- Email: [email protected]
The effects of acetic acid, gluconic acid and trisodium citrate treatment of beef trimmings on microbial, color and odor characteristics of ground beef through simulated retail display
Meat Science 60 (2002) 245–252 www.elsevier.com/locate/meatsci
The effects of acetic acid, gluconic acid and trisodium citrate treatment of beef trimmings on microbial, color and odor characteristics of ground beef through simulated retail display M.R. Stivariusa, F.W. Pohlmanb,*, K.S. McElyeab, J.K. Appleb a Griffith Laboratories, Griffith Center, Alsip, IL 60658, USA Department of Animal Science, University of Arkansas, Fayetteville, AR 72701, USA
b
Received 5 March 2001; received in revised form 4 May 2001; accepted 4 May 2001
Abstract Antimicrobial effects of selected acidulants in a ground beef production system were studied. Lean beef trimmings were inoculated with Escherichia coli (EC) and Salmonella Typhimurium (ST) then treated with either 5% acetic acid, 5% gluconic acid (GA) or 5% trisodium citrate and then compared with an untreated control (C). Trimmings were ground, packaged and sampled at 0, 1, 2, 3 and 7 days of display for EC, ST, coliforms, aerobic plate counts (APC), sensory color and odor as well as instrumental color traits. Acetic acid reduced (P< 0.05) all bacterial types evaluated, but caused changes (P< 0.05) in ground beef color (L*, a* and b* values) and odor characteristics. Conversely, although GA reduced (P< 0.05) EC and APC, it had little effect on color or odor characteristics as compared with C. Trisodium citrate did not affect (P>0.05) microbial populations, color or odor characteristics of ground beef. # 2002 Elsevier Science Ltd. All rights reserved. Keywords: Acetic acid; Gluconic acid; Trisodium citrate; Ground beef; Antimicrobial; Bacteria
1. Introduction The use of acidulants on beef carcasses before processing has been shown to reduce, but not eliminate microbial pathogens on carcass surfaces (Dorsa, Cutter, & Siragusa, 1997). Emswiler, Kotula, and Rough (1976) suggested that whenever beef carcasses are fabricated into retail cuts, any carcass surface microbial contamination is inoculated onto newly exposed surfaces. Dorsa, Cutter, and Siragusa (1998) demonstrated that the most important factor contributing to source and level of microbial contamination of ground beef was the microbial state of raw beef materials used for grinding. Recently researchers have begun to refocus attention on the use of acidulants in the processing of beef trimmings destined for ground beef, as a means to provide additional safety for the ground beef product (Dorsa et al., 1998; Ellebracht, Castillo, Lucia, Miller, & Acuff, 1999). An organic acid that has been frequently studied * Corresponding author. Tel.: +1-501-575-5634; fax: +1-501-5757294. E-mail address: [email protected] (F.W. Pohlman).
on beef tissues is acetic acid (Anderson & Marshall, 1990; Hardin, Acuff, Lucia, Oman, & Savell, 1995). Acetic acid, a preservative commonly known as vinegar (Luck & Jager, 1998), has effective antimicrobial capabilities due to its ability to lower the pH and cause instability of bacterial cell membranes (Luck & Jager, 1998). Specifically, acetic acid has been shown to be effective against Escherichia coli O157:H7 and Salmonella Typhimurium, reducing these pathogens by 0.1 log colony forming units (CFU)/g to 4.67 log CFU/cm2 for E. coli O157:H7 (Conner, Kotrola, Mikel, & Tamblyn, 1997; Cutter, Dorsa, & Siragusa, 1997) and by 0.73 log CFU/cm2 to 2.8 log CFU/cm2 for Salmonella Typhimurium, respectively, on carcass tissue surfaces (Bell, Marshall, & Anderson, 1986; Kochevar, Sofos, LeValley, & Smith, 1997). While acetic acid is effective for reducing bacterial populations on meat surfaces, mixed results for color deterioration of whole muscle cut surfaces have been reported (Bell et al., 1986; Mikel, Goddard, & Bradford, 1996). Currently, it is unclear what effect organic acids such as acetic acid might have on meat color when used as an antimicrobial agent in a ground meat system.
0309-1740/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved. PII: S0309-1740(01)00130-9
246
M.R. Stivarius et al. / Meat Science 60 (2002) 245–252
Gluconic acid, an organic acid commonly used as a preservative, has been studied briefly as an antimicrobial decontaminant for beef steaks (Garcia-Zepeda et al., 1994). Garcia-Zepeda et al. (1994) found that the use of 1.5% gluconic acid was effective against lactic acid bacteria and did not affect lightness (L*) color values but increased meat redness (a*) and decreased yellowness (b*) during initial display. Garcia-Zepeda et al. (1994) also demonstrated that the combination of gluconic acid and lactic acid at varied rates reduced psychrotrophic bacterial populations as well as lactic acid bacteria levels without adversely affecting color. Citric acid is commonly used to adjust pH of foods and as an antioxidant, which allows for its use as a preservative and possibly as an antimicrobial agent. Anderson et al. (1990) showed that citric acid in combination with acetic acid, lactic acid and ascorbic acid reduced aerobic plate counts, E. coli and Salmonella Typhimurium approximately 1, 0.5 and 1 log10, respectively. Arganosa and Marriott (1989) used citric acid in the production of restructured beef steaks and found that it caused steaks to become lighter (L*) and less red (a*) in color compared to controls. Unfortunately, little is known regarding the effects of trisodium citrate, a sodium salt of citric acid, on microbial reductions and color of meat when used as a decontamination agent. Trisodium citrate is generally recognized as safe (GRAS) when used as a food additive and is commonly used in soft drink and processed dairy product manufacturing. Therefore, due to the lack of documentation on the effectiveness of acetic and gluconic acids as well as trisodium citrate as microbial inhibitors in ground beef processing systems, the objective of the current research was to evaluate the effectiveness of acetic and gluconic acids and trisodium citrate on microbial, instrumental color and sensory color and odor characteristics of ground beef treated before grinding.
2. Materials and methods 2.1. Bacterial preparation and inoculation Inoculums were prepared from frozen ( 80 C) stock cultures of E. coli (ATCC No. 11775; EC) and a nalidixic acid resistant strain of Salmonella Typhimurium (ATCC No. 1769NR; ST). E. coli was maintained by brain heart infusion (BHI; Difco Laboratories, Detroit, MI, USA) broth with glycerol (20%) and Salmonella Typhimurium was maintained by BHI broth containing nalidixic acid (86 mmol; Fisher Scientific, Fairlawn, NJ, USA) with glycerol (20%). Frozen cultures of E. coli and Salmonella Typhimurium were thawed, and 0.1 ml of E. coli suspension was inoculated into separate 40-ml aliquots of BHI, and 0.1 ml of Salmonella Typhimurium suspension was inoculated into separate 40-ml aliquots
of BHI with nalidixic acid (86 mmol). After 18 h of incubation at 37 C, the bacteria were harvested by centrifugation (3649 g for 20 min at 37 C; Beckman GS-6 series, Fullerton, CA, USA), re-suspended in the same volume of 0.1% buffered peptone water (BPW; Difco Laboratories, Detroit, MI, USA) and then pooled together (1600 ml of E. coli and 1600 ml of Salmonella Typhimurium) to make a bacterial cocktail. The cocktail (3200 ml; log 107 CFU/ml E. coli and log 107 CFU/ ml Salmonella Typhimurium) was cooled to 4 C and combined with thawed, boneless cow beef trimmings (12.8 kg) and allowed to attach for 1 h under refrigeration (4 C). The meat was then drained, and separated into 3.2-kg batches and placed in a 4 C cooler for 12–14 h to allow further microbial attachment. 2.2. Antimicrobial treatment application and sample processing Antimicrobial treatments for this study included: (1) 5% (vol/vol) acetic acid solution (Shurfine Inc., Northlake, IL, USA; AA); (2) 5% (wt/vol) gluconic acid solution (PMP Fermentation Products, Peoria, IL, USA; GA); (3) 5% (wt/vol) trisodium citrate solution (ADM, Decatur, IL, USA; TSC); and (4) an untreated control (C). All antimicrobial treatments were prepared using deionized water with the exception of acetic acid, which was commercially prepared. For all antimicrobial applications, beef trimmings were placed into a Lyco meat tumbler (Model 4Q, Lyco Inc., Janesville, WI, USA) with 400 ml of the selected antimicrobial treatment, and aerobically tumbled for 3 min (16 rpm). Upon completion of the antimicrobial application phase, beef trimmings were removed from the tumbler and ground twice using a Hobart grinder (Model 310, Hobart Inc., Troy, OH, USA) with a 3.2-mm plate. The ground beef was then divided into 454-g samples and packaged on styrofoam trays with absorbent diapers. The trays were overwrapped with polyvinyl chloride film with an oxygen transmission rate of 1400 cc/m2/24 h/1 atm (Borden Inc., Dallas, TX, USA) and stored under simulated retail conditions (4 C; deluxe warm white fluorescent lighting, 1630 lx, Phillips Inc., Somerset, NJ, USA). Multiple trays of ground beef from each treatment were packaged to allow for independent package use for microbial, instrumental color and sensory color and odor analysis on each sampling day of display (days 0, 1, 2, 3, and 7). Fat content for all treatments were standardized to 10% and validated using a Hobart Fat Analyzer (Model F101, Hobart Inc. Troy, OH, USA). Treated ground beef pH was also sampled immediately after grinding by homogenizing a 1.8-g portion of ground beef in 18 ml of distilled water and evaluated using an Orion Model 420A pH meter with a ROSS electrode (Model 8165BN, Orion Research, Inc., Beverly, MA, USA).
