SciELO - Scientific Electronic Library Online

vol.47 issue3Estimate of genetic parameters for carcass traits and visual scores inmeat sheep using Bayesian inference via threshold and linear modelsBeverages formulated with whey protein and added lutein author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand




Related links


Ciência Rural

On-line version ISSN 1678-4596

Cienc. Rural vol.47 no.3 Santa Maria  2017  Epub Dec 12, 2016 


Microbial contamination in industrial tofu

Contaminação microbiana em tofu industrial

Thaís Teresa Brandão Cavalheiro Ribeiro1 

Géssica Costa1 

Marisa da Costa1  * 

1Programa de Pós-graduação em Microbiologia Agrícola e do Ambiente, Instituto de Ciências Básicas da Saúde (ICBS), Universidade Federal do Rio Grande do Sul (UFRGS), Rua Sarmento Leite, 500, Sala 216, 90050-170, Porto Alegre, RS, Brasil.


This study aimed to evaluate the microbiological quality of tofu sold in supermarkets in Porto Alegre/Brazil. Bacteria counts were performed for Bacillus cereus , mesophilic, coliforms and Staphylococcus coagulase positive and negative. The presence of Listeria sp. was also evaluated. Two different brands of tofu (A and B) were collected, one lot per month, for six months. Five samples from each lot were analyzed. All lots presented mesophilic aerobic counts above 4.3x105CFU g-1. Four of the six lots from brand A and all lots from brand B showed E. coli and/or Staphylococcus coagulase positive counts above the Brazilian law accepted limits. The Staphylococcus coagulase negative counts were higher than those of coagulase positive in all lots. In all lots where Staphylococcus coagulase positive counts were above the legal limit, there were counts of coagulase negative above 104CFU g-1. B. cereus and Listeria sp. were not found in either brand. The majority of lots of brand A and all lots of brand B were unsuitable for human consumption. Our results showed that there are problems in tofu manufacturing in both industries analyzed. There is a need of improvement on its microbial quality to avoid problems of food-borne illness, and finally the need of a better control by the Brazilian inspection services.

Key words: soy cheese; mesophile; coliforms; Staphylococcus; coagulase


Este estudo teve como objetivo avaliar a qualidade microbiológica de queijo tofu comercializados em um supermercado da cidade de Porto Alegre, RS. Foram realizadas contagens de Bacillus cereus , bactérias mesófilas, coliformes, Staphylococcus coagulase positive e negativo e, a pesquisa de Listeria sp. Foi realizada a coleta de duas diferentes marcas (A e B), com um lote por mês, durante seis meses; sendo analisadas cinco amostras de cada lote. Todos os lotes apresentaram contagens de mesófilos aeróbios acima de 4,3x105UFC g-1. Quatro dos seis lotes da marca A e todos os lotes da marca B apresentaram contagens de E. coli e/ou Staphylococcus coagulase positive acima do permitido pela legislação vigente. As contagens de Staphylococcus coagulase negativos foram superiores às contagens dos coagulase positivos em todos os lotes. Nos lotes em que foram encontradas contagens fora do padrão estabelecido para os Staphylococcus coagulase positivos, havia contagens dos coagulase negativos acima de 104UFC g-1. Não foram isolados B. cereus e Listeria sp. Ambas as marcas apresentaram sérios problemas sanitários, pois a maioria dos lotes da marca A e todos os lotes da marca B estavam impróprios para o consumo mesmo na ausência dos outros patógenos pesquisados. Nossos resultados mostram que há problemas sérios na produção de tofu em ambas as indústrias analisadas. Existe a necessidade de melhorias no sistema de produção para evitar a ocorrência de doenças causadas por este alimento e, também, que deve haver um maior controle da qualidade deste alimento pelos serviços de fiscalização.

