Raw well-chilled (1 ± 1°C) ground whole-hog prerigor meat (30 to 45 min postmortem, containing 1.5% salt to delay rigor onset) was supplied by a commercial pork processing plant, and made into sausages within 7 d of slaughter per manufacturer suggestion. The meat was stored in a walk-in cooler (1 ± 1°C) and processed within the next 24 h. Synthetic antioxidants were included in all treatments, which consisted of a proprietary combination of butylated hydroxyanisole (BHA), butylated hydroxytoluene (BHT), and propyl gallate (PG) at approximately 0.02% based on fat composition, which is within the legal limits described by the USDA (9 C.F.R. §319.141, 9 C.F.R. §424.21; Code of Federal Regulations, 2015). A total of 30 (10 treatments × 3 replications) 36.3-kg batches of fresh pork sausage were manufactured in the study. Ten batches of fresh pork sausage were formulated and randomly assigned to one of the following treatment combinations (Table 1) that also contained the synthetic antioxidants described above: 1500 ppm rosemary extract (R) (FORTIUM Brand, Kemin Food Technologies Inc., Des Moines, IA) + 100 ppm green tea (G) (FORTIUM Brand, Kemin Food Technologies Inc.), (1500 R + 100 G), R1500 + G200, R1500 + G300, R2000 + G100, R2000 + G200, R2000 + G300, R2500 + G100, R2500 + G200, R2500 + G300, and control (synthetic antioxidants in reps 1 and 2, R1500 and synthetic antioxidants in rep 3 since it represented commercial formulations at the time that rep 3 was conducted). Each batch of pre-rigor meat with the addition of the spice blend, corn syrup solids, chilled water, synthetic and an antioxidant combination were blended for 3 min in a paddle mixer (Model 150; Butcher Boy Limited, Ayshire, Scotland, UK) and stored in a walk-in cooler (1 ± 1°C) for 48 h. This was followed by grinding with a commercial mixer-grinder (Model 80055; Hollymatic Co., Countryside, IL) equipped with a 4 mm-hole diameter plate. Aliquots of ground meat were collected prior to stuffing and analyzed for fat, moisture, and protein contents (Method 2007.04; AOAC, 2007) using a FOSS FoodScan Meat Analyzer Near-Infrared (NIR) Spectrophotometer (Model 78810; Foss Co., Hillerød, Denmark). Fat, moisture, and protein contents were 28% ± 0.70, 53% ± 0.56, and 14% ± 0.36, respectively for the 3 replications (data not shown).

Before filling, the natural casings were rinsed to remove salt, treated for tenderization (proprietary treatment) and kept hydrated in warm water (40 ± 1°C) until stuffed with the sausage blends. Total plate counts (TPC) were determined for the untreated and treated casings using 3M Petrifilm (3M Global Headquarters, St. Paul, MN). Casings that were not treated contained 3.3 to 4.1 log cfu/g and treated casings contained between 1.4 and 2.6 log cfu/g. The ground meat was stuffed into the casing with a filling machine (Model RS1040C; Risco Vacuum Stuffer, Thiene, Italy) and hand-linked at 12.7 ± 1-cm intervals with a diameter of approximately 34 ± 1 mm (6-Inch Dial Caliper, Pittsburgh-Item 66541; Cen-Tech, Harbor Freight Tools, Camarillo, CA).

