Inactivation of Salmonella on In-Shell Pecans during Conditioning Treatments Preceding Cracking and Shelling

Transcription

1 588 Journal of Food Protetion, Vol. 74, No. 4, 2011, Pages doi: / x.jfp Copyright G, International Assoiation for Food Protetion Inativation of Salmonella on In-Shell Peans during Conditioning Treatments Preeding Craking and Shelling LARRY R. BEUCHAT* AND DAVID A. MANN Center for Food Safety and Department of Food Siene and Tehnology, University of Georgia, 1109 Experiment Street, Griffin, Georgia , USA MS : Reeived 24 September 2010/Aepted 19 November 2010 ABSTRACT Studies were done to determine the effetiveness of onditioning treatments for killing Salmonella in and on immersioninoulated and surfae-inoulated in-shell peans. Treatment of immersion-inoulated, dried, stored peans in hlorinated water (400 mg/ml) redued Salmonella by not more than 1.6 log CFU/g. Treatment of immersion-inoulated, dried, stored peans in hlorinated water (200 mg/ml, 1 min) followed by soaking in water for 2 h at 21uC and treating for 10 min in water at 85 to 95uC redued Salmonella by.5.12 log CFU/g; treatment of nuts ontaining a low population of Salmonella (,0.60 log for 15 min at 90uC failed to eliminate the pathogen. s of $6.42 log CFU/g were ahieved by treating surfae-inoulated nuts in water at 90 or 95uC for 80 s; treatment of nuts ontaining 1.78 log CFU/g at 95uC for 10 min did not eliminate the pathogen. Salmonella on surfae-inoulated in-shell peans (kernel moisture, 4.75%; water ativity, 0.62) that had been dried and stored at 4uC for 3 to 5 weeks was more resistant to onditioning treatments than was Salmonella on surfae-inoulated peans (kernel moisture, 5.60%; water ativity, 0.73) that were not thoroughly dried. Conditioning treatments were less effetive for killing Salmonella on immersion-inoulated peans than on surfae-inoulated peans. Response of Salmonella to onditioning treatments varied, depending on the method of inoulation and whether nuts were dried and stored between the time of inoulation and treatment, whih emphasizes the importane of following praties ommonly used by ommerial pean shellers when validating the lethality of onditioning treatments. Outbreaks of salmonellosis have been assoiated with onsumption of ontaminated almonds (7, 18, 23), peanuts (22, 25), a peanut snak (20, 37), peanut butter (8, 9, 30, 34), and desiated oonut (31, 33, 35, 40). Salmonella has been isolated from almonds (7, 10, 11, 14), Brazil nuts (15, 27), ashew nuts (15), maadamia nuts (38), pistahios (26), walnuts (32), and prepakaged mixed nuts (Brazil nuts, ashew nuts, almonds, peanuts, and walnuts) (27). Nuts and other low-water-ativity (a w ) foods and food ingredients may harbor Salmonella for long periods of time and have been doumented as soures of the pathogen in outbreaks of infetions (16). In part beause of outbreaks of salmonellosis assoiated with almonds and peanut butter, the Groery Manufaturers Assoiation (17) prepared a handbook for safe proessing of nuts. Guidane for developing safety plans and validating proesses for tree nuts and peanuts is provided. Results from studies on thermal tolerane harateristis of Salmonella on almond kernels were used to develop guidelines for proess validation (1). Several studies desribing inativation of Salmonella on almond kernels upon treatment with dry heat (2, 21), moist heat (19), steam (24), hot water (2, 39), infrared heat (2, 6), hot oil (13), and propylene oxide (12) have been published. * Author for orrespondene. Tel: ; Fax: ; lbeuhat@uga.edu. Researh in the early 1970s revealed that treating immersion-inoulated in-shell peans ontaining Salmonella at a population of 5.8 log CFU/g for 2 min in water at 99uC failed to eliminate the pathogen (3). Treatment of the inshell peans in water at 93uC for 2 min did not eliminate Salmonella that had infiltrated the nuts and reahed the kernel. s in Salmonella on Nonpareil almond kernels resulting from hot water treatment at 70, 80, and 88uC have been reported to be 1.7, 3.0, and 4.6 log CFU/g, respetively, within 30 s (39). Treatment of almonds in water at 85 and 88uC for 40 and 20 s, respetively, redued Salmonella by at least 5.73 log CFU/g (2). Salmonella in peanut butter also is known to have a high tolerane to heat. Treatment at 90uC for less 30 min is not suffiient to kill large populations (5 log (28, 36). Aside from studies on peans, almonds, and peanuts, little is known about the fate of Salmonella on nuts and in nut produts during exposure to proess treatments. Salmonella an infiltrate in-shell peans during immersion in ontaminated water (4) andisabletogrowonhigh-a w nutmeats (5). Orhard ontamination of peans with Esherihia oli has been reported (29). In-shell peans and pean nutmeats are treated with hlorinated or nonhlorinated water one or more times during the leaning, onditioning, and shelling sequene. In-shell peans may be immersed in hot water as a onditioning treatment, and nutmeats are dried with fored hot air after separation from nonedible omponents of

2 J. Food Prot., Vol. 74, No. 4 BEHAVIOR OF SALMONELLA ON PECANS 589 the nut after raking. These proesses represent potential points of deontamination. Proess validation is typially done using in-shell peans that have been surfae inoulated with Enteroous faeium, a surrogate for Salmonella. The method for preparing inoulated in-shell peans has been adapted from a method developed to validate the effiay of proesses for killing Salmonella on almond kernels (1). We onduted a study foused on determining the effetiveness of onditioning treatments for killing Salmonella on and/or in immersion-inoulated and surfaeinoulated in-shell peans. The goal was to determine whether onditioning treatments used in pean shelling operations are suffiient to kill large populations of Salmonella (5 log and whether these treatments result in different lethalities, depending on the method used to inoulate nuts and whether they were dried and stored before treatment. MATERIALS AND METHODS Soure of peans. In-shell peans of the Desirable and Seedling varieties (Carya illinoinensis) were obtained from ommerial shellers in Georgia during Otober 2009 through January Unless stated otherwise, Desirable variety peans were used in experiments. Nuts were stored at 4uC until used. Serotypes and soures of salmonellae. A five-serotype mixture of Salmonella enteria was used as an inoulum. Salmonella serotypes were Anatum (strain 6802) isolated from raw peanuts, Enteritidis (strain ATCC BAA-1045) from raw almonds, Oranienburg (strain 1839) and Sundsvall (strain 1659) from peans, and Tennessee (strain K4643), whih is a linial isolate from a patient in an outbreak of salmonellosis assoiated with onsumption