Palavras-chave: queijo de soja; mesófilos; coliformes; Staphylococcus; coagulase


Brazil is the second largest producer of soybeans in the world (EMBRAPA SOJA, 2016). The consumption of soy and its products is still low in Brazil, although the consumption of soy grains and products has been encouraged since the 1980s due to its high protein content (BEHRENS & DA SILVA, 2004; EMBRAPA, 2015). Another incentive to its consumption has been the positive correlations between the consumption of soybased foods with a reduction in the risk of developing certain diseases (FRITZ et al., 2013; LAI & YEW, 2015; TAKAGI et al., 2015; JAMILIAN & ASEMI, 2016).

Tofu, a soy based cheese, has a high water content which, combined with its high protein content, makes it susceptible to the growth of microorganisms, especially if proper manufacturing and/or storage procedures are not followed (CAI et al., 1999; PAULETTO & FOGAÇA, 2012). Microbiological quality of this product must be checked to avoid infections or food poisoning. But, few reports about the microbiological quality of tofu have been published (van KOOIJ & DE BOER, 1985; ASHENAFI, 1994; MALLET et al., 2007; DANIYAN et al., 2011; ANANCHAIPATTANA et al., 2012; ROSSI et al., 2016). Therefore the aim of this study was to evaluate the bacteriological quality of two different brands of industrialized tofu sold in a supermarket over a period of six months in 2015, at Porto Alegre city, Brazil. For this propos we searched for coliforms, mesophyles bacteria, Staphylococcus coagulase positive and negative, Bacillus cereus and Listeria sp.


Collection and dilution of samples

Two traditional tofu brands (hereinafter referred to as brands A and B) were collected monthly. Every month five samples of the same lot of each brand were purchased in a supermarket. At the time of collection cooling conditions, the date of manufacture, expiry date and batch number were observed and recorded. Only samples within the legal expiry date were collected. Samples were immediately sent to the laboratory under refrigeration and analyzed within a maximum of 24 hours. Sampling and analyses were carried out over six months (November to April).

Counting of aerobic mesophilic bacteria, coliforms, Staphylococcus sp. and Bacillus cereus

Each sample had the outermost layer of the product aseptically removed. An amount of 25g of the inner part of each tofu sample was first homogenised in 225mL of peptone saline. From this dilution subsequent dilutions (10-2 to 10-5) were performed, all in duplicate (SILVA et al., 2007).

Counting of mesophilic aerobic bacteria

All dilutions were spread onto a plate counting agar (PCA-Kasvi) as described by SILVA et al. (2007). The counts were expressed in colony forming units per gram (CFU g-1) of the product.

Coliform counting and Escherichia coli identification

Total coliform counting was performed with violet red bile agar (VRBA-Fluka Analytica). Confirmation of total and fecal coliforms and E. coli identification were carried out according to Silva and co-workers (2007) using brilliant green bile broth and EC broth (Himedia). Counts were expressed in CFU g-1 of the product.

Counting and identification of Staphylococcus coagulase positive and negative

All dilutions were spread in Baird Parker agar (Acumedia) supplemented with egg yolk solution and potassium tellurite according to SILVA et al. (2007). Typical and atypical colonies present were counted and expressed in CFU g-1 of the product. Four to five typical and atypical Staphylococcus colonies were isolated and tested by Gram staining, catalase test, glucose oxidation and fermentation, mannitol fermentation, thermonuclease production and coagulase test (MACFADDIN, 2000; SILVA et al., 2007). Isolates were identified as Staphylococcus coagulase positive (SCP) or Staphylococcus coagulase negative (SCN).

Counting of B. cereus

The quantitative evaluation was performed by spreading all dilutions in cereus selective agar (Acumedia) (SILVA et al., 2007).

Detection of Listeria monocytogenes

Detection was carried out according to HITCHINS (2014) consisting of the enrichment of the food in Listeria enrichment broth (Acumedia) and isolation in Oxford Agar (Acumedia) and Palcam agar (Himedia).

Statistical analysis

Results from all bacterial counts were expressed in CFU g-1 and analyzed using Tukey's test. A value of P<0.05 was considered to be statistically significant.


