Comunicao Breve/Brief Communication

Enterobacteriaceae in processed cocoa products

Enterobactrias em produtos processados de cacau
RIALA6/1348

Maristela da Silva do NASCIMENTO1*, rika Marques REOLON1, Aline Regina Barbosa

SANTOS1, Vanessa Eliana MOREIRA1, Ivone Francisca da SILVA2, Neusely da SILVA1
*Endereo para correspondncia: 1Instituto de Tecnologia de Alimentos, Campinas, SP, Brasil, tel: 19 3743 1815, fax: 19 3743
1812, e-mail: [email protected].
2
Departamento de Cincias de Alimentos, Faculdade de Engenharia de Alimentos, Universidade Estadual de Campinas,
Campinas/SP, Brasil.
Recebido: 22.03.2010 Aceito para publicao: 24.02.2011

ABSTRACT

Little is known of the presence of Salmonella in Brazilian cocoa, which justifies the present work that had the aim
of checking the presence of total Enterobacteriaceae, coliforms, Escherichia coli and Salmonella in semi-processed
cocoa products. A total of 150 samples of cocoa products from two different cocoa-processing manufacturers
were analyzed (30 samples of nibs, 30 of liquor, 30 of cocoa cake, 30 of cocoa butter and 30 of cocoa powder).
The samples of processed cocoa products had pH values between 5.31 and 7.35, and water activity ranging
from 0.29 to 0.52. E. coli and Salmonella were not detected in any of the samples analyzed. Regarding the other
analyzed microorganisms, 17% of the nibs contained total Enterobacteriaceae, 20 % showed total coliforms,
while thermotolerant coliforms were detected in one sample (0.6 log MPN/g). Seven percent of liquor and cocoa
powder samples showed total coliforms. In cocoa cake, the same percentage in regard to total Enterobacteriaceae
was observed. Total Enterobacteriaceae and total coliforms were detected in one sample of cocoa butter. Despite
the low contamination observed, these results indicate failure in quality programs of the manufactures studied,
since these bacteria are easily inactivated by thermal and sanitizer process.
Keywords. cocoa, Salmonella, coliform, Escherichia coli, chocolate.

RESUMO

Pouco se sabe sobre a presena de Salmonella no cacau brasileiro, o que justifica o presente trabalho cujo
objetivo foi verificar a presena de enterobactrias totais, coliformes, Escherichia coli e Salmonella em derivados
processados do cacau. Foram analisadas 150 amostras de produtos derivados do cacau provenientes de duas
indstrias processadoras (30 amostras de nibs, 30 de liquor, 30 de torta, 30 de manteiga de cacau e 30 de
cacau em p). As amostras apresentaram valores de pH entre 5,31 e 7,35 e atividade de gua entre 0,29 e 0,52.
No foi detectada E. coli e Salmonella em nenhuma amostra analisada. Com relao aos demais grupos de
microrganismos analisados, 17% das amostras de nibs apresentaram enterobactrias totais, 20% apresentaram
coliformes totais e em uma amostra foram detectados coliformes termotolerantes (0,6 log NMP/g). Sete por
cento das amostras de liquor e cacau em p apresentaram coliformes totais. Nas amostras de torta foi observada
a mesma porcentagem para enterobactrias totais. Em uma amostra de manteiga de cacau foram detectados
enterobactrias totais e coliformes totais. Apesar da baixa contaminao observada, esses resultados indicam
falhas no programa de qualidade das indstrias avaliadas, j que essas bactrias so facilmente inativadas por
processo trmico e sanitizante.
Palavras-chave. Salmonella, coliforme, Escherichia coli, cacau, chocolate.

Rev Inst Adolfo Lutz. 2011; 70(1):81-85

81

Nascimento MS, Reolon EM, Santos ARB, Moreira VE, Silva IF, Silva N. Enterobacteriaceae in processed cocoa products.
Rev Inst Adolfo Lutz. So Paulo, 2011; 70(1):81-85.