M.R. Stivarius et al. / Meat Science 60 (2002) 245–252
2.3. Microbial sampling On days 0, 1, 2, 3, and 7 of simulated retail display, 25 g of ground beef was aseptically removed from the packages and placed into whirlpack bags (Nasco, Ft. Atkinson, WI, USA) with 225 ml of 0.1% buffered peptone water and buffered to a pH of 7 with sodium hydroxide. Samples were then stomached in a Model 400 Lab Stomacher (Seward, London, UK) for 2 min and serial dilutions made. Subsequent duplicate platings were made on Salmonella Shigella agar (Difco Laboratories, Detroit, MI, USA) containing nalidixic acid (86 mmol), Petrifilm1 (3M Corp., St. Paul, Minnesota, USA) aerobic plate count (APC) plates and Petrifilm1 E. coli/coliform plate count plates. Plates were then incubated at 37 C in an aerobic incubation chamber (either VWR Model 5015 or Model 3015 incubators, VWR Scientific, West Chester, Pennsylvania, USA) and APC along with Salmonella Shigella agar plates were read at 48 h, whereas E. coli/coliform plates were read at 24 h. Counts were recorded as colony forming units per gram (CFU/g). 2.4. Instrumental color On days 0, 1, 2, 3 and 7 of simulated retail display, instrumental color was measured using a HunterLab MiniScan XE Spectrocolorimeter, Model 4500L (Hunter Associates Laboratory Inc., Reston, WV, USA). Samples were read using illuminant A/10 observer and evaluated for CIE (L*, a* and b*) color values. In addition, reflectance measurements were taken in the visible spectrum from 580 to 630 nm. The reflectance ratio of 630/580 nm was calculated and used to estimate the oxymyoglobin proportion of the myoglobin pigment (Strange, Benedict, Gugger, Metzger, & Swift, 1974; Hunt et al., 1991). In addition, hue angle, which describes the hue or color of ground beef was calculated [tan 1(b*/a*)], as was the saturation index [(a*2+b*2)0.5], which describes the brightness or vividness of color (Hunt et al., 1991). Before use, the Spectrocolorimeter was standardized using white tile, black tile, and working standards. Eight measurements were taken of each sample and averaged for statistical analysis. 2.5. Sensory color and odor A six member trained sensory panel was used to evaluate sensory color and odor characteristics of ground beef samples through display. Panelists were selected and trained by an experienced panel leader according to the American Meat Science Association guidelines (AMSA, 1978; Hunt et al., 1991). On days 0, 1, 2, 3 and 7 of simulated retail display, sensory panelists evaluated overall color and worst point color (5=bright purplish red, 4=dull purple red, 3=slightly
247
brownish red, 2=moderately brownish red, and 1=brown) and percentage surface discoloration [7=no discoloration (0%), 6=slight discoloration (1–20%), 5=small discoloration (20–39%), 4=modest discoloration (40–59%), 3=moderate discoloration (60–79%), 2=extensive discoloration (80–95%), 1=total discoloration (96–100%)]. In addition, panelists evaluated beef odor (8=extremely beef like, 7=very beef like, 6=moderately beef like, 5=slightly beef like, 4=slightly non-beef like, 3=moderately non-beef like, 2=very non-beef like, and 1=extremely non-beef like) and off odor characteristics (5=no off odor, 4=slight off odor, 3=small off odor, 2=moderate off odor, and 1=extreme off odor; Hunt et al., 1991). For visual evaluation, packages were first viewed under simulated retail lighting (deluxe warm white fluorescent lighting, 1630 lx) conditions for overall color, worst point color and percentage discoloration. Packages were then taken to a static pressure room, opened, and evaluated by panelists for beef odor and off odor characteristics. 2.6. Statistical analysis The experiment was replicated three times. The randomized complete block factorial experiment was analyzed using the GLM procedure of SAS (1988). For sensory panel data, a panelist term was added to the model to account for sensory panelist variation. Treatments were blocked by replicate then analyzed for the main effects of antimicrobial type, day of display and appropriate interactions. For variables confounded by interactions, interaction means were generated, separated using the PDIFF option of SAS (1988), and then plotted. Least square means for all other variables were generated and separated using the PDIFF option of SAS (1988).
3. Results and discussion 3.1. Impact of antimicrobial treatments on microbial populations and instrumental color Although the effectiveness of organic acids against microbial pathogens on carcasses has been researched and reviewed (Dickson & Anderson, 1992; Siragusa, 1995), little is known regarding their effects in ground meat systems. The results for the effects of antimicrobial treatment application before grinding on ground beef microbial characteristics are presented in Table 1. Acetic acid (AA) treatment of beef trimmings before grinding reduced (P< 0.05) ground beef E. coli (EC), Salmonella Typhimurium (ST), coliforms (CO) and aerobic plate count (APC) by 0.9, 1.47, 1.25 and 1.25 log colony forming units (CFU)/g, respectively, compared with C. Acetic acid (AA) has previously been shown to be successful against Salmonella
248
M.R. Stivarius et al. / Meat Science 60 (2002) 245–252
L* and b* values are summarized in Table 2. Across days of display, ST was reduced in ground beef by 0.37 log CFU/g while EC, CO and APC remained unchanged (P>0.05). Also, across 7 days of display, ground beef became lighter (P < 0.05; L*) and less (P < 0.05) yellow (b*) in color. Ellebracht et al. (1999) found that ground beef did not change in lightness (L*) or yellowness (b*) with increased storage times, a finding that disagrees with Gill and Bandoni (1997) who found that beef was lighter and more yellow in color after 24 h of storage.
