Mohamed Kamel

The aim of this study was to determine the course of effects on the microflora on beef carcasses of a commercial dry chilling process in which carcasses were dry chilled for 3 days. Groups of 25 carcasses selected at random were sampled when the chilling process commenced and after the carcasses were chilled for 1, 2, 4, 6, 8, 24, and 67 h for determination of the numbers of aerobes, coliforms, and Escherichia coli. The temperatures of the surfaces and the thickest part of the hip (deep leg) of carcasses, as well as the ambient air conditions, including air temperature, velocity, and relative humidity (RH), were monitored throughout the chilling process. The chiller was operated at 0°C with an off-coil RH of 88%. The air velocity was 1.65 m/s when the chiller was loaded. The initial RH levels of the air in the vicinity of carcasses varied with the locations of carcasses in the chiller and decreased rapidly during the first hour of chilling. The average times for shoulder surfaces, rump surfaces, and the deep leg of carcasses to reach 7°C were 13.6 ± 3.1, 16.0 ± 2.4 and 32.4 ± 3.2 h, respectively. The numbers of aerobes, coliforms, and E. coli on carcasses before chilling were 5.33 ± 0.42, 1.95 ± 0.77, 1.42 ± 0.78 log CFU/4,000 cm(2), respectively. The number of aerobes on carcasses was reduced by 1 log unit each in the first hour of chilling and in the subsequent 23 h of chilling. There was no significant difference (P > 0.05) between the numbers of aerobes recovered from carcasses after 24 and 67 h of chilling. The total numbers (log CFU/100,000 cm(2)) on carcasses before chilling and after the first hour of chilling were 3.86 and 2.24 for coliforms and 3.30 and 2.04 for E. coli. The subsequent 23 h of chilling reduced the numbers of both groups of organisms by a further log unit. No coliforms or E. coli were recovered after 67 h of chilling. The findings show that the chilling regime investigated in this study resulted in significant reductions of all three groups of indicator organism

The objective of this study was to determine the immediate source of Escherichia coli on beef trimmings produced at a large
packing plant by analyzing the E. coli on trimmings at various locations of a combo bin filled on the same day and of bins filled
on different days. Ten 2,000-lb (907-kg) combo bins (B1 through B10) of trimmings were obtained from a large plant on 6 days
over a period of 5 weeks. Thin slices of beef with a total area of approximately 100 cm2 were excised from five locations (four
corners and the center) at each of four levels of the bins: the top surface and 30, 60, and 90 cm below the top. The samples were
enriched for E. coli in modified tryptone soya broth supplemented with 20 mg/liter novobiocin. The positive enrichment cultures,
as determined by PCR, were plated on E. coli/coliform count plates for recovery of E. coli. Selected E. coli isolates were
genotyped using multiple-locus variable-number tandem repeat analysis (MLVA). Of the 200 enrichment cultures, 43 were
positive by PCR for E. coli, and 32 of these cultures yielded E. coli isolates. Two bins did not yield any positive enrichment
cultures, and three PCR-positive bins did not yield any E. coli isolates. MLVA of 165 E. coli isolates (30, 62, 56, 5, and 12 from
B6 through B10, respectively) revealed nine distinct genotypes. MLVA types 263 and 89 were most prevalent overall and on
individual days, accounting for 49.1 and 37.6% of the total isolates, respectively. These two genotypes were also found at
multiple locations within a bin. All nine genotypes belonged to the phylogenetic group A0 of E. coli, suggesting an animal origin.
The finding that the trimmings carried very few E. coli indicates an overall effective control over contamination of beef with E.
coli at this processing plant. The lack of strain diversity of the E. coli on trimmings suggests that most E. coli isolates may have
come from common sources, most likely equipment used in the fabrication process.