INTRODUCTION
The first step in conventional cocoa processing
consists of roasting the cleaned, whole beans in their shells
at a temperature between 120 and 150 C, for 5 to 120
minutes. During roasting, a series of chemical reactions
critical to the development of aroma, taste and color of the
chocolate take place. The time/temperature combination
will depend on several factors, including the origin and
harvesting time within the year, pre-roasting treatments,
the moisture level, the size of the beans and the desired
flavor characteristics. Upon completion of roasting, the
beans are cooled and cracked. Next, the shell and germ
are removed (winnowing) to obtain the nibs1.
The nibs are ground to a fine particle size, called
cocoa mass or cocoa liquor. Cocoa mass is subjected
to hydraulic pressing, which separates some of the
cocoa butter from the solid cocoa mass, resulting in
compressed cocoa cake. Cocoa butter is the most valuable
by-product of the pressing process. Cocoa cake is normally
submitted to one more grinding process to obtain a fine
powder of different particle sizes. Cocoa powder can be
used in the manufacture of chocolate-flavored beverages,
confectionery and bakery fillings and ice cream1.
From a public health standpoint, the
microorganisms most likely to adversely affect the
quality of chocolate are enteropathogens like Salmonella.
With regard to Salmonella, the European Union2 lists
chocolate among the products associated with major
salmonellosis outbreaks in humans that spread across
several countries and affected large numbers of people.
Although cocoa products are not the only ingredients
that may introduce Salmonella into chocolate, they
have been implicated as the most prominent potential
source of some outbreaks (dried cocoa beans, cocoa
powder). Cordier3 points out cocoa beans as a major
source of Salmonella contamination throughout the
manufacturing of chocolate and cocoa-based ingredients.
Consequently, it is imperative that in-process products
be routinely screened for the presence of Salmonella as
part of an integrated control process that begins with
the selection of reliable suppliers4.
Indicator microorganisms are generally used
to measure the quality of the practices used to ensure
proper processing. Total Enterobacteriaceae are used as
an indicator of hygiene practices since they are easily
inactivated by sanitizers, and are able to colonize several
niches of food processing plants5.
82

The coliform group is a subgroup of

Enterobacteriaceae, and the most used as an indicator
in the food industry. They are aerobic or facultative
anaerobic Gram-negative, nonspore-forming rods that
ferment lactose, forming acid and gas. Representatives
of 20 or more species may conform to criteria for this
group, including fecal (Escherichia coli) and non fecal
origins (Citrobacter, Enterobacter, etc)5.
Nonetheless, there is still a huge gap in the scientific
data and information available concerning contamination
of primary ingredients. Thus, the implementation of any
prevention measures should be preceded by detailed
studies that allow determining prevalence and the main
points of entrance of these microorganisms into the
chocolate production chain. For this reason, the aim
of this study was to investigate the presence of total
Enterobacteriaceae, coliforms, Escherichia coli and
Salmonella in semi-processed cocoa products.
MATERIAL AND METHODS
Sampling
A total of 150 samples divided into five groups of
semi-finished processed cocoa products from two different
processing Brazilian manufacturers were analyzed. Each
group consisted of 30 samples of each product type: nibs,
cocoa liquor, cocoa cake, cocoa butter and cocoa powder.
Test methods
Water activity was measured at 25 C with a
hygrometer Aqua Lab 3TE (Braseq, Brazil). The pH was
determined according to the method developed by the
Instituto Adolfo Lutz6. Salmonella analysis was performed
according to the method of the Bacteriological Analytical
Manual (BAM/FDA)7. The pre-enrichment phase was
performed in reconstituted skimmed milk to 10%,
supplemented with brilliant green. After that, the sample
materials were enriched in RappaportVassiliadis and
tetrathionate broth and subsequently plated onto Bismuth
Sulphite Agar, Hecktoen enteric agar, and Xylose Lysine
Deoxycholate agar. Suspect colonies were confirmed by
biochemical and serological tests. Enterobacteriaceae were
enumerated on Violet Red Bile Glucose agar by the plate
count method. Total coliforms, thermotolerant coliforms
and E. coli counts were determined by the Most Probable
Number (MPN) technique, described in the Compendium
of methods for microbiological examination of foods8.
The presumptive test was performed using Lauryl Sulfate

Nascimento MS, Reolon EM, Santos ARB, Moreira VE, Silva IF, Silva N. Enterobacteriaceae in processed cocoa products.
Rev Inst Adolfo Lutz. So Paulo, 2011; 70(1):81-85.

Tryptose broth. Confirmation of total coliforms was carried

out in Brilliant Green Bile broth and that of thermotolerant
coliforms in EC broth. Confirmation of E. coli was followed
by isolation on EMB agar and confirmation by biochemical
tests (Indole, VogesProskauer, Methyl Red and citrate).