Typhimurium and E. coli O157:H7 for decontaminating beef carcasses (Dickson, 1992). Our findings support work by Dorsa et al. (1998) who used 2% acetic acid to reduce Salmonella Typhimurium and E. coli O157:H7 in ground beef by over 2.5 log CFU/g through 7 days of display. These results are also in agreement with Bell et al. (1986) and Kotula and Thelappurate (1994), who found similar reductions in aerobic bacteria on muscle surfaces after treatment with AA. Gluconic acid (GA) treatment of beef trimmings also reduced (P < 0.05) ground beef EC and APC by 0.25 and 0.52 log CFU/g, but had no effect (P>0.05) on ST or CO when compared with C. Trisodium citrate (TSC) did not reduce (P>0.05) any microorganisms studied, which reaffirms the review of Siragusa (1995) on the antimicrobial properties of TSC when used on meat. Ground beef from the GA and AA treatments were lighter (P < 0.05; L*) in color compared with C and TSC. However, AA was less (P < 0.05) yellow (b*) in color than C, GA or TSC treatments, which were not different (P>0.05) from each other (Table 1). Arganosa et al. (1989) reported restructured beef steaks treated with acetic or citric acids were lighter (L*) less red (a*) and not different in yellowness (b*) than controls. Similarly, Kotula et al. (1994) found beef steaks treated with 1.2% AA were lighter (L*) in color, but no difference in a* or b* values were apparent when compared with a control. Therefore, although AA treatment of beef trimmings before grinding was effective for microbe reductions in ground beef, high concentrations such as the ones used in this study, resulted in ground beef color changes.
3.3. Effect of antimicrobial treatments and duration of display on instrumental color and sensory color and odor characteristics Fig. 1 shows the day of display by antimicrobial treatment interaction for instrumental color measures. Ground beef from either AA or GA treatments tended to be less (P < 0.05) red (a*) and contain less oxymyoglobin (630/580 nm) than C (Fig. 1, panels a and b). Ground beef from the TSC treatment was similar (P>0.05) to C after day 0 of display for redness (a*) and oxymyoglobin content (630/580 nm), which is in agreement with Arganosa et al. (1989). Mikel et al. (1996) indicated that using organic acids caused darker colored lean, especially at lower pH values, in vacuum packaged beef steaks. They concluded these differences in results might be due to the amount, type and/or concentration of acid used. Ground beef pH in this study was 5.55 for C, 4.41 for AA, 5.42 for GA and 5.26 for TSC. Since ground beef pH was lower for the AA and GA treatments than C, this may have caused the lower redness values and oxymyoglobin contents for both of these treatments. In addition, the low pH value for ground beef from the AA treatment most likely caused this treatment to discolor sooner than any of the other antimicrobial treatments.
3.2. Effects of display on microbial growth and instrumental color The effect of duration of display, pooled across antimicrobial treatments, on microbial populations and CIE
Table 1 Effect of antimicrobial treatmentsa applied to beef trimmings prior to grinding on least square mean ( S.E.) log CFUb/g E. coli, coliform, Salmonella Typhimurium, aerobic plate count (APC), and CIE L*c and b*c values of ground beef through simulated retail display Treatment C Microorganism E. coli Coliform Salmonella Typhimurium APC Instrumental color CIE L* CIE b* a b c d
6.510.08zd 5.890.10z 5.700.11z 7.220.10z 46.240.24x 21.000.18z
AA
GA
5.610.08x 4.640.10y 4.230.11y 5.970.11x
6.22 0.08y 5.700.10z 5.600.11z 6.700.11y
48.220.24y 17.230.18y
50.120.24z 20.480.18z
C=Control; AA=5% acetic acid; GA=5% gluconic acid; TSC=5% trisodium citrate. Colony forming unit. L*: 0=black and 100=white; b*: 60=blue and +60=yellow. Least square means within a row bearing different letters are different (P< 0.05).
TSC 6.370.08yz 5.840.10z 5.520.11z 6.990.10z 45.590.24x 20.790.18z
249
M.R. Stivarius et al. / Meat Science 60 (2002) 245–252
Acetic acid (AA) treated ground beef maintained a larger (P < 0.05) hue angle than C through display (Fig. 1, panel c). Likewise, ground beef from the GA treatment tended to have a larger numeric hue angle value, yet was not different (P>0.05) until day 7 of display. In addition, the TSC treatment never differed (P>0.05) from C for hue angle value through display. This may be explained by the lack of differences (P>0.05) in a* (Fig. 1, panel a) or b* (Table 1) values used to calculate hue angle, when comparing TSC to C. Fig. 1, panel d shows the day of display by antimicrobial treatment
interaction on the saturation index of ground beef. Ground beef from the AA treatment was less (P < 0.05) vivid in color (saturation index), whereas ground beef from the GA treatment was similar in vividness (P>0.05) on days 0, 2 and 3 of display when compared with C. Likewise, ground beef from the TSC treatment was never different (P < 0.05) in saturation index compared to C through display. Day of display by antimicrobial treatment interaction for sensory evaluated color and odor characteristics are presented in Fig. 2. Sensory panelists found that ground
Table 2 Effect of duration of display, pooled across antimicrobial treatments, on least square mean ( S.E.) log CFUa/g E. coli, coliform, Salmonella Typhimurium, aerobic plate count (APC), CIE L*b and b*b values of ground beef Day of display 0 Microorganism E. coli Coliform Salmonella Typhimurium APC Instrumental color CIE L* CIE b* a b c
6.19 0.09 5.54 0.14 5.32 0.12yc 6.66 0.10 46.190.27x 21.020.21z
1
2
3
7
6.12 0.09 5.42 0.16 5.44 0.12y 6.71 0.10
6.23 0.09 5.64 0.15 5.47 0.12y 6.92 0.11
6.25 0.09 5.61 0.14 5.13 0.12xy 6.70 0.12
6.10 0.09 5.38 0.14 4.95 0.12x 6.62 0.10
47.440.27y 21.520.21z
48.220.27z 20.010.21y
48.110.27yz 18.230.21x
47.770.77yz 18.590.21x
Colony forming units. L*: 0=black and 100=white; b*: 60=blue and +60=yellow. Least square means within a row bearing different letters are different (P<0.05).
Fig. 1. Day of display by antimicrobial treatment interaction effect on the least square mean ( S.E.) CIE a* value (a), 630/580 nm reflectance ratio (b), hue angle (c) and saturation index (d) of ground beef through simulated display. abcLeast square means within a day bearing different superscripts are different (P<0.05). dC=control, eAA=5% acetic acid, fGA=5% gluconic acid and gTSC=5% trisodium citrate. h 60=green and +60=red. iCalculated as 630/580 nm reflectance. jCalculated as tan 1(b*/a*). kCalculated as (a*2+b*2)0.5.
250
M.R. Stivarius et al. / Meat Science 60 (2002) 245–252
beef treated with AA was less (P < 0.05) bright purple red in overall color (Fig. 2, panel a) and worst point color (Fig. 2, panel b), and maintained a higher (P < 0.05) percentage discoloration (Fig. 2, panel c) than C through display. These findings are in agreement with Mikel et al. (1996), who reported visual appraisal of acid treated strip loins were browner in color when compared with a control. Acetic acid (AA) treated ground beef lost the desirable red color faster than any other treatment, which is in agreement with Bell et al. (1986) who found that beef cubes treated with 1.2% AA discolored immediately. Kotula et al. (1994), however, found no difference in beef rib steaks treated with 1.2% AA when looking at sensory evaluated color. The variation in
results may be explained by the difference in surface area exposed to AA treatment, with beef trimmings and beef cubes having a larger surface area per unit mass than beef rib steaks. Gluconic acid (GA) treated ground beef was similar (P>0.05) in overall color to C on days 0, 1 and 3 of display (Fig. 2, panel a). However, sensory panelists detected (P < 0.05) small overall color differences between GA and C treatments on days 2 and 7 of display. Likewise, sensory panelists found small differences (P < 0.05) in worst point color between GA and C treatments on days 0, 2 and 7 of display (Fig. 2, panel b). However, panelists were unable to detect any difference (P>0.05) in percentage discoloration between GA
Fig. 2. Day of display by antimicrobial treatment interaction effect on the least square mean ( S.E.) sensory evaluated (a) overall color, (b) worst point color, (c) percentage discoloration, (d) beef odor and (e) off odor characteristics of ground beef through simulated display. abcLeast square means within a day bearing different superscripts are different (P<0.05). dC=control, eAA=5% acetic acid, fGA=5% gluconic acid and g TSC=5% trisodium citrate. hColor score: 1=brown and 5=bright purple red. iPercentage discoloration: 1=total discoloration (96–100%) and 7=no discoloration (0%). jBeef odor score: 1=extreme non-beef like and 8=extreme beef like. kOff odor score:1=extreme off odor and 5=no off odor.