The possible origin of Escherichia coli found on cuts and trimmings in the breaking facility of a beef packing plant was examined using multiple-locus variable-number tandem repeat analysis. Coliforms and E. coli were enumerated in samples obtained from 160 carcasses that would enter the breaking facility when work commenced and after each of the three production breaks throughout the day, from the conveyor belt before work and after each break, and from cuts and trimmings when work commenced and after each break. Most samples yielded no E. coli, irrespective of the surface types. E. coli was recovered from 7 (<5%) carcasses, at numbers mostly ≤1.0 log CFU/160,000 cm(2). The log total numbers of E. coli recovered from the conveyor belt, cuts, and trimmings were mostly between 1 and 2 log CFU/80,000 cm(2). A total of 554 E. coli isolates were recovered. Multiple-locus variable-number tandem repeat analysis of 327 selected isolates identified 80 distinct genotypes, with 37 (46%) each containing one isolate. However, 28% of the isolates were of genotypes that were recovered from more than one sampling day. Of the 80 genotypes, 65 and 2% were found in one or all four sampling periods throughout the day. However, they represented 23 and 14% of the isolates, respectively. Of the genotypes identified for each surface type, at least one contained ≥9 isolates. No unique genotypes were associated with carcasses, but 10, 17, and 19 were uniquely associated with cuts, trimmings, and the belt, respectively. Of the isolates recovered from cuts, 49, 3, and 19% were of genotypes that were found among isolates recovered from the belt, carcasses, or both the belt and carcasses, respectively. A similar composition was found for isolates recovered from trimmings. These findings show that the E. coli found on cuts and trimmings at this beef packing plant mainly originated from the conveyor belt and that small number of E. coli strains survived the daily cleaning and sanitation process, thus persisting in the plant.

The aim of the study was to determine the effects of meat pH on the abilities of 11 psychrotolerant Clostridium spp. to grow on, and to possibly cause blown pack spoilage of vacuum packaged beef. Beef steaks of pH 5.4–5.6, 5.7–5.9 or ≥6.0, i.e. of normal, intermediate or high pH were prepared and vacuum packaged. Groups of 3 steaks of the same pH range were inoculated with log phase cultures of Clostridium algoriphilum, Clostridium algidixylanolyticum, Clostridium bowmanii, Clostridium estertheticum, Clostridium frigoris, Clostridium frigidicarnis, Clostridium gasigenes, Clostridium lacusfryxellense, Clostridium psychrophilum, Clostridium tagluense or Clostridium vincentii. Each pack was resealed immediately after the steak was inoculated, and pack volumes were determined by water displacement, immediately after resealing and at intervals during storage at 2 °C for 56 days. All of the clostridia grew in packs of high pH beef but none caused pack swelling. Packs of intermediate pH beef inoculated with C. estertheticum began to swell after 14 days, with a mean rate of increase of pack volumes of 6.80 ml/day. One pack of intermediate pH beef inoculated with C. frigoris was swollen after 37 days. Packs of normal pH beef that had been inoculated with C. estertheticum began swelling after 14 days with a mean rate of increase of pack volumes of 7.70 ml/day. Packs of normal or intermediate pH beef inoculated with other clostridia did not swell. After storage, the numbers of most Clostridium spp., as determined by real-time PCR were greater on beef of high pH than of lower pH values, but the numbers of C. frigidicarnis and C. lacusfryxellense were highest on intermediate pH meat, the numbers of C. estertheticum were higher on meat of lower than of high pH, and the numbers of C. tagluense were the same on meat of all pH values. With high pH meat, glucose was reduced to very low level in rinse fluids from packs that had been inoculated with any Clostridium sp. With intermediate and normal pH meat, glucose was reduced to very low concentrations in only rinse fluids from beef that had been inoculated with C. estertheticum.

Beef steaks (2 cm thick) were each inoculated at three sites in the central plane with Escherichia coli O157:H7 at 5.9 ± 0.3 log CFU per site. Temperatures at steak centers were monitored during cooking on a hot plate or the grill of a gas barbeque. Steaks were cooked in groups of five using the same procedures and cooking each steak to the same temperature, and surviving E. coli O157:H7 at each site was enumerated. When steaks cooked on the hot plate were turned over every 2 or 4 min during cooking to between 56 and 62°C, no E. coli O157:H7 was recovered from steaks cooked to ≥58 or 62°C, respectively. When steaks were cooked to ≤71°C and turned over once during cooking, E. coli O157:H7 was recovered from steaks in groups turned over after ≤8 min but not from steaks turned over after 10 or 12 min. E. coli O157:H7 was recovered in similar numbers from steaks that were not held or were held for 3 min after cooking when steaks were turned over once after 4 or 6 min during cooking. When steaks were cooked on the grill with the barbeque lid open and turned over every 2 or 4 min during cooking to 63 or 56°C, E. coli O157:H7 was recovered from only those steaks turned over at 4-min intervals and cooked to 56°C. E. coli O157:H7 was recovered from some steaks turned over once during cooking on the grill and held or not held after cooking to 63°C. E. coli O157:H7 was not recovered from steaks turned over after 4 min during cooking to 60°C on the grill with the barbeque lid closed or when the lid was closed after 6 min. Apparently, the microbiological safety of mechanically tenderized steaks can be assured by turning steaks over at intervals of about 2 min during cooking to ≥60°C in an open skillet or on a barbecue grill. When steaks are turned over only once during cooking to ≥60°C, microbiological safety may be assured by covering the skillet or grill with a lid during at least the final minutes of cooking.

