Inactivation of Salmonella on Pecan Nutmeats by Hot Air Treatment and Oil Roasting

Transcription

1 1441 Journal of Food Protection, Vol. 74, No. 9, 2011, Pages doi: / x.jfp Copyright G, International Association for Food Protection Inactivation of Salmonella on Pecan Nutmeats by Hot Air Treatment and Oil Roasting LARRY R. BEUCHAT* AND DAVID A. MANN Center for Food Safety and Department of Food Science and Technology, University of Georgia, 1109 Experiment Street, Griffin, Georgia , USA MS : Received 11 February 2011/Accepted 27 April 2011 ABSTRACT Studies were done to determine the effectiveness of hot air drying, dry roasting, and oil roasting in killing Salmonella on pecan nutmeats. Pecan halves and pieces were inoculated by immersion in a five-serotype suspension of Salmonella or by surface application of powdered chalk containing the pathogen. Hot air treatment of low-moisture (2.8 to 4.1%) and high-moisture (10.5 to 11.2%) immersion-inoculated nutmeats (initial population, 6.18 to 7.16 log CFU/g) at 120uC for 20 min reduced the number of Salmonella by 1.18 to 1.26 and 1.89 to 2.04 log CFU/g, respectively. However, regardless of the moisture content, hot air treatment of pecan halves containing 0.77 log CFU/g at 120uC for 20 min failed to eliminate Salmonella. Reductions were.7 log CFU/g when dry pieces were dry roasted at 160uC for 15 min. Treatment of halves at 140uC for 20 min, 150uC for 15 min, or 170uC for 10 min reduced Salmonella by 5 log CFU/g. The pathogen was slightly more heat resistant in immersion-inoculated nutmeats than on surface-inoculated nutmeats. Exposure of immersion-inoculated pieces to peanut oil at 127uC for 1.5 min or 132uC for 1.0 min reduced the number of Salmonella by 5 log CFU/g. Treatment of halves at 138uC for 2.0 min reduced Salmonella by 5 log CFU/g; treatment at 132uC for 2.5 to 4.0 min did not always achieve this reduction. Hot air treatment cannot be relied upon to reduce Salmonella by 5 log CFU/g of raw pecan nutmeats without changing sensory qualities. Treatment temperatures and times typically used to oil roast nutmeats appear to be sufficient to reduce Salmonella by 5 log CFU/g. Salmonella is known to have exceptional heat resistance when cells are embedded in food matrices with low water activity (a w ). Average decimal reduction times at 90uC (D 90uC -values) for Salmonella serotypes Anatum, Senftenberg, and Typhimurium in milk chocolate have been reported to range from 11 to 75 min (3, 17). Salmonella was reduced by only 3 log CFU/g during conching of 72% chocolate at 75uC for 22 h (20). Thermal tolerance of Salmonella Typhimurium and Salmonella Alachua has been reported to increase with an increase in percent solids in milk (14). The D 75 77uC -value for Salmonella Weltevreden increased from about 30 min in wheat flour at about a w 0.57 to 150 min in flour at about a w 0.26 (1). D 49uC -values for eight Salmonella serotypes in corn flour ranged from 18 to 594 min (40). Habituation of Salmonella to reduced a w induces increased heat tolerance (30). Outbreaks of salmonellosis implicating peanuts (27), a peanut snack (24, 36), peanut butter (10, 11, 34), and almonds (9, 21, 32) as vehicles of Salmonella have raised interest in better understanding thermal tolerance characteristics of the pathogen in nuts and nut products. Heat inactivation of a three-serotype mixture of Salmonella in peanut butter used to produce a peanut butter coated snack associated in an outbreak of salmonellosis has been studied. Even by heating peanut butter at 90uC for 50 min, only a * Author for correspondence. Tel: ; Fax: ; lbeuchat@uga.edu. 3.2-log CFU/g reduction was observed (35). A study using a three-strain mixture of Salmonella Tennessee isolated from peanut butter associated with an outbreak revealed that older cells were more resistant than young cells in peanut butter (a w, 0.45) heated at 90uC for up to 50 min (29). The minimum time to obtain a 7-log reduction at 90uC was significantly greater for Salmonella Tennessee isolated from peanut butter than for clinical isolates of Salmonella Tennessee from sporadic cases or for other serotypes. Several studies have focused on determining thermal resistance of Salmonella Enteritidis on almond kernels. A 5.47-log CFU/g reduction was achieved by dry heat treatment of almonds at 149uC for 16 min (12). Dry heat treatment of almonds inoculated with a four-strain mixture of Salmonella Enteritidis at 55 or 60uC for 4 days resulted in a maximum inactivation of 1.30 log CFU/g (2). Treatment at 60uC for 4 days followed by catalytic infrared (IR) heat treatment for 70 s reduced the population by an additional 1.0 log. Exposure of almond kernels to IR heat for 30, 35, and 45 s followed by cooling at room temperature yielded reductions in Salmonella Enteritidis of 0.63, 1.03, and 1.51 log CFU/g, respectively (8). Holding kernels at warm temperatures for 60 min after IR treatment resulted in a.7.5-log reduction. Pretreatment of almonds with water, citric acid, or lactic acid before dry roasting provides a means to increase antimicrobial efficacy (25). Increased relative humidity of hot air used to treat almonds enhances the rate of inactivation of Salmonella Enteritidis (22). Steam