M.R. Stivarius et al. / Meat Science 60 (2002) 245–252
and C treatments until day 7 of display (Fig. 2, panel c). These findings tend to disagree with Garcia-Zepeda et al. (1994) who found that 3% GA treated beef was significantly browner in color on day 0 than the control. Conversely, trisodium citrate showed no detrimental effects on ground beef overall color (Fig. 2, panel a), worst point color (Fig. 2, panel b) or percentage discoloration (Fig. 2, panel c) and was equal to or superior in color characteristics when compared with C. Sensory panelists also evaluated beef odor characteristics and found that TSC and GA treatments were not different (P>0.05) for this attribute when compared with C, but that ground beef from the AA treatment had a less (P < 0.05) beef like odor when compared with all other treatments through display (Fig. 2, panel d). Panelists also reported that ground beef from the AA treatment possessed more (P < 0.05) off odor than did all other treatments through display, and that TSC was not different (P>0.05) in off odor than C through 7 days of display (Fig. 2, panel e). Likewise, GA was not different from C in off odor (P>0.05) through 3 days of display, but did display less (P < 0.05) off odor than C by day 7 of display.
4. Conclusion Consumer demand for a safe meat supply obviates the need for antimicrobial measures in the manufacturing process, as well as more research on the color and odor stability of ground beef. Results from this study indicate that AA treatment of beef trimmings before grinding can reduce EC, CO, ST and APC in ground beef by approximately 1 log each. Likewise, GA is also effective for reducing EC and APC in ground beef when applied to beef trimmings before grinding, but these decreases are not sufficient to warrant industry adoption. Although 5% GA had little impact on ground beef color and odor characteristics, the use of 5% AA did elicit color and odor changes. Therefore, while 5% GA can be expected to slightly reduce EC and APC in ground beef with minimal quality changes, additional work is required to determine the optimal level of acetic acid usage in order to maintain ground beef color and odor characteristics, while reducing microbial pathogens.
Acknowledgements Appreciation is expressed to the Arkansas Beef Council for funding this research. The authors would also like to thank J. Davis, L. Rakes, A. Ivey, L. McBeth, R. Story, and E. Kroger for their assistance in conducting these trials, and special appreciation is extended to Z. Johnson for assistance in data analysis.
251
References AMSA. (1978). Guidelines for cookery, sensory evaluation of meat. Chicago, IL: Am. Meat Sci. Assoc. and National Live Stock and Meat Board. Anderson, M. E., & Marshall, R. T. (1990). Reducing microbial populations on beef tissues: concentration and temperature of an acid mixture. Journal of Food Science, 55(4), 903–905. Arganosa, G. C., & Marriott, N. G. (1989). Organic acids as tenderizers of collagen in restructured beef. Journal of Food Science, 54(5), 1173–1176. Bell, M. F., Marshall, R. T., & Anderson, M. E. (1986). Microbiological and sensory tests of beef treated with acetic and formic acids. Journal of Food Protection, 49(3), 207–210. Conner, D. E., Kotrola, J. S., Mikel, W. B., & Tamblyn, K. C. (1997). Effect of acetic-lactic acid treatments applied to beef trim on populations of Escherichia coli O157:H7 and Listeria monocytogenes in ground beef. Journal of Food Protection, 60(12), 1560–1563. Cutter, C. N., Dorsa, W. J., & Siragusa, G. R. (1997). Parameters affecting the efficacy of spray washes against Escherichia coli O157:H7 and fecal contamination. Journal of Food Protection, 60(6), 614–618. Dickson, J. S. (1992). Acetic acid action on beef tissue surfaces contaminated with Salmonella typhimurium. Journal of Food Science, 57(2), 297–301. Dickson, J. S., & Anderson, M. E. (1992). Microbiological decontamination of food animal carcasses by washing and sanitizing systems: a review. Journal of Food Protection, 55(2), 133–140. Dorsa, W. J., Cutter, C. N., & Siragusa, G. R. (1997). Effects of acetic acid, lactic acid and trisodium phosphate on the microflora of refrigerated beef carcass surface tissue inoculated with Escherichia coli O157:H7, Listeria innocua, and Clostridium sporogenes. Journal of Food Protection, 60(6), 610–624. Dorsa, W. J., Cutter, C. N., & Siragusa, G. R. (1998). Bacterial profile of ground beef made from carcass tissue experimentally contaminated with pathogenic and spoilage bacteria before being washed with hot water, alkaline solution, or organic acid and then stored at 4 or 12 C. Journal of Food Protection, 61(9), 1109–1118. Ellebracht, E. A., Castillo, A., Lucia, L. M., Miller, R. K., & Acuff, G. R. (1999). Reduction of pathogens using hot water and lactic acid on beef trimmings. Journal of Food Science, 64(6), 1094– 1099. Emswiler, B. S., Kotula, A. W., & Rough, D. K. (1976). Bactericidal effectiveness of three chlorine sources used in beef carcass washing. Journal of Animal Science, 42(6), 1445–1450. Gill, C. O., & Bandoni, M. (1997). The effects of hot water pasteurizing treatments on the appearances of pork and beef. Journal of Meat Science, 46(1), 77–87. Garcia-Zepeda, C. M., Kastner, C. L., Willard, B. L., Phebus, R. K., Schewnke, J. R., Fijal, B. A., & Prasai, R. K. (1994). Gluconic acid as a fresh beef decontaminant. Journal of Food Protection, 57(11), 956–962. Hardin, M. D., Acuff, G. R., Lucia, L. M., Oman, J. S., & Savell, J. W. (1995). Comparison of methods for decontamination from beef carcass surfaces. Journal of Food Protection, 58(4), 368–374. Hunt, M. C., Acton, J. C., Benedict, R. C., Calkins, C. R., Cornforth, D. P., Jeremiah, L. E., Olson, D. G., Salm, C. P., Savell, J. W., & Shivas, S. D. (1991). AMSA guidelines for meat color evaluation. In E. Schofield (Ed.), Proceedings of the Forty-Fourth Annual Reciprocal Meat Conference (pp. 3–17). Manhattan, Kansas: Kansas State University (9–12 July 1991). Kochevar, S. L., Sofos, J. N., LeValley, S. B., & Smith, G. C. (1997). Effect of water temperature, pressure and chemical solution of removal of fecal material and bacteria from lamb adipose tissue by spray-washing. Meat Science, 45(3), 377–388. Kotula, K. L., & Thelappurate, R. (1994). Microbiological and sensory
252
M.R. Stivarius et al. / Meat Science 60 (2002) 245–252
attributes of retail cuts of beef treated with acetic and lactic acid solution. Journal of Food Protection, 57(8), 665–670. Luck, E., & Jager, M. (1998). In G.F. Edwards (Ed.), Antimicrobial food additives (2nd ed.). New York: Springer-Verlag (pp. 137–144, 116–119, 42–43,239, 98–101). Mikel, W. B., Goddard, B. L., & Bradford, D. D. (1996). Muscle microstructure and sensory attributes of organic acid treated beef strip loins. Journal of Food Science, 61(5), 1058–1061,1093.
SAS. (1988). SAS/STAT user’s guide: release 6.12. Cary, NC: SAS Institute, Inc. Siragusa, G. R. (1995). The effectiveness of carcass decontamination systems for controlling the presence of pathogens on the surfaces of meat animal carcasses. Journal of Food Safety, 15, 229–238. Strange, E. D., Benedict, R. C., Gugger, R. E., Metzger, V. G., & Swift, C. E. (1974). Simplified methodology for measuring meat color. Journal of Food Science, 39, 988–992.