The objective of the study was to identify factors affecting the fractions of the bacteria naturally present on surfaces of beef cuts that are carried into deep tissues when the meat is mechanical tenderized by piercing with banks of thin blades. The surfaces and ten strips of meat from the deep tissues of beef primal cuts tenderized first and last on each of five days at a retail store meat fabrication facility were sampled for enumeration of total aerobic counts. Each strip was excised from the whole thickness of a cut after surfaces were sterilized. The mean log numbers of total aerobes recovered from the surfaces of cuts tenderized first or last each day were 2.18 and 1.57 log cfu cm−2, respectively. The mean log numbers recovered per strip from individual cuts tenderized first or last each day ranged from 0.30 to 1.45 and from 0.03 to 1.04 log cfu, respectively. These findings indicate that bacteria from the tenderizing equipment augmented the numbers of aerobes on the surfaces of cuts tenderized first each day, with some of the additional aerobes being carried into deep tissues. Subsequently, pieces of cuts stored in air at 2 °C were tenderized at a laboratory using commercial equipment. Each cut was divided into three pieces with one piece being not treated, one being sprayed with water and one being sprayed with 5% lactic acid. The mean log numbers of total aerobes recovered from the surfaces of not treated pieces of cut stored for 0, 2, 4, 6, 10 or 14 days were 0.6, 0.8, 2.6, 4.2, 8.5 and 8.9 log cfu cm−2, respectively. No aerobes were recovered from the deep tissues of any of the pieces of cuts tenderized on day 0. Mean log numbers recovered from the deep tissues of not treated tenderized pieces of cuts stored for 2, 4, 6, 10 or 14 days were 0.3, 0.3, 2.2, 7.8 and 8.1 log cfu per strip, respectively. Spraying with 5% lactic acid reduced the mean log numbers of aerobes on pieces of cuts stored for 2, 4, or 6 days by 1, 2 or 2 log units, respectively, but did not reduce the numbers on pieces of cuts stored for 10 or 14 days. Mean log numbers recovered from the deep tissues of tenderized pieces of cuts sprayed with 5% lactic acid were not significantly different (P > 0.05) from the mean log numbers recovered from the corresponding, tenderized not treated pieces of cuts. These findings showed that the fraction of the total aerobes on cut surfaces that are carried into deep tissues during mechanical tenderizing can vary with the stage of growth of the spoilage flora; and that reduction of numbers of aerobes on the surface by treatment with lactic acid before tenderizing does not necessarily reduce the numbers carried into deep tissues during tenderizing.

BACKGROUND: The microbiological condition of beef produced at North American plants has been improved as a result of the use of effective carcass-decontaminating treatments. The effect of these treatments on the storage life of beef has not been established. In this study, beef primal cuts in vacuum packs stored at −1.5 or 2 ∘C for up to 160 days were assessed for their microbiological and organoleptic properties.
RESULTS: The odours of boneless cuts were acceptable after storage at either temperature for ≤160 days; and the flavours of steaks from boneless cuts stored at 2 or −1.5 ∘C for ≤70 or ≤120 days, respectively, were acceptable. The storage life of bone-in cuts stored at 2 or −1.5 ∘C was, respectively, shorter or the same as that of boneless cuts stored at the same temperature. More than 20 microbial species thatwere mostly obligate aerobeswere present on both types of cuts before storage. After storage for ≥30 days, the microflora was dominated by carnobacteria and Enterobacteriaceae were present in the flora from early storage times.
CONCLUSIONS: A storage life of 120–140 days was attained by vacuum-packaged beef primals from decontaminated carcasses stored at −1.5 ∘C. The bone-in cuts stored at 2 ∘C were spoiled at earlier times, probably by Enterobacteriaceae.