Fresh pork sausages (5 links per tray, 300 links per treatment) were packaged in expanded polystyrene trays (20S Yellow Tray; Cascades Plastics Inc., Warrenton, MO) and overwrapped with PVC stretch film with an oxygen permeability of 121 cc/100cm² per d and water vapor permeability of 2.2 g/100cm² per d (Premium-LT 80 718; LINPAC Packaging-Filmco Inc., Aurora, OH), and stored in an air-blast freezer overnight at –30°C. The frozen pork sausages were then stored in a –20°C walk-in freezer (Model TK-3476-WF-L, K Thermo-Kool, Mid-South Industries Inc., Laurel, MS) for 0.5, 3, and 6 mos. Packages (16 to 20 per treatment) were randomly selected and arranged in a walk-in cooler (3 ± 1°C, Model TK-3476-WF-L; K Thermo-Kool, Mid-South Industries Inc., Laurel, MS) where all treatments were subjected to continuous cool white fluorescent illumination (Cool White 34 Watt; Sylvania Supersaver Ecologic, Danvers, MA) of 800 lux at the package surface for 21 d. Samples were evaluated for instrumental color, lipid oxidation, and sensory descriptive analysis after 0, 7, 14, and 21 d. Consumer acceptability was performed on d 8. Packages for psychrotrophic plate counts (PPC), pH, and proximate analysis were vacuum packaged with 3-mil standard barrier vacuum pouches (Prime Source 75001826, 0.75 gauge nylon/2.25 gauge polyethylene, OTR = 0.6 cm³/100m² per d at 0°C, MVTR = 0.6g/100m² per d at 38°C, and 100% relative humidity; Rebel Butcher Supply Company, Flowood, MS) and frozen at –72°C in an ultra-cold freezer (Model, Lo-Cold Freezer, ScienTemp Corp, Adrian, MI) until subsequent analyses could be conducted.

A quadratic relationship existed between storage time and TBARS throughout simulated retail display following 0.5 and 3 mo of frozen storage, where the TBARS values for all treatments increased from d 0 to 7 and then decreased from d 14 to 21 (Fig. 1A and B). A linear relationship was observed after 6 mo of frozen storage where maximum values for TBARS were reached on d 0, after which a decline was observed (Fig. 1C). In addition, both rosemary and green tea extracts reduced TBARS values (P < 0.05), particularly in treatment combinations with higher concentrations of rosemary extract for up to 6 mo of storage at –20°C (Fig. 1C). The lowest (P < 0.05) TBARS values of the sausages were observed when treatment combinations with 2500 ppm of rosemary extract were added while the highest (P < 0.05) TBARS values were obtained for the control. Treatment combinations with 2500 ppm of rosemary extract also provided significant reduction in lipid oxidation (P < 0.05) which ranged from 20 to 62%, 43 to 59%, and 40 to 48% for 0.5, 3, and 6 mo, respectively (Fig. 1).

Major volatile decomposition products are regarded as good indicators of the oxidative state of meat lipids such as 2,4-decadienal which is one of the main volatiles derived from the oxidation of oleic and linoleic acid (Wood et al., 2008). On the basis of 2,4-decadienal formation, the order of concentration was control > R1500 > R2000 > R2500 following 3 mo of frozen storage and throughout 21 d of retail display (Table 3). Hexanal concentration increased over refrigerated storage time but did not differ between antioxidant treatments (data not shown).

Lipid oxidation was negatively correlated with natural plant extract concentration, a finding in agreement with other studies that reported higher levels of rosemary or green tea extracts aid in preventing lipid oxidation. Özyurt et al. (2010) reported that the addition of rosemary extract (300 ppm) inhibited the formation of TBARS in fried sea bream fillets over a 4-mo frozen storage (–18°C) period. Similar results were obtained by Sebranek et al. (2005) who revealed that sausage patties containing 2,500 ppm rosemary extract and stored at –20°C had significantly lower TBARS values over the 4 mo sampling period when compared to the control. A recent study by Perlo et al. (2018) indicated that rosemary extract sprayed at 4,000 ppm on 2.54 cm thick pork longissimus chops resulted in an effective delay of lipid oxidation without affecting pH, color or water content stored at MAP conditions at 4°C for 45 d. Jongberg et al. (2018) recently reported that there was a dose-dependent concentration of green tea extract (25 to 160 ppm) that lowered TBARS values in brine-injected pork chops stored in high-oxygen atmosphere packaging over 7 d storage (5°C).

The efficacy of the phenolic diterpenes in rosemary extract such as carnosic acid and carnosol is presumably due to the greater retention of these lipophilic antioxidants in the fat globule-protein interface (Berdahl and McKeague, 2015). The incorporation of a proprietary carrier system transports it to the interface where it increases the interfacial concentrations. This results in greater association with the interfacial protein film and thus its ability to scavenge free radicals produced from the decomposition of interfacial lipid hydroperoxides, which leads to increased oxidative stability especially in sausages with higher concentrations of rosemary extract. The polar and hydrophilic green tea catechin gallates or epicatechins protect the continuous phase from lipid oxidation. The antioxidative effect of added tea catechins in meats is related to chelating free iron ions which are major lipid oxidation catalysts in muscle foods (Tang et al., 2001). These catechins are also known to trap superoxide, hydroxyl, and peroxyl radicals and to end free-radical chain reactions in the continuous phase thereby preventing lipid oxidation in meat products (Vuong et al., 2011).