2 1442 BEUCHAT AND MANN J. Food Prot., Vol. 74, No. 9 treatment at 93 1uC for 65 s has been shown to reduce Salmonella Enteritidis by 4.1 and 5.8 log CFU/g of Mission and Nonpareil almond kernels, respectively (28). Salmonella Enteritidis and Salmonella Senftenberg can be reduced by.5 log CFU/g of almond kernels by treatment in hot oil at 127uC for 1.5 min (16). Less is known about thermal resistance of salmonellae on pecans. Salmonella Senftenberg is reduced by 4 log CFU/ g on moist, immersion-inoculated in-shell pecans treated in water at 99uC for 2 min (4). An initial in-shell Salmonella population of 6 log CFU/g was reduced to 2.5 and 2.7 log CFU/g of shells and nutmeats, respectively, after treatment at 93uC for 2 min. Five-log reductions of salmonellae can be achieved by applying conditioning treatments typically used in the pecan industry, but the magnitude of reduction is dependent on the method of inoculation and storage conditions between inoculation and conditioning (7). Thermal inactivation of Salmonella on pecan nutmeats upon exposure to hot air and hot oil has not been described. A recall of pecans found to contain Salmonella (39), coupled with the ability of the pathogen to infiltrate in-shell nuts (5) and survive for at least 18 months (6), raises interest in determining the effects of heat treatment of pecan nutmeats on the viability of Salmonella they may contain. We undertook a series of studies to determine the effectiveness of dry hot air treatment in killing Salmonella on pecan nutmeats. Treatment temperatures and times typically used in the pecan industry to dry nutmeats after conditioning and shelling as well as those used to dry roast nutmeats were tested. The efficacy of dry air heat treatment in killing Salmonella in and on immersion-inoculated and on surfaceinoculated nutmeats was determined. In addition, thermal inactivation of Salmonella in oil-roasted nutmeats was studied. MATERIALS AND METHODS Pecans. Desirable variety pecan (Carya illinoinensis) nutmeats obtained from a commercial pecan sheller in Georgia were used. Mammoth halves (250 halves or fewer per lb [551 or fewer per kg]) and medium pieces (maximum diameter, 6/16 in. [0.95 cm] and minimum diameter, 3/16 in. [0.48 cm]; i.e., pieces can pass through a round opening 0.95 cm in diameter but not a round opening 0.48 cm in diameter) (38) were used in all experiments. Nutmeats were stored at 4uC until used. Measurement of moisture content. Nutmeats were chopped to granule size in a One-touch Chopper (model HC306, Black and Decker, Towson, MD). The moisture content of 5-g samples was determined with a Mettler Toledo Moisture Analyzer (model HB43- S, Mettler Toledo, Greinfensee, Switzerland). Samples were dried at 130uC for 5 to 7 min (nutmeats) or 8 to 10 min (inedible nut components), depending on the initial moisture content. Weight loss was attributed to removal of water during drying. The percent moisture in nutmeats and inedible materials was calculated. Measurement of a w. Portions of the nutmeats prepared for moisture analysis were used to determine a w. Measurement of a w was made using 3-g samples and an AquaLab Model CX2 Water Activity Meter (Decagon Devices, Inc., Pullman, WA). Salmonellae used. A five-serotype mixture of Salmonella enterica was used as an inoculum: Anatum, strain 6802, isolated from raw peanuts; Enteritidis, strain ATCC BAA-1045, from raw almonds; Oranienburg, strain 1839, from pecans; Sundsvall, strain 1659, from pecans; and Tennessee, strain K4643, clinical isolate from a patient in an outbreak of salmonellosis associated with consumption of peanut butter. Preparation of immersion inoculum. Salmonellae were grown at 37uC for 24 h in tryptic soy broth (TSB; Becton, Dickinson and Company [BD], Sparks, MD) supplemented with nalidixic acid (50 mg/ml) (TSBN). One milliliter of culture of each serotype was spread on the surface of TSAN (TSBN supplemented with agar [15 g/liter], 60 ml per plate) in large petri plates (150 by 15 mm). Four to eight plates were prepared, depending on the amount of inoculum needed for each experiment. Plates were incubated at 37uC for 24 to 26 h. Cells were grown on an agar medium rather than in broth because at least one of the serotypes (Salmonella Enteritidis) used in the study appears to have increased resistance to drying on almonds when cells for inoculum are grown on TSA rather than in TSB (37). To harvest cells, 5 to 6 ml of sterile 0.1% peptone water was deposited on the lawn that had developed on the surface of each plate and cells were suspended in the peptone by gently rubbing the lawn with a sterile glass rod. Cell suspensions harvested from plates inoculated with a given serotype were pooled and analyzed for populations of Salmonella (see procedure described below). Equal volumes (20 to 40 ml, depending on the experiment) of each serotype suspension were combined to give 100 to 200 ml of a fiveserotype mixture. The population of Salmonella in this mixture was also determined. Undiluted cell suspensions were used to inoculate chalk that was subsequently dried and used to surface inoculate pecan nutmeats (see procedures described below). Suspensions diluted in sterile deionized water were used for immersioninoculation studies within 1 h after preparation. Procedure for immersion inoculation. Nutmeats (mammoth halves and medium pieces) were inoculated by immersing in diluted (10 22 and ) five-serotype suspensions of salmonellae prepared as described above. Nutmeats (1,400 g at 21uC) were placed in bags fabricated from fiberglass insect screen (Phifer, Inc., Tuscaloosa, AL), immersed in suspensions (21uC) at a 1:2 ratio (wt/vol), and gently agitated for 30 s. After removing from the suspension, nutmeats were drained, removed from bags, placed in an aluminum screen basket, dried at 30uC in a forced-air convection oven for 23 1 h, and deposited in a 1-qt (0.94-liter) Snap n Seal freezer bag (Kroger Co., Cincinnati, OH) (25 g/bag). Bags containing nutmeats for each replicate trial were placed in a 1-gal (3.79-liter) Snap n Seal freezer bag and stored at 4uC for 3 to 5 weeks before being used in hot air convection heating and oil roasting experiments. Preparation of surface inoculum. A five-serotype suspension of salmonellae was prepared as described above. Nontoxic white Crayola chalk (Code # , Binney and Smith, Easton, PA) was used as a carrier for salmonellae. Chalk sticks (200 g) were immersed in 200 ml of Salmonella suspension at 4uC for 24 h. The suspension-saturated chalk was removed from the excess suspension, placed on an aluminum screen, and dried at 30uC for 24 h. The chalk was pulverized with a food processor (model 70590, type FP11, Hamilton Beach, Southern Pines, NC); particle size was subsequently reduced by using a mortar and pestle. The powdered inoculum was deposited in a glass bottle, sealed, and stored at 4uC until used to inoculate nutmeats. The moisture content and a w, as well as the number of Salmonella in powdered inoculum, were determined immediately before surface inoculating nutmeats.

