Bacteriology of Indole Production in Shrimp Homogenates Held at Different Temperatures

Abstract

Homogenized, head-on, white shrimp (Penaeus setiferus) were held at 4, 12 and 22°C until putrefactive spoilage occurred. Repetitive bacterial sampling was performed and 1647 bacterial isolations were made from the shrimp homogenates. Of these, 42 isolates (2.6%) produced indole. Isolates that produced indole belonged to the genera Flavobacterium (52.4%), Aeromonas (23.8%), Proteus (21.4%) and Yersinia (2.5%). No Escherichia coli were isolated. Aeromonas and Proteus exhibited proteolysis and were able to produce indole in shrimp extracts without added L-tryptophan. These organisms favored higher growth temperatures. The majority of the Flavobac-terium isolates were psychrotrophic, non-proteolytic and could not produce indole in shrimp extracts without added L-tryp-tophan. Suppression of bacterial reproduction with antibacterial compounds inhibited indole production. Two paths of indole production are suggested based on temperature of decomposition. Due to lack of proper handling and processing in the country of origin, spoiled or decomposed shrimp are often encountered in shipments entering domestic ports. To ensure product safety and quality, the U.S. Food and Drug Administration (FDA) constantly evaluated imported shrimp for Salmonella, filth and decomposition. Although determined subjectively by sensory evaluation, decomposition is substantiated objectively, when needed, by determining the concentration of indole in the product (7). In a previous report (3), we established that temperature of storage was the primary factor affecting the formation of indole in raw shrimp. Even though indole was also present in shrimp stored over prolonged time periods on ice or refrigerated, the results in that report suggested that indole formation is most likely related to high temperature abuse, presumably because the ability to convert tryptophan to indole is common among mesophilic microorganisms. This is in agreement with the results of Ponder (6) who suggested that organisms such as Proteus 'Texas A&M University. 2 National Fisheries Institute. morganii or Escherichia coli were major indole producers in decomposing shrimp. However, since low levels of in-dole were also encountered in iced or refrigerated shrimp, psychrotrophic organisms or endogenous enzyme systems may play a role in the conversion of L-tryptophan to in-dole. The purpose of this study was to elucidate the mechanism of indole formation during spoilage of common shrimp harvested from the Gulf of Mexico. Specific objectives were to: (a) isolate and identify indole-producing microorganisms from shrimp homogenates undergoing spoilage under different conditions, (b) demonstrate the potential of these organisms to produce indole in sterile shrimp extracts, (c) determine the temperature tolerance of the indole-producing organisms, and (d) determine if indole can be produced by endogenous shrimp enzyme systems. MATERIALS AND METHODS Raw material and bacterial isolates White shrimp (Penaeus setiferus), of size equivalent to approximately 40 tails per pound, were obtained from commercial shrimp trawlers operating along the Texas coast. Extreme care was taken to minimize external contamination when the shrimp were separated from the incidental catch, packed on ice and transported to the Seafood Technology Laboratory at Texas A&M University in College Station. Upon arrival, the shrimp were divided in six lots each weighing ca. 1.8 kg. Each lot (head-on) was homogenized with two parts of sterile distilled water using a sterile stainless steel blender. The shrimp were homogenized to achieve even distribution of the different additives and to determine if endogenous indole-producing enzymes may be associated with the gut content or digestive system. L-tryptophan (Sigma Chemical Co., St. Louis, MO) was added to three of six lots at a concentration of 10 mg/100 g shrimp at the time of homogenization. Lots were paired as shrimp only and shrimp plus added tryptophan and stored at 4, 12 or 22°C until severely decomposed. The homogenates were initially sampled at the time of homogeniza-tion then at 3-h intervals at 4 and 12°C. Bacterial numbers were determined by the spread-plate method using tryptic soy agar (TSA, Difco). Appropriate dilutions using 0.1% peptone water in 0.1-ml quantities were spread evenly over the surface of prepoured agar plates. Plates were incubated at 22°C for 48 h for samples from all of the homoge-nates, and additionally at 12°C for 4 d and 4°C for 7 d for samples stored at those temperatures. After the plates were counted, representative colony types appearing on countable plates (30 to 300 colonies) were picked and transferred