The CIE L* (lightness) values were higher (P < 0.05) for the control followed by samples containing lower concentrations of rosemary extract during 0.5 and 3 mo of frozen storage (Tables 2 and 3). The models for fresh pork sausage also demonstrate a gradual decline in the lightness values across all treatments as retail display time progressed. The CIE a* values (redness) were higher in fresh pork sausages with increased (P < 0.05) levels of rosemary extract compared with the control after 0.5 mo of frozen storage (Fig. 2A, Table 2). Green tea extract was a significant variable following 6 mo of frozen storage where redness values were higher for samples with 300 ppm of green tea extract especially on d 7 and 14 when compared to the other treatments (P < 0.05; Fig. 2B, Table 4). The CIE a* values decreased (P < 0.05) as frozen storage time progressed regardless of natural plant extract addition. However, it declined more rapidly in the control samples. Also noteworthy was the color reversion from brown to red across all sausage treatments during the last stages of retail display. This trend was accompanied by a gradual decline in the redness values as refrigerated storage progressed followed by an increase on d 21 as the storage period ended. This could be attributed to metmyoglobin reduction by aerobic bacterial proliferation where the release of metabolic reductants such as pyridine nucleotides [i.e., NAD(P)H] may have converted metmyoglobin to its reduced myoglobin forms in the presence of metmyoglobin reductase (Renerre, 2000). The highest chroma (saturation index) values at 3 mo of frozen storage were found in samples containing 2,500 ppm of rosemary extract and the lowest were recorded in the control (Fig. 3A, Table 3). The trend for the chroma values was similar to that for redness and yellowness values as retail display progressed. Surface hue angles (discoloration) were lower (P < 0.05) for samples with higher concentrations of rosemary extract compared with the control after 6 mo of frozen storage (Fig. 3B, Table 4).

The best synergistic effect was obtained from treatment combinations with higher concentrations of rosemary extract which exhibited better color retention as well as an improvement in color relative to the control samples. Similar observations have been made by Sebranek et al. (2005) who demonstrated that rosemary extract added at a level of 2,500 ppm in pork sausage patties was more effective in maintaining color than the control and BHA/BHT treatment after 3 mo of frozen storage. Green tea extract (600 ppm) improved the red color of omega-3 fortified surimi gels at the beginning of storage; however, no significant changes during frozen storage at –18°C for 9 mo were observed (Pérez-Mateos et al., 2006). In contrast, in the current study, green tea extract addition positively affected redness values following 6 mo of frozen storage. The variation in the efficiency of green tea extract between different studies could be related to differences in the amount of polyphenols in the extract, the meat species studied, the synergistic effects brought about by the addition of other ingredients, or the oxidation pattern of oxymyoglobin under conditions of reduced enzymatic activity where storage temperature, packaging method, and light intensity are major influencing factors (Georgantelis et al., 2007).

