain, m-calpain and calpastatin has been shown to be similar or even higher in red than white muscle [46]. Also, it has been demonstrated that p o s t m o r t e m proteolysis and tenderization process do not occur in the red muscle. Therefore, it has been argued that the calpain content does not agree with the proposed role of this proteolytic system in post m o r t e m proteolysis [47]. However, Cassens et a l [48] demonstrated that the slow-twitch red muscle has a three- to four-fold higher zinc

content than fast-twitch white muscle. They reported that the largest portion of muscle zinc (64 and 86% in the white and red muscle, respectively) was found in the fraction composed primarily of myofibrils and nuclei [48]. Their findings have recently been substantiated by Kondo et a! [49] who reported that red muscle contained 4.3 times higher zinc than white muscle. Therefore, higher content of zinc could be one of the reasons for the lack of post mortem proteolysis and tenderization in red muscle [32, 50]. In addition to zinc, other factors could also be involved in the regulation of calpains in post mortem muscle, such as a calpain activator. In 1982, DeMartino and Blumenthai, while examining the role of calmodulin on the activities of calpains in brain, identified a protein which was capable of stimulating the activities of both la-calpain and m-calpain I511. It stimulated the activities of both calpains up to 25-fold but it did not alter their calcium requirement for activity. Recently, Pontremoli et a l [52] reported successful isolation of an activator from rabbit skeletal muscle. Whether calpain activator exists in skeletal muscle of meat producing animals (ie, cattle, sheep and pigs) remains to be determined. Yet, other possible regulators of calpains are camosine, anserine and u-l-methyl-histidine. In muscle these dipeptides are found at relatively high (mmol) concentration [53, 54]. It has been reported that these dipeptides are mild activators of calpains [551. In addition, carnosine increased the inhibitory activity of calpastatin, whereas anserine and 1,-Imethyi-histidine reduced the inhibitory effect of calpastatin [55]. These are some examples that should clearly indicate that while two tissues (eg, red and white muscle) might have identical calpain content, they might not have the same proteolytic capacity and thus we should be careful in interpreting quantitative data. In summary, based on information currently available, we conclude that post mortem changes associated with meat tenderization are the result of proteolysis of key myofibrillar proteins. Experimental evidence indicates that the calpain proteolytic system is probably responsible for this post mortem proteolysis. There is no doubt that factors other than proteolysis will affect meat tenderness (elevated ionic strength during post mortem storage, connective tissue, etc). However, we believe that the principal reasons for the observed differences in the rates of post mortem tenderization (eg, among species or muscles) are differences in the rate of degradation of key myofibrillar proteins, which is probably mediated by the calpain proteolytic system. Also: 'The ultimate test of any theory's validity must be its capacity to explain the wide tenderness variability found among carcasses that have undergone normal post mortem treatment: muscles that remain attached to the carcass or side,

244 sides that remain suspended during cooling, and cooling that remains within the usual rate limits o f meat-industry practice. Without this vital extension to w o r k s ' operating conditions, a laboratory-generated hypothesis will necessarily remain o f only a c a d e m i c interest' [56].