of peanut butter. Measurement of moisture ontent. The moisture ontent of pean nutmeats in and out of the shell was determined as previously desribed (4) before and after subjeting peans to various treatments. In-shell peans were raked with a mehanial raker, and nutmeats were separated from the shells and paking material. Nutmeats were hopped in a One-Touh Chopper (model HC306, Blak and Deker, Towson, MD). The moisture ontent was determined with a moisture analyzer (model HB43-S, Mettler Toledo, Greinfensee, Switzerland). Samples (5 g) were dried at 130uC for 5 to 7 min, depending on the initial moisture ontent. Weight loss was attributed to removal of water during drying. The perent moisture in nutmeats was alulated. Measurement of a w. A portion of the nutmeats prepared for moisture analysis was used to determine the a w. Measurement of a w was made using 3-g samples and an AquaLab water ativity meter (model CX2, Deagon Devies, In., Pullman, WA). Preparation of hlorinated water. Chlorinated water ontaining free hlorine at onentrations of 0, 100, 200, and 400 mg/ml was prepared by ombining NaOCl (Aldrih, Milwaukee, WI) with 0.05 M potassium phosphate buffer (ph 6.8, 21uC). The free hlorine ontent of the water was determined with an amperometri titrator (Hah, Ames, IA) immediately before use. Preparation of inoula for immersion inoulation. The five Salmonella serotypes were separately grown at 37uC for 24 h in trypti soy broth (TSB; Difo, BD, Sparks, MD) supplemented with 50 mg/ml nalidixi aid (TSBN). After two onseutive transfers of ultures at 24-h intervals, 1 ml of ulture was surfae spread on eah of four large (150 by 15 mm) petri plates ontaining TSBN supplemented with 15 g/liter agar (TSAN). Plates were inubated at 37uC for 24 to 26 h. To harvest ells, 6 ml of sterile 0.1% peptone was deposited on the lawn that had developed on the surfae of eah plate, and ells were suspended in the peptone by gently rubbing the lawn with a sterile glass rod. Cell suspensions harvested from four plates of eah serotype were pooled. Equal volumes (17 ml) of eah serotype suspensions were ombined to give 85 ml of the five-serotype mixture. Populations of Salmonella in single-serotype and five-serotype suspensions were determined by serially diluting suspensions in sterile 0.1% peptone and surfae plating samples (0.1 ml, in dupliate) on TSAN. Plates were inubated at 37uC for 24 h before olonies were ounted. A high-population inoulum was prepared by adding 50 ml of the five-serotype ell suspension to 5 liters of sterile deionized water. A low-population inoulum was prepared by diluting the suspension by in sterile 0.1% peptone and ombining 50 ml of the diluted suspension with 5 liters of sterile deionized water. Populations of Salmonella in the high- and low-population suspensions were determined as desribed above. Preparation of inoula for surfae inoulation. The five Salmonella serotypes were grown in TSBN at 37uC for 24 h as desribed above. Six large petri plates ontaining TSAN were surfae inoulated with 1 ml of 24-h TSBN ulture and inubated at 37uC for 24 to 26 h. Cells were harvested as desribed above, populations in serotype suspensions were determined, and 32-ml aliquots of eah suspension were pooled to give a high-population inoulum. A low-population inoulum was prepared by diluting this suspension (10 27 ) in sterile deionized water. The populations of Salmonella in high- and low-population inoula were determined. Proedure for immersion inoulation. Cleaned in-shell peans (2,500 g) at 21uC were immersed, with oasional agitation, in 5 liters of inoulum (21uC) for 5 h. Peans were then removed from the ell suspension and drained, and weight and a w were determined. Dupliate samples (five peans per sample; approximately 50 g) were analyzed for populations of Salmonella. The immersion-inoulated peans were plaed in aluminum mesh baskets and dried, with oasional mixing, in a fored-air Isotemp oven (model 851, Fisher Sientifi, Dubuque, IA) at 30uC for 21 to 23 h. The moisture ontent and a w of kernels were determined. Dried nuts were plaed in 1-gal (3.79-liter) Snap N Seal freezer bags (Kroger Co., Cininnati, OH), double bagged in a seond 1- gal Snap N Seal freezer bag, and stored a 4uC for 3 to 5 weeks before using in leaning and onditioning treatment studies. Tripliate samples, eah onsisting of five inoulated peans, were analyzed for populations of Salmonella before and after drying and after storing at 4uC. Counts obtained from peans after storing at 4uC are referred to as initial ounts in experiments foused on determining inativation harateristis of Salmonella after treatment with hlorinated water, after soaking in water, and/or after treatment with hot water. Proedures for surfae inoulation. Two proedures were used to surfae inoulate leaned in-shell peans before treatment with hlorinated water and onditioning. The first proedure followed, with minor modifiation, that used to validate the effiay of treatments used to redue Salmonella (by 4 or 5 log on almond kernels (1). Forty undamaged in-shell peans (a. 400 g, 21uC) and 25 ml of the five-serotype inoulum (21uC)

3 590 BEUCHAT AND MANN J. Food Prot., Vol. 74, No. 4 were plaed in a plasti bag (30 by 30 m), the bag was sealed, and the ontents were ontinuously mixed by hand for 1 min, resulting in absorption of essentially all of the inoulum. Inoulated nuts were plaed on a double layer of filter paper (P8 grade, Fisherbrand Qualitative, Pittsburgh, PA) on a rak in a plasti tub, loosely overed with a lid, and held at 21uC for 23 1 h. Three bathes (120 nuts) were ombined to represent one repliate bath (a. 1,200 g) before subjeting nuts to hlorinated water and onditioning treatments. The seond proedure for preparing surfae-inoulated inshell peans was similar to the first, with one exeption. Instead of holding inoulated nuts at 21uC for 23 1 h before treatment, inoulated nuts were plaed in an aluminum mesh basket and dried in a fored-air oven (Fisher Sientifi Isotemp) at 30uC for 18 h. Dry nuts were then stored in 1-gal Snap N Seal freezer bags at 4uC for 3 to 5 weeks before being exposed to hlorinated water and onditioning treatments. Moisture ontent, a w, and populations of Salmonella were measured after inoulation, after drying, and after storage at 4uC. Preliminary hlorinated water treatment studies. Conditioning leaned in-shell peans just before raking and shelling may involve spraying or soaking nuts in hlorinated water or nonhlorinated water for up to several hours, depending on the desired moisture ontent needed in kernels to minimize shattering during the rakling and shelling proess. The wetting treatment may be followed by immersing nuts in hot water. Preliminary experiments were