The effects of acetic acid, gluconic acid and trisodium citrate treatment of beef trimmings on microbial, color and odor characteristics of ground beef through simulated retail display M.R. Stivariusa, F.W. Pohlmanb,*, K.S. McElyeab, J.K. Appleb a Griffith Laboratories, Griffith Center, Alsip, IL 60658, USA Department of Animal Science, University of Arkansas, Fayetteville, AR 72701, USA
b
Received 5 March 2001; received in revised form 4 May 2001; accepted 4 May 2001
Abstract Antimicrobial effects of selected acidulants in a ground beef production system were studied. Lean beef trimmings were inoculated with Escherichia coli (EC) and Salmonella Typhimurium (ST) then treated with either 5% acetic acid, 5% gluconic acid (GA) or 5% trisodium citrate and then compared with an untreated control (C). Trimmings were ground, packaged and sampled at 0, 1, 2, 3 and 7 days of display for EC, ST, coliforms, aerobic plate counts (APC), sensory color and odor as well as instrumental color traits. Acetic acid reduced (P< 0.05) all bacterial types evaluated, but caused changes (P< 0.05) in ground beef color (L*, a* and b* values) and odor characteristics. Conversely, although GA reduced (P< 0.05) EC and APC, it had little effect on color or odor characteristics as compared with C. Trisodium citrate did not affect (P>0.05) microbial populations, color or odor characteristics of ground beef. # 2002 Elsevier Science Ltd. All rights reserved. Keywords: Acetic acid; Gluconic acid; Trisodium citrate; Ground beef; Antimicrobial; Bacteria
1. Introduction The use of acidulants on beef carcasses before processing has been shown to reduce, but not eliminate microbial pathogens on carcass surfaces (Dorsa, Cutter, & Siragusa, 1997). Emswiler, Kotula, and Rough (1976) suggested that whenever beef carcasses are fabricated into retail cuts, any carcass surface microbial contamination is inoculated onto newly exposed surfaces. Dorsa, Cutter, and Siragusa (1998) demonstrated that the most important factor contributing to source and level of microbial contamination of ground beef was the microbial state of raw beef materials used for grinding. Recently researchers have begun to refocus attention on the use of acidulants in the processing of beef trimmings destined for ground beef, as a means to provide additional safety for the ground beef product (Dorsa et al., 1998; Ellebracht, Castillo, Lucia, Miller, & Acuff, 1999). An organic acid that has been frequently studied * Corresponding author. Tel.: +1-501-575-5634; fax: +1-501-5757294. E-mail address: [email protected] (F.W. Pohlman).
on beef tissues is acetic acid (Anderson & Marshall, 1990; Hardin, Acuff, Lucia, Oman, & Savell, 1995). Acetic acid, a preservative commonly known as vinegar (Luck & Jager, 1998), has effective antimicrobial capabilities due to its ability to lower the pH and cause instability of bacterial cell membranes (Luck & Jager, 1998). Specifically, acetic acid has been shown to be effective against Escherichia coli O157:H7 and Salmonella Typhimurium, reducing these pathogens by 0.1 log colony forming units (CFU)/g to 4.67 log CFU/cm2 for E. coli O157:H7 (Conner, Kotrola, Mikel, & Tamblyn, 1997; Cutter, Dorsa, & Siragusa, 1997) and by 0.73 log CFU/cm2 to 2.8 log CFU/cm2 for Salmonella Typhimurium, respectively, on carcass tissue surfaces (Bell, Marshall, & Anderson, 1986; Kochevar, Sofos, LeValley, & Smith, 1997). While acetic acid is effective for reducing bacterial populations on meat surfaces, mixed results for color deterioration of whole muscle cut surfaces have been reported (Bell et al., 1986; Mikel, Goddard, & Bradford, 1996). Currently, it is unclear what effect organic acids such as acetic acid might have on meat color when used as an antimicrobial agent in a ground meat system.
0309-1740/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved. PII: S0309-1740(01)00130-9
246
M.R. Stivarius et al. / Meat Science 60 (2002) 245–252
Gluconic acid, an organic acid commonly used as a preservative, has been studied briefly as an antimicrobial decontaminant for beef steaks (Garcia-Zepeda et al., 1994). Garcia-Zepeda et al. (1994) found that the use of 1.5% gluconic acid was effective against lactic acid bacteria and did not affect lightness (L*) color values but increased meat redness (a*) and decreased yellowness (b*) during initial display. Garcia-Zepeda et al. (1994) also demonstrated that the combination of gluconic acid and lactic acid at varied rates reduced psychrotrophic bacterial populations as well as lactic acid bacteria levels without adversely affecting color. Citric acid is commonly used to adjust pH of foods and as an antioxidant, which allows for its use as a preservative and possibly as an antimicrobial agent. Anderson et al. (1990) showed that citric acid in combination with acetic acid, lactic acid and ascorbic acid reduced aerobic plate counts, E. coli and Salmonella Typhimurium approximately 1, 0.5 and 1 log10, respectively. Arganosa and Marriott (1989) used citric acid in the production of restructured beef steaks and found that it caused steaks to become lighter (L*) and less red (a*) in color compared to controls. Unfortunately, little is known regarding the effects of trisodium citrate, a sodium salt of citric acid, on microbial reductions and color of meat when used as a decontamination agent. Trisodium citrate is generally recognized as safe (GRAS) when used as a food additive and is commonly used in soft drink and processed dairy product manufacturing. Therefore, due to the lack of documentation on the effectiveness of acetic and gluconic acids as well as trisodium citrate as microbial inhibitors in ground beef processing systems, the objective of the current research was to evaluate the effectiveness of acetic and gluconic acids and trisodium citrate on microbial, instrumental color and sensory color and odor characteristics of ground beef treated before grinding.
2. Materials and methods 2.1. Bacterial preparation and inoculation Inoculums were prepared from frozen ( 80 C) stock cultures of E. coli (ATCC No. 11775; EC) and a nalidixic acid resistant strain of Salmonella Typhimurium (ATCC No. 1769NR; ST). E. coli was maintained by brain heart infusion (BHI; Difco Laboratories, Detroit, MI, USA) broth with glycerol (20%) and Salmonella Typhimurium was maintained by BHI broth containing nalidixic acid (86 mmol; Fisher Scientific, Fairlawn, NJ, USA) with glycerol (20%). Frozen cultures of E. coli and Salmonella Typhimurium were thawed, and 0.1 ml of E. coli suspension was inoculated into separate 40-ml aliquots of BHI, and 0.1 ml of Salmonella Typhimurium suspension was inoculated into separate 40-ml aliquots
of BHI with nalidixic acid (86 mmol). After 18 h of incubation at 37 C, the bacteria were harvested by centrifugation (3649 g for 20 min at 37 C; Beckman GS-6 series, Fullerton, CA, USA), re-suspended in the same volume of 0.1% buffered peptone water (BPW; Difco Laboratories, Detroit, MI, USA) and then pooled together (1600 ml of E. coli and 1600 ml of Salmonella Typhimurium) to make a bacterial cocktail. The cocktail (3200 ml; log 107 CFU/ml E. coli and log 107 CFU/ ml Salmonella Typhimurium) was cooled to 4 C and combined with thawed, boneless cow beef trimmings (12.8 kg) and allowed to attach for 1 h under refrigeration (4 C). The meat was then drained, and separated into 3.2-kg batches and placed in a 4 C cooler for 12–14 h to allow further microbial attachment. 2.2. Antimicrobial treatment application and sample processing Antimicrobial treatments for this study included: (1) 5% (vol/vol) acetic acid solution (Shurfine Inc., Northlake, IL, USA; AA); (2) 5% (wt/vol) gluconic acid solution (PMP Fermentation Products, Peoria, IL, USA; GA); (3) 5% (wt/vol) trisodium citrate solution (ADM, Decatur, IL, USA; TSC); and (4) an untreated control (C). All antimicrobial treatments were prepared using deionized water with the exception of acetic acid, which was commercially prepared. For all antimicrobial applications, beef trimmings were placed into a Lyco meat tumbler (Model 4Q, Lyco Inc., Janesville, WI, USA) with 400 ml of the selected antimicrobial treatment, and aerobically tumbled for 3 min (16 rpm). Upon completion of the antimicrobial application phase, beef trimmings were removed from the tumbler and ground twice using a Hobart grinder (Model 310, Hobart Inc., Troy, OH, USA) with a 3.2-mm plate. The ground beef was then divided into 454-g samples and packaged on styrofoam trays with absorbent diapers. The trays were overwrapped with polyvinyl chloride film with an oxygen transmission rate of 1400 cc/m2/24 h/1 atm (Borden Inc., Dallas, TX, USA) and stored under simulated retail conditions (4 C; deluxe warm white fluorescent lighting, 1630 lx, Phillips Inc., Somerset, NJ, USA). Multiple trays of ground beef from each treatment were packaged to allow for independent package use for microbial, instrumental color and sensory color and odor analysis on each sampling day of display (days 0, 1, 2, 3, and 7). Fat content for all treatments were standardized to 10% and validated using a Hobart Fat Analyzer (Model F101, Hobart Inc. Troy, OH, USA). Treated ground beef pH was also sampled immediately after grinding by homogenizing a 1.8-g portion of ground beef in 18 ml of distilled water and evaluated using an Orion Model 420A pH meter with a ROSS electrode (Model 8165BN, Orion Research, Inc., Beverly, MA, USA).