The models for fresh pork sausage microbial growth at each frozen storage period showed a quadratic relationship between PPC and storage time following 0.5, 3, and 6 mo of frozen storage where the PPC for all treatments increased from d 0 to 21. The lowest (P < 0.05) PPC of the sausages were observed when treatment combinations with 100 ppm of green tea extract were added while the highest (P < 0.05) PPC were obtained when the control (synthetic antioxidants only) was used. Also noteworthy were the higher counts of psychrotrophic bacteria at the beginning of the frozen storage period which ranged from 4.10 to 5.03 log₁₀ CFU/g compared with 2.9 to 3.8 log₁₀ CFU/g and 3.7 to 4.4 log₁₀ CFU/g for 3 and 6 mo, respectively (Fig. 4ABC, Tables 3 and 4). After 14 d of storage, levels of psychrotrophic bacteria continued to increase for all treatments where the control was highest (P < 0.05) and ranged from 7.00 to 7.90 log₁₀ CFU/g, 6.9 to 7.4 log₁₀ CFU/g, and 7.3 to 8.0 log₁₀ CFU/g for 0.5, 3, and 6 mo, respectively. PPC higher than 7 log₁₀ CFU/g have been reported as threshold values for spoilage in meat, poultry, and seafood products (Jay et al., 2005). Perhaps the increased PPC could have been due to the presence of lactic acid bacteria (LAB) which indicates that the sausages may not be spoiled when PPC reaches 7 log₁₀ CFU/g (Kin et al., 2011; Silva et al., 1995). In the present study, the co-antioxidative effect between the natural plant extracts and the synthetic antioxidants increased the microbial stability of the pork sausages allowing for frozen storage to last for up to 6 mos. Although microbial stability varied with natural plant extract combinations, it was evident that longer frozen storage was dependent on green tea extract concentration inhibiting psychrotrophic microbial growth. Our data show that natural plant extract combinations with 100 ppm of green tea extract resulted in lower PPC and were more effective in reducing microbial growth.

After 3 mos of frozen storage, the pH decreased (P < 0.05) as a function of storage time throughout 21 d of display (Fig. 5A, Table 3) and appear to coincide with an increase in PPC (Fig. 4B). Reduction in pH has been mostly associated with increased Gram-positive bacteria populations such as LAB or through lipolysis which liberates free fatty acids in the meat system. Fermentation is promoted by the microorganisms present, which results in the production of organic acids and a lower product pH (Papadima et al., 1999). This is also supported by the lower pH values of control samples, especially after 14 d (P < 0.05) of refrigerated storage (Fig. 5AB). Control samples also had the highest populations for psychrotrophic bacterial growth during the entire storage period (Fig. 4ABC). A similar effect on the pH of fresh pork sausages due to the addition of rosemary extract has been reported by Martínez et al. (2006).

Following 0.5 mo of frozen storage, all treatments exhibited a decrease (P < 0.05) in intensity ratings for descriptors regarded as positive drivers of liking for this product such as “pork complex”, “spice complex”, “ginger”, “nutmeg”, “rosemary”, “browned”, “caramelized” aromas, “flavor intensity”, “fat”, “lean”, “spice complex”, “ginger”, “nutmeg”, “black pepper”, “caramelized”, “sweet”, “salty”, “savory” flavors and juicy texture with increased retail display time (Table 2). The data also showed that the sausages were mainly differentiated (P < 0.05) by “spice complex”, “nutmeg”, “rosemary” aromas, “fat, “bitter” flavors, and hardness and chewiness textural attributes following 3 mo of frozen storage (Table 3). The intensities of most of the sensory descriptive notes remained the same for all treatments within the same frozen storage period except for “pork complex”, “spice complex”, “ginger”, “nutmeg”, “rosemary”, “browned” aromas, “flavor intensity”, “fat”, “spice complex”, “ginger”, “nutmeg”, “black pepper”, “savory” flavors and “juicy” which decreased as retail display time increased (data not shown). The sausages were characterized (P < 0.05) by their “pork complex”, flavor intensity and oily following 6 mo of frozen storage (Table 4, data not shown for textural attributes). Also, the mean attribute intensity ratings of “pork complex”, “spice complex”, “ginger”, “nutmeg”, “rosemary”, “copper herbal”, “browned”, “caramelized” aromas, “flavor intensity”, “fat”, “lean”, “spice” “complex”, “nutmeg”, “black pepper”, “rosemary”, “caramelized”, “salty”, “savory” flavors, “astringent”, “oily” feeling factors and “juicy” texture gradually decreased (P < 0.05) over the 21 d display period (data not shown). In contrast, the mean attribute intensity ratings of “off-odor”, liver-organy, “rancid”, “fruity” aromas, “off-flavor”, “liver-organy”, “metallic”, “rancid”, “fruity”, “sour” flavors, “astringent” feeling factor and “hard” texture gradually increased (P < 0.05) over the 21 d retail display period and ranged between 0 and 1.5 on the 0- to 15- point scale across all frozen storage periods. The samples were considered inedible and were not evaluated for their flavor and textural attributes following 21 d of refrigerated storage due to the appearance of surface growth following all frozen storage periods.