3 J. Food Prot., Vol. 74, No. 9 INACTIVATION OF SALMONELLA ON PECANS 1443 Procedure for surface inoculation. Medium pecan pieces and mammoth halves (1,800 g at 21uC) were separately combined with 18 g of powdered chalk inoculum in a 1-gal (3.79-liter) Snap n Seal freezer bag (Kroger Co.) and gently tumbled for 2 min. Nutmeats were placed on a screen and gently shaken to remove nonadhering chalk inoculum from the nutmeats. Inoculated nutmeats (25-g samples) were placed in 1-qt (0.947-liter) Snap n Seal freezer bags and sealed. Bags containing samples for each replicate trial were placed in 1-gal freezer bags and stored at 4uC for 3 to 5 weeks before subjecting nutmeats to hot air convection heat treatments. A limited study was done to determine the efficacy of oil roasting in killing Salmonella. Moisture content, a w, and populations of Salmonella on the inoculated nutmeats were determined before and after storage. Hot air treatment of immersion-inoculated nutmeats. Immersion-inoculated, dried nutmeats were removed from storage at 4uC and adjusted to 21uC over a 2- to 3-h period. Nutmeats with no moisture adjustment are referred to as dry. Wet nutmeats (21uC) were prepared by immersing 25 g in 50 ml of sterile deionized water in a 150-ml beaker for 30 s, with gentle agitation, followed by decanting the water, draining the nutmeats, and placing them on paper towels for 3 to 5 min. Wet nutmeats were subjected to heat treatment within 5 min after wetting. Reductions in the number of Salmonella caused by hot air treatment of dry nutmeats (2.8 to 4.1% nutmeat moisture; a w, 0.52 to 0.61) and wet nutmeats (10.5 to 11.2% nutmeat moisture; a w, 0.94 to 0.96) were determined. Dry or wet nutmeats (25 g, dry weight basis) were arranged in a single layer in aluminum screen trays (ca. 11 cm in diameter by 2 cm in height) and placed in a forced-air Fisher Scientific Isotemp oven (control sensitivity, 0.25uC; model 851F, Fisher Scientific, Dubuque, IA). Dry and wet nutmeats (25-g samples) were exposed to forced dry air at 60, 70, 80, 90, 100, 110, and 120uC for 0, 5, 10, 15, and 20 min. In addition, dry nutmeats were also treated at 130, 140, 150, 160, and 170uC for 0, 5, 10, 15, and 20 min. The lower temperature-time treatments are in the range of or exceed those used to dry nutmeats after shelling. The higher temperature-time treatments are in the range of or exceed dry roasting conditions. Treatment temperatures were measured with two T-type thermocouples (Omega Engineering, Stamford, CT) placed approximately 1 cm above and below the single layer of 25-g sample and recorded on a Hotmux Temperature Recorder (DCC Corp., Pennsauken, NJ). Immediately after heat treatments, nutmeats were placed in a Stomacher 400 bag (Seward Medical Ltd., London, UK) containing 100 ml of cold (4uC) lactose broth (BD) supplemented with nalidixic acid (50 mg/ml) (LBN). Samples were analyzed for the presence (by enrichment) and number of Salmonella (in CFU per gram) using the methods described below. Hot air treatment of surface-inoculated nutmeats. Dry, surface-inoculated medium pecan pieces and mammoth halves were removed from storage at 4uC and adjusted to 21uC before exposing to heat in a forced-air oven as described above for immersion-inoculated nutmeats, with the exception that not all of the treatment temperature-time combinations were tested. For surface inoculation studies, 25-g samples of nutmeats were treated at 60, 90, 120, 130, 140, 150, 160, and 170uC for 0, 5, 10, 15, and 20 min. Samples were analyzed for the presence (by enrichment) and number of Salmonella. Oil roasting. The effectiveness of hot oil treatment in killing Salmonella on immersion-inoculated, dried, stored pecan pieces and halves was determined. Inoculated nutmeats were removed from storage at 4uC and adjusted to 21uC. Samples (25 g) were deposited in bags (9 by 20 cm) fabricated from fiberglass insect screen and immersed in hot peanut oil in a Neslab RTE 10 bath (Thermo Fisher Scientific, Newington, NH), which is stated by the manufacturer to have a temperature stability of 0.01uC. In our study, the oil temperature fluctuated 1uC from the target temperature. Single bags containing samples of nutmeats were immersed in peanut oil (Kroger Co.) at 110, 116, 121, 127, 132, and 138uC for 0, 0.5, 1.0, 1.5, 2.0, 2.5, 3.0, 3.5, and 4.0 min. In a limited study, surface-inoculated pieces and mammoth halves were oil roasted for 1.0 and 1.5 min, respectively, at 127, 132, and 138uC. The bags were held, with gentle agitation, 6 to 7 cm below the surface of the oil, which was 13 cm deep. The temperature in this central location of the oil bath was measured and recorded as described above. Immediately after hot oil treatment, nutmeats (25- g samples) were deposited in Stomacher 400 bags containing 100 ml of cold (4uC) LBN and analyzed for presence and number of Salmonella. Detection and enumeration of Salmonella. Stomacher 400 bags containing 25 g of treated nutmeats and 100 ml of LBN were pummeled for 1 min at normal speed. Duplicate samples of homogenates were spiral plated (WASP2, Microbiology International, Frederick, MD) or surface spread (quadruplicate 0.25-ml samples and duplicate 0.1-ml samples) on TSAN and bismuth sulfite agar (BSA; BD) supplemented with nalidixic acid (50 mg/ ml) (BSAN). Samples (0.1 ml, in duplicate) of LBN and nutmeat homogenate serially diluted in sterile 0.1% peptone water were also surface plated on TSAN and BSAN. Bags containing the preenriched homogenate, as well as the TSAN plates on which samples had been applied, were incubated at 37uC for 24 h; BSAN plates were incubated at 37uC for 48 h. Colonies formed on TSAN and BSAN that were presumptive positive for Salmonella were counted. If colonies presumptive for Salmonella did not develop on TSAN, the preenriched LBN was streaked on BSAN. Plates were incubated at 37uC for 48 h before examining for colonies presumptive for Salmonella. For samples anticipated to have low numbers of Salmonella, 1 ml of preenriched LBN homogenate was added to 10 ml of tetrathionate broth (BD) and 0.1 ml was added to 10 ml of Rappaport-Vassiliadis broth (BD). Enrichment broths were incubated for 24 h at 37 and 42uC, respectively, before streaking on BSAN. Presumptive-positive colonies that formed on BSAN plates within 48 h at 37uC were randomly selected for confirmation. Cells from these colonies were subjected to confirmation tests using BBL Enterotube II (BD) or API 20E (biomérieux Vitek, Hazelwood, MO) assays, and a Salmonella latex agglutination test (Oxoid, Basingstoke, UK). The detection limit for enumerating Salmonella by direct plating was 0.60 log CFU/g (4 CFU/g) of nutmeat. The detection limit by enrichment was 1 CFU/25 g of nutmeat. Statistical analysis. All experiments were replicated three to six times. Values from duplicate or triplicate samples representing each test parameter combination in each replicate trial were analyzed with a general linear model on SAS software (version 9.1, SAS Institute, Cary, NC). The least significant difference test was used to determine significant differences (a ~ 0.05) in mean values. RESULTS AND DISCUSSION Higher numbers of Salmonella were more often recovered from inoculated nutmeats when samples were spread on TSAN than on BSAN. In some instances, counts on TSAN were significantly higher (a ~ 0.05). Only counts obtained from samples plated on TSAN are reported.