A total of six lots of tofu from each manufacturer were collected, resulting in a total of 60 samples being analyzed. Distribution of the sampling over the different seasons was as follows: four (33%) in spring, six (50%) in summer and two (17%) in winter. The season in which the samples were obtained showed no effect on the bacterial counts that may indicate that probably no oscillation in the refrigeration temperature occurred during exposition of those cheeses on the shelves. All samples remained under refrigeration at the time of sampling. At the time of sampling, all samples had been on the shelf no more than three weeks and all were within three months of their expiry date. All samples showed the same characteristics: good-looking, white cheese with slightly yellowish liquid, sweet smell and soft texture.

Counting of mesophilic aerobic bacteria

All samples showed aerobic mesophilic bacteria above 4.3x105CFU g-1 (Tables 1 and 2). There was no significant difference between the means of the lots for the two brands. There is no specific legislation outlining the maximum allowable mesophilic bacteria counts in tofu in Brazil. But the amount of these bacteria obtained in many batches showed values around 106CFU g-1, which is higher than the limits for many other foods. For example, the limits for mesophilic bacteria in banked breast milk is 102CFU mL-1, and for pasteurized milk drinks is 1.5x105CFU mL-1 (BRAZIL, 2001). Also, the values reported in our study are higher than those recommended by "The Soy Food Association of America" for soy milk, where counts of mesophilic bacteria should not exceed 2x104CFU mL-1 (TSAA, 1996). Our results are in agreement with van KOOIJ & DE BOER (1985) and ASHENAFI (1994) who also reported high scores of mesophilic bacteria in tofu. ASHENAFI (1994) demonstrated a correlation between the high bacterial counts in tofu and the handling throughout its production. Working with tofu produced in a small industrial scale, ROSSI et al. (2016) reported mesophilic bacteria in all samples analyzed of fresh tofu, and observed that this counts were one to three log higher than those observed with soybean samples obtained in the same industry. This observation showed that along the process of tofu manufacturing, conditions are being created (addition of water, grinding, coagulation, molding, etc) that favour a naturally enhancement of the microbial load. Fails in the process certainly will favour the development of unsuitable microorganisms also.

Table 1 Bacterial counts of mesophilic bacteria, coliforms, coagulase positive and negative Staphylococci in tofu brand A. 

a: Mean of five samples per lot. b: CPS: coagulase positive Staphylococci . c: CNS: coagulase negative Staphylococci. d : In brackets: minimal and maximal values. Bold type indicates lots that exceed the limits of Brazilian legislation to tofu. NF: not found.

Table 2 Bacterial counts of mesophilic bacteria, coliforms, coagulase positive and negative Staphylococci in tofu brand B. 

a: Mean of five samples per lot. b: CPS: coagulase positive Staphylococci . c: CNS: coagulase negative Staphylococci . d: In brackets: minimal and maximal values. Bold type indicates lots that exceed the limits of Brazilian legislation to tofu. NF: not found.

Coliform counting and E. coli identification

Tables 1 and 2 show the averages of fecal and total coliforms, and E. coli identification in all lots tested. Of the six lots analyzed for brand A, all presented total and fecal coliforms and three lots showed fecal coliform counts above the legal limit of 102CFU g-1. All samples of brand B showed total and fecal coliform levels above the established limits (BRAZIL, 2003). The presence of E. coli was confirmed in the three (50%) lots of brand A and in the six lots (100%) of brand B. These results demonstrated that there were failures in the production of tofu. ANANCHAIPATTANA et al. (2012) also reported coliforms and E. coli in 67% and 28%; respectively, in tofu samples, and highlighted the need of the improvement of hygienic practices during tofu production. E. coli, besides having toxigenic strains (not specifically searched in this research), is an indicator of possible contamination by other harmful microorganisms as Salmonella sp. and Campylobacter sp. (ANANCHAIPATTANA et al., 2012).

