Tenderization of beef by lactic acid injected atdi�erent times post mortem

Philippe Berge a,*, Per Ertbjerg b, Lone Melchior Larsen c, Thierry Astruc c,Xavier Vignon a, Anders J. Mùller b

aINRA, Station de Recherches sur la Viande, Centre de Clermont-Fd/Theix, F-63122 St-GeneÁs-Champanelle, FrancebDepartment of Dairy and Food Science, The Royal Veterinary & Agricultural University, Rolighedsvej 30, DK-1958 Frederiksberg C, Denmark

cChemistry Department, The Royal Veterinary & Agricultural University, Rolighedsvej 30, DK-1958 Frederiksberg C, Denmark

Received 4 August 1999; received in revised form 20 April 2000; accepted 20 August 2000

Abstract

The potential to tenderize beef muscles by the injection of lactic acid (0.5 M, 10% w/w) was studied using the pectoralis profundus

muscle from cull cows. The injection was performed either 1 h (pre rigor) or 24 h (post rigor) post mortem, and the meat was storedfor 2 or 14 days post mortem. Both treatments caused a rapid pH drop to around 5.0 within 4 h of injection. Other e�ects were: (1)an accelerated release of lysosomal enzymes into the cytosol; (2) a greater degradation of myosin heavy chains; (3) ultrastructural

alterations of the myo®brils which included a general weakening or rupture in the M-lines and, to a lesser extent, in the I-bands; (4)a decreased heat stability of perimysial collagen indicated by a lower insoluble collagen content, lower di�erential scanning calori-metry transition temperature, and lower transition temperatures in isometric tension tests on muscle strips. The lactic acid injections

improved signi®cantly the textural traits of the meat (shear value, tensile strength, sensory scores) at 2 days post mortem with littlefurther improvement when storage was extended to 14 days post mortem. Changes in texture were of similar amplitude at both postmortem injection times. The tenderization mechanisms of lactic acid injection are discussed. # 2001 Elsevier Science Ltd. All rightsreserved.

Keywords: Beef; Lactic acid injection; Marination; Tenderness; Collagen; Proteases; Ultrastructure

1. Introduction

Collagen-rich beef muscles are generally tough, evenunder the most favourable ageing conditions. Long heattreatments decrease to some extent the toughness ofsuch muscles (Bouton & Harris, 1972) through the par-tial heat denaturation and subsequent solubilization oftheir connective tissue components. However, theadvantage of this e�ect is limited by the tougheninge�ect of the heat denaturation of the myo®brillar pro-teins and the high levels of moisture loss during pro-longed cooking (Martens, Stabussvik, & Martens, 1982).Marination in acidic solutions, e.g. acetic or lactic acid,have been traditionally used as a means of softening and¯avouring meats. These treatments decrease themechanical resistance of meats, including those of highconnective tissue content (Gault, 1985; Kijowski &

Mast, 1993; O�er & Knight, 1988; Wenham & Locker,1976). However, extended periods are needed for fullmarination due to the slow penetration of exogenousacids, which may limit the e�ciency of this treatment(Kotula & Thelappurath, 1994; Seuss & Martin, 1993;Wenham & Locker, 1994). Therefore, the injection ofthe solution allows more rapid di�usion of the acid intothe muscle and results in a faster decrease in mechanicalstrength and increase in tenderness (Cannon, McKeith,Martin, Novakofski, & Carr, 1996b; Eilers et al., 1994;Ertbjerg, Larsen, & Mùller, 1995; Ertbjerg, Mielche,Larsen, & Mùller, 1999). The e�ects of such acids onmeat texture depend on pH drop following treatment.They have been related to the combination of threemechanisms: (1) dilution of the load-bearing materialdue to the swelling of the myo®bres across their mainaxis (Gault, 1985; Rao & Gault, 1990); (2) direct weak-ening of the perimysial connective tissue (Ertbjerg et al.,1995; Lewis & Purslow, 1991; Oreskovich, Bechtel,McKeith, Novakofski, & Basgall, 1992; Stanton & Light,1990) (3) acceleration of the post mortem tenderization

0309-1740/01/$ - see front matter # 2001 Elsevier Science Ltd. All rights reserved.

PI I : S0309-1740(00 )00110-8

Meat Science 57 (2001) 347±357

www.elsevier.com/locate/meatsci

* Corresponding author. Tel.: +33-473-624-000; fax: +33-473-

624-089.

E-mail address: berge@clermont.inra.fr (P. Berge).

by proteases (viz cathepsins) whose activity is optimal atlow pH values (O�er &Knight, 1988;Wenham&Locker,1976). Most of the works published on acid treatmentsapplied to beef are on marination in acetic acid. The pHvalues obtained in the meat were generally low (between 5and 3 depending upon the acid concentration used) lead-ing to considerable swelling of the meat. In the few worksusing lactic acid, it was injected pre rigor and the ®nalmeat pH values were generally higher than those obtainedwith acetic acid (4.5 or more), suggesting meat swellingwas limited. However, lactic acid injections considerablyimproved the texture of tough beef (Eilers et al., 1994;Ertbjerg,Mielche et al., 1999). This was associated with anacceleration of post mortem release of lysosomal pro-teases (Ertbjerg, Larsen, & Mùller, 1999; Ertbjerg,Mielche et al., 1999; éstdal, Larsen, & Mùller, 1992)which could be thus responsible for the greater postmortem solubilization of perimysial collagen (Ertbjerget al., 1995; Stanton & Light, 1990). The objective of thisstudy was to investigate the tenderizing potential of lacticacid injection in collagen-rich beef muscles. The mod-i®cations brought about by this treatment on the activityof lysosomal proteases, the heat stability of the collagenfraction, the myo®brillar structure and the texture of thecooked meat were evaluated according to the time postmortem of injection and the storage period.