done to determine the effetiveness of hlorinated water (0, 100, 200, and 400 mg/ml) for killing Salmonella on and/or in immersion-inoulated nuts that had been dried and stored at 4uC for 3 to 5 weeks, as desribed above. Thirty peans (a. 300 g) with high- or low-population inoula and adjusted to 4 or 21uC were immersed in 600 ml of sterile water (21uC) or water (21uC) initially ontaining free hlorine at onentrations of 100, 200, or 400 mg/ml. Five-nut samples (a. 50 g) were removed from the hlorinated water after immersion with oasional agitation for 1, 3, 8, 16, and 24 h, plaed in a sterile plasti bag, and rushed with a hammer. The rushed shells and nutmeats were immediately transferred to a Stomaher 400 bag (Seward Medial Ltd., London, UK) ontaining 200 ml of old (4uC) latose preenrihment broth (Difo, BD) supplemented with 50 mg/ml nalidixi aid (LBN). Proedures for determining the presene (by enrihment) and population of Salmonella on nuts are desribed below. Conditioning treatments. Immersion- and surfae-inoulated peans at 4 and 21uC were treated in hot water or sequentially in hlorinated water, soak water, and hot water. For treatment involving hot water only, 25 nuts were immersed in a 10-liter hot water bath (Neslab RTE 10, Thermo Eletron Corp., Newington, NH) at 75, 80, 85, 90, or 95uC( 0.01uC). After heating for 40 s to 20 min, depending on the water temperature, 5-nut samples of Desirable nuts (a. 50 g) or 11-nut samples of Seedling nuts (a. 50 g) were removed from the water, plaed in a sterile plasti bag, rushed with a hammer, and immediately transferred to a Stomaher 400 bag ontaining 200 ml of old (4uC) LBN, as desribed above. In a seond set of experiments, inoulated nuts (1,250 g) at 4uC were immersed in 2,500 ml of hlorinated (200 mg/ml) water (21uC) for 1 min and then immersed in 2,500 ml of sterile deionized water (21uC) and soaked with oasional agitation for 2 h. Nuts were then treated in hot water as desribed above. The presene (by enrihment) and populations of Salmonella on nuts were determined. Detetion and enumeration of Salmonella. Dupliate samples of in-shell peans and ell suspensions were analyzed for the presene (by enrihment) and population of Salmonella. Immediately after rushed nuts that had been subjeted to various treatments were plaed in the old LBN, the mixture was vigorously shaken by hand for 30 s. After 3 to 5 min without shaking, the mixture was again vigorously shaken for 30 s, and samples were removed for spiral plating (WASP2, Mirobiology International, Frederik, MD) and surfae plating (quadrupliate 0.25-ml samples and dupliate 0.1-ml samples) on TSAN and bismuth sulfite agar (Difo, BD) supplemented with 50 mg/ml nalidixi aid (BSAN). Samples of the LBN wash (0.1 ml, in dupliate) were serially diluted in 0.1% peptone water and surfae plated on TSAN and BSAN. Bags ontaining the preenrihment mixture of rushed nuts and LBN and the TSAN plates on whih samples had been inoulated were inubated at 37uC for 24 h; BSAN plates were inubated at 37uC for 48 h. Colonies on TSAN and BSAN that were presumed positive for Salmonella were ounted. When olonies presumptive for Salmonella did not develop on TSAN, the preenrihed LBN was streaked on BSAN. Plates were inubated at 37uC for 48 h before examining for olonies presumptive for Salmonella. For samples antiipated to have low numbers of Salmonella, 1 ml of preenrihed LBN homogenate was added to 10 ml of tetrathionate broth (Difo, BD) and 0.1 ml was added to 10 ml of Rappaport-Vassiliadis broth (Difo, BD). Enrihment broths were inubated at 37uC for 24 h and 42uC for 24 h, respetively, and then streaked on BSAN. Presumptive-positive olonies that formed on BSAN plates within 48 h at 37uC were randomly seleted and onfirmed as Salmonella with the BBL Enterotube II (BD, Franklin Lakes, NJ) or API 20E (biomérieux Vitek, Hazelwood, MO) assays and a Salmonella latex agglutination test (Oxoid, Basingstoke, UK). The detetion limit for enumerating Salmonella by diret plating was 4 CFU/g of in-shell peans. The detetion limit by enrihment was 1 CFU per approximately 50 g of peans. Statistial analysis. All experiments were repliated three to six times. Values (n ~ 6 to 12) from dupliate samples representing eah test parameter ombination in eah repliate trial were analyzed with a general linear model of the SAS software (version 9.1, SAS Institute, Cary, NC). The least signifiant differene test was used to determine signifiant differenes (a ~ 0.05) among mean values. RESULTS AND DISCUSSION In all experiments, higher populations of Salmonella were more often reovered on TSAN than on BSAN. In some ases, these differenes were signifiant (a ~ 0.05). For this reason, only mean values alulated using ounts obtained by plating samples on TSAN are reported. Treatment with hlorinated water. Debris is separated from in-shell peans upon delivery from the orhard to storage failities, aumulators, or shellers (Fig. 1). The leaning proess involves separation of nuts from nonedible materials either by immersion in water or by fored air. Water used to lean nuts is typially hlorinated, but beause of soil, dust, and other materials with high organi ontent that are ommingled with peans, the onentration of free hlorine rapidly dereases. Nevertheless, any redution in the number of Salmonella on in-shell peans during the leaning proess would ontribute to the desired

4 J. Food Prot., Vol. 74, No. 4 BEHAVIOR OF SALMONELLA ON PECANS 591 FIGURE 1. Flow diagram showing general steps in harvesting and proessing peans. aumulative log redution resulting from the ontinuum of leaning, onditioning, and shelling operations. Cleaned inshell peans may or may not be treated with hlorinated water just before spraying with or soaking in hlorinated or nonhlorinated water as part of the onditioning proess. These treatments would also be expeted to ause a redution in Salmonella. In the first series of experiments, immersion-inoulated nuts were used to simulate ontamination that may our as a result of exposure of nuts to water ontaining Salmonella, either in preharvest environments or during storage or leaning. We examined the effiay of hlorinated water treatment for killing Salmonella on and/or in leaned, immersion-inoulated in-shell peans that had been dried and stored at 4uC for 3 to 5 weeks. Results from studies using nuts adjusted to 21uC before treatment in hlorinated water at 21uC are shown in Figure 2. The behavior of Salmonella on peans initially at 4uC (not shown) was similar. An initial population of 0.63 log CFU/g was not signifiantly redued after peans were immersed in solutions ontaining hlorine initially at onentrations up to 400 mg/ml. Compared with the initial ount, signifiantly higher populations (a ~ 0.05) of the pathogen were reovered from nuts after treatment in hlorinated water for 16 h and again after 24 h. Signifiantly more salmonellae were reovered from the ontrol peans (0 mg/ml hlorine) ompared with peans treated in hlorinated water at eah sampling time between 1 and 24 h. The population of Salmonella reovered inreased by 2.20 log CFU/g for peans soaked in water ontaining no FIGURE 2. Population of Salmonella reovered from immersioninoulated, dried, stored (4uC, 3 to 5 weeks) in-shell peans treated with hlorinated water for up to 24 h at 21uC. added hlorine. The population of Salmonella reovered inreased as the soak time progressed, regardless of the presene of hlorine. This inrease ould be attributed in part to inreased release of ells that had been initially attahed to shell and internal tissue surfaes. When peans initially ontained a higher population of Salmonella (5.62 log, signifiant dereases of 0.41 to 0.98 log CFU/g ourred within 1 h after immersion in water ontaining 100 to 400 mg/ml hlorine. Thereafter, gradual inreases in populations ourred during the subsequent 23-h soak period. The number of Salmonella ells reovered from peans soaked for 24 h in water ontaining no added hlorine was 0.56 log CFU/g higher than the number reovered from peans before soaking. Conentrations of free hlorine in the soak water were not maintained during the 24-h treatment but would likely have dereased substantially. Lethality attributable to hlorine was evident only within the first hour of treatment of high-population inoulum nuts, whih suggests that salmonellae that survived treatment for 1 h were released from the surfae of external and internal pean tissues during the 24-h soak, resulting in more aurate ounts. In any ase, treatment of leaned, immersion-inoulated, dried peans with initial onentrations of free hlorine up to 400 mg/ml annot be relied upon to eliminate Salmonella. Maintaining the onentration of free hlorine at a lethal level throughout the soak period is predited to result in greater redutions in

5 592 BEUCHAT AND MANN J. Food Prot., Vol. 74, No. 4 Salmonella but may not be suffiient to kill large numbers (4 to 5 log of the pathogen ells, partiularly those that have infiltrated internal tissues. In a previous study, we observed that Salmonella rapidly infiltrates undamaged inshell peans when nuts are immersed in a suspension of the pathogen (4). Salmonella ells that reah the kernel an survive for several months one the nuts are dried. Any hlorine in water that infiltrates undamaged in-shell peans would likely not be in a lethal form upon reahing the kernel. Treatment of immersion-inoulated, dried, stored nuts. In the next series of experiments, we foused on determining the effetiveness of treatments with hot water (75 to 95uC) for up to 20 min or with hlorinated water (200 mg/ml) for 1 min followed by soaking in tepid water (21uC) for 2 h and hot water (75 to 95uC) for up to 20 min for killing Salmonella on and/or in immersion-inoulated inshell nuts that had been dried (kernel moisture, 3.50%; a w, 0.51) and stored at 4uC for 3 to 5 weeks These treatments simulate onditioning steps to whih in-shell peans may be exposed on a ommerial sale. The populations of Salmonella reovered from nuts after treatment with hot water for up to 20 min are shown in Table 1. Rates of inativation were not markedly different in nuts initially kept at 4uC ompared with nuts initially kept at 21uC. s of $4.82 log CFU/g were ahieved by immersing nuts in water at 95uC for as little as 5 min. Beause the limit of detetion was 0.60 log CFU/g by diret plating, redutions of.4.82 log CFU/g ould not be alulated. However, Salmonella was deteted by enrihment in many samples in whih redutions were.4.82 log CFU/g, indiating that treatment at 80 to 95uC for as long as 20 min was not suffiient to eliminate the pathogen when initially present at 5.42 log CFU/g. Means of values from repliate trials are presented in this and other tables. These means have standard deviations, thus refleting values from one or more repliate experiments in whih the redution(s) was less than the alulated mean and therefore less than the log redutions reported. Treatment at 85 to 95uC for 15 or 20 min was subjetively judged to ause the testa on the kernel to darken, thereby ompromising sensory quality. Table 2 ontains the results from hot water treatment of immersion-inoulated, dried in-shell peans initially ontaining a low population of Salmonella (,0.60 log. The detetion limit was 4 CFU/g (0.60 log by diret plating and 1 CFU/50 g by enrihment. As with the studies using peans with a higher level inoulum (Table 1), the initial temperature of the nut did not have a marked effet on survival of Salmonella upon immersion in hot water. Even with initial ounts of,0.60 log CFU/g, the pathogen survived in 50-g samples treated at 75 to 95uC for up to 20 min. Experiments were done to determine the extent of inativation of Salmonella on and/or in immersion-inoulated, dried in-shell peans sequentially treated in hlorinated water (200 mg/ml) at 21uC for 1 min, soaked in water at 21uC for 2 h, and treated in hot water. s in Salmonella aused by treatment with hlorinated water were TABLE 1. of Salmonella from immersion-inoulated, dried, stored in-shell peans (Desirable variety) after treatment in hot water for up to 20 min Salmonella reovered or deteted after hot water treatment a 5 min 10 min 15 min 20 min Detetion Detetion Detetion b Detetion 0 min reovery Water temp (uc) Initial nut temp (uc) A A3.87 B 1.55 A 2.24 C 3.18 A 2.19 C C A B 4.22 A 1.20 A 1.85 B 3.57 A 1.89 B 1/1 3.53,0.60 3/ A B2.58 B 2.84 A 1.57 C 1/1 3.85,0.60 1/3.4.82,0.60 2/ A B 2.73 B 2.69,0.60 2/3.4.82,0.60 3/3.4.82,0.60 2/ ,0.60 3/3.4.82,0.60 2/3.4.82,0.60 3/3.4.82,0.60 1/ A A3.51 B 1.91 A 2.93 BC 2.49 A 1.92 C 3.50 A 2.19 C A A3.46 B 1.96 AB 2.43 BC 2.99 A 0.95 D 1/ A 1.56 CD 1/ A A 3.38 B 2.04 B 0.60 C 2/2 4.82,0.60 2/ A 1.38 C 2/ A B1.65 B 3.77,0.60 1/3.4.82,0.60 0/ A 1.20 B 1/ A B0.60 B 1/ B 0.60 B 0/2 4.82,0.60 0/3.4.82,0.60 2/ a Within the same row, reovery values that are not followed by the same letter are signifiantly different (a ~ 0.05). Within the same initial nut temperature and the same olumn, reovery values that are not preeded by the same letter are signifiantly different (a ~ 0.05). Limit of detetion by diret plating was 4 CFU/g (0.60 log. b Number of samples positive for Salmonella after enrihment/number of samples analyzed. Values are shown only for the samples (one, two, or three samples) from three repliate trials for whih no ounts were obtained by diret plating. Limit of detetion by enrihment was 1 CFU per five-pean sample (1 CFU/50 g). ompared with the initial population (0 min).