Counting and identification of Staphylococcus coagulase positive and negative

Two lots of brand A and four lots of brand B showed levels of SCP above the official limits of 5.3x103CFU g-1 (Tables 1 and 2). These results show, once again, the likely inadequate hygiene in the production process of those cheeses or in the cooling conditions during storage. Contamination with SCP is a significant indicator of improper handling of the product during manufacturing. Staphylococcus sp. can be found in the microbiota of the handlers and in the absence of good hygiene practices may be transferred to food (SOARES et al., 2012). SCP counts as high as 5.5x104CFU g-1 were observed, which are similar to the levels demonstrated by van KOOIJ & De BOER (1985) in Nigeria and by MALLET et al. (2007) in the city of Lavras, Brazil. The presence of SCP was also confirmed in tofu by DANIYAN et al. (2011).

SCN counts ranged from 3.4x101 to 6.8x104CFU g-1 in brand A and from 3.7x101 to 6.3x104CFU g-1 in brand B. Levels of SCN were higher than those of SCP in all samples. In all lots where SCP counts were above the legal limit, there were counts of SCN above 104CFU g-1. In a comparison between packaged and unpackaged tofu ANANCHAIPATTANA et al. (2012) showed a high contamination rate, of both kinds of samples, contaminated with SCN, but no bacterial counts were discriminated. FLEMING et al. (2010), studying bovine cheese, reported higher SCN counts in 50% of those samples where counts of SCP were also above the limits of legislation. The SCN may also hold enterotoxin genes and many reports of food poisoning from SCN through ingestion of raw cow's milk have been made (CARMO et al., 2002; CUNHA et al., 2006; PODKOWIK et al., 2013; MELLO et al., 2014). Despite the possible ability of SCN to produce enterotoxins, most reports of staphylococcal poisoning have been predominantly related to the presence of SCP. However, future studies should examine a possible role of SCN in staphylococcal food poisoning.

Counting of B. cereus and Detection of Listeria monocytogenes : B. cereus and Listeria sp. were not isolated in any of the tofu samples analyzed. Both bacteria are ubiquitously distributed, and can be present even in the food-processing environments (RYSER & MARTH, 2007; LOGAN, & De VOS, 2009). The search of these bacteria are important because they may cause food borne illness. Presence of L. monocytogenes may cause death in immunologically deficient persons. Unexpectedly there are no obligation to search Listeria sp. in tofu, despite their higher nutritive and water contends, conditions that favour bacterial growth, Listeria sp. particularly, by its ability of growing at cold temperatures (BRAZIL, 2001). But, the presence of some microorganisms in food at high counts is able to inhibit the growth of other microorganisms. For example, the presence of E. coli in high counts inhibits the growth of L. monocytogenes in food (DAILEY et al., 2014). In our study, the presence of high counts of mesophilic aerobic bacteria, coliforms and Staphylococcus sp. may have been an obstacle to the growth of B. cereus and Listeria sp. Other authors have been describing B. cereus and L. monocytogenes presence in tofu and recalls of contaminated tofu have been published (SCHAEFER, 2007; ANANCHAIPATTANA et al., 2012; UFDA, 2015).


Based on the levels of coliforms and SCP presented in the analyzed tofu, four of the six lots of brand A and all six lots of brand B were unsuitable for human consumption because of the high counts of these microorganisms. Contamination of fecal coliforms and SCP were most likely due to the absence of good manufacturing practices during the production of these cheeses, taking into account that these microorganisms are natural human microbiota. The high counts of mesophilic bacteria and SCN are also the reflection of the lack of hygiene in the process. Concluding, it seems that there are fails in the control of the good practices in the industries and that, this food, is not receiving attention by the Brazilian inspection services


This work was supported by Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) fellowship and Programa de Apoio a Pós-graduação (PROAP)/ Universidade Federal do Rio Grande do Sul (UFRGS)


ANANCHAIPATTANA, C. et al. Bacterial contamination of soybean curd (Tofu) sold in Thailand. Food Science and Technology Research, v.18, p.843-848, 2012. Available from: >. Accessed: Mar. 31, 2016. doi: 10.3136/fstr.18.843. [ Links ]

ASHENAFI, M. Microbiological evaluation of tofu and tempeh during processing and storage. Plant Foods for Human Nutrition, v.45, p.183-189, 1994. Available from: >. Accessed: Mar. 31, 2016. doi:10.1007/BF01088476. [ Links ]