2. Materials and methods

2.1. Animals and sampling procedure

Eight Friesian cull cows (3±4 years of age) wereslaughtered at an average carcass weight of 256 (�39)kg. The M. pectoralis profundus was excised from bothsides of each carcass immediately after slaughter. Thismuscle was chosen for its relatively high collagen con-tent (on average 1.2% wet wt. in this experiment) andtoughness (Drans®eld, 1977; Dumont, 1979). A squaresample was cut from the central part of the muscle ofeach side, discarding the anterior and posterior ends,and the thinner areas at the dorsal and ventral sides.The sample obtained from each side was subdividedinto six subsamples (approximately 70�100 mm), i.e. intotal 12 subsamples from each animal. The six sub-samples from the anterior location were used for theprotease assay (determination of lysosomal enzymeactivities), for the study of muscle ultrastructure and fortexture measurements (shear force and sensory evalua-tions). The six subsamples from the posterior locationwere used for the study of collagen properties.

2.2. Experimental design

The subsamples of meat were allocated randomly tosix experimental treatments according to a 3�2 factorial

design. The treatments were: non-injected control (C);lactic acid injected at 1 h (LA-1); or 24 h post mortem(LA-24), each group being aged for 2 or 14 days.

2.3. Injection procedure

The samples were injected with a 0.5 M lactic acidsolution (10%w/w) using an Eppendorf 4780 multi-pipetteequipped with a Plus/8 adaptor with ®xed needles. Theinjection was performed at 3 depths (50 ml per level) withapproximately 5 mm between injection points. The con-trols were only punctured. All samples were vacuumpacked at 4 h post mortem and stored for 24 h post mor-tem at 15�C. After the post rigor injection (LA-24) wasperformed, all samples were stored at 4�C until completionof their respective ageing period. Subcellular fractionationfor determination of lysosomal enzyme activities, and ®xa-tion of samples for muscle ultrastructure were performedon fresh samples as described below. Samples for the eva-luation of texture and collagen properties were stored atÿ20�C and analysed after thawing overnight at 4�C.

2.4. Measurements

2.4.1. pHThe muscle pH was determined at 4 and 48 h post

mortem (ultimate pH value) using a glass electrode(mod. PHM 62, Radiometer, Copenhagen, DK), the pHvalue being the average of four measurements at di�er-ent locations within each sample.

2.4.2. Myo®brillar subcellular fractionsSubcellular fractionation was carried out at days 2 and

14 post mortem by di�erential centrifugation of musclehomogenates according to the method of Ertbjerg,Mielche et al. (1999). Myo®brillar, mitochondrial, lyso-somal and microsomal fractions were obtained by suc-cessive centrifugations of supernatants at 1100, 3000,27 000 and 100 000 g, respectively. Fluorimetry was usedto measure the combined activities of cathepsins B andL (Kirschke, Wood, Roisen, & Bird, 1983) and theactivity of b-glucuronidase (Moeller, Fields, Dutson,Landmann, & Carpenter, 1976) in the membrane(mitochondrial+lysosomal+microsomal) and solublefractions. Myo®brillar protein degradation was fol-lowed by sodium dodecyl sulphate-polyacrylamide gelelectrophoresis (SDS-PAGE) as described by Ertbjerg,Mielche et al. (1999). The gels were stained with Coo-masie Brilliant Blue and scanned to determine the den-sity of each individual band using the CREAMscanning system (Kem-En-Tec A/S, Copenhagen, Den-mark) as described by Ertbjerg, Mielche et al. (1999).

2.4.3. Myo®brillar ultrastructure andimmunolocalization of cathepsinsThe myo®brillar ultrastructure was studied by trans-

mission electron microscopy using a Philips EM400

348 P. Berge et al. /Meat Science 57 (2001) 347±357

microscope under an acceleration voltage of 80 kV.Meat samples (2�2.5 mm in section�10 mm in the longaxis of ®bres) were ®xed in a solution of 4% paraf-ormaldehyde, 0.1% glutaraldehyde in 0.1 M cacodylatebu�er (pH 7.4) for 1 h, then diced into 2 mm3 cubesand rinsed three times for 15 min in a cacodylate buf-fer. For ultrastructural observations, the blocks werepost®xed in 1% OsO4 for 1 h at 4�C. After dehydra-tion in ethanol, the samples were embedded in epoxyresin and cut into ultra-thin sections (80 nm) collectedon copper grids and stained with uranyl acetate (satu-rated aqueous solution) then with lead citrate asdescribed by Reynolds (1963). For the immunolocali-zation of cathepsins, the blocks were directly dehy-drated through a cold ethanol gradient (20±85%) afterthe paraformaldehyde ®xation, then progressively in®l-trated with LR (London-Resin) White resin (TAAB-Eurobio, France) and ®nally embedded in LR Whiteresin in closed gelatin capsules. Polymerization wasperformed at 60�C for 48 h. Ultra thin sections weremounted on nickel grids and processed for immunocy-tochemistry.Immunocytochemistry was performed using the

immuno-gold technique by ¯oating the grids on dropsof reagent at room temperature. Brie¯y, after 20 min in1% skimmed milk, 0.5% bovine serum albumin (BSA,Sigma) in phosphate bu�er saline (PBS), the grids wereincubated in rabbit antibodies to bovine cathepsin Ddiluted 1:50 in PBS for 2 h at 37�C. The antibodies werekindly supplied by Dr. L.B. Larsen, Danish Institute ofAgricultural Sciences, Research Centre Foulum, Den-mark. The grids were then washed in PBS±BSA andincubated in gold-protein A complexes (10 nm, Sigma)diluted 1:50 in PBS for 1 h at room temperature. Thegrids were further washed in PBS, rinsed with distilledwater and dried. The speci®city of the immunolabellingwas tested in control sections by replacing the anti-cathepsin D antiserum with non-immune rabbit serum.The sections were directly observed without counter-staining.