6 J. Food Prot., Vol. 74, No. 4 BEHAVIOR OF SALMONELLA ON PECANS 593 TABLE 2. Detetion of Salmonella in immersion-inoulated, dried, stored in-shell peans (Desirable variety) after treatment in hot water for up to 20 min No. of samples positive for Salmonella a Initial nut temp (uc) Water temp (uc) 0 min 5 min 10 min 15 min 20 min a Number of samples positive for Salmonella by enrihment from six five-pean (50-g) samples in six repliate trails. Initial population (0 min) was,4 CFU/g (,0.60 log. Limit of detetion by enrihment was 1 CFU per five-pean sample (1 CFU/50 g). TABLE 3. in the populations of Salmonella in and on immersion-inoulated, dried, stored and surfae-inoulated (not dried or stored) in-shell peans (Desirable variety) aused by treatment in hlorinated water followed by soaking in water Salmonella Method of inoulation Inoulum level Treatment a Detetion b Immersion High Sequene 1 Initial population 5.72 A After treatment Chlorine (200 mg/ml, 1 min) 4.18 B 1.54 Chlorine (200 mg/ml, 1 min) z water (2 h) 4.39 B 1.33 Sequene 2 (ontrol) Initial population 5.72 A After treatment Water (1 min) 5.45 A 0.27 Water (1 min) z water (2 h) 5.44 A 0.28 Surfae High Sequene 1 Initial population 8.28 A After treatment Chlorine (200 mg/ml, 1 min) 7.06 B 1.22 Chlorine (200 mg/ml, 1 min) z water (2 h) 6.45 C 1.83 Sequene 2 (ontrol) Initial population 8.28 A After treatment Water (1 min) 7.82 B 0.46 Water (1 min) z water (2 h) 7.38 C 0.90 Low Sequene 1 Initial population 1.78 After treatment Chlorine (200 mg/ml, 1 min) / Chlorine (200 mg/ml, 1 min) z water (2 h),0.60 2/ Sequene 2 (ontrol) Initial population 1.78 After treatment Water (1 min) / Water (1 min) z water (2 h) / a Within the same method of inoulation, inoulum level, and treatment sequene, reovery values that are not followed by the same letter are signifiantly different (a ~ 0.05). Limit of detetion by diret plating was 4 CFU/g (0.60 log. b Number of samples positive for Salmonella by enrihment/number of samples analyzed. Values are shown only for the samples (one, two, or three samples) from three repliate trials for whih no ounts were obtained by diret plating. Limit of detetion by enrihment was 1 CFU per five-pean sample (1 CFU/50 g). ompared with initial population.

7 594 BEUCHAT AND MANN J. Food Prot., Vol. 74, No. 4 signifiant (a ~ 0.05) (Table 3). Subsequent soaking in old water did not result in an additional signifiant derease. s of 1.33 to 1.54 log CFU/g were ahieved by these treatments ompared with redutions of 0.27 to 0.28 log CFU/g using water without added hlorine as an initial treatment. Table 4 ontains the populations of Salmonella reovered from immersion-inoulated peans that were dried and stored at 4uC for 3 to 5 weeks and then treated in hlorinated water for 1 min, soaked in old water for 2 h, and treated in hot water. Beause the results of previous experiments indiated that the initial temperature of the nuts (4 or 21uC) did not markedly affet the extent of Salmonella inativation, only nuts initially kept at 21uC were used in this and subsequent studies. s in Salmonella populations (ompared with the initial population) are also presented. Treatment in hot water (75 to 95uC) for 5 min after treatment in hlorinated water and soaking in water was not suffiient to redue Salmonella by 5 log CFU/g. However, with the possible exeption of a 10-min treatment at 80uC, treatment at 75 to 95uC for 10 min or longer resulted in redutions of.5.12 log CFU/g. Treatment with hlorinated water, soaking in water, and then treatment in hot water, partiularly for 5 and 10 min, inreased redutions ompared with those obtained for nuts treated with only hot water (Table 1). The same experiments were done using nuts ontaining low populations of Salmonella (,0.60 log. Results are shown in Table 5. Although treatment in hlorinated water and then soaking in water before treatment in hot water appears to be more effetive for eliminating low populations of Salmonella ompared with treatment with only hot water (Table 2), viable Salmonella ells ould still be deteted in peans treated in hot (90uC) water for 15 min. Treatment with water (ontrol) instead of hlorinated water for 1 min followed by soaking in water for 2 h and treating in hot water (75 and 80uC) for 20 min failed to eliminate the pathogen. Regardless of the treatment parameters, salmonellae that may have infiltrated the nuts during immersion inoulation are apparently proteted against inativation by hlorine and heat. Treatment of surfae-inoulated in-shell peans that were not dried and stored. The method developed to validate the effiay of treatments for inativating Salmonella on almond kernels (1) is urrently used, with minor modifiations, by proess authorities to validate the effiay of onditioning treatments for killing the pathogen on inshell peans. This method involves surfae inoulation, whih simulates ontamination resulting from ontat of nuts with a soure ontaining the pathogen but results in minimal infiltration of Salmonella into the internal tissues as would be the ase, for example, if in-shell peans were immersed in water ontaining Salmonella in either pre- or postharvest environments. Although the method involves holding nuts at ambient temperature for 24 h after surfae inoulation, the nuts are not dried to a kernel moisture typially desired for long-term storage and may therefore not be representative of all nuts entering the onditioning TABLE 4. of Salmonella from immersion-inoulated, dried, stored in-shell peans (Desirable variety) after treatment in hlorinated water, soaking in water, and treatment in hot water for up to 20 min a Salmonella after hlorinated water treatment and soaking b Salmonella after hot water treatment Chlorinated water Soaking in water 5 min 10 min 15 min 20 min Detetion Detetion Detetion Detetion d Initial Salmonella population Hot water temp (uc) A 4.18 B B 1.33 A 3.23 C 2.49,0.60 3/ / / A 4.18 B B 1.33 AB 2.25 C 1/ / / / A 4.18 B B 1.33 AB 1.14 C 0/1 4.58,0.60 0/ / / A 4.18 B B 1.33 AB 1.82 C 1/1 4.64,0.60 0/ / / A 4.18 B B 1.33 B 0.90 C 2/2 4.82,0.60 0/ / / a Within the same row, values that are not followed by the same letter are signifiantly different (a ~ 0.05). Within the same olumn, values that are not preeded by the same letter are signifiantly different (a ~ 0.05). Limit of detetion by diret plating was 4 CFU/g (0.60 log. b Peans (21uC) were treated with hlorinated water (200 mg/ml) for 1 min at 21uC, soaked in water at for 2 h at 21uC, and then treated in hot water. ompared with initial population. d Number of samples positive for Salmonella by enrihment/number of samples analyzed. Values are shown only for the samples (one, two, or three samples) from three repliate trials for whih no ounts were obtained by diret plating. Limit of detetion by enrihment was 1 CFU per five-pean sample (1 CFU/50 g).