The majority of the antioxidant combinations resulted in a decrease in aroma intensity from d 0 to 7 and then increased from d 14 to 21 following 0.5 mo of frozen storage (Fig. 6A, Table 2). Ginger aroma decreased as retail display progressed with intensity increasing (P < 0.05) as rosemary concentration increased (Fig. 6C, Table 2). Treatment combinations with higher concentrations of green tea extract resulted in sausages with more “caramelized” (Fig. 6B, Table 2) aroma and were associated with fewer negative drivers of liking such as “off-odor” (Fig. 6D, Table 2), “rancid” flavor (Fig. 7A, Table 2) and “off-flavor” (Fig. 7B, Table 2) as retail display progressed. Pork sausages with G300 had the highest intensity in “rancid” flavor at d 7 but it ultimately resulted in the lowest throughout the rest of the retail display period. These results were in agreement with the findings of Martínez et al. (2006) who reported a close relationship between the less intense rancidity notes of fresh pork sausages formulated with green tea extract (1,000 ppm) with those of lipid oxidation during the refrigerated storage of the product.

Pork sausages with higher concentrations of rosemary extract were associated with stronger “spice complex”, “nutmeg”, “rosemary” aromas and “fat” flavor following 3 mo of frozen storage (Table 3). The “pork complex” flavor continued to be a positive attribute for the pork sausage flavor with higher intensities found in treatment combinations with G300 for d 14 of the retail display period. Sausages formulated with G300 also displayed significantly higher (P < 0.05) oily feeling factor through 21 d of refrigerated storage after 3 mo of frozen storage (data not shown) and 6 mo of frozen storage (Table 4). The panelists identified the predominant off-odor at the end of shelf-life during simulated retail display (3 ± 1°C) as that of spoilage due to microbial growth and to a lesser extent deterioration related to oxidative processes. This deterioration was largely attributed to increased psychrotrophic growth for all treatments, which exceeded 7 log CFU/g after 21 d of simulated retail display. Off-flavors generated by lipid oxidation is, apart from microbial spoilage, the primary source of off-flavors in muscle foods (Gray et al., 1996). These results suggest that sausages could be successfully incorporated with rosemary and green tea extracts without significantly changing its initial positive sensory attributes. Both natural plant extracts provided protection to the product, inhibiting the formation of undesirable characteristics. From these results, it was apparent that the antioxidant effect of rosemary extract was significantly enhanced by green tea extract in preventing oxidation. This was probably due to the lipid antioxidant activity of rosemary extract and the ability of the green tea extract to regenerate rosemary extract (Grün, 2009). Also, green tea extract seemed to exert enhanced protection once the product had reached 6 mo of frozen storage and then was displayed under retail conditions.

On average, consumers (n = 175) preferred (P < 0.05) sausages formulated with natural plant extracts after 3 mo of storage when compared to the control, but almost all sausages had average consumer scores between 6 and 7, signifying “like slightly” to “like moderately” (Table 5). Overall, the panelists did not identify any differences (P > 0.05) between the sausages in terms of aroma, flavor, texture, and overall acceptability following 0.5 (n = 166) and 6 (n = 177) mo of frozen storage. However, this is misleading since the control treatment was already beyond shelf-life after 8 d under retail conditions when stored frozen for 6 mos. Instead, a more accurate interpretation of these results indicate that the R and G treatments were not different from the control after 0.5 mo of storage, were preferred (P < 0.05) over the control treatment after 3 mo of frozen storage, and were still acceptable after 6 mo of frozen storage and 8 d of retail display, while the control was already spoiled. According to the descriptive analyses performed by the trained panel, sausages which received higher consumer acceptability scores were characterized by “spice complex”, “ginger”, “nutmeg”, “rosemary”, “copper-herbal” aromas, “ginger”, “fat”, “bitter”, “metallic” flavors and “hard”, “chewy” textures. Descriptive attributes regarded as positive drivers for the liking of the product showed a gradual decrease in their mean attribute intensity ratings with increased frozen storage time. However, the effects of such a decrease in these attributes were not as apparent in the consumer acceptability ratings of the product as sausages with combinations of rosemary and green tea extracts were rated as “liked” even after 6 mos of frozen storage followed by 8 d of retail display (Table 5). In general, rosemary and green tea extracts were able to increase the shelf-life of pork sausage links without affecting the sensory qualities that are important to regular consumers of the product. It should be noted that rosemary extract (1500 ppm) was added to the control samples for the third replication of the experiment and was still not acceptable for consumer testing after 6 mos frozen storage and 8d retail display due to spoilage.