2.4.4. Collagen: heat stability and di�erential scanningcalorimetry (DSC) measurementsThe total collagen content of raw meat was derived

from the hydroxyproline concentration (collagen=7.5�HyPro) determined on ®ve replicates using the methodof Bergman and Loxley (1963) adapted by Bonnet andKopp (1984). The insoluble collagen content was deter-mined following the same procedure on the residueobtained after heating the meat sample in a Tris±HCl0.02M, NaCl 0.23M, pH 7.4 bu�er solution (1:5 w/v) at90�C for 2 h and subsequently discarding the heat-solu-ble fraction as described by Bonnet and Kopp (1992).The large variations in total collagen content foundbetween animals and between samples within each ani-mal made it necessary to adjust the values of insoluble

collagen to a constant muscle total collagen contentwhich was taken as the average value over the popula-tion of samples analysed.DSC was performed on perimysial tissue extracted

from 100 g of meat. Meat cubes were homogenized inice-cooled water (1:8 w/v) and sieved (0.5-mm mesh).Then the residue was homogenized in 200 ml ice-cooledwater and sieved (0.5-mm mesh). The process was repe-ated four times. The ®nal residue was defatted anddehydrated in acetone as described by Bonnet andKopp (1992). Samples of isolated tissue (20 mg dry wt.)were allowed to rehydrate for 16 h at 4�C in waterbefore the transition temperature of collagen was mea-sured by DSC using a DSC-121 (Setaram, Lyon,France) at a constant heating rate of 3�C/min from 10to 100�C.

2.4.5. Mechanical measurementsThe isometric tension developed during heating was

measured on muscle samples (40�5�5 mm, approxi-mately) stored for 14 days, using an Instron universaltesting machine (model 4302) as described by Kopp,Sale , and Bonnet (1977). The samples were cut so thatthe longest dimension was parallel to the main muscle®ber axis. They were restrained at their resting length(isometric conditions) while heated from 15 to 97�C atthe rate of 3�C/min in an isotonic bu�er solution similarto that used for the collagen heat solubilization. Aftercompletion of the isometric tension test, the sampleswere vacuum packed and kept frozen at ÿ20�C formeasurement of their tensile strength. The tensile testwas performed using the same testing machine at a dis-placement rate of 50 mm/min after the samples hadbeen thawed at room temperature.The shear force measurements were performed using

a Warner-Bratzler (WB) shear blade with a triangularslot angular cutting edge adapted to an an Instron uni-versal testing machine (model 4301). Vacuum-packedmeat samples (approximately 250 g) were heated at 60�Cfor 120 min to ensure a uniform end-point temperaturein all samples. Initial yield (myo®brillar toughness, M-force) and ®nal yield (connective tissue toughness, C-force) were determined from the shear deformationcurves obtained on meat cubes (10�10�50 mm) asreported by Mùller (1981).

2.4.6. Sensory evaluationSensory evaluation of texture was performed in two

sessions on di�erent days by a ®ve-member semi-trainedpanel using meat from four cows. The cooked meatsamples were scored on a descriptive scale that extendedfrom extremely tough (score 1) to extremely tender(score 10). Tenderness was evaluated as the ease of ®rstbite (i.e. resistance to initial chewing) and chewiness(i.e. amount of chewing to get the sample ready forswallowing).

P. Berge et al. /Meat Science 57 (2001) 347±357 349

2.4.7. Statistical analysis of dataData were subjected to a two-way analysis of variance

using the General Linear Models (GLM) procedure ofSAS (1989±1996). The model included the main e�ectsof lactic acid treatment and duration of storage, and oftheir interaction. The Student t-test was used to assessthe signi®cance of the di�erences between treatmentmeans.

3. Results

The post mortem pH values for the di�erent treat-ments are presented in Table 1. In the LA-1 treatment,the pH value dropped to 5.1 within 4 h post mortemand stabilized at 5.1 at 28 h post mortem; at that time,the pH of the control was 5.5. The LA-24 injectionresulted in a decline to pH 4.9 as measured at 28 and 48h post mortem (data not shown).Both LA-1 and LA-24 injections increased con-

siderably the cathepsin B+L activity in the solublefraction obtained after subcellular fractionation (Fig. 1).Analysis of variance showed that this increase was highlysigni®cant (P<0.001) and was similar in amplitude(approximately two-fold) in both lactic acid treatmentswhatever the storage period (2 or 14 days). Lactic acidinjection reduced the cathepsin B+L activity in themembrane fraction by approximately 30% (P<0.001).Similarly the b-glucuronidase activity increased (P<0.01)in the soluble fraction and decreased (P<0.001) in themembrane fraction as a result of lactic acid injection. Themain e�ect of extending the storage period from 2 to 14days was a small increase in cathepsin B+L (P < 0.001)and b-glucuronidase (P=0.06) activities in the solublefraction. Enzyme activities in the membrane fractionwere una�ected by storage period (P>0.05).The most important changes in the bands obtained