4 1444 BEUCHAT AND MANN J. Food Prot., Vol. 74, No. 9 FIGURE 1. Inactivation of Salmonella in initially wet (10.5 to 11.2% moisture) and dry (2.8 to 4.1% moisture) pecan pieces and halves exposed to hot air treatment at 60 to 120uC for up to 20 min. We observed in another study (7) that thermal resistance of Salmonella in or on in-shell pecans is affected by the method of inoculation and conditions to which pecans are exposed between the time of inoculation and treatment with hot water. Salmonella that survived drying and storage of nuts at 4uC for 3 to 5 weeks were more heat resistant than were cells on nuts that were not dried or stored after inoculation. An increase in heat tolerance of Salmonella after exposure to low-a w environments has been reported by others (26, 30). In all studies reported here, immersion-inoculated nutmeats were dried after inoculation, and both immersion- and surfaceinoculated nutmeats were stored at 4uC for 3 to 5 weeks before being used in hot air, dry roast, and oil roast experiments. These conditions mimicked those to which contaminated nutmeats may be exposed in the pecan industry. Hot air treatment of immersion-inoculated nutmeats. Thermal inactivation curves for Salmonella in initially wet pecan pieces (11.2% moisture; a w, 0.96) and halves (10.5% moisture; a w, 0.94) and initially dry pieces (2.8% moisture; a w, 0.52) and halves (4.1% moisture; a w, 0.61) heated at 60 to 120uC for up to 20 min are shown in Figure 1. Initial populations were 7.16 and 6.18 log CFU/g of pieces and halves, respectively. Significant reductions (a ~ 0.05) of 1.89 and 2.04 log CFU/g of wet pieces and halves, respectively, were obtained by heating nutmeats at 120uC for 20 min. Salmonella was more resistant in dry nutmeats; significant reductions in dry pieces and halves heated at 120uC for 20 min were 1.26 and 1.18 log CFU/g, respectively. Reductions (0.02 to 0.10 log CFU/g) in wet and dry halves heated for up to 20 min at 60uC were not significant (a ~ 0.05), whereas heating wet and dry pieces for 20 min at 60uC resulted in significant reductions of 0.74 and 0.53 log CFU/g, respectively. A correlation between increased heat resistance of salmonellae and decreased a w in chocolate (3), wheat flour (1), and eggs (31) has been reported. Others have cited this trend in reviews of the behavior of salmonellae in these and other low-moisture foods (13, 15, 18, 33). Shown in Figure 2 are changes in the moisture content and a w of immersion-inoculated pecan pieces and halves that were dried, stored at 4uC for 3 to 5 weeks, and wetted before exposure to hot air at 60 to 120uC for up to 20 min. The shaded area indicates a moisture content in a range (3.5 to 5.5%) generally desired in nutmeats stored at refrigerated or subfreezing temperatures. Regardless of the air temperature, the moisture content in wet pieces and halves was reduced to less than 5.5% within 10 and 15 min, respectively. The dashed line indicates the a w (0.70) below which growth of molds on nutmeats does not occur or is greatly retarded. Subsequent to cracking and shelling pecans, hot air treatment for sufficient times to achieve desired moisture content or a w of nutmeats (Fig. 2) is clearly not sufficient to kill large numbers of Salmonella (Fig. 1). When dry immersion-inoculated pieces and halves were exposed to air temperatures of 60, 90, and 120uC for 5 to 20 min, the moisture content and a w were significantly reduced (Fig. 3). Reductions of Salmonella in dry nutmeats were less than those in wet nutmeats exposed to the same treatment (Fig. 1). Heating wet or dry nutmeats at temperatures as low as 80uC resulted in darkening the color of kernel testa and an increased brittleness in texture. More severe temperature-time treatments caused changes in sensory qualities that were subjectively judged to compromise the overall acceptability of the nutmeats without substantially reducing Salmonella counts. The next series of experiments was focused on determining the effectiveness of hot air in killing Salmonella initially at lower numbers on immersion-inoculated pecan pieces and halves. Shown in Table 1 are numbers of Salmonella recovered from wet and dry pieces initially

5 J. Food Prot., Vol. 74, No. 9 INACTIVATION OF SALMONELLA ON PECANS 1445 FIGURE 2. Moisture content and a w of initially wet (10.5 to 11.2% moisture) pecan pieces and halves exposed to hot air treatment at 60 to 120uC for up to 20 min. The shaded area indicates the moisture content in a range generally desired in nutmeats stored at refrigerated or subfreezing temperatures. A maximum a w of 0.70 (dashed line) prevents or greatly minimizes the growth of molds on nutmeats. containing the pathogen at populations of 2.09 and 1.20 log CFU/g, respectively. Treatment of nutmeats at 60 to 120uC for up to 20 min failed to eliminate the pathogen, regardless of the initial moisture content or count. Shown in Table 2 are the numbers of samples (of six analyzed) of wet and dry immersion-inoculated pecan halves that were positive for Salmonella after treatment with hot air at 60 to 120uC for up to 20 min. The initial population was 0.77 log CFU/g (5.9 CFU/g). Regardless of the moisture content or treatment temperature-time combination, Salmonella was detected in at least one of six samples in six replicate trials. As observed in experiments using nutmeats with much higher counts (Fig. 1), the range of air temperature-time treatments typically used by commercial pecan shellers to dry nutmeats after shelling cannot be relied upon to eliminate Salmonella, even when the pathogen is present at low populations. In an attempt to obtain greater reductions of Salmonella on immersion-inoculated nutmeats, the air temperature was increased to 130 to 170uC, a range more typically used for dry roasting. Reductions resulting from treatments at these temperatures for up to 20 min are shown in Figure 4. Dashed lines indicate 4- and 5-log CFU/g reductions. The lowest limit of detection by direct plating was 0.60 log CFU/g (shaded area). An initial population of 7.62 log CFU/g of pieces (3.5% moisture; a w, 0.63) was reduced by 2.89 log CFU/g when nutmeats were heated at 130uC for 20 min and by.7 log CFU/g within 15 min when the air temperature was 160 or 170uC. However, the pathogen was detected in 3 of 6 and 2 of 6 samples of pieces heated for 20 min at 160 and 170uC, respectively (data not shown), indicating that these harsh treatments did not eliminate the pathogen. Treatment at 130 or 140uC for 20 min failed to reduce Salmonella by 5 log CFU/g of pecan pieces. FIGURE 3. Moisture content and a w of initially dry (2.8 to 4.1% moisture) pecan pieces and halves exposed to hot air treatment at 60, 90, and 120uC for up to 20 min.