Overseas customers for Canadian vacuum packaged beef require assurance of the product's storage life, but little information on the storage life of vacuum package beef produced in Canada, or North America generally is available. Therefore, vacuum packaged halved strip loin and halved bone-in loin cuts were obtained from a beef packing plant and stored at 2 °C or − 1.5 °C. Three cuts of each type were sampled when cuts were received and at sequential intervals of 10, 20 or 30 days, as was appropriate, up to 160 days of storage. The odour of each cut was assessed at pack opening, then the cut within the pack was sampled by rinsing. Numbers of aerobes and coliforms in the rinse fluid were determined. Steaks prepared from each cut were displayed and assessed daily for appearance and odour. A steak from each cut was frozen and subsequently assessed for flavour. On both boneless and bone-in cuts the maximum numbers of aerobes and coliforms were attained after about 40 days at 2 °C or 80 days at − 1.5 °C. The maximum numbers of aerobes were about 0.5 log units less, and the maximum numbers of coliforms were up to 2 log unit less on cuts stored at − 1.5 °C than on cuts stored at 2 °C. Boneless or bone-in cuts stored at − 1.5 °C and boneless cuts stored at 2 °C were spoiled by acid/dairy odours and flavours. Bone-in cuts stored at 2 °C were spoiled by putrid odours associated with the bone marrow. Storage lives of strip loins were 80 days at 2 °C or 130 days at − 1.5 °C; and storage lives of bone-in loins were 50 days at 2 °C or 140 days at − 1.5 °C.

Spraying carcasses with solutions of lactic acid is a decontaminating treatment widely used at North American beef packing plants. For HACCP purpose, the efficacy of such treatment should be validated. The utility of validation by reference to reduction in numbers of Escherichia coli naturally present on beef has been questioned because reductions of E. coli might be greater than reductions of E. coli O157:H7, which may be more acid tolerant than E. coli generally. To investigate the effects of lactic acid sprays on the two types of E. coli on beef surfaces, slices of beef with cut muscle, fat or membrane surfaces were inoculated with a five strain cocktail of acid-adapted E. coli O157:H7 or with a strain of E. coli that was not adapted to acid, at number of 5, 1 or − 1 log cfu/cm2. Inoculated slices were not treated or were sprayed with water or solutions of lactic acid at concentrations of 2% or 5%. The volume of the fluids sprayed on surfaces had been shown to be in excess of the volume required for maximum reduction of the numbers of E. coli on surfaces. For each slice of lactic acid-treated meat the numbers of E. coli or E. coli O157:H7 recovered on agars that either allowed recovery of injured cells or recovery of only cells that had not been injured by the treatment were similar. The differences in the log mean numbers (log A) and in the log total numbers (N) of E. coli and E. coli O157:H7 recovered from slices that were not treated or were treated with acid were calculated. Also, differences in values for log A and N for the numbers of those bacteria recovered from slices treated with water or with acid were calculated. Means of the differences showed that the acid treatments gave similar reductions in the numbers of E. coli or E. coli O157:H7, but those reductions were somewhat greater with 5% than with 2% lactic acid. The findings indicate that the E. coli or E. coli O157:H7 that survived acid treatments of meat surfaces were protected from exposure to injurious concentrations of undissociated acid. Therefore, tolerance of acid conditions does not substantially increase the ability of E. coli O157:H7 to survive treatment of beef surfaces with solutions of lactic acid as compared with E. coli generally.

Vacuum-packaged top butt cuts from a beef packing plant that does not use any carcass decontaminating interventions were assessed for their organoleptic and microbiological properties during storage at 2 or -1.5°C. Cuts stored at 2°C were acceptable after storage for 140 days but were unacceptable after 160 days because of persistent sour, acid odors. Odors of cuts stored at -1.5°C for 160 days were acceptable. The numbers of aerobes on cuts increased from <1 log CFU/cm(2) to 7 or 6 log CFU/cm(2) for cuts stored at 2 or -1.5°C, respectively. The numbers of Enterobacteriaceae increased from <-1 log CFU/cm(2) to 5 or 3 log CFU/cm(2) for cuts stored at 2 or -1.5°C, respectively. Bacteria recovered from initial microflora were, mainly, strictly aerobic organisms. Bacteria recovered from cuts stored for 160 days were mainly Carnobacterium spp. that grew on an acetate-containing agar generally selective for lactic acid bacteria other than Carnobacterium. C. divergens and C. maltaromaticum were recovered from cuts stored at 2°C, but C. maltaromaticum was the only species of Carnobacterium recovered from cuts stored at -1.5°C. No lactic acid bacteria of genera that usually predominate in the spoilage microflora of vacuum-packaged beef at late storage times were recovered from the spoilage microflora. The findings indicate that carnobacteria, initially present at very small numbers, grew exponentially to persistently dominate the spoilage microflora of vacuum-packaged beef cuts of unusually long storage life.