after SDS-PAGE electrophoresis of the myo®brils withor without lactic acid treatment are shown in Table 2.Only bands showing detectable changes are shown. Thee�ects of lactic acid were generally similar at bothinjection times. The major e�ect of lactic acid injection onmyo®brillar proteins was a marked decrease in the den-sity of the band migrating at the position of the myosin

heavy chain (MHC) and an increase in the density of the150 kDa band. This e�ect was signi®cant (P<0.05) at 2days post mortem and still perceptible after 14 days.Actin degradation by lactic acid was small as the densityof this band was only slightly less in injected sampleswhen compared with controls. The 31-kDa band, whichis assumed to be a product of the degradation oftroponin T (Ertbjerg, Henckel, Karlsson, Larsen, &Mùller, 1999; Ho, Stromer, & Robson, 1994;Hu�-Lonergan, Mitsuhashi, Beckman, Parrish, Olson,& Robson, 1996), increased with storage but remainedvirtually una�ected by the lactic acid treatments. Theband migrating at the position of a-actinin showed sig-ni®cantly lower densities after the lactic acid treatment(P<0.05) only in LA-1 samples stored for 2 days and inLA-24 samples stored for 14 days. The density of a 95-kDa band at 2 days post mortem increased after bothlactic acid treatments but this e�ect was no moreapparent after 14 days of storage.The injections, particularly the post rigor injection,

caused dramatic but heterogeneous changes in themyo®brillar ultrastructure. The myo®brils exhibited anamorphous (or coagulated) appearence (Fig. 2A±C). Anoticeable number of I-bands were weakened, and occa-sionally ruptured, near the Z-lines with gaps appearing inthe Z-lines (Fig. 2B, C). The M-lines appeared exten-sively weakened at day 2 post mortem (Fig. 2B) andruptured at day 14 post mortem, leading to an apparentincrease in sarcomere length in these samples (Fig. 2C).These e�ects were observed in some LA-1 samples andin all LA-24 samples, both at days 2 and 14 post mor-tem. Some parts of the muscle sections exhibited a nor-mal appearance, particularly in the LA-1 samples,showing that they had probably not been a�ected uni-formly by the treatment. As a general rule, the extent ofthe structural alterations caused by lactic acid injectionwas greater in the LA-24 treatment (Fig. 2B, C).The antibodies to bovine cathepsin D allowed the

detection of large vesicles, usually in a subsarcolemmalpart of the muscle cells, that were identi®ed as second-ary lysosomes (Fig. 2D). In the meat samples stored for2 days post mortem, the vesicular localization of thecathepsin D seemed to be well preserved whatever thetime of injection (Fig. 2E). Although the lysosomalmembranes may have been a�ected by the acid treat-ments, the di�usion of the enzymes from the vesiclesappeared to be limited. Immunolabelling was lowest inthe LA-24 samples. In the samples stored for 14 days,the vesicles were not as well de®ned as in the samplesstored for 2 days, indicating that di�usion of cathepsinswas greater as storage time increased from 2 to 14 dayspost mortem (Fig. 2F) in the lactic acid-injected sam-ples. But the immunolabelling was always observed inthe vicinity of the lysosomes, with the exception of LA-24 samples. In the latter samples, the labelling was veryfaint, suggesting that the enzymes were no longer con-

Table 1

Post mortem pH values in the beef samples as a�ected by lactic acid

injected either at 1 h (LA-1) or at 24 h (LA-24) post mortem

Time post mortem Lactic acid treatmenta S.E.M.b

Control LA-1 LA-24

4 h 6.4 a 5.1 b 6.4 a 0.2

28 h 5.5 a 5.1 b 4.9 c 0.2

a Means values; control=non-injected; values in the same row with

the same letter are not signifcantly di�erent.b Standard error of means.

350 P. Berge et al. /Meat Science 57 (2001) 347±357

centrated in the lysosomal area, but had di�used in thecytosol.Lactic acid injections reduced (P<0.001) the meat-

insoluble collagen content (Table 3). The decrease aver-aged 6 and 10% in LA-1 and LA-24 treatments,

respectively, corresponding to values of collagen heatsolubility (or proportion of soluble collagen in the totalcollagen) of 16 and 18% compared to 11% in the con-trols. The insoluble collagen content increased slightlywith storage. Although this change was signi®cant

Fig. 1. Combined activities of cathepsins B and L (top graphs) and activity of b-glucuronidase (bottom graphs) at two times post mortem in two

subcellular fractions of beef muscle injected with lactic acid either at 1 h (LA-1) or at 24 h (LA-24) post mortem. (a, b, c) values at the same storage

time di�er signi®cantly when accompanied by di�erent superscripts (P<0.05).

P. Berge et al. /Meat Science 57 (2001) 347±357 351

(P<0.001), its practical implication was probably neg-ligible. There was no signi®cant interaction betweenlactic acid treatment and storage period (P>0.05).The isometric tension curves showed that the heat-

induced contraction of the muscle tissue started at alower temperature in the 60�C temperature region in thelactic acid injected samples (di�erences of ÿ2.0 andÿ1.5�C with controls for LA-1 and LA-24, respectively)(Table 3). Two other transitions could be observed atabout 65 and 73±76�C (Fig. 3). In the LA-1- and LA-24-injected samples, the average temperatures correspond-ing to these transitions were 1.2±2.6�C lower (P<0.001)compared with the control samples. The decrease in thetransition temperature of collagen was con®rmed by theDSC measurements on the isolated perimysium. Theonset temperature was on average 0.8 and 1.5�C lower(P<0.001) in LA-1 and LA-24 treatments, respectively,compared with controls. From the results, there was noclear evidence of the e�ect of time post mortem ofinjection; the DSC temperature of the collagen transi-tion was signi®cantly lower in LA-24 samples comparedto LA-1 samples while the temperature of start of iso-metric contraction was slightly, but signi®cantly lower(P<0.01) in LA-1 samples compared to LA-24 samples.Tensile strength of cooked meat was 13% (LA-1;