References I 2

3 4 5 6 7

8

9

10

!1 12

13

14

15 16

Lehmann KB (1907) Studies of the causes for the toughness of meats. Arch F Hyg 63, 134 (Quoted by reference number 6) Goll DE. Otsuka Y, Nagainis PA, Shannon JD, Sathe SK Muguroma M (1983) Role of muscle proteinases in maintenance of muscle integrity and mass. J Food Biochem 7, 137-177 Robson RM, Huiatt TW (1983) Roles of the cytoskeletal proteins desmin, titin and nebulin in muscle. Rec Meat ConfProc 36, 116-124 Koohmaraie M (1988) The role of endogenous proteases in meat tenderness. Rec Meat ConfProc 41, 89-100 Hoagland R, McBryde CN, Powick WC (1917) Changes in fresh beef during cold storage above freezing. USDA Bull No 433. Washington, DC Penny IF (1980) The enzymology of conditioning. In: Developments in Meat Science-I (Lawrie R, ed) Applied Science Publishers, Ltd, London, 115-143 Koohmaraie M, Babiker AS, Merkel RA, Dutson TR (1988) Role of Ca2+ dependent proteases and lysosomal enzymes in post mortem changes in bovine skeletal muscle. J Food Sci 53, 1253-1257 Koohmaraie M, Babiker AS, Schroeder AL, Merkel RA, Dutson TR (1988) Acceleration of post mortem tenderization in ovine carcasses through activation of Ca2+dependent proteases. J Food Sci 53, 1638-1641 Koohmaraie M, Crouse JD, Mersmann HJ (1989) Acceleration of post mortem tenderizalion in ovine carcasses through infusion of calcium chloride: Effect of concentration and ionic strength. J Anita Sci 67, 934-942 Koohmaraie M, Whipple G, Crouse JD (1990) Acceleration of post mortem tenderization in lamb and Brahman-cross beef carcasses through infusion of calcium chloride. J Anita Sci 68, 1278-1283 Koohmaraie M (1990) Inhibition of post mortem tenderization in ovine carcasses through infusion of zinc. J Anita Sci 68, 1476--1483 Fiems LO, Buts B, Boucque CV, Demeyer DI, Cottyn BG (1990) Effect of a [3-agonist on meat quality and myofibrillar protein fragmentation in bulls. J Meat Sci 27, 29-39 Kretchmar DH, Hathaway MR, Epley RJ, Dayton WR (1990) Alterations in post mortem degradation of myofibrillar proteins in muscle of lambs fed a [3-adrenergic agonist. J A n i m Sci 68, 1760-1772 Koohmaraie M, Shackelford SD, Muggli-Cockett NE, Stone RT (1991) Effect of a ~-adrenergic agonist, I.,6~.~9, on muscle growth, endogenous proteinase activities and post mortem proteolysis in wether lambs. J Anita Sci (in press) Koohmaraie M, Shackelford S (1991) Effect of calcium chloride infusion on the tenderness of lambs fed a I~-adrenergic agonist. J A n i m Sci 69, 2463-2471 Whipple G, Koohmaraie M, Dikeman ME, Crouse JD, Hunt MC, Klemm RD (1990) Evaluation of attributes that

affect longissimus muscle tenderness in Bos taurus and Bos indicus cattle. J Anim Sci 68, 2716-2728 17 Shackelford S, Koohmaraie M, Miller ME Crouse JD, Reagan JO (I 991) An evaluation of tenderness of the longissimus muscle of Angus by Hereford versus Brahman crossbred heifers. J Anim Sci 69, 17 !-177 18 Tarrant PV (1987) Review: Muscle biology and biochemistry. Proc Eur Meet o f Meat Res Workers, !-5 19 Rivett AJ (1989) The multicatalytic proteinase of mammalian cells. Arch Biochem Biophys 268, 1-8 20 LaCourt AA, Obled A, Deval C, Ouali A, Valin C (1986) Post mortem localization of lysosomal peptide hydrolase. Cathepsin B. In: Cysteine Proteinases and Their lnhibitots (Turk V, ed) Walter de Gruyter and Co, New York, 239-248 21 Mikami M. Whiting AH, Taylor MAJ, Maciewicz RA, Etherington DJ (1987) Degradation of myofibrils from rabbit, chicken and beef by catbepsin L and lysosomal lysates. Meat Sci 21, 81-97 22 Goll DE, Thompson VF, Taylor RG, Christiansen JA (1992) Properties and regulation of the calpain system and its role in muscle growth. Biochimie 74, 225-237 23 Davey CL, Gilbert KV (1969) Studies in meat tenderness. 7. Changes in the fine structure of meat during aging. J Food Sci 34, 69-74 24 Busch WA, Stromer MH, Goll DE, Suzuki A (1972) Ca 2+specific removal of Z-lines from rabbit skeletal muscle. J C e l l Bio152, 367-381 25 Taylor MAJ, Etherington DJ (1991) The solubilization of myofibrillar proteins by calcium ions. J Meat Sci 29, 211-219 26 Hattori A, Takahashi K (1979) Studies on the post mortem fragmentation of myofibrils. J Biochem 85, 47-56 27 Hattori A, Takahashi K (1982) Calcium-induced weakening of skeletal muscle Z-disks. J Biochem 92, 381-390 28 Takahashi K, Kim OH, Yano K (1987) Calcium-induced weakening of Z-disks in post mortem skeletal muscle. J Biochem 101,767-773 29 Guroff G (1964) A neutral, calcium-activated proteinase from the soluble fraction of rat brain. J Biol Chem 239, 149-155 30 Alarcon-Rojo A, Dransfield E (1989) Effect of calcium ions on texture of beef during conditioning. Proc lnt Cong Meat Sci and Technol, Copenhagen, 35, 1141-1144 31 Koohmaraie M, Seideman SC, Schollmeyer JE, Dutson TR, Crouse JD (1987) Effect of post mortem storage on Ca2+-dependent proteases, their inhibitor and myofibril fragmentation. Meat Sci 19, 187-196 32 Koohmaraie M, Seideman SC, Schollmeyer JE, Dutson TR, Babiker AS (1988) Factors associated with the tenderness of three bovine muscles. J Food Sci 53, 407-4 !0 33 Koohmaraie M, Whipple G, Kretchmar DH, Crouse JD, Mersmann HJ (1991) Post mortem proteolysis in longissimus muscle from beef, lamb and pork carcasses. J Anita Sci 69, 617-624 34 George SM, Calkins CR, Koohmaraie M (1991) Inhibition of crude and purified ~t-calpain by calpastatin. J Anita Sci 69 (Suppl !), 336 35 Vidalenc P, Cottin P, Merdaci N, Ducastaing A (1983) Stability of two Caa+-dependent neutral proteinases and their specific inhibitor during post mortem storage of rabbit skeletal muscle. J Sci Food Agric 34, 1241-1250 36 Ducastaing A, Valin C, Schollmeyer .I, Cross R (1985) Effects of electrical stimulation on post mortem changes in the activities of two Ca-dependent neutral proteinases and their inhibitor in beef muscle. Meat Sci 15, 193-202