M.R. Stivarius et al. / Meat Science 60 (2002) 245–252
2.3. Microbial sampling On days 0, 1, 2, 3, and 7 of simulated retail display, 25 g of ground beef was aseptically removed from the packages and placed into whirlpack bags (Nasco, Ft. Atkinson, WI, USA) with 225 ml of 0.1% buffered peptone water and buffered to a pH of 7 with sodium hydroxide. Samples were then stomached in a Model 400 Lab Stomacher (Seward, London, UK) for 2 min and serial dilutions made. Subsequent duplicate platings were made on Salmonella Shigella agar (Difco Laboratories, Detroit, MI, USA) containing nalidixic acid (86 mmol), Petrifilm1 (3M Corp., St. Paul, Minnesota, USA) aerobic plate count (APC) plates and Petrifilm1 E. coli/coliform plate count plates. Plates were then incubated at 37 C in an aerobic incubation chamber (either VWR Model 5015 or Model 3015 incubators, VWR Scientific, West Chester, Pennsylvania, USA) and APC along with Salmonella Shigella agar plates were read at 48 h, whereas E. coli/coliform plates were read at 24 h. Counts were recorded as colony forming units per gram (CFU/g). 2.4. Instrumental color On days 0, 1, 2, 3 and 7 of simulated retail display, instrumental color was measured using a HunterLab MiniScan XE Spectrocolorimeter, Model 4500L (Hunter Associates Laboratory Inc., Reston, WV, USA). Samples were read using illuminant A/10 observer and evaluated for CIE (L*, a* and b*) color values. In addition, reflectance measurements were taken in the visible spectrum from 580 to 630 nm. The reflectance ratio of 630/580 nm was calculated and used to estimate the oxymyoglobin proportion of the myoglobin pigment (Strange, Benedict, Gugger, Metzger, & Swift, 1974; Hunt et al., 1991). In addition, hue angle, which describes the hue or color of ground beef was calculated [tan 1(b*/a*)], as was the saturation index [(a*2+b*2)0.5], which describes the brightness or vividness of color (Hunt et al., 1991). Before use, the Spectrocolorimeter was standardized using white tile, black tile, and working standards. Eight measurements were taken of each sample and averaged for statistical analysis. 2.5. Sensory color and odor A six member trained sensory panel was used to evaluate sensory color and odor characteristics of ground beef samples through display. Panelists were selected and trained by an experienced panel leader according to the American Meat Science Association guidelines (AMSA, 1978; Hunt et al., 1991). On days 0, 1, 2, 3 and 7 of simulated retail display, sensory panelists evaluated overall color and worst point color (5=bright purplish red, 4=dull purple red, 3=slightly
247
brownish red, 2=moderately brownish red, and 1=brown) and percentage surface discoloration [7=no discoloration (0%), 6=slight discoloration (1–20%), 5=small discoloration (20–39%), 4=modest discoloration (40–59%), 3=moderate discoloration (60–79%), 2=extensive discoloration (80–95%), 1=total discoloration (96–100%)]. In addition, panelists evaluated beef odor (8=extremely beef like, 7=very beef like, 6=moderately beef like, 5=slightly beef like, 4=slightly non-beef like, 3=moderately non-beef like, 2=very non-beef like, and 1=extremely non-beef like) and off odor characteristics (5=no off odor, 4=slight off odor, 3=small off odor, 2=moderate off odor, and 1=extreme off odor; Hunt et al., 1991). For visual evaluation, packages were first viewed under simulated retail lighting (deluxe warm white fluorescent lighting, 1630 lx) conditions for overall color, worst point color and percentage discoloration. Packages were then taken to a static pressure room, opened, and evaluated by panelists for beef odor and off odor characteristics. 2.6. Statistical analysis The experiment was replicated three times. The randomized complete block factorial experiment was analyzed using the GLM procedure of SAS (1988). For sensory panel data, a panelist term was added to the model to account for sensory panelist variation. Treatments were blocked by replicate then analyzed for the main effects of antimicrobial type, day of display and appropriate interactions. For variables confounded by interactions, interaction means were generated, separated using the PDIFF option of SAS (1988), and then plotted. Least square means for all other variables were generated and separated using the PDIFF option of SAS (1988).
3. Results and discussion 3.1. Impact of antimicrobial treatments on microbial populations and instrumental color Although the effectiveness of organic acids against microbial pathogens on carcasses has been researched and reviewed (Dickson & Anderson, 1992; Siragusa, 1995), little is known regarding their effects in ground meat systems. The results for the effects of antimicrobial treatment application before grinding on ground beef microbial characteristics are presented in Table 1. Acetic acid (AA) treatment of beef trimmings before grinding reduced (P< 0.05) ground beef E. coli (EC), Salmonella Typhimurium (ST), coliforms (CO) and aerobic plate count (APC) by 0.9, 1.47, 1.25 and 1.25 log colony forming units (CFU)/g, respectively, compared with C. Acetic acid (AA) has previously been shown to be successful against Salmonella
248
M.R. Stivarius et al. / Meat Science 60 (2002) 245–252
L* and b* values are summarized in Table 2. Across days of display, ST was reduced in ground beef by 0.37 log CFU/g while EC, CO and APC remained unchanged (P>0.05). Also, across 7 days of display, ground beef became lighter (P < 0.05; L*) and less (P < 0.05) yellow (b*) in color. Ellebracht et al. (1999) found that ground beef did not change in lightness (L*) or yellowness (b*) with increased storage times, a finding that disagrees with Gill and Bandoni (1997) who found that beef was lighter and more yellow in color after 24 h of storage.