P<0.01) to 20% (LA-24; P<0.001) less in the injectedsamples stored for 14 days post mortem compared withcontrols. Correspondingly, lactic acid injection reduced(P<0.001) the WB shear force components (Fig. 4).After 2-days storage, the reduction was 34±39% for

both M- and C-force, with no e�ect of injection time.Further ageing to 14 days post mortem reduced M-forceonly in control samples (P < 0.05), but not in theinjected samples. After 14 days, M-force in controlsamples was still higher than in injected samples. Sto-rage did not change C-force in any of the treatments.Texture sensory scores were also markedly improved bylactic acid injection (P<0.05), especially the ease of ®rstbite (Fig. 5), but they were not in¯uenced signi®cantlyby storage P>0.05). Whatever the texture parametermeasured (tensile strength, shear force, sensory attri-butes) the values obtained for LA-1 and LA-24 sampleswere not signi®cantly di�erent (P>0.05), showing thatlactic acid was equally e�cient in tenderizing the meatat both injection times.

4. Discussion

The changes observed after the injection of lactic acidinto the meat clearly demonstrate that this treatmentwas e�cient in reducing the toughness of a collagen-richbeef muscle such as M. pectoralis profundus. Previousworks have already shown the positive e�ect of injectinglactic acid on beef texture (Eilers et al., 1994, Ertbjerg etal., 1995; Ertbjerg, Mielche et al., 1999). The e�ciencyof acid treatment depends not only on the intrinsictoughness of the meat (Wenham & Locker, 1976) butalso on the type and the concentration of acid used.Gault (1985) examined the e�ects of low pH, obtainedby the addition of acetic acid solutions, on cooked meattenderness. Peak force was maximum at pH valuesaround 5.0 (range 4.5±5.5) and decreased sharply as thepH decreased from 4.6 to values less than 4.1. Thisimprovement of tenderness was related to a markedincrease in both raw and cooked meat swelling, anindication of increased water-holding capacity of themeat. Similarly, Ertbjerg et al. (1995) showed that lacticacid injected at low concentration (0.3 M), leading to apH value of 5.2 which is close to the isoelectric point ofthe major myo®brillar proteins, did not improve beeftexture, while injection at 1.0 M resulted in a meat pHof 4.6 and decreased meat toughness. However, in thepresent experiment with 0.5 M lactic acid, the meat pHdid not drop below 4.9 and there was no signi®cantchange in the moisture content of the raw meat (datanot presented). This suggests that the swelling of muscle®bres was not the main cause of the tenderizationachieved.The data presented here strongly suggest that the

mechanism of meat tenderization by lactic acid is acombination of weakening of the myo®brillar structureand, to a lesser extent, that of the connective tissue. The®rst evidence was supplied by the instrumental mea-surements of texture. Both components of the WB sheartest (C- and M-force values) and the tensile strenghth of

Table 2

Density of protein bands obtained on SDS-PAGE of myo®brils of

beef injected with lactic acid either at 1 h (LA-1) or at 24 h (LA-24)

post mortem (the density of bands is expressed as peak area in per cent

of total peak area)

Band position Storage

durationa

(days)

Lactic acid treatmentb S.E.M.c

Control LA-1 LA-24

Myosin 2 28.1 a 19.9 b 20.3 b 2.9

heavy chain 14 18.2 a 15.8 a 13.2 a 5.1

150 kDa 2 7.6 a 11.6 b 10.6 b 3.1

14 10.9 a 12.0 a 15.8 b 3.2

a-actinin 2 5.3 a 3.5 b 3.9 ab 1.2

14 4.6 a 3.9 a 2.7 b 0.9

95 kDa 2 0.6 a 1.5 b 2.3 c 0.3

14 1.7 a 2.3 a 1.8 a 0.8

Actin 2 15.5 a 13.5 b 13.9 ab 1.7

14 13.7 a 11.9 b 11.5 b 1.6

31 kDa 2 0.9 a 1.1 a 1.1 a 0.4

14 1.8 a 1.7 a 2.4 a 0.8

a E�ect of storage duration signi®cant for myosin heavy chain, a-actinin (P< 0.001), the 150 and 95-kDa bands and actin (P< 0.01),

and non-signi®cant for the 31-kDa band (P=0.24).b Mean values; control=non-injected.c Standard error of means.

352 P. Berge et al. /Meat Science 57 (2001) 347±357

cooked meat were markedly reduced by lactic acidtreatment which is consistent with the signi®cantimprovement in sensory scores for ease of ®rst bite andchewiness. At the ultrastuctural level, evidence of theweakening of the myo®brillar structure was found in therupture of M-lines and I-bands as well as in theincreased amorphous appearance of the myo®bres of

the lactic acid- injected samples. These alterations couldbe attributed to the acceleration of the release of thelysosomal proteases and to the increase of post mortemdegradation of myo®brillar proteins observed after thelactic acid injection. The higher activity of cathepsinsB+L in the soluble fraction after pre and post rigorinjections, indicated an accelerated release of these

Fig. 2. Ultrastructure (photographs A, B and C) and immuno localization of cathepsin D (photographs D, E and F) in ultrathin sections of post

mortem bovine pectoralis profundus muscle. (A) Control section of muscle ®xed at day 2 post mortem. (B) Section of LA-24 muscle ®xed at day 2

post mortem. (C) Section of LA-24 muscle ®xed at day 14 post mortem. An increasing weakening of the M-lines (M) is observed with post mortem

time leading to extensive rupture, as well as a weakening of I-bands (I and arrows) with occasional ruptures near the Z-lines and gaps in the Z-lines.