Microbiological samples were obtained from the hands of workers, personal equipment and conveyor belts, in the carcass breaking facility of a beef packing plant where chilled carcasses carry coliforms and Escherichia coli at numbers < 1 cfu/10,000 cm2. Before work started, steel mesh gloves carried coliform and E. coli at numbers of 4 and 3 log cfu/25 items, respectively, while their numbers on conveyor belts were >2 and <2 log cfu/2500 cm2, respectively. After a period of work the numbers of both coliforms and E. coli on steel mesh gloves and conveyor belts were reduced by about 1 log unit, but the numbers on cotton gloves worn under steel mesh gloves were 5 and >4 log cfu/25 items, respectively. The findings indicate that the proximate source of most coliforms and E. coli that contaminate cuts and trimmings are, respectively, cotton gloves and conveyors. Microbiological samples were similarly obtained at a second plant where drying of carcass surfaces during chilling gave carcasses with coliforms and E. coli at numbers about 1 cfu/1000 cm2. Before work started, steel mesh gloves were free of coliforms and E. coli; and the uniform wearing of rubber gloves over cotton ones precluded contamination of cotton gloves. The conveyor belt carried coliforms and E. coli at numbers about 1 cfu/cm2 and >1 cfu/100 cm2, respectively, before work started; and at numbers < 1 cfu/cm2 and about 1 cfu/100 cm2, respectively, after a period of work. The findings indicate that at the second plant both cuts and trimmings are contaminated with coliforms and E. coli from the conveyor.

Decontamination of beef by spraying with solutions of lactic acid is common practice in North America and is to be permitted in the EU. Validation of each such treatment is necessary for HACCP purposes. The utility of validation by reference to reduction in numbers of Escherichia coli naturally present on beef is questioned. This is because reductions of E. coli generally might be greater than reductions of the main target organism, E. coli O157:H7, which can be more acid tolerant. To investigate the effects of lactic acid sprays on the two types of E. coli on beef surfaces, slices of beef with cut muscle, fat or membrane surfaces were prepared. The surfaces were inoculated with a five strain cocktail of acid-adapted E. coli O157:H7 or with a not-acid-adapted strain of E. coli at number of 5, 1 or −1 log cfu/cm2. Inoculated slices were not treated or were sprayed with water or 2% or 5% lactic acid at 0.5 ml/cm2. For each slice of lactic acid-treated meat the numbers of E. coli or E. coli O157:H7 recovered on agars that allowed resuscitation of injured cells or did not allow resuscitation were similar. The differences in the log mean numbers (log A) and in the log total numbers (N) of bacteria recovered from slices that were not treated or treated with acid were calculated. Differences in log A and N for bacteria recovered from slices treated with water or acid were calculated also. Means of the differences indicated that each acid treatment gave similar reductions in the numbers of E. coli or E. coli O157:H7. However, reductions were somewhat greater with 5% than with 2% lactic acid. The findings suggest that the E. coli or E. coli O157:H7 that survived acid treatments of meat surfaces were protected from exposure to injurious concentrations of undissociated acid. Consequently, strains of E. coli that are or are not relatively tolerant of acid conditions will be inactivated to similar extents by solutions of lactic acid applied to beef surfaces.

Membrane, fat and cut muscle surfaces of beef were inoculated with Escherichia coli at numbers about 4, 1 or −1 log cfu/cm2. The inoculated meat was sprayed with water or 5% lactic acid at volumes of 0.5, 0.1 or 0.02 ml/cm2. Spraying with water reduced the numbers of E. coli on membrane surfaces by up to 1 log unit, but had little effect on the numbers of E. coli on fat or cut muscle surfaces. Spraying with 5% lactic acid reduced the highest numbers of E. coli on membrane surfaces by up to 4 log units; but those numbers on fat or cut muscle surfaces were reduced by ≤1.5 log unit, and the reductions declined with decreasing volumes of 5% lactic acid. With inocula of 1 log cfu/cm2, spraying lactic acid in any volume reduced the numbers of E. coli on membrane or fat surfaces by about 1 log unit, and the numbers on cut muscle surfaces by between 0.8 and 0.2 log unit. E. coli were detected in enrichment cultures of samples from all surfaces inoculated with E. coli at −1 log cfu/cm2 and sprayed with 5% lactic acid at 0.5 ml/cm2. The findings indicate that spraying relatively heavily contaminated cuts or trimmings with 5% lactic acid at ≥0.1 ml/cm2 can be expected to reduce numbers of E. coli and, presumably, associated pathogens by between 0.5 and 1 log unit. However, such a treatment is likely to be at best marginally effective for reduce the numbers of these organisms on lightly contaminated product