Typhimurium and E. coli O157:H7 for decontaminating beef carcasses (Dickson, 1992). Our findings support work by Dorsa et al. (1998) who used 2% acetic acid to reduce Salmonella Typhimurium and E. coli O157:H7 in ground beef by over 2.5 log CFU/g through 7 days of display. These results are also in agreement with Bell et al. (1986) and Kotula and Thelappurate (1994), who found similar reductions in aerobic bacteria on muscle surfaces after treatment with AA. Gluconic acid (GA) treatment of beef trimmings also reduced (P < 0.05) ground beef EC and APC by 0.25 and 0.52 log CFU/g, but had no effect (P>0.05) on ST or CO when compared with C. Trisodium citrate (TSC) did not reduce (P>0.05) any microorganisms studied, which reaffirms the review of Siragusa (1995) on the antimicrobial properties of TSC when used on meat. Ground beef from the GA and AA treatments were lighter (P < 0.05; L*) in color compared with C and TSC. However, AA was less (P < 0.05) yellow (b*) in color than C, GA or TSC treatments, which were not different (P>0.05) from each other (Table 1). Arganosa et al. (1989) reported restructured beef steaks treated with acetic or citric acids were lighter (L*) less red (a*) and not different in yellowness (b*) than controls. Similarly, Kotula et al. (1994) found beef steaks treated with 1.2% AA were lighter (L*) in color, but no difference in a* or b* values were apparent when compared with a control. Therefore, although AA treatment of beef trimmings before grinding was effective for microbe reductions in ground beef, high concentrations such as the ones used in this study, resulted in ground beef color changes.
3.3. Effect of antimicrobial treatments and duration of display on instrumental color and sensory color and odor characteristics Fig. 1 shows the day of display by antimicrobial treatment interaction for instrumental color measures. Ground beef from either AA or GA treatments tended to be less (P < 0.05) red (a*) and contain less oxymyoglobin (630/580 nm) than C (Fig. 1, panels a and b). Ground beef from the TSC treatment was similar (P>0.05) to C after day 0 of display for redness (a*) and oxymyoglobin content (630/580 nm), which is in agreement with Arganosa et al. (1989). Mikel et al. (1996) indicated that using organic acids caused darker colored lean, especially at lower pH values, in vacuum packaged beef steaks. They concluded these differences in results might be due to the amount, type and/or concentration of acid used. Ground beef pH in this study was 5.55 for C, 4.41 for AA, 5.42 for GA and 5.26 for TSC. Since ground beef pH was lower for the AA and GA treatments than C, this may have caused the lower redness values and oxymyoglobin contents for both of these treatments. In addition, the low pH value for ground beef from the AA treatment most likely caused this treatment to discolor sooner than any of the other antimicrobial treatments.
3.2. Effects of display on microbial growth and instrumental color The effect of duration of display, pooled across antimicrobial treatments, on microbial populations and CIE
Table 1 Effect of antimicrobial treatmentsa applied to beef trimmings prior to grinding on least square mean ( S.E.) log CFUb/g E. coli, coliform, Salmonella Typhimurium, aerobic plate count (APC), and CIE L*c and b*c values of ground beef through simulated retail display Treatment C Microorganism E. coli Coliform Salmonella Typhimurium APC Instrumental color CIE L* CIE b* a b c d
6.510.08zd 5.890.10z 5.700.11z 7.220.10z 46.240.24x 21.000.18z
AA
GA
5.610.08x 4.640.10y 4.230.11y 5.970.11x
6.22 0.08y 5.700.10z 5.600.11z 6.700.11y
48.220.24y 17.230.18y
50.120.24z 20.480.18z
C=Control; AA=5% acetic acid; GA=5% gluconic acid; TSC=5% trisodium citrate. Colony forming unit. L*: 0=black and 100=white; b*: 60=blue and +60=yellow. Least square means within a row bearing different letters are different (P< 0.05).
TSC 6.370.08yz 5.840.10z 5.520.11z 6.990.10z 45.590.24x 20.790.18z
249
M.R. Stivarius et al. / Meat Science 60 (2002) 245–252
Acetic acid (AA) treated ground beef maintained a larger (P < 0.05) hue angle than C through display (Fig. 1, panel c). Likewise, ground beef from the GA treatment tended to have a larger numeric hue angle value, yet was not different (P>0.05) until day 7 of display. In addition, the TSC treatment never differed (P>0.05) from C for hue angle value through display. This may be explained by the lack of differences (P>0.05) in a* (Fig. 1, panel a) or b* (Table 1) values used to calculate hue angle, when comparing TSC to C. Fig. 1, panel d shows the day of display by antimicrobial treatment
interaction on the saturation index of ground beef. Ground beef from the AA treatment was less (P < 0.05) vivid in color (saturation index), whereas ground beef from the GA treatment was similar in vividness (P>0.05) on days 0, 2 and 3 of display when compared with C. Likewise, ground beef from the TSC treatment was never different (P < 0.05) in saturation index compared to C through display. Day of display by antimicrobial treatment interaction for sensory evaluated color and odor characteristics are presented in Fig. 2. Sensory panelists found that ground
Table 2 Effect of duration of display, pooled across antimicrobial treatments, on least square mean ( S.E.) log CFUa/g E. coli, coliform, Salmonella Typhimurium, aerobic plate count (APC), CIE L*b and b*b values of ground beef Day of display 0 Microorganism E. coli Coliform Salmonella Typhimurium APC Instrumental color CIE L* CIE b* a b c
6.19 0.09 5.54 0.14 5.32 0.12yc 6.66 0.10 46.190.27x 21.020.21z
1
2
3
7
6.12 0.09 5.42 0.16 5.44 0.12y 6.71 0.10
6.23 0.09 5.64 0.15 5.47 0.12y 6.92 0.11
6.25 0.09 5.61 0.14 5.13 0.12xy 6.70 0.12
6.10 0.09 5.38 0.14 4.95 0.12x 6.62 0.10
47.440.27y 21.520.21z
48.220.27z 20.010.21y
48.110.27yz 18.230.21x
47.770.77yz 18.590.21x
Colony forming units. L*: 0=black and 100=white; b*: 60=blue and +60=yellow. Least square means within a row bearing different letters are different (P<0.05).
Fig. 1. Day of display by antimicrobial treatment interaction effect on the least square mean ( S.E.) CIE a* value (a), 630/580 nm reflectance ratio (b), hue angle (c) and saturation index (d) of ground beef through simulated display. abcLeast square means within a day bearing different superscripts are different (P<0.05). dC=control, eAA=5% acetic acid, fGA=5% gluconic acid and gTSC=5% trisodium citrate. h 60=green and +60=red. iCalculated as 630/580 nm reflectance. jCalculated as tan 1(b*/a*). kCalculated as (a*2+b*2)0.5.
250
M.R. Stivarius et al. / Meat Science 60 (2002) 245–252
beef treated with AA was less (P < 0.05) bright purple red in overall color (Fig. 2, panel a) and worst point color (Fig. 2, panel b), and maintained a higher (P < 0.05) percentage discoloration (Fig. 2, panel c) than C through display. These findings are in agreement with Mikel et al. (1996), who reported visual appraisal of acid treated strip loins were browner in color when compared with a control. Acetic acid (AA) treated ground beef lost the desirable red color faster than any other treatment, which is in agreement with Bell et al. (1986) who found that beef cubes treated with 1.2% AA discolored immediately. Kotula et al. (1994), however, found no difference in beef rib steaks treated with 1.2% AA when looking at sensory evaluated color. The variation in
results may be explained by the difference in surface area exposed to AA treatment, with beef trimmings and beef cubes having a larger surface area per unit mass than beef rib steaks. Gluconic acid (GA) treated ground beef was similar (P>0.05) in overall color to C on days 0, 1 and 3 of display (Fig. 2, panel a). However, sensory panelists detected (P < 0.05) small overall color differences between GA and C treatments on days 2 and 7 of display. Likewise, sensory panelists found small differences (P < 0.05) in worst point color between GA and C treatments on days 0, 2 and 7 of display (Fig. 2, panel b). However, panelists were unable to detect any difference (P>0.05) in percentage discoloration between GA
Fig. 2. Day of display by antimicrobial treatment interaction effect on the least square mean ( S.E.) sensory evaluated (a) overall color, (b) worst point color, (c) percentage discoloration, (d) beef odor and (e) off odor characteristics of ground beef through simulated display. abcLeast square means within a day bearing different superscripts are different (P<0.05). dC=control, eAA=5% acetic acid, fGA=5% gluconic acid and g TSC=5% trisodium citrate. hColor score: 1=brown and 5=bright purple red. iPercentage discoloration: 1=total discoloration (96–100%) and 7=no discoloration (0%). jBeef odor score: 1=extreme non-beef like and 8=extreme beef like. kOff odor score:1=extreme off odor and 5=no off odor.