6 1446 BEUCHAT AND MANN J. Food Prot., Vol. 74, No. 9 TABLE 1. Recovery of Salmonella from immersion-inoculated Desirable var. pecan nutmeats (medium pieces) after treatment in hot air for up to 20 min Salmonella population recovered or detected after hot air treatment a 0 min 5 min 10 min 15 min 20 min log CFU/g log CFU/g En b Re c log CFU/g En Re log CFU/g En Re log CFU/g En Re Treatment temp (uc) Moisture content (%) of nutmeats A A 2.13 A z0.04 AB 1.68 A 0.41 A 1.70 A 0.39 A 1.71 A A A 2.25 A z0.16 A 1.80 A 0.29 B 1.03 B 1.06 B 0.90 B A A1.94 AB 0.15 AB 1.52 AB 1/ AB 1.20 AB 0.89 B 0.97 B A A2.20 A z0.11 BC 1.04 B 1/ BC 0.65 B 1.44 B 0.62 B 4/ A A 1.93 A 0.16 C 0.75 B 1.34 BC 0.69 B 1.40 B 0.61 B 3/ A A 1.91 A 0.18 C 0.63 B 3/ C,0.60 B 4/ C,0.60 B 5/ A A 1.38 AB 0.71 C 0.62 BC 3/ C,0.60 C 5/ C,0.60 C 3/ A A 1.22 A z0.02 A 1.08 A 0.12 A 1.07 A 0.13 A 1.00 A A A 1.07 A 0.13 A 0.99 A 1/ A 1.00 A 0.20 AB 0.76 A 2/ A A 0.76 B 1/ A 0.87 AB 1/ A 0.64 B 4/ B,0.60 B 3/ a Values (log CFU per gram) in the same row that are not followed by the same letter are significantly different (a ~ 0.05). Within the same moisture content, values (log CFU per gram) in the same column that are not preceded by the same letter are significantly different. The limit of detection by direct plating was 4 CFU/g (0.60 log CFU/g). b En, number of samples positive for Salmonella by enrichment/number of samples analyzed. Values are shown only for the samples (one to six samples) from six replicate trials not giving counts by direct plating. The limit of detection by enrichment was 1 CFU/50 g. Re, reduction (log CFU per gram) compared with initial population (0 min). c An initial population of 7.01 log CFU/g of halves (3.9% moisture; a w, 0.66) was reduced by 3.15 log CFU/g when treatment was at 130uC for 20 min (Fig. 4). Treatment at 170uC for 10 min, 150 or 160uC for 15 min, or 140uC for 20 min reduced the Salmonella count by.5 log CFU/g. As with pecan pieces, however, heating pecan halves at 160 or 170uC for 20 min did not eliminate Salmonella in 3 of 6 and 2 of 6 samples, respectively, in six replicate trials. Overall, thermal resistance of Salmonella appears to be slightly greater in dry halves than in dry pieces upon exposure to hot air treatments at 60 to 120uC (Fig. 1) and 130 to 170uC (Fig. 4). This may be attributable in part to the larger mass in halves compared to pieces, thereby presumably resulting in a reduced rate of heat penetration (particularly in the early seconds of heat treatment) into the internal high-fat (70 to 72%) tissues where some of the Salmonella may have infiltrated. The protection of microorganisms against thermal inactivation in high-fat matrices has been known for many years (19). Hot air treatment of surface-inoculated nutmeats. Thermal inactivation curves for Salmonella on dry (3.9% moisture; a w, 0.66) surface-inoculated pecan pieces and halves exposed to hot air (60 to 170uC) for up to 20 min are shown in Figure 5. Initial populations were 6.73 and 6.67 log CFU/g, respectively. Salmonella was reduced by 0.69 and 0.47 log CFU/g of pieces and halves, respectively, heated at 90uC for 20 min. Significant reductions (a ~ 0.05) of 2.36 log CFU/g of pieces and 2.43 log CFU/g of halves occurred when nutmeats were heated at 120uC for 20 min. Heating at 130uC for 20 min reduced Salmonella on pieces and halves by 3.00 and 3.30 log CFU/g, respectively. Treatment at 160 or 170uC reduced populations by.6 log CFU/g within 15 min. Salmonella was recovered by enrichment (1 of 6 replicate samples) of pieces heated at 170uC for 20 min but not in halves heated at 160uC for 20 min or 170uC for 15 or 20 min. A comparison of log reductions on dry surface-inoculated nutmeats (Fig. 5) with reductions on or in immersion-inoculated nutmeats (Figs. 1 and 4) reveals that, for nutmeats exposed to the same temperature-time combinations, overall, reductions of Salmonella were slightly greater on surface-inoculated nutmeats. This would be expected since some of the cells in immersion-inoculated nutmeats were likely to have infiltrated nutmeats, thereby affording protection against heat and, consequently, a slower rate of inactivation compared to cells on the surface of external tissues. Higher reductions of Salmonella may be achievable by exposing pecan nutmeats to other treatments, either before or after hot air treatment. Bari et al. (2) reported that hot water treatment of almond kernels inoculated with Salmonella (5.73 log CFU/g) at 88uC for 20 s, followed by IR heat treatment for 70 s, reduced the population to,1 log CFU/g. Holding almonds at 60uC for 4 days resulted in a reduction of 1.30 log CFU/g; subsequent IR treatment for 70 s reduced the pathogen by an additional 1.23 log CFU/g. IR treatment of almond kernels followed by holding kernels at warm temperatures results in substantial reductions of Salmonella Enteritidis (8). The lethality of IR treatment was greatly

