**2. Description of methods**

To increase the reliability of the results obtained, all tests should be performed in triplicate, following the methods described below.

#### **2.1. Total and thermotolerant coliforms**

The bacteria are in Gram-negative bacilli form, are facultative, not sporogenic anaerobes, capable of fermenting lactose with gas production, and are temperature dependent. Total coliforms, also known as coliforms at 35°C, are a sub-group of the *Enterobacteriaceae* family. The second edition of Bergey's Manual of Systematic Bacteriology [22] includes 44 genera and 176 species in this sub-group. The total coliform group includes only enterobacteria that can ferment lactose with the production of gas, for 24–48 hours at 35°C. More than 20 species fall into this category, including bacteria originating from the gastrointestinal tract of humans and other warm blooded animals such as *Escherichia coli*, and non-enteric bacteria such as *Citrobacter, Enterobacter, Klebsiella* and *Serratia*, among others [23].

not absence but values within a range between *m* and *M*. In the two-class plan, *M* separates acceptable from unacceptable lots. In a three-class plan this value separates an acceptable lot

Samples of food concentrates such as honey are microbiologically