M.R. Stivarius et al. / Meat Science 60 (2002) 245–252
and C treatments until day 7 of display (Fig. 2, panel c). These findings tend to disagree with Garcia-Zepeda et al. (1994) who found that 3% GA treated beef was significantly browner in color on day 0 than the control. Conversely, trisodium citrate showed no detrimental effects on ground beef overall color (Fig. 2, panel a), worst point color (Fig. 2, panel b) or percentage discoloration (Fig. 2, panel c) and was equal to or superior in color characteristics when compared with C. Sensory panelists also evaluated beef odor characteristics and found that TSC and GA treatments were not different (P>0.05) for this attribute when compared with C, but that ground beef from the AA treatment had a less (P < 0.05) beef like odor when compared with all other treatments through display (Fig. 2, panel d). Panelists also reported that ground beef from the AA treatment possessed more (P < 0.05) off odor than did all other treatments through display, and that TSC was not different (P>0.05) in off odor than C through 7 days of display (Fig. 2, panel e). Likewise, GA was not different from C in off odor (P>0.05) through 3 days of display, but did display less (P < 0.05) off odor than C by day 7 of display.
4. Conclusion Consumer demand for a safe meat supply obviates the need for antimicrobial measures in the manufacturing process, as well as more research on the color and odor stability of ground beef. Results from this study indicate that AA treatment of beef trimmings before grinding can reduce EC, CO, ST and APC in ground beef by approximately 1 log each. Likewise, GA is also effective for reducing EC and APC in ground beef when applied to beef trimmings before grinding, but these decreases are not sufficient to warrant industry adoption. Although 5% GA had little impact on ground beef color and odor characteristics, the use of 5% AA did elicit color and odor changes. Therefore, while 5% GA can be expected to slightly reduce EC and APC in ground beef with minimal quality changes, additional work is required to determine the optimal level of acetic acid usage in order to maintain ground beef color and odor characteristics, while reducing microbial pathogens.
Acknowledgements Appreciation is expressed to the Arkansas Beef Council for funding this research. The authors would also like to thank J. Davis, L. Rakes, A. Ivey, L. McBeth, R. Story, and E. Kroger for their assistance in conducting these trials, and special appreciation is extended to Z. Johnson for assistance in data analysis.
251
References AMSA. (1978). Guidelines for cookery, sensory evaluation of meat. Chicago, IL: Am. Meat Sci. Assoc. and National Live Stock and Meat Board. Anderson, M. E., & Marshall, R. T. (1990). Reducing microbial populations on beef tissues: concentration and temperature of an acid mixture. Journal of Food Science, 55(4), 903–905. Arganosa, G. C., & Marriott, N. G. (1989). Organic acids as tenderizers of collagen in restructured beef. Journal of Food Science, 54(5), 1173–1176. Bell, M. F., Marshall, R. T., & Anderson, M. E. (1986). Microbiological and sensory tests of beef treated with acetic and formic acids. Journal of Food Protection, 49(3), 207–210. Conner, D. E., Kotrola, J. S., Mikel, W. B., & Tamblyn, K. C. (1997). Effect of acetic-lactic acid treatments applied to beef trim on populations of Escherichia coli O157:H7 and Listeria monocytogenes in ground beef. Journal of Food Protection, 60(12), 1560–1563. Cutter, C. N., Dorsa, W. J., & Siragusa, G. R. (1997). Parameters affecting the efficacy of spray washes against Escherichia coli O157:H7 and fecal contamination. Journal of Food Protection, 60(6), 614–618. Dickson, J. S. (1992). Acetic acid action on beef tissue surfaces contaminated with Salmonella typhimurium. Journal of Food Science, 57(2), 297–301. Dickson, J. S., & Anderson, M. E. (1992). Microbiological decontamination of food animal carcasses by washing and sanitizing systems: a review. Journal of Food Protection, 55(2), 133–140. Dorsa, W. J., Cutter, C. N., & Siragusa, G. R. (1997). Effects of acetic acid, lactic acid and trisodium phosphate on the microflora of refrigerated beef carcass surface tissue inoculated with Escherichia coli O157:H7, Listeria innocua, and Clostridium sporogenes. Journal of Food Protection, 60(6), 610–624. Dorsa, W. J., Cutter, C. N., & Siragusa, G. R. (1998). Bacterial profile of ground beef made from carcass tissue experimentally contaminated with pathogenic and spoilage bacteria before being washed with hot water, alkaline solution, or organic acid and then stored at 4 or 12 C. Journal of Food Protection, 61(9), 1109–1118. Ellebracht, E. A., Castillo, A., Lucia, L. M., Miller, R. K., & Acuff, G. R. (1999). Reduction of pathogens using hot water and lactic acid on beef trimmings. Journal of Food Science, 64(6), 1094– 1099. Emswiler, B. S., Kotula, A. W., & Rough, D. K. (1976). Bactericidal effectiveness of three chlorine sources used in beef carcass washing. Journal of Animal Science, 42(6), 1445–1450. Gill, C. O., & Bandoni, M. (1997). The effects of hot water pasteurizing treatments on the appearances of pork and beef. Journal of Meat Science, 46(1), 77–87. Garcia-Zepeda, C. M., Kastner, C. L., Willard, B. L., Phebus, R. K., Schewnke, J. R., Fijal, B. A., & Prasai, R. K. (1994). Gluconic acid as a fresh beef decontaminant. Journal of Food Protection, 57(11), 956–962. Hardin, M. D., Acuff, G. R., Lucia, L. M., Oman, J. S., & Savell, J. W. (1995). Comparison of methods for decontamination from beef carcass surfaces. Journal of Food Protection, 58(4), 368–374. Hunt, M. C., Acton, J. C., Benedict, R. C., Calkins, C. R., Cornforth, D. P., Jeremiah, L. E., Olson, D. G., Salm, C. P., Savell, J. W., & Shivas, S. D. (1991). AMSA guidelines for meat color evaluation. In E. Schofield (Ed.), Proceedings of the Forty-Fourth Annual Reciprocal Meat Conference (pp. 3–17). Manhattan, Kansas: Kansas State University (9–12 July 1991). Kochevar, S. L., Sofos, J. N., LeValley, S. B., & Smith, G. C. (1997). Effect of water temperature, pressure and chemical solution of removal of fecal material and bacteria from lamb adipose tissue by spray-washing. Meat Science, 45(3), 377–388. Kotula, K. L., & Thelappurate, R. (1994). Microbiological and sensory
252
M.R. Stivarius et al. / Meat Science 60 (2002) 245–252
attributes of retail cuts of beef treated with acetic and lactic acid solution. Journal of Food Protection, 57(8), 665–670. Luck, E., & Jager, M. (1998). In G.F. Edwards (Ed.), Antimicrobial food additives (2nd ed.). New York: Springer-Verlag (pp. 137–144, 116–119, 42–43,239, 98–101). Mikel, W. B., Goddard, B. L., & Bradford, D. D. (1996). Muscle microstructure and sensory attributes of organic acid treated beef strip loins. Journal of Food Science, 61(5), 1058–1061,1093.
SAS. (1988). SAS/STAT user’s guide: release 6.12. Cary, NC: SAS Institute, Inc. Siragusa, G. R. (1995). The effectiveness of carcass decontamination systems for controlling the presence of pathogens on the surfaces of meat animal carcasses. Journal of Food Safety, 15, 229–238. Strange, E. D., Benedict, R. C., Gugger, R. E., Metzger, V. G., & Swift, C. E. (1974). Simplified methodology for measuring meat color. Journal of Food Science, 39, 988–992.