7 J. Food Prot., Vol. 74, No. 9 INACTIVATION OF SALMONELLA ON PECANS 1447 TABLE 2. Detection of Salmonella in immersion-inoculated Desirable var. pecan halves (mammoth halves) after treatment in hot air for up to 20 min Moisture content (%) of nutmeats Treatment temp (uc) No. of samples positive for Salmonella a 0 min 5 min 10 min 15 min 20 min a Number of samples positive for Salmonella by enrichment of six 5-pecan (50-g) samples from six replicate trials. Initial population (0 min) in high-moisture (10.5%) pecan halves was,0.60 log CFU/g (,4 CFU/g); initial population in low-moisture (4.1%) halves was 0.77 log CFU/g (5.89 CFU/g). The limit of detection by enrichment was 1 CFU/50 g. influenced by the maximum surface temperature of kernels. Kim and Harris (25) found that pretreatment of almond kernels with citric or lactic acids decreased the number of Salmonella Enteritidis by approximately 1 log CFU/g, and additional reductions of 4.7 and 5.3 log CFU/g occurred when almonds were subsequently heated at 135uC for 40 min. Physical or chemical treatments in combination with hot air treatment may also prove to be more efficacious than hot air alone in killing Salmonella on pecan nutmeats. FIGURE 4. Inactivation of Salmonella in dry immersion-inoculated pecan pieces (3.5% moisture; a w, 0.63) and halves (3.9% moisture; a w, 0.66) exposed to hot air treatment at 130 to 170uC for up to 20 min. Dashed lines indicate 4- and 5-log CFU reductions of Salmonella per g. The limit of detection of Salmonella by direct plating was 0.60 log CFU/g (4 CFU/g) (shaded area). FIGURE 5. Inactivation of Salmonella on dry (3.9% moisture; a w, 0.66) surface-inoculated pecan pieces and halves exposed to hot air treatment at 60 to 170uC for up to 20 min. Dashed lines indicate 4- and 5-log CFU reductions of Salmonella per g. The limit of detection by direct plating was 0.60 log CFU/g (shaded area).

8 1448 BEUCHAT AND MANN J. Food Prot., Vol. 74, No. 9 observed during longer exposure times. Upwardly concave inactivation curves for salmonellae also have been observed for oil-roasted almonds (16). Possible explanations for the rapid initial reductions were that loosely attached Salmonella cells were washed off almonds when they were placed in oil, the presence of a less protected and more sensitive outer layer of cells, and a rapid decrease in a w at the almond surface. The same factors may have contributed to more rapid reductions of Salmonella on pecan nutmeats during the first minute or so of exposure to hot oil compared with slower rates of reduction thereafter. Cells that survived initial exposure to hot oil were likely afforded protection against heat inactivation as a result of decreased moisture in subsurface nutmeat tissue where they may have infiltrated. Pronounced tailing of thermal inactivation curves also has been observed for Salmonella on peanut butter. Shachar and Yaron (35) reported a 2.5-log CFU/g reduction in peanut butter heated at 90uC for 5 min, followed by an additional 0.30-log reduction after 20 min. Ma et al. (29) observed that tailing of inactivation curves for Salmonella in peanut butter may be affected by serotype and strain. FIGURE 6. Inactivation of Salmonella in dry (3.9% moisture; a w, 0.64) immersion-inoculated pecan pieces and halves exposed to oil at 110 to 138uC for up to 4 min. Dashed lines indicate 4- and 5-log CFU reductions of Salmonella per g. The limit of detection by direct plating was 0.60 log CFU/g (shaded area). Oil roast treatment of immersion-inoculated nutmeats. Immersion-inoculated, dried pecan pieces and halves stored at 4uC for 3 to 5 weeks were adjusted to 21uC before exposure to hot (110 to 138uC) oil for up to 4 min. Inactivation curves are shown in Figure 6. Less than a 5-log CFU/g reduction of Salmonella was obtained by oil roasting pieces at 110 to 121uC for 4.0 min. However, a 5-log reduction was obtained by heating pieces at 127uC for 1.5 min or at 132uC for 1.0 min. Four-log reductions were obtained in pieces roasted at 110 or 116uC for 2.5 min, 121uC for 1.5 min, or 127uC for 1.0 min. With the exception of treatment for 0.5 min, at the same temperature-time treatment combination, inactivation of Salmonella tended to be greater in pecan pieces than in halves. As with pieces, 5- log reductions of Salmonella did not occur in halves roasted at 110 to 121uC for 4.0 min. Treatment of halves at 132uC for 2.5 to 4.0 min did not always result in 5-log reductions. A 5-log reduction was obtained when halves were oil roasted at 138uC for 2.0 min. Four-log reductions were obtained in halves oil roasted at 110uC for 3.0 min, 121uC for 2.0 min, or 127uC for 1.5 min. The rate of inactivation of Salmonella on pecan pieces and halves was rapid during the first 1.0 to 1.5 min of exposure to hot oil (Fig. 6). Slower inactivation rates were Oil roast treatment of surface-inoculated nutmeats. A limited study was done to determine the efficacy of oil roasting in killing Salmonella on surface-inoculated nutmeats. Dry (3.9% moisture; a w, 0.64) surface-inoculated nutmeats that had been stored at 4uC for 3 to 5 weeks were adjusted to 21uC and oil roasted at 127, 132, and 138uC. Reductions of Salmonella on pieces exposed to oil at these temperatures for 1.0 min were 4.67 to 4.91 log CFU/g; reductions on halves roasted for 1.5 min were 4.31 to 5.24 log CFU/g. These reductions are similar to those resulting from oil roasting immersion-inoculated nutmeats at the same temperature-time combinations (Fig. 6). A comparison of our observations on thermal resistance of Salmonella on oil roasted pecan nutmeats to those reported by Du et al. (16) on almond kernels shows that the pathogen is inactivated at a somewhat slower rate on pecan nutmeats. A Weibull model was used in the almond study to predict 4- and 5-log reductions of Salmonella Enteritidis on almonds roasted in safflower oil at 127uC to