BEHRENS, J. H.; DA SILVA, M. A. A. P. Atitude do consumidor em relação a soja e produtos derivados. Ciência e Tecnologia de Alimentos, v.24, p.431-439, 2004. Available from: >. Accessed: Mar. 31, 2016. doi: 10.1590/S0101-20612004000300023. [ Links ]

BRAZIL. Agencia Nacional de Vigilância Sanitária. Resolução RDC 12, de 2 de janeiro de 2001. Available from: >. Accessed: Jan. 15, 2016. [ Links ]

BRAZIL. Ministério da Agricultura, Pecuária e Abastecimento. Instrução normativa n.62, de 26 de agosto de 2003, p.1-21. Available from: >. Accessed: May 15, 2014. [ Links ]

CAI, T. et al. Processing effect on soybean storage proteins and their relationship with tofu quality. Journal of Agricultural and Food Chemistry, v.47, p.720-727, 1999. Available from: >. Accessed: Mar. 31, 2016. doi: 10.1021/jf980571z. [ Links ]

CARMO, L. S. et al. Food poisoning due to enterotoxigenic strains of Staphylococcus present in Minas cheese and raw milk in Brazil. Food Microbiology, v.19, p.9-14, 2002. Available from: >. Accessed: Mar. 31, 2016. doi: 10.1006/fmic.2001.0444. [ Links ]

CUNHA, M. L. R. S. et al. Detection of enterotoxins genes in coagulase-negative Staphilococci isolated from foods. Brazilian Journal of Microbiology, v.37, p.70-74, 2006. Available from: >. Accessed: Mar. 31, 2016. doi: 10.1590/S1517-83822006000100013. [ Links ]

DANIYAN, S. Y. et al. Microbiological and physiochemical assessment of street vended soyabean cheese sold in Minna, Nigeria. International Journal of Biomedical and Advance Research, v.2, p.25-31, 2011. Available from: >. Accessed: Mar. 31, 2016. doi: 10.7439/ijbar.v2i1.19. [ Links ]

DAILEY, R. C. et al. The effects of competition from non-pathogenic foodborne bacteria during the selective enrichment of Listeria monocytogenes using buffered Listeria enrichment broth. Food Microbiology , v.44, p.173-179, 2014. Available from: >. Accessed: Mar. 31, 2016. doi: 10.1016/ [ Links ]

EMBRAPA. Programa de incentivo de utilização da soja na alimentação humana. Agência Embrapa de Informação tecnológica. Available from: >. Accessed: Aug. 30, 2015. [ Links ]

EMBRAPA SOJA. Soja em números. Available from: >. Accessed: Feb. 16, 2016. [ Links ]

FLEMING, L. R. et al. Quantification and antibiotic resistance of Staphylococci isolated from cheeses. Perspectivas da Ciência e Tecnologia, v.2; n.1/2, p.13-19, 2010. Available from: >. Accessed: Mar. 31, 2016. [ Links ]

FRITZ, H. et al. Soy, red clover, and isoflavones and breast cancer: a systematic review. PLoS ONE, v.8, p.e8, 1968. Available from: >. Accessed: Mar. 31, 2016. doi: 10.1371/journal.pone.0081968, 2013. [ Links ]

HITCHINS, A. D. Detection and enumeration of Listeria monocytogenes in foods. In: US Food and Drug Administration (FDA). Bacteriological analytical manual online. Available from: >. Accessed: June, 10, 2014. [ Links ]

JAMILIAN, M.; ASEMI, Z. The effects of soy isoflavones on metabolic status of patients with polycystic ovary syndrome. Journal of Clinical Endocrinology & Metabolism, v.101, n.9, p. 3386-3394. Available from: >. Accessed: Mar. 31, 2016. doi: 10.1210/jc.2016-1762, 2016. [ Links ]

LAI, Y. C.; YEW, Y. W. Tofu, urinary phytoestrogens, and melanoma: An analysis of a national database in the United States. Dermatologica Sinica, v.33, p.210-214, 2015. Available from: >. Accessed: Mar. 31, 2016. doi: 10.1016/j.dsi.2015.05.003. [ Links ]