Sections of muscle immuno-labelled for cathepsin. Control (D) and LA-1 (E) muscle sections ®xed at day 2 post mortem show lysosomal structures

containing gold particles (L). In the LA-1 muscle ®xed at day 14 post mortem (F), the labelling decreased but was still observed around the lyso-

somes. Z, Z-lines; N, nucleus.

P. Berge et al. /Meat Science 57 (2001) 347±357 353

enzymes from the lysosomes into the cytosol, con®rm-ing works of Ertbjerg, Larsen et al. (1999) and Ertbjerg,Mielche et al. (1999). The MHC was degraded more bylactic acid than the other myo®brillar proteins exam-ined. SDS-PAGE analysis also indicated some increasein the degradation of actin and a-actinin. These resultsare supported by the two transitions in the isometrictension curves at about 65 and 75�C found duringheating of the muscle samples. From the results of pre-vious works by Martens et al. (1982) and Beilken andHarris (1987), the ®rst transition could be assigned tothe denaturation of myosin and collagen, and the sec-ond to actin. The decrease in the temperatures at which

these transitions were observed shows also that themajor myo®brillar proteins were a�ected at the mole-cular level by the treatments and that these changes mayaccount for the lower strength of the myo®brillar struc-ture of the lactic acid-treated meat.The decrease in the temperature of the collagen tran-

sition by DSC con®rms the ®ndings of Kijowski (1993)on di�erent types of connective tissue treated with lacticor acetic acid. It is consistent with the decrease in thetemperature of the start of contraction at around 60�Cfound during the isometric tension test and also with theincrease of the heat solubility of this protein. Increasedcollagen heat solubility was also reported by Oreskovichet al. (1992) after marinating beef in acetic acid. How-ever, the meat pH value reported by these authors wasquite low (pH 3.1), suggesting a direct e�ect of this acidon collagen, probably including swelling of perimysium.Acid solubilization of collagen was unlikely at the muchhigher pH values achieved, with lactic acid, in the pre-sent experiment. Instead, the changes found in collagenstability were probably the result of an increased col-lagenolytic activity of the cathepsins due to their earlierrelease from the lysosomes and the favourable condi-tions due to acidi®cation of the medium after injection.Our results, together with those of Stanton and Light(1990), bring evidence of the weakening of the perimy-sum by injection of lactic acid, and the greater pro-pensity of its collagen to heat solubilization, thuscontributing to the overall improvement in meat tex-ture.There was no apparent advantage in injecting the

meat either pre or post rigor with lactic acid in terms ofcooked meat toughness (shear value, sensory scores).The increase in the activity of cathepsins B+L in thesoluble fraction was similar in LA-1 and LA-24 treat-

Fig. 3. Typical isometric tension curve developed by the beef muscle

samples during heating (arrows indicate thermal transitions).

Table 3

Heat stability, calorimetric and tensile characteristics of collagen from beef injected with lactic acid either at 1 h (LA-1) or at 24 h (LA-24) post

mortem (least square means)a

Lactic acid treatmentb Storage duration S.E.M.c

Control LA-1 LA-24 2 days 14 days

Insoluble collagen contentd (mg/g fresh muscle) 10.4 a 9.7b 9.3 c 9.6 a 10.0 b 1.4

DSC measurements on perimysium

Temperature of collagen transition (�C) 64.9 a 64.1 b 63.4 c 64.1 a 64.1 a 0.6

Isometric tension test on musclee

Temperature of start of contraction (�C) 61.1 a 59.1 b 59.6 c nd nd 1.0

Temperature at ®rst transition (�C) 66.1 a 64.5 b 64.6 b nd nd 0.9

Temperature at second transition (�C) 75.8 a 74.6 b 73.2 c nd nd 1.4

Tensile strength on cooked meatd,e (N/g dry matter/cm). 6.9 a 6.0 b 5.5 b nd nd 1.3

a Values within the same row and corresponding to the same source of variation are signi®cantly di�erent when accompanied by di�erent

(P<0.05). Nd, Not determined.b Control=non-injected; LA-1=injected at 1 h post mortem; LA-24=injected at 24 h post mortem.c Standard error of means.d After adjustment to a constant muscle total collagen content.e Performed only on meat stored for 14 days.

354 P. Berge et al. /Meat Science 57 (2001) 347±357

Fig. 5. Sensory evaluation at two times post mortem of the texture of a beef muscle injected with lactic acid either at 1 h (LA-1) or at 24 h (LA-24)

post mortem (scale from 1=extremely tough to 10=extremely tender). (a, b, c) values at the same storage duration di�er signi®cantly when

accompanied by di�erent superscripts (P<0.05).

Fig. 4. Components of the shear force at two times post mortem in a beef muscle injected with lactic acid either at 1 h (LA-1) or at 24 h (LA-24) post

mortem. (a, b, c) values at the same storage time di�er signi®cantly when accompanied by di�erent superscripts (P<0.05).