be 0.74 and 1.3 min, respectively. Reductions of 2.9, 3.0, and 3.6 log CFU/g occurred within 30 s when almonds were oil roasted at 116, 121, and 127uC, respectively. This compares to reductions of 1.13, 1.21, and 1.38 log CFU/g of immersioninoculated pecan pieces and 1.87, 1.85, and 2.07 log CFU/g of immersion-inoculated halves roasted in peanut oil at 116, 121, and 127uC, respectively, for 30 s. In the Du et al. (16) study, neither Salmonella Enteritidis nor Salmonella Senftenberg was recovered by enrichment of 1-g samples after almonds inoculated at 5 log CFU/g were exposed to oil at 127uC for 1.5 min. They concluded that standard oil roasting times and temperatures should result in much greater than a 5-log reduction. The same conclusion can be made for pecan nutmeats. The observation that Salmonella is more resistant on pecan pieces than on halves upon exposure to hot oil at 110 to 138uC is contrary to observations that the pathogen is slightly more resistant on halves than on pieces exposed to

9 J. Food Prot., Vol. 74, No. 9 INACTIVATION OF SALMONELLA ON PECANS 1449 hot air at temperatures at 60 to 170uC. Differences in the rate of heat penetration resulting from exposure of nutmeats to hot air versus oil, coupled with differences in the surface/ volume ratio of the two types of nutmeats and the extent of infiltration of cells into nutmeat tissue, are factors which may have contributed to these findings. Differences in moisture content of nutmeats used in hot air and oil roasting studies may also have affected the rate of inactivation of Salmonella. The moisture contents of dry and wet nutmeats used in the hot air study were 2.8 to 4.1% and 10.5 to 11.2%, respectively. The moisture content of nutmeats used in the oil roast study was 3.7 to 5.0%. The rate of inactivation of Salmonella Enteritidis on almond kernels exposed to hot oil has been shown to be influenced by the moisture content (23). It was reported that treatment of kernels containing 5.1 and 8.6% moisture in oil at 121uC for 1 min reduced Salmonella by 2.2 and 5.1 log CFU/g, respectively. Our studies on hot air treatment of pecan halves and pieces are in agreement with these observations in that the heat resistance of Salmonella was less in wet nutmeats than in dry nutmeats. The effectiveness of hot oil treatment in killing Salmonella on or in pecan nutmeats would also be expected to be influenced by moisture content. ACKNOWLEDGMENTS This research was funded in part by the U.S. Department of Agriculture, Integrated Research, Education, and Extension Competitive Grants Program National Integrated Food Safety Initiative, National Pecan Shellers Association, National Pecan Growers Council, Georgia Pecan Growers Association, and Georgia Agricultural Commodity Commission for Pecans. We thank Larry Willson at Sunnyland Farms, Inc., Albany, GA; Tom Stevenson at Orchard Management Services, LLC, Albany, GA; and Joe Graziano at Terri Lynn Inc., Cordele, GA, for supplying in-shell pecans and pecan nutmeats for this study. We also thank Jessica Tatum and Kia Stringer for technical assistance in the laboratory. REFERENCES 1. Archer, J., E. T. Jervis, J. Bird, and J. E. Gaze Heat resistance of Salmonella Weltevreden in low-moisture environments. J. Food Prot. 61: Bari, M. L., D. Nei, I. Sotome, I. Nishina, L. Isobe, and S. Kawamoto Effectiveness of sanitizers, dry heat, hot water, and gas catalytic infrared heat treatments to inactivate Salmonella on almonds. Foodborne Pathog. Dis. 6: Barrile, J. C., and J. F. Cone Effect of added moisture on the heat resistance of Salmonella Anatum in milk chocolate. Appl. Microbiol. 19: Beuchat, L. R., and E. K. Heaton Salmonella survival on pecans as influenced by processing and storage conditions. Appl. Microbiol. 29: Beuchat, L. R., and D. A. Mann Factors affecting infiltration and survival of Salmonella in in-shell pecans and nutmeats. J. Food Prot. 73: Beuchat, L. R., and D. A. Mann Survival and growth of Salmonella in high-moisture pecan nutmeats, in-shell pecans, inedible nut components, and orchard soil. J. Food Prot. 73: Beuchat, L. R., and D. A. Mann Inactivation of Salmonella on in-shell pecans during conditioning treatments preceding cracking and shelling. J. Food Prot. 74: Brandl, M. T., Z. Pan, Z. Huynh, Y. Zhu, and T. H. McHugh Reduction of Salmonella Enteritidis population sizes on almond kernels with infrared heat. J. Food Prot. 71: Centers for Disease Control and Prevention Outbreak of Salmonella serotype Enteritidis infections associated with raw almonds United States and Canada, Morb. Mortal. Wkly. Rep. 53: Centers for Disease Control and Prevention Multistate outbreak of Salmonella serotype Tennessee infections associated with peanut butter United States, Morb. Mortal. Wkly. Rep. 56: Centers for Disease Control and Prevention Multistate outbreak of Salmonella infections associated with peanut butter and peanut butter-containing products United States, Morb. Mortal. Wkly. Rep. 58: Ceylan, E., G. Huang, and A. M. McNamara Dry heat treatment for non-pathogenic surrogate cultures for Salmonella Enteritidis on whole almonds, p Abstr. 93rd Annu. Meet. IAFP International Association for Food Protection, Des Moines, IA. 13. Chen, Y., V. N. Scott, F. A. Freier, J. Kuehm, M. Moorman, J. Meyer, T. Morille-Hinds, L. Post, L. Smoot, S. Hood, J. Shebuski, and J. Banks Control of Salmonella in low-moisture foods. III. Process validation and environmental monitoring. Food Prot. Trends 29: Dega, C. A., J. M. Goepfert, and C. H. Amundson Heat resistance of salmonellae in concentrated milk. Appl. Microbiol. 