LOGAN, N. A.; De VOS, P. Genus I. Bacillus Cohn 1872, 174AL. In: De VOS et al. The Firmicutes. Athens: Springer, 2009. V.3, p.21-128. [ Links ]

MACFADDIN, J. F. Biochemical tests for identification of medical bacteria. Philadelphia: Lippincott Willians & Wilkins, 2000. 912p. [ Links ]

MALLET, A. C. T. et al. Estudo da qualidade sanitária do queijo de soja "TOFU". In: CONGRESSO DE PÓS GRADUAÇÃO DA UFLA, 16., 2007, Universidade Federal de Lavras/UFLA. Anais...UFLA, 2007. p.1-6. Available from: >. Accessed: Mar. 31, 2016. [ Links ]

MELLO, J. F. et al. Sanitary quality, occurrence and identification of Staphylococcus sp. in food services. Brazilian Journal of Microbiology , v.45, p. 1031-1037, 2014. Available from: >. Accessed: Mar. 31, 2016. [ Links ]

PAULETTO, F. B.; FOGAÇA, A. O. Avaliação da composição centesimal de tofu e okara. Disciplinarum Scientia, Série: Ciências da Saúde, v.13, p.85-95, 2012. Available from: >. Accessed: Mar. 31, 2016. [ Links ]

PODKOWIK, M. et al. Enterotoxigenic potential of coagulase-negative Staphylococci. International Journal ofFood Microbiology , v.163, p.34-40, 2013. Available from: >. Accessed: Mar. 31, 2016. doi: 10.1016/j.ijfoodmicro.2013.02.005. [ Links ]

ROSSI, F. et al. Microbiological characteristics of fresh tofu produced in small industrial sacale and identification of specific spoiling microorganisms (SSO). Food Science and Technology, v.70, p.280-285, 2016. Available from: >. Accessed: Mar. 31, 2016. doi: 10.1016/j.lwt.2016.02.057. [ Links ]

SCHAEFER, P. Organic Tofu Recall, Listeria Discovered. 2007. Available from: >. Accessed: Aug. 31, 2016. [ Links ]

SILVA, N. et al. Manual de Métodos de Análise Microbiológica de Alimentos. São Paulo; Varela, 2007. 536p. [ Links ]

SOARES, L. S. et al. Knowledge, attitudes and practices in food safety and the presence of coagulase-positive Staphylococci on hands of food handlers in the schools of Camaçari, Brazil. Food Control, v.27, p.206-213, 2012. Available from: <>. Accessed: Mar. 31, 2016. doi: 10.1016/j.foodcont.2012.03.016. [ Links ]

RYSER, E. T.; MARTH, E. H. Listeria, listeriosis, and food safety. Boca Raton: CRC, 2007. 873p. [ Links ]

TAKAGI, A. et al. Possibility of breast cancer prevention: use of soy isoflavones and fermented soy beverage produced using probiotics. International Journal of Molecular Sciences, v.16, p.10907-10920, 2015. Available from: >. Accessed: Mar. 31, 2016. doi: 10.3390/ijms160510907. [ Links ]

TSAA (THE SOYFOODS ASSOCIATION OF AMERICA). Voluntary standards for the composition and labeling of soymilk in the United States. 1996. 16p. Available from: >. Online. Accessed: Feb. 16, 2016. [ Links ]

UFDA (U.S. FOOD AND DRUG ADMINISTRATION). Amy's kitchen recalls various products because of possible health risk. Available from: >. Accessed: Aug. 31, 2016. [ Links ]

van KOOIJ, J. A. V; De BOER, E. A survey of the microbiological quality of commercial tofu in the Netherlands. Food Microbiology , v.2, p.349-354, 1985. Available from: >. Accessed: Mar. 31, 2016. doi: 10.1016/0168-1605(85)90025-X [ Links ]


Received: March 06, 2016; Accepted: October 10, 2016; Revised: November 28, 2016

E-mail: *Corresponding author

Creative Commons License This is an open-access article distributed under the terms of the Creative Commons Attribution License