8 J. Food Prot., Vol. 74, No. 4 BEHAVIOR OF SALMONELLA ON PECANS 595 TABLE 5. Detetion of Salmonella in immersion-inoulated, dried, stored in-shell peans (Desirable variety) after treatment in hlorinated water, soaking in water, and treatment in hot water for up to 20 min No. of samples positive for Salmonella a Treatment Water temp (uc) 0 min 5 min 10 min 15 min 20 min Water for 1 min and then water for 2 h (ontrol) Chlorinated water (200 mg/ml) for 1 min and then water for 2 h a Number of samples positive for Salmonella from six 50-g samples in six repliate trials. Initial population was,0.60 log CFU/g (,4. Limit of detetion by enrihment was 1 CFU per five-pean sample (1 CFU/50 g). step of raking and shelling operations. The kernel moisture in surfae-inoulated in-shell peans prepared using essentially the same method reommended for almond kernels (1) was 5.60% (a w, 0.73). We onduted studies to determine and ompare the effiay of onditioning treatments for killing Salmonella on surfae-inoulated in-shell peans that were or were not dried and stored at refrigeration temperature after inoulation. We also ompared the effiay of onditioning treatments for killing Salmonella on immersion-inoulated, dried, stored nuts with the effiay of treatments for killing the pathogen on surfae-inoulated nuts that were or were not dried and stored before treatment. Table 6 provides the results from studies designed to determine the effetiveness of hot water treatment for killing Salmonella on surfae-inoulated in-shell peans that were not thoroughly dried after inoulation (kernel moisture, 5.60%; a w, 0.73 ), i.e., using the method typially followed by pean proess authorities. An initial population of 8.33 log CFU/g was redued by $6.43 log CFU/g by heating nuts at 80 to 95uC for 2 min; treatment at 75 to 95uC for 5 min aused redutions of $7.13 log CFU/g. However, heat treatment for up to 15 min at these temperatures did not always eliminate Salmonella, as indiated by enrihment results. Even when the initial population was 1.78 log CFU/ g, treatment at 95uC for 10 min or 75 and 80uC for 15 min failed to eliminate the pathogen (Table 7). These studies reveal that inativation of large numbers of Salmonella ells ours within 2 min when nuts are treated in hot water, but extended treatment is insuffiient to eliminate the pathogen when initially present at a population of 1.78 log CFU/g. Beause.6-log redution was ahieved within 2 min by treating surfae-inoulated peans in hot water, further studies were onduted to determine minimum heating temperature and time ombinations needed to kill 5 log CFU/g. Table 8 provides results of studies in whih surfaeinoulated in-shell peans were treated in hot water for 40 to 240 s. An initial population of 7.47 log CFU/g was redued by $6.42 log CFU/g in nuts treated at 90 or 95uC for $80 s and by $5.75 log CFU/g in nuts treated for 240 s at 85 to 95uC. Compared with 2-min values shown in Table 6, redutions resulting from 120-s treatments (Table 8) are less overall. We attribute this finding, in part, to differenes in higher initial populations on nuts used in the two experiments. The higher the initial Salmonella population, the greater the redution, partiularly early in the treatment. Inativation urves were typially upwardly onave, with a preipitous derease in population during the initial seonds of exposure to hot water. Considering all results shown in Tables 6 and 8, redutions of 5 log CFU/g an be ahieved within 80 s by treatment in hot water at 90 or 95uC and perhaps at lower temperatures or shorter heating times if the initial population were higher. Studies on the effetiveness of hot water treatment for killing Salmonella on surfae-inoulated in-shell peans were followed by studies to determine whether treatment in hlorinated water and soaking in water before treatment in hot water would inrease total log redutions. With the same method used to validate inativation of Salmonella in some ommerial onditioning operations, an initial population of 8.28 log CFU/g was signifiantly redued by 1.22 and 1.83 log CFU/g, respetively, on surfae-inoulated peans by treating nuts with hlorinated water (200 mg/ml) at 21uC for 1 min or treating nuts with hlorinated water and then soaking them in water at 21uC for 2 h (Table 3). Treatments with nonhlorinated water or nonhlorinated water followed by soaking in water aused redutions of 0.46 and 0.90 log CFU/g, respetively. The water soaking resulted in signifiant redutions in ounts on nuts, regardless of pretreatment in hlorinated water or nonhlorinated water, indiating that detahment of ells ourred during the soaking period. This redution did not our on immersion-inoulated peans (Table 3); a higher proportion of ells were assumed to be present on internal tissues in these peans and less likely to be released from nuts during the soaking period. With an initial lower population on surfae-inoulated peans (1.78 log, redutions of 1.18 and 0.88 log CFU/g were ahieved by treating nuts in hlorinated or nonhlorinated water, respetively. Table 9 lists the redutions in Salmonella populations on surfae-inoulated in-shell peans treated in hlorinated