Table 1. Results of pH and water activity of processed cocoa

products
Product

pH

Water activity

Nibs

4.98 to 5.67

0.42 to 0.57

Liquor

5.54 to 7.12

0.22 to 0.80

Cocoa cake

5.40 to 8.48

0.24 to 0.48

Cocoa butter

5.51 to 7.87

0.41 to 0.56

Cocoa powder

5.57 to 8.24

0.28 to 0.58

RESULTS
Tables 1 and 2 show the results for water activity,
pH and microbiological analyses of 150 samples of
Table 2. Results of microbiological analyses of processed cocoa products

Product

Cocoa nibs

Cocoa liquor

Cocoa butter

Cocoa cake

Cocoa powder

Contamination
parameter

Total Enterobacteriaceae
(Log CFU/g)*

Total coliforms
(Log MPN/g)**

Thermotolerant
coliforms
(Log MPN/g)**

Positive samples (%)

05 (17%)

06 (20%)

01 (3%)

Counts in positive
samples (mean)

1.0 to 2.0 (1.2)

0.6 to 1.4 (0.9)

0.6

Positive samples (%)

02 (7%)

Counts in positive
samples (mean)

0.6 to 1.0 (0.8)

Positive samples (%)

01 (3%)

01 (3%)

Counts in positive
samples (mean)

1,0

0.6

Positive samples (%)

02 (7%)

Counts in positive
samples (mean)

1.0 to 1.3 (1.2)

Positive samples (%)

02 (7%)

Counts in positive
samples (mean)

0.6 to 0.9 (0.7)

*Colony forming units, detection limit 1.0 Log CFU/g. **Most probable number, detection limit 0.5 Log MPN/g

processed cocoa products from two cocoa processing

manufacturers. Salmonella and E. coli were not detected
in any of the 150 samples (date not shown).
The cocoa nib samples had water activity values
between 0.422 and 0.567 and the pH varied between 4.98
and 5.67. Five (17%) of the 30 samples were found to be
contaminated with total Enterobacteriaceae, with counts

varying from 1.0 to 2.0 Log CFU/g. Total coliforms were

detected in six (20%) of the 30 samples, with mean of
0.9 Log MPN/g, while thermotolerant coliforms were
detected in one sample (0.6 Log MPN/g).
The samples of cocoa liquor had water activity
values between 0.22 and 0.80 and the pH varied between
5.54 and 7.12. Total Enterobacteriaceae were not detected
83

Nascimento MS, Reolon EM, Santos ARB, Moreira VE, Silva IF, Silva N. Enterobacteriaceae in processed cocoa products.
Rev Inst Adolfo Lutz. So Paulo, 2011; 70(1):81-85.

by the plate count method. However, total coliforms

were isolated in two (7%) of the 30 samples by the MPN
technique, with counts of 0.6 and 1.0 Log MPN/g.
The samples of cocoa butter had water activity
values ranging between 0.41 and 0.56 and the pH varied
between 5.40 and 8.48. One sample (3%) was found to
be contaminated with total Enterobacteriaceae (1.0 Log
CFU/g). In another sample (3%) total coliforms were
detected, with a count of 0.6 Log MPN/g.
The cocoa cake samples had water activity
ranging between 0.24 and 0.35. The pH varied between
6.60 and 7.35. Total Enterobacteriaceae were observed
in two samples (7%), with counts of 1.0 and 1.3 Log
CFU/g. However, total coliforms were not isolated in
any analyzed sample.
The samples of cocoa powder had water activity
between 0.28 and 0.58. The pH varied between 5.57
and 8.24. Just as occurred for liquor samples, total
Enterobacteriaceae were not detected in cocoa powder
by the plate count method. Nevertheless, in two samples
(7%), using the MNP technique, contamination by total
coliforms was observed, with counts varying from 0.6
to 0.9 Log MPN/g.
The fact of total Enterobacteriaceae not being
detected in samples which showed contamination
by total coliforms was possibly due to the difference
between methods and detection limit used in this study.
Total Enterobacteriaceae were analyzed by the plate
count method, whose detection limit is 1.0 Log CFU/g.
In contrast, the MPN method was used to isolate total
coliforms which a part from having a lower detection
limit (0.5 Log MPN/g), also has a step in nonselective
broth that allows for the recovery of possibly injured cells.
Furthermore, it is worth pointing out that the total coliform
count observed in liquor and cocoa powder samples were
below the detection limit of the total Enterobacteriaceae
method (1.0 Log CFU/g).
DISCUSSION
The processing of cocoa beans into semi-processed
products consists basically of the following steps: bean
selection and cleaning, roasting, grinding, pressing
and pulverizing. Roasting is the main step capable of
significantly reducing the microbial counts, even though
alkalinization also serves as a barrier function.
The low water activity and the high fat level of
some semi-processed cocoa products may influence the
84