P. Berge et al. /Meat Science 57 (2001) 347±357 355

ments at day 2 post mortem. The immunolabelling ofcathepsin D showed that although the di�usion tendedto be greater in the LA-24-treated meat, it was limitedto the vicinity of the lysosomes, whatever the treatmentat day 2 post mortem. Lacourt, Obled, Deval, Ouali andValin (1986) showed by indirect immuno¯uorescencethat there was no change in the post mortem localiza-tion of cathepsin B and no apparent di�usion of thisenzyme in the bovine muscle, even after a prolongedstorage (1 month at 4�C). However, it is possible thatthe sensitivity of the present immuno-staining procedureallowed detection of only the relatively high concentra-tions of cathepsin D but not the lower concentrationslikely after di�usion into the sarcoplasm. Consideringthe small number of gold particles remaining in thelysosomal compartment, it can be speculated that theenzyme had, at least partially, left this organelle after14-day storage, but no de®nite conclusion can be drawnon the extent of the di�usion of the enzyme into thecytosol. Low pH for only 24 h, as observed in the LA-24samples stored for 2 days, was su�cient not only toinduce pronounced e�ects on the distribution of lyso-somal enzyme activity in the subcellular fractions, butalso on myo®brillar protein degradation. The timenecessary to reach a meat pH value of 6.0 was approxi-mately 1.5 h in LA-1 samples compared to 9 h in thecontrols. The much faster rate of pH fall post mortem inthe pre rigor injected meat is likely to have caused anearlier activation of cathepsins. Nevertheless, the postrigor treatment had slightly more pronounced e�ects onsome of the measured traits. This may have been aresult of the slightly lower ultimate pH after post rigortreatment as compared to the pre rigor treatment. Fur-thermore, Ertbjerg, Mielche et al. (1999) suggested thatthe sharp pH drop following a pre rigor injection oflactic acid combined with a still relatively high muscletemperature could cause an early loss of calpain activ-ity. Thus, it can be suggested that the observed weak-ening of the muscle structure and the associatedtenderization following lactic acid injection were pre-dominantly caused by an earlier or an increased releaseof the lysosomal proteases.

5. Conclusion

The injection of 0.5 M lactic acid into a collagen-richbeef muscle (M. pectoralis profundus) reduced markedlythe toughness and increased the tenderness score of themeat as soon as 2 days post mortem. Both the myo®-brillar toughness and, to a lesser extent, the backgroundtoughness (connective tissue) were favourably a�ected.The time post mortem of injection (pre vs. post rigor)had only a minor in¯uence on the short-term as well ason the long-term e�ciency of the treatment on meattexture. However, one should not underestimate possible

side e�ects of lactic acid such as the development of asour ¯avour and the colour of the raw meat turning greywhen the meat pH drops below 5.0 (Eilers et al., 1994;Seuss and Martin, 1993). Further investigation is nee-ded to determine the extent of these detrimental e�ectsof lactic acid injection on meat colour and ¯avour thatcould limit the acceptability of the treated meat.

Acknowledgements

This work was part of a collaborative research pro-gram (AIR CT92-0521) funded by the European Union.The authors are very grateful to Dr. L.B. Larsen, Dan-ish Institute of Agricultural Sciences, Research CentreFoulum, Denmark, who kindly supplied the antibodiesfor cathepsin D localization.

References

Beilken, S. L., & Harris, P. V. (1987). A comparison of isometric ten-

sion and di�erential scanning calorimetry measurements. Journal of

Food Science, 52, 1455±1460,1499.

Bergman, I., & Loxley, R. (1963). Two improved and simpli®ed

methods for the spectrophotometric determination of hydroxypro-

line. Analytical Chemistry, 35, 1961±1965.

Bonnet, M., & Kopp, J. (1984). Dosage du collageÁ ne dans les tissus

conjonctifs, la viande et les produits carne s. Cahiers Techniques de

l'INRA, 5, 19±30.

Bonnet, M., & Kopp, J. (1992). Pre paration des e chantillons pour le

dosage et la caracte risation du collageÁ ne musculaire. Viandes et

Produits CarneÂs, 13, 87±91.

Bouton, P. E., & Harris, P. V. (1972). The e�ects of cooking tem-

perature and time on some mechanical properties of meat. Journal

of Food Science, 37, 140±144.

Cannon, J. E., McKeith, F. K., Martin, S. E., Novakofski, J., & Carr,

T. R. (1993). Acceptability and shelf-life of marinated fresh and

precooked pork. Journal of Food Science, 58, 1249±1253.

Drans®eld, E. (1977). Intramuscular composition and texture of beef

muscles. Journal of the Science of Food and Agriculture, 28, 833±842.

Dumont, B. L. (1979). ProbleÁ mes pose s par la de ®nition et l'appre cia-

tion des qualite s de la viande de boeuf. Cahiers de Nutrition et de

DieÂteÂtique, 14, 167±169.

Eilers, J. D., Morgan, J. B., Martin, A. M., Miller, R. K., Hale, D. S.,

Acu�, G. R., & Savell, J. V. (1994). Evaluation of calcium chloride

and lactic acid injection on chemical, microbiological and descrip-

tive attributes of mature cow beef. Meat Science, 38, 443±451.

Ertbjerg, P., Larsen, L. M., & Mùller, A.J. (1995). Lactic acid treat-

ment for upgrading low quality beef. In Proceedings of the 41st

international congress of meat science and technology (Vol. 2) (pp.

670±671), 20±25 August 1995, San Antonio, USA.

Ertbjerg, P., Larsen, L. M., & Mùller, A. J. (1999). E�ect of prerigor

lactic acid treatment on lysosomal enzyme release in bovine muscle.

Journal of the Science of Food and Agriculture, 79, 95±100.

Ertbjerg, P., Mielche, M. M., Larsen, L. M., & Mùller, A. J. (1999).