9 596 BEUCHAT AND MANN J. Food Prot., Vol. 74, No. 4 TABLE 6. of Salmonella from surfae-inoulated in-shell peans (Desirable variety) that were not dried or stored but were treated in hot water for up to 15 min Salmonella reovered or deteted after hot water treatment a 2 min 5 min 10 min 15 min Detetion Detetion Detetion b Detetion 0 min, reovery Water temp (uc) A A 4.64 B 3.69,0.60 3/3.7.73,0.60 2/ / A B 1.90 B /2 7.25,0.60 2/ / A B 1.75 B / / / A B 0.60 B 2/2 7.73,0.60 3/3.7.73,0.60 2/ / A B 0.82 B 7.51,0.60 3/3.7.73,0.60 2/ / a Within the same row, values that are not followed by the same letter are signifiantly different (a ~ 0.05). Within the same olumn, values that are not preeded by the same letter are signifiantly different (a ~ 0.05). Limit of detetion by diret plating was 4 CFU/g (0.60 log. b Number of samples positive for Salmonella by enrihment/number of samples analyzed. Values are shown only for the samples (one, two, or three samples) from three repliate trials for whih no ounts were obtained by diret plating. Limit by enrihment was 1 CFU per five-pean sample (1 CFU/50 g). ompared with initial population (0 min). TABLE 7. Detetion of Salmonella on surfae-inoulated in-shell peans (Desirable variety) that were not dried or stored but were treated in hot water for up to 15 min Water temp (uc) No. of samples positive for Salmonella a 0 min 2 min 5 min 10 min 15 min a Number of samples positive for Salmonella by enrihment from three five-pean (50-g) samples in three repliate trials. Initial population (0 min) was 1.78 log CFU/g. Limit of detetion by enrihment was 1 CFU per five-pean sample (1 CFU/50 g). water (200 mg/ml, 1 min) and soaked in water (2 h) at 21uC before treatment with hot water. Hot water treatment for 2 min at 80 to 95uC after treatment in hlorinated water and soaking in water aused $5.16-log redutions in Salmonella. Treatment at 75 to 95uC for 5 min redued populations by 6.79 to.7.68 log CFU/g, but treatment for 15 min did not always eliminate the pathogen. s on surfaeinoulated nuts treated with hlorinated water, soaking, and then hot water (Table 9) were not markedly different from redutions on nuts treated with only hot water (Table 6). When nuts with initial low populations of Salmonella (,0.60 log were treated with hlorinated water, soaked in water, and then treated with hot water, the pathogen was eliminated in two-thirds of the samples (Table 10). However, as with nuts ontaining an initial high population, hot water treatment for up to 15 min annot be relied upon to eliminate Salmonella initially present in low numbers. Observations on treatments ausing redutions in Salmonella of more than 5 log CFU/g on surfae-inoulated peans within 2 min (Table 9) were followed by studies in whih nuts were subjeted to treatments with hlorinated water and soaked in water before treatment with hot water for shorter times. An initial Salmonella population of 7.95 log CFU/g was signifiantly redued by 0.67 and 1.19 log CFU/g by sequential treatments with hlorinated water and soak water, respetively (Table 11). s using nonhlorinated water or nonhlorinated water followed by soaking in water were 0.52 and 0.59 log CFU/g, respetively (values not shown). As with studies using only a hot water treatment (Table 8),.5-log redutions in Salmonella were ahieved by heating nuts at 90 or 95uC for 80 s. Comparison of redutions aused by sequential treatments with hlorinated water, soak water, and hot water (Table 11) versus treatment with only hot water (Table 8) revealed similar redutions at various treatment temperature and time ombinations. The average weight of Desirable variety in-shell peans used in this study was 9.7 g. Seedling peans are muh smaller (a. 4.6 g per nut) and therefore had a higher surfae area/weight ratio than did the Desirable peans. We investigated whether this differene in surfae:weight ratio should be onsidered when validating the effiay of

10 J. Food Prot., Vol. 74, No. 4 BEHAVIOR OF SALMONELLA ON PECANS 597 TABLE 8. of Salmonella from surfae-inoulated in-shell peans (Desirable variety) that were not dried or stored but were treated in hot water for up to 240 s Salmonella reovered or deteted after hot water treatment a Water temp (uc) 0 s, reovery 40 s 80 s 120 s 240 s b Detetion Detetion Detetion A A5.02 B 2.45 A 4.33 BC 3.14 A 3.38 CD 4.09 A 3.03 D 1/ A AB 4.45 B 3.02 A 3.57 B 3.90 A 3.62 B 3.85 A 3.83 B 2/ A A4.74 B 2.73 A 4.03 B 1/ A 4.36 B 1/ B 1.72 C 1/ A AB4.27 B 3.20 B 1.05 C 1/3 6.42,0.60 1/3.6.87,0.60 1/ A B3.17 B 4.30,0.60 3/ B 1.30 C 0/1.6.87,0.60 0/ a Within the same row, values that are not followed by the same letter are signifiantly different (a ~ 0.05). Within the same olumn, values that are not preeded by the same letter are signifiantly different (a ~ 0.05). Limit of detetion by diret plating was 4 CFU/g (0.60 log. b ompared with initial population (0 min). Number of samples positive for Salmonella by enrihment/number of samples analyzed. Values are shown only for the samples (one, two, or three samples) from three repliate trials for whih no ounts were obtained by diret plating. Limit of detetion by enrihment was 1 CFU per five-pean sample (1 CFU/50 g). TABLE 9. of Salmonella from surfae-inoulated in-shell peans (Desirable variety) that were not dried or stored but were treated in hlorinated water, soaked in water, and treated in hot water for up to 15 min a Salmonella after hlorinated water treatment and soaking b Salmonella reovered or deteted after hot water treatment Hot water temp (uc) Initial Salmonella population Chlorinated water Soaking in water 2 min 5 min 10 min 15 min Detetion d Detetion Detetion A 7.06 B C 1.83 A 3.71 D E 2/ E 1/ E A 7.06 B C 1.83 A 3.12 D E 2/2 7.68,0.60 3/ A 7.06 B C 1.83 B 1.37 D D 2/2 7.68,0.60 3/3.7.68,0.60 0/ A 7.06 B C 1.83 B 1.78 D D 1/1 7.44,0.60 2/3.7.68,0.60 0/ A 7.06 B C 1.83 B 1.38 D D 2/2 7.64,0.60 1/ / a Within the same row, values that are not followed by the same letter are signifiantly different (a ~ 0.05). Within the same olumn, values that are not preeded by the same letter are signifiantly different (a ~ 0.05). Limit of detetion by diret plating was 4 CFU/g (0.60 log. b Peans (21uC) were treated with hlorinated water (200 mg/ml) for 1 min at 21uC and soaked in water for 2 h at 21uC before treating with hot water. ompared with initial population (0 min). d Number of samples positive for Salmonella by enrihment/number of samples analyzed. Values are shown only for the samples (one, two, or three samples) from three repliate trials for whih no ounts were obtained by diret plating. Limit of detetion by enrihment was 1 CFU per five-pean sample (1 CFU/50 g).