245 37 38

39

40 41 42

43

44 45

Suzuki K, Tsuji S, Ishiura S, Kimura Y, Kubota S, lmahori K (1981) Autolysis of calcium-activated neutral protease of chicken skeletal muscle. J Biochem 90, 1787-1793 Suzuki K, Tsuji S, Kubota S, Kimura Y, Imahori K (1981) Limited autolysis of Ca2+-activated neutral protease (CANP) changes its sensitivity to Ca 2+ ions. J Biochem 90, 275-278 Hathaway DR, Werth D, Haeberle JR (1982) Limited autolysis reduces the Ca 2+ requirement of a smooth muscle Ca2+-activated protease. J Biol Chem 257, 9072-9077 Mellgren RL, Repetti A. Muck TC, Reimann EM (1982) Rabbit skeletal muscle calcium-dependent protease requiring millimolar Ca 2+. J Biol Chem 257, 7203-7209 Parkes C, Kembhavi AA, Barratt AJ (1985) Calpain inhibition by peptide epoxides. Biochem J 230, 509516 DeMartino GN, Huff CA, Croall DE (1986) Autoproteolysis of the small subunit of calcium-dependent protease il activates and regulates protease activity. J Biol Chem 261, 12047-12052 lmajoh S, Kawasaki H, Suzuki K (1986) Limited autolysis of calcium-activated neutral protease (CANP) reduction of the Ca2÷-requirement is due to the NH,-terminal processing of the large subunit. J Biochem 100, 633642 Inomata M, Imahori K, Kawashima S (1986) Autolytic activation of calcium-activated neutral protease. Biochem Biophys Res Commun 138, 638--643 Crawford C, Willis AC, Gagnon J (1987) The effects of autolysis on the structure of chicken calpain II. Biochern J 248, 579-588

46 Ouali A, Talmant A (1990) Calpains and calpastatin distribution in bovine, porcine and ovine skeletal muscles. Meat Sci 28, 331-348 47 Ouali A (1990) Meat tenderization: possible causes and mechanisms. A review. J Muscle Foods I, 129-165 48 Cassers RG, Hoekstra WG, Faltin EC, Briskey EJ (1967) Zinc content and subcellular distribution in red vs white porcine skeletal muscle. Am J Physiol 212, 668--692 49 Kondo H, Kimura M, Itokawa Y (1991) Manganese, copper, zinc, and iron concentrations and subcellular distribution in two types of skeletal muscle. Proc Sot" Exp Biol Med 196, 83-88 50 Olson DG, Parrish FC Jr, Stromer MH (1976) Myofibril fragmentation and shear resistance of three bovine muscles during post mortem storage. J Food Sci 41, 1036--1041 51 DeMartino GN, Blumenthal DK (1982) Identification and partial purification of a factor that stimulates calciumdependent proteases. Biochemistry 21,4297--4303 52 Pontremoli S, Viotti PL, Michetti M, Sparatore B, Salamino F, Melloni E (1990) Identification of an endogenous activator of calpain in rat skeletal muscle. Biochem Biophys Res Comm 17 I, 569-574 53 Crush KG (1970) Camosine and related substances in animal tissues. Comp Biochem Physiol 34, 3-30 54 Margolis F (1974) Camosine in the primary olfactory pathway. Science 184, 909-911 55 Johnson PJ, Hammer JL (1989) Effect of L-l-methyl-histidine and the muscle dipeptides carnosine and anserine on the activities of muscle calpains. Comp Biochem Physiol 94B, 45--48 56 Marsh BB, Russell RL, Swartz DR, Ringkob TP (1988) Letters to the Editor. Meat Sci 24, 227-232