increase in heat resistance of the pathogens4. According

to Jay5, the optimum pH value for Enterobacteriaceae
growth falls within the neutral range (6.6 to 8.2); values
above 9 and lower than 4 are considered bactericidal.
With regard to water activity, growth inhibition of most
members of this family occurs at values lower than
0.94. All the products tested had a pH falling within
the development range for Enterobacteriaceae such as
Salmonella, but at the same time had water activity values
lower than 0.60 which is an adverse growth condition.
However, the fact that the semi-processed cocoa product
exhibited low water activity values does not prevent the
contamination with and the subsequent survival of these
microorganisms, what was observed in this study.
In all 150 samples of semi-processed cocoa products
analyzed, Salmonella and E. coli were not detected. However,
18.5% of the samples showed total Enterobacteriaceae and/or
total coliforms, microorganisms routinely used as hygiene
indicator of process5. Hence, despite the low contamination
(2 Log CFU or MPN/g), this result suggests failure in
quality programs, since these bacteria are easily inactivated
by thermal and sanitizer process.
It is worrying from public health point of view
because a possible contamination by Salmonella and E.
coli could occur in these conditions. As these products
are used as raw material for chocolate manufacturing, the
control must be rigorous, seeking the absence of indicator
organisms. This concern is corroborated by the fact that
two salmonellosis outbreaks associated with chocolate
have been reported, in which the source of contamination
was traced back to the cocoa beans or the cocoa powder9.
In addition, epidemiological surveys have confirmed that
the Salmonella infecting dose in this type of product is
very low, sometimes even lower than 1 CFU/g9.
Therefore, temperature monitoring during
cocoa roasting, use of quality raw material, use of
good manufacturing practices, employee training, pest
control, and adequate, well-executed hygiene program
are prime factors throughout the supply chain of cocoa
and chocolate to obtain products microbiologically safe.
Acknowledgments
The authors wish to thank Fundao de Amparo
Pesquisa do Estado de So Paulo (FAPESP) and Conselho
Nacional de Desenvolvimento Cientfico e Tecnolgico
(CNPq) for financial support.

Nascimento MS, Reolon EM, Santos ARB, Moreira VE, Silva IF, Silva N. Enterobacteriaceae in processed cocoa products.
Rev Inst Adolfo Lutz. So Paulo, 2011; 70(1):81-85.

REFERENCES

1.

Beckett ST. The science of chocolate, 2nd ed. London (UK):

Royal Society of Chemistry Paperbak; 2008.

2.

European Commission. Salmonella in Foodstuffs. Opinion of the

Scientific Committee on Veterinary Measures Relating to Public
Health. [cited 2007 nov 22]. Avaliable from: http://ec.europa.eu/
food/fs/sc/scv/out66_en.pdf.

3.

Cordier JL. HACCP in the chocolate industry. Food Contr.

1994;5:171-5.

4.

International Commission on Microbiological Specifications for

Foods. Cocoa, chocolate and confectionery. In: Microorganisms
in Foods 6: Microbiological Ecology of Food Commodities. 2nd
ed. New York (NY): Kluwer Academic; 2005. p. 467-479.

5.

Jay JE. Microbiologia de alimentos. 6th ed. Porto Alegre (RS):

Artmed, 2005.

6.

Instituto Adolfo Lutz (So Paulo Brasil). Mtodos FsicoQumicos para Anlise de Alimentos: Normas Analticas do
Instituto Adolfo Lutz. 4th ed. Braslia (DF): ANVISA; 2005.

7.

Andrews WH, Hammack TS. Salmonella. In: Bacteriological

analytical manual online. Food and Drug Administration. [cited
2007 nov 22]. Avaliable from: http://www.cfsan.fda.gov/ebam/
bam-5.html.

8.

Kornacki JL, Johnson JL. Enterobacteriaceae, coliforms and

Escherichia coli quality and safety indicators. In: Downes, FP,
Ito K, editors. Compendium of methods for the microbiological
examination of foods. Washington (DC): American Public
Health Association; 2001. p.69-82.

9.

Werber D, Dreesman J, Feil F, van Treeck U, Fell G, Ethelberg S

et al. International outbreak of Salmonella Oranienburg due to
German chocolate. BMC Infect Dis. 2005;5(7):7-17.

85