Relationship between proteolytic changes and tenderness in prerigor

lactic acid marinated beef. Journal of the Science of Food and Agri-

culture, 79, 970±978.

Ertbjerg, P., Henckel, P., Karlsson, A., Larsen, L. M., & Mùller, A. J.

(1999). Combined e�ect of epinephrine and exercise on calpain/cal-

pastatin and cathepsin B and L activity in porcine longissimus

muscle. Journal of Animal Science, 77, 2428±2436.

356 P. Berge et al. /Meat Science 57 (2001) 347±357

Gault, N. F. S. (1985). The relationship between water-holding capa-

city and cooked meat tenderness in some beef muscles as in¯uenced

by acidic conditions below the ultimate pH.Meat Science, 15, 15±30.

Ho, C. Y., Stromer, M. H., & Robson, R. M. (1994). Identi®cation of

the 30 kDa polypeptide in post mortem skeletal muscle as a degra-

dation product of troponin-T. Biochimie, 76, 369±375.

Hu�-Lonergan, E., Mitsuhashi, T., Beekman, D. D., Parrish, J. F. C.,

Olson, D. G., & Robson, R. M. (1996). Proteolysis of speci®c mus-

cle structural proteins by m-calpain at low pH and temperature is

similar to degradation in postmortem bovine muscle. Journal of

Animal Science, 74, 993±1008.

Kijowski, J. (1993). Thermal transition temperature of connective tis-

sues from marinated spent hen drumsticks. International Journal of

Food Science and Technology, 28, 587±594.

Kijowski, J., & Mast, M. G. (1993). Tenderization of spent fowl

drumsticks by marination in weak organic solutions. International

Journal of Food Science and Technology, 28, 337- 342.

Kirschke, H., Wood, L., Roisen, F. J., & Bird, J. W. C. (1983).

Activity of lysosomal cysteine proteinase during di�erentiation of

rat skeletal muscle. Biochemistry Journal, 214, 814±877.

Kopp, J., Sale , P., & Bonnet, Y. (1977). ContractomeÁ tre pour l'e tude

des proprie te s physiques des ®bres conjonctives: tension iso-

me trique, degre de re ticulation, relaxation. Canadian Institute of

Food Science and Technology Journal, 10, 69±72.

Kotula, K. L., & Thelappurath, R. (1994). Microbiological and sen-

sory attributes of retail cuts of beef treated with acetic and lactic

acid solutions. Journal of Food Protection, 57, 665±670.

Lacourt, A., Obled, A., Deval, C., Ouali, A., & Valin, C. (1986). Post

mortem localization of lysosomal peptide hydrolase, cathepsin B.

In V. Turk, Cysteine proteinases and their inhibitors (pp. 239±248).

Berlin: Walter de Gruyter.

Lewis, G. J., & Purslow, P. P. (1991). The e�ect of marination and

cooking on the mechanical properties of intramuscular connective

tissue. Journal of Muscle Foods, 2, 177±195.

Martens, H., Stabussvik, E., & Martens, M. (1982). Texture and col-

our changes in meat during cooking related to thermal denaturation

of muscle proteins. Journal of Texture Studies, 13, 291±309.

Mùller, A. J. (1981). Analysis of Warner-Bratzler shear pattern with

regard to myo®brillar and connective tissue components of tender-

ness. Meat Science, 5, 247±260.

Moeller, P. W., Fields, P. A., Dutson, T. R., Landmann, A., & Car-

penter, Z. L. (1976). E�ect of high temperature conditioning on

subcellular distribution and levels of lysosomal enzymes. Journal of

Food Science, 41, 216±217.

O�er, G., & Knight, P. (1988). The structural basis of water-holding in

meat Ð Part 1. General principles and water uptake in meat pro-

cessing. In R. Lawrie, Developments in meat science-4 (pp. 63±171).

Barking, UK: Elsevier Science.

Oreskovich, D. C., Bechtel, P. J., McKeith, F. K., Novakofski, J., &

Basgall, E. J. (1992). Marinade pH a�ects textural properties of

beef. Journal of Food Science, 57, 305±311.

éstdal, H., Larsen, L. M., & Mùller, A. J. (1992). Lactic acid treat-

ment of pre-rigor beef and distribution of cathepsin B+L activity.

In Proceedings of the 38th international congress of meat science and

technology (Vol. 3) (pp. 403±406), 23±28 August 1992, Clermont-

Ferrand, France

Rao, M. V., & Gault, N. F. S. (1990). Acetic acid marinating Ð the

rheological characteristics of some raw and cooked beef muscles

which contribute to changes in meat tenderness. Journal of Texture

Studies, 21, 455±477.

Reynolds, E. S. (1963). The use of lead citrate at high pH as an elec-

tron opaque stain in electron microscopy. Journal of Cell Biology,

17, 208±212.

SAS (1989±1996). Software release 6.12. Cary, NC, USA.: SAS Insti-

tute

Seuss, I., & Martin, M. (1993). The in¯uence of marinating with food

acids on the composition and sensory properties of beef. Fleisch-

wirtschaft, 73, 292±295.

Stanton, C., & Light, N. D. (1990). The e�ects of conditioning on

meat collagen Ð Part 4. The use of pre-rigor lactic acid injection to

accelerate conditioning in bovine meat. Meat Science, 27, 141±159.

Wenham, L. M., & Locker, R. H. (1976). The e�ect of marinading

on beef. Journal of the Science of Food and Agriculture, 27, 1079±

1084.

P. Berge et al. /Meat Science 57 (2001) 347±357 357