Impact of Roller Type on Salmonella Transfer during Simulated Commercial Conveyance of Tomatoes

Introduction:  Different types of roller conveyors are used in the commercial tomato packinghouses with Salmonellacross-contamination of tomatoes occurring via direct contact with the surface of these rollers during product conveying. 

Purpose:  The goal of this study was to quantify the transfer of Salmonellafrom inoculated tomatoes to three different roller surfaces and then from inoculated roller surfaces to uninoculated tomatoes using a pilot-scale packing line.         

Methods: Five lbs (2.3 kg) of greenhouse-grown red round tomatoes were dip-inoculated to contain Salmonella Typhimurium LT2 (avirulent) at 4 log CFU/g, air-dried for 2 h in a biosafety cabinet and then washed in sanitizer-free water for 2 min. Inoculated tomatoes were passed single file over a specially designed 1.5-m conveyor equipped with polyethylene, foam or brush rollers followed by 25 uninoculated tomatoes. Two tomatoes (~500 g) were collected after both 2 min of washing and roller conveying. All 25 uninoculated tomatoes were individually collected after conveying. Six polyethylene (118 cm²), three foam (362 cm²) and three brush roller (4 bunches of bristles) surface samples were collected using Kimwipes^® before and after conveying the uninoculated tomatoes. Tomatoes were hand-rubbed in 100 ml of phosphate buffer in a Whirl-pak bag for 2 min and then surface-plated on Xylose Lysine Tergitol-4 agar (XLT-4) with or without 0.45 μm membrane filtration to quantify Salmonella. Surface samples were homogenized by stomaching in 15 ml of phosphate buffer for 1 min and then similarly examined for salmonellae.

Results:  Regardless of the roller type, Salmonella populations on inoculated tomatoes did not significantly (P < 0.05) decrease during contact with the roller conveyors. After conveying uninoculated tomatoes over the product-inoculated foam rollers, all 25 previously uncontaminated tomatoes were cross-contaminated with Salmonella at >100 CFU/tomato. Using polyethylene rollers, 24 and 76% of the 25 tomatoes were cross-contaminated with Salmonella at 10 - 100 and 1 - 10 CFU/tomato, respectively. In contrast, only 8% of 25 tomatoes were cross-contaminated using brush rollers with Salmonella populations of 1 - 10 CFU/tomato. Overall, cross-contamination was greatest using foam (P < 0.05), followed by polyethylene (P < 0.05) and brush rollers (P< 0.05).   

Significance:  These findings, which clearly demonstrate the ability of Salmonella to transfer from roller conveyors to uncontaminated tomatoes during processing, are critical to the development of science-based transfer models for risk assessments.