Swab samples were obtained from groups of 25 carcasses at various stages of processing at a large beef packing plant. The log mean number of aerobes recovered from carcasses after skinning was 2.2 log CFU/cm(2). Spraying the uneviscerated carcasses with 5% lactic acid reduced the numbers of aerobes by about 1 log unit; but subsequent carcass dressing operations, a second treatment with 5% lactic acid, pasteurizing, and carcass cooling had no substantial effect upon the number of aerobes on carcasses. The total numbers of coliforms or Escherichia coli cells recovered from skinned carcasses were <2 log CFU/2,500 cm(2). The numbers were reduced by the washing of uneviscerated carcasses but increased after evisceration operations. The numbers were reduced by spraying with lactic acid and pasteurizing, with no coliforms or E. coli being recovered from pasteurized carcass sides. No coliforms or E. coli cells were recovered from the forequarters of cooled carcass sides, but E. coli cells were recovered from the hindquarters of 1 of 50 cooled carcass sides, at 1.4 log CFU/1,000 cm(2). The numbers of aerobes on conveyor belts in the carcass breaking facility were similar to the numbers on cooled carcass, but the numbers of aerobes on cuts and trimmings and the number of coliforms and E. coli cells on the products and belts were higher than the numbers on carcasses. The findings indicate that most cooled carcasses produced at the plant carry E. coli at numbers <1 CFU/10,000 cm(2) but that product can be contaminated with small numbers of E. coli (<1 CFU/100 cm(2)) during carcass breaking.

The effects of pre-emulsified beef fat and canola oil (CO) (25%) with Tween 80 (T-80) or sodium caseinate (SC) were studied in beef meat batters prepared at three protein levels (9%, 12% and 15%). Raising meat protein level to 15% resulted in low emulsion stability of products prepared with CO. Using pre-emulsified beef fat with Tween 80 (BF-T80) showed significantly higher fat and water losses at all protein levels. There were no differences in fat and water losses between pre-emulsified beef fat and CO when SC was used at the 9% and 12% protein levels compared to the controls (non pre-emulsification). Light microscopy revealed fat globule coalescence in the CO meat batters prepared with 15% protein and BF-T8 treatments, as well as formation of fat channels and more protein aggregation; both resulted in lower emulsion stability. Using SC to emulsify fat/oil produced a finer dispersion of fat globules compared to all the other treatments.

The effects of substituting 1.5% of the meat proteins with low gelling soy protein isolate (LGS), high gelling soy protein isolate (HGS), native whey protein isolate (NWP), and preheated whey protein isolate (PWP) were compared at varying levels of proteins (12, 13 and 14%), with all meat control batters prepared with canola oil. Cooking losses were lower for all the non-meat protein treatments compared to the all meat controls. When raising the protein level from 12 to 14%, cooking losses increased in all treatments except for the NWP treatments. Using LGS increased emulsification and resulted in a more stable meat batters at the 13 and 14% protein treatments. Textural profile analysis results showed that elevating protein level increased hardness and cohesiveness. The highest hardness values were obtained for the PWP treatments and the lowest for the HGS, indicating a strong non-meat protein effect on texture modification. Non-meat protein addition resulted in lighter and less red products (i.e., lower red meat content) compared to the all meat controls; color affected by non-meat protein type. Light microscopy revealed that non-meat proteins decreased the frequency of fat globules' agglomeration and protein aggregation. The whey protein preparations and HGS formed distinct “islands” within the meat batters' matrices, which appeared to interact with the meat protein matrix.