For 0.5 mos frozen samples, cluster analysis indicated that 92% of the respondents found treatment combinations R1500 + G100, R1500 + G300, and R2000 + G200, to be acceptable (≥ 6, like slightly) after 8 d of retail storage, followed by R2000 + G100, R2500 + G200, R2500 + G300 and control which were liked by 88, 84, 81, and 77% of the respondents, respectively (Table 6). Cluster analysis revealed that a greater number of consumers liked sausages formulated with natural plant extracts when compared to the control following 3 mo of frozen storage. After 8 d of retail display, 98% of the respondents found treatment combinations R1500 + G300, R2000 + G100, R2000 + G200, and R2500 + G300 to be acceptable (≥ 6, like slightly) followed by R1500 + G100, R2500 + G200, and control which were liked by 95, 85 and 71% of the respondents, respectively (Schilling et al., 2018). After 6 mo of storage at –20°C, the organoleptic characteristics of all the sausage treatments remained acceptable both after freezing and after refrigerated storage (Table 7). However, the control sausages displayed spoilage and detectable rancidity by d 8 of the retail display period following 6 mo of frozen storage and were not evaluated by consumers. The largest consumer group (50% of the consumers) rated all sausages as “like very much” and was partial (P < 0.05) to R1500 + G100, R1500 + G300, and R2500 + G300 sausages over R2000 + G100 sausages (Table 7). Cluster 2 which comprised of 28% of the consumers ranged from “like slightly” to “like moderately” in their liking of the products and was partial to sausages formulated with R2000 + G100. Cluster 3 (21% of the consumers preferred (P < 0.05) sausages with R2500 + G300 as “like moderately” over those from the R2000 + G100 and R2000 + G200 sausages which were rated as “neither like nor dislike”. After 8 d of retail display, all of the respondents found treatment combinations R1500 + G100, R1500 + G300, R2500 + G200, and R2500 + G300 to be acceptable followed by R2000 + G100 and R2000 + G200 which were liked by 79% of the respondents.

An estimation of the samples’ shelf-life in the present study was made based on the detection of off-odors or off-flavors by the descriptive panelists or when the product was deemed unacceptable when compared with the gold standard. Combinations of at least 2500 ppm of rosemary extract and green tea extract at 100 to 300 ppm as well as R2000 + G300 increased the shelf-life of fresh pork sausages to 16 to 18 d of storage after 0.5 mo of frozen storage in comparison to less than 14 d for the control treatment with synthetic antioxidants and 1500 ppm rosemary. Accordingly, for 3 mo of frozen storage, all treatment combinations retained the shelf-life of fresh pork sausages to 15 d of storage compared with the control whose shelf-life was limited to less than 14 d (between 8 and 13 d). After 6 mo of frozen storage, all sausage samples reached the end of their flavor shelf-life following 14 d of retail display compared with the control which displayed spoilage and detectable rancidity by d 7 under similar storage conditions. The panelists identified the predominant off-odor as that of spoilage due to microbial growth and to a lesser extent deterioration related to oxidative processes.

The addition of combined phenolic-rich rosemary and green tea extracts with synthetic antioxidants increased effectiveness in slowing down long-term frozen storage oxidation reactions, reduced psychrotrophic bacterial growth and maintained quality during simulated retail display better than using synthetic antioxidants alone. Treatment with 2,500 ppm rosemary extract significantly reduced lipid oxidation. These results correlated positively with the protective effects against oxidative rancidity as the addition of these natural plant extracts helped preserve the intensity ratings of the positive sensory attributes and instrumental color in the product without compromising other qualities that are important to consumers. Evaluation of color showed rosemary and green tea extracts were sufficient to achieve the desired red color in the product.