11 598 BEUCHAT AND MANN J. Food Prot., Vol. 74, No. 4 TABLE 10. Detetion of Salmonella in surfae-inoulated in-shell peans (Desirable variety) that were not dried or stored but were treated in hlorinated water, soaked in water, and treated in hot water for up to 15 min No. of samples positive for Salmonella a Treatment Water temp (uc) 0 min 2 min 5 min 10 min 15 min Water for 1 min and then water for 2 h (ontrol) Chlorinated water (200 mg/ml) for 1 min and then water for 2 h a Number of samples positive for Salmonella from three 50-g samples in three repliate trials. Initial population was,0.60 log CFU/g (,4. Limit of detetion by enrihment was 1 CFU per five-pean sample (1 CFU/50 g). onditioning treatments for killing Salmonella on surfaeinoulated in-shell peans. Seedling peans were surfae inoulated using the method followed to validate the effiay of ommerial pean onditioning treatments for killing Salmonella. An initial population of 8.26 log CFU/g was redued by 1.59 and 2.09 log CFU/g after treatment with hlorinated water and soak water, respetively (Table 12); treatment with nonhlorinated water and soak water aused redutions of 0.83 and 1.23 log CFU/g, respetively (values not shown). These redutions are similar to those ahieved with surfae-inoulated Desirable peans. Subsequent treatment of Seedling peans with hot water aused redutions in Salmonella similar to those resulting from treatment of Desirable peans (Table 9). Regardless of the pean variety, redutions exeeding 6 log CFU/g an be ahieved by treating surfae-inoulated peans with hlorinated water and soak water followed by treatment with hot water (75 to 95uC) for 5 to 15 min. However, temperature and time treatment ombinations tested in this study annot be relied on to eliminate Salmonella initially present at high populations (8.26 to 8.28 log. Treatment of surfae-inoulated, dried, stored inshell peans. The surfae-inoulation method used to validate onditioning treatments applied by pean shellers does not mimi situations in whih immersion- or surfaeinoulated nuts are dried and held at refrigeration temperature before onditioning. Drying and storing peans after inoulation would preditably result in death of a portion of the salmonellae that infiltrate nuts and/or attah to the shell surfae during inoulation. In our study, Salmonella was redued by 0.42 log CFU/g when nuts were held for 24 h at 21uC between the time of inoulation and treatment, whereas a redution of 1.23 log CFU/g was obtained by drying nuts at 30uC for 24 h followed by storing at 4uC for 3 to 5 weeks before treatment. The Salmonella ells that survived drying and storage would, overall, be assumed to have a higher level of resistane to physial or hemial assaults, thereby resulting in lower numbers of ells being inativated upon exposure to onditioning treatments. In studies of surfae-inoulated in-shell peans that were not subsequently dried and stored, the kernel moisture was 5.6% and the a w was 0.73 at the time nuts were treated with hlorinated water, soaked in water, and treated with hot water. We investigated whether drying surfae-inoulated nuts and storing them at a refrigerated temperature (4uC) before treatment with hot water or hlorinated water, soak water, and hot water would result in hanges in inativation harateristis of Salmonella. Table 13 provides the results from experiments designed to determine the effetiveness of hot water treatment for killing Salmonella on surfaeinoulated in-shell peans that had been dried (kernel moisture, 4.75%; a w, 0.62) and stored at 4uC for 3 to 5 weeks. Compared with surfae-inoulated nuts that had not been dried and stored at 4uC (Table 6), redutions in Salmonella on surfae-inoulated, dried peans (initial population, 7.67 log using the same treatment time and temperature ombinations were substantially less, partiularly in nuts treated at 80 to 95uC for 2 min, 75 to 85uC for 5 min, or 75uC for 10 or 15 min (Table 13). s in Salmonella on surfae-inoulated in-shell peans that were dried and stored at 4uC for 3 to 5 weeks before treating with hlorinated water, soaking in water, and treating in hot water are shown in Table 14. An initial ount of 7.48 log CFU/g was redued by 0.90 and 1.09 log CFU/g by treating nuts in hlorinated water for 1 min and soaking in water for 2 h, respetively. s of 0.12 and 0.17 log CFU/g were ahieved by treating nuts in water for 1 min (ontrol) followed by soaking in water for 2 h, respetively (values not shown). Treatment in water for 95uC for 2 min or 75 to 95uC for 5 min redued Salmonella ounts by 5.64 to.6.88 log CFU/g ompared with ounts before hlorine and water soak treatments. s on surfae-inoulated nuts that were dried and stored at 4uC before treating in hlorinated water, soaking in water, and treating in hot water (Table 14) were not as large as those redutions surfae-inoulated nuts that were not dried and stored before treatment (Table 9). The number of Salmonella on surfaeinoulated in-shell peans dereased by 0.98 log CFU/g during drying at 30uC and by an additional 0.08 log CFU/g