The effects of meat protein substitution (2%) with sodium caseinate, milk protein isolates or whey protein isolates (WPI) at varying levels of meat protein (13–15%) were studied in emulsified beef meat batters prepared with canola oil (25%). Increasing meat protein to 14 and 15% resulted in less stable emulsions (higher cook loss) compared with the 13% protein. At an equal protein level, all dairy proteins reduced cook loss (P < 0.05) compared with the all meat protein treatments. Overall, WPI provided the best emulsifying and moisture retention. Increasing the protein content also resulted in higher hardness and springiness values in all treatments. The addition of dairy protein resulted in softer, lighter and less red products compared with the all meat controls; colors being affected by the dairy protein source and protein content. Light microscopy revealed that increasing protein content caused some fat globule coalescence and more protein aggregation. WPI formed distinct dairy protein gel regions within the meat batter matrix, which appeared to interact with meat protein matrix.

The effects of fat reduction (25.0%, 17.5%, and 10.0%) and substituting beef fat with canola oil or pre-emulsified canola oil (using soy protein isolate, sodium caseinate or whey protein isolate) on cooking loss, texture and color of comminuted meat products were investigated. Reducing fat from 25 to 10% increased cooking loss and decreased hardness. Canola oil or pre-emulsified treatments showed a positive effect on improving yield and restoring textural parameters. Using sodium caseinate to pre-emulsify the oil resulted in the highest hardness value. Cohesiveness was affected by fat type and level. The color of reduced fat meat batters was darker for all, except the beef fat treatments. Using canola oil or pre-emulsified oil resulted in a significant reduction in redness. The results show that pre-emulsification can offset some of the changes in reduced fat meat products when more water is used to substitute for the fat and that pre-emulsification can also help to produce a more stable meat matrix.

The main goal of this research was to investigate the mechanisms involved in meat emulsion stabilization. The effects of substituting beef fat (25%) with rendered beef fat, canola oil, palm oil, or hydrogenated palm oil at varying meat protein levels (8 to 14%) were studied. There was no significant difference in fat loss among meat batters prepared with beef fat, rendered beef fat, or palm oil. Hydrogenated palm oil provided the most stable batters at all protein levels. Canola oil showed significant fat and water separation when the protein level was raised to and above 14%; this did not occur in the other treatments. The difference was related to the physicochemical characteristics of the canola oil. Light microscopy revealed coalescence of fat globules in the canola oil meat batters prepared with ≥ 14% protein, as well as formation of fat channels and more protein aggregation. Substituting 1.5% of meat proteins with non-meat proteins (low gelling soy protein isolate, high gelling soy protein isolate, native and preheated whey protein isolates) caused a reduction in cooking losses of the canola oil meat batters. Non-meat proteins with low gelling ability and high emulsifying capacity provided the most stable meat batters. Light microscopy revealed that non-meat proteins decreased the amount of fat agglomeration and protein aggregation. Using pre-emulsified beef fat and canola oil (25%) with Tween 80 or sodium caseinate, at different protein levels (9, 12, 15%), were studied. Replacing beef fat with pre-emulsified beef fat (BF-T80) showed significant fat and water losses during cooking. Light microscopy revealed a widespread of irregular fat globules and more protein aggregation in BF-T80 treatment. However, pre-emulsified canola oil with Tween 80 resulted in reduced fat losses at the 15% protein level compared to the control (no pre-emulsification). Pre-emulsified canola oil, using sodium caseinate, resulted in higher fat and water binding capacity than the control and meat batters prepared with Tween 80. The results indicate an important difference in the lipid holding mechanism between beef fat and canola oil.

The effects of beef fat (25%) substitution with rendered beef fat, canola oil, palm oil, or hydrogenated palm oil at varyingmeat protein levels (8%, 11%, and 14%) were studied in emulsified beef meat batters. There was no significant difference in fat loss among meat batters made with beef fat, rendered beef fat, or palm oil. Hydrogenated palm oil provided the most stable batters at all protein levels. Increasing meat protein to 14% resulted in high fat loss in batters prepared with canola oil, which did not occur in the other formulations. This indicates that the physicochemical characteristics of fat/oil affect emulsion stability. Cooked batter hardnesswas higher (P <0.05) when protein level was raised; highest in hydrogenated palm oil batters when compared at similar protein levels. As protein level was raised springiness values were increased in all the meat treatments. Springiness was higher in the canola oil treatments. Light microscopy revealed fat globule coalescence in canola oil meat batters prepared with 14% protein, as well as the development of fat channels andmore protein aggregation; both seem to result in lower emulsion stability. Hydrogenated palm oil batters showed fat particles with sharp edges as opposed to the round ones seen in all other treatments.