23: Doyle, M. E., and A. S. Mazzotta Review of studies on the thermal resistance of salmonellae. J. Food Prot. 63: Du, W.-X., S. J. Abd, K. L. McCarthy, and L. J. Harris Reduction of Salmonella on inoculated almonds exposed to hot oil. J. Food Prot. 73: Goepfert, J. M., and R. A. Biggie Heat resistance of Salmonella Typhimurium and Salmonella Senftenberg 775W in milk chocolate. Appl. Microbiol. 16: Grocery Manufacturers Association Control of Salmonella in low-moisture foods. Available at: SalmonellaControlGuidance.pdf. Accessed 3 May Hansen, N. H., and H. Riemann Factors affecting the heat resistance of non-sporing organisms. J. Appl. Bacteriol. 26: Haxgart, S., N. Christensen, S. Knochel, M. Ronsbo, D. Nielsen, and H. Heimdal Survival of Salmonella during storage and heating in chocolate, p In Final Program and Abstr. Book, 22nd Int. ICFMH Symp., FoodMicro 2010, Copenhagen, 30 August to 3 September Isaacs, S., J. Aramini, B. Ciebin, J. A. Farrar, R. Ahmed, D. Middleton, A. U. Chandran, L. J. Harris, M. Howes, E. Chan, A. S. Pichette, K. Campbell, A. Gupta, L. Y. Lior, M. Pearce, C. Clark, F. Rodgers, F. Jamieson, I. Brophy, and E. Ellis An international outbreak of salmonellosis associated with raw almonds contaminated with a rare phage type of Salmonella Enteritidis. J. Food Prot. 68: Jeong, S., B. P. Marks, and A. Orta-Ramirez Thermal inactivation kinetics for Salmonella Enteritidis PT30 on almonds subjected to moist-air convection heating. J. Food Prot. 72: Kaur, H., and L. J. Harris The impact of almond moisture on the survival of Salmonella Enteritidis PT30 after exposure to hot oil. Abstr. P3-47. J. Food Prot. 73(Suppl. A): Killalea, D., L. R. Ward, D. Roberts, J. de Louvois, F. Sufi, J. M. Stuart, P. G. Wall, M. Susman, M. Schweiger, P. J. Sanderson, J. S. T. Fisher, P. S. Mead, O. N. Gill, C. L. R. Bartlett, and B. Rowe International epidemiological and microbiological study of outbreak of Salmonella agona infection from ready to eat savoury snack. I. England and Wales and the United States. Br. Med. J. 313: Kim, B. U., and L. J. Harris The effect of pre-treatments on the reduction of Salmonella Enteritidis PT30 on almonds during dry roasting, p Abstr. 93rd Annu. Meet. IAFP International Association for Food Protection, Des Moines, IA. 26. Kirby, R. M., and R. Davies Survival of dehydrated cells of Salmonella typhimurium LT2 at high temperatures. J. Appl. Bacteriol. 68: Kirk, M. D., C. L. Little, M. Lem, M. Fyfe, D. Genobile, A. Tan, J. Threlfall, A. Paccagnella, D. Lightfoot, H. Lyi, L. McIntyre, L. Ward,

10 1450 BEUCHAT AND MANN J. Food Prot., Vol. 74, No. 9 D. J. Brown, S. Surnam, and I. S. T. Fisher An outbreak due to peanuts in the shell caused by Salmonella enterica serotypes Stanley and Newport sharing molecular information to solve international outbreaks. Epidemiol. Infect. 132: Lee, S.-Y., S.-W. Oh, H.-J. Chung, J. I. Reyes-de-Corcuera, J. R. Powers, and D.-H. Kang Reduction of Salmonella enterica serovar Enteritidis on the surface of raw shelled almonds by exposure to steam. J. Food Prot. 69: Ma, L., G. Zhang, P. Gerner-Smidt, V. Mantripragada, I. Ezeoke, and M. P. Doyle Thermal inactivation of Salmonella in peanut butter. J. Food Prot. 72: Mattick, K. L., F. Jorgensen, J. D. Legan, M. M. Lappin-Scott, and T. J. Humphrey Habituation of Salmonella spp. at reduced water activity and its effect on heat resistance. Appl. Environ. Microbiol. 66: McBee, L. E., and O. J. Cotterill High temperature storage of spray-dried eggs. 3. Thermal resistance of Salmonella oranienberg. Poult. Sci. 50: Muller, L. L., M. Hjertgvist, L. Payne, H. Patterson, A. Olssen, L. P. Sorshell, and Y. Anderson Cluster of Salmonella Enteritidis in Sweden 2005 to 5006 suspect source: almonds. Eurosurveillance 12: Podolak, R., E. Enache, W. Stone, D. G. Black, and P. H. Elliott Sources and risk factors for contamination, survival, persistence, and heat resistance of Salmonella in low-moisture foods. J. Food Prot. 73: Scheil, W., S. Cameron, C. Dalton, C. Murray, and D. Wilson A South Australian Salmonella Mbandaka outbreak investigation using a database to select controls. Aust. N. Z. J. Public Health 22: Shachar, D., and S. Yaron Heat tolerance of Salmonella enterica serovars Agona, Enteritidis, and Typhimurium in peanut butter. J. Food Prot. 69: Shobat, T., M. G. Green, D. Merom, O. N. Gill, A. Reisfeld, A. Matas, D. Blau, N. Gal, and P. E. Slater International epidemiological and microbiological study of outbreak of Salmonella agona infection from a ready to eat savoury snack. II. Israel. Br. Med. J. 313: Uesugi, A. R., M. D. Danyluk, and L. J. Harris Survival of Salmonella Enteritidis phage type 30 on inoculated almonds stored at 220, 4, 23, and 35uC. J. Food Prot. 69: U.S. Department of Agriculture United States standards for grades of shelled pecans. Effective July 15, 1969 (reprinted January 1997). Available at: pdf. Accessed 8 October U.S. Food and Drug Administration Recall firm press release. American pecan co. recalls pecans because of possible health risk. Available at: htm. Accessed 8 October Van Cauwenberge, J. E., R. J. Bothast, and W. F. Kwolek Thermal inactivation of eight Salmonella serotypes on dry corn flour. Appl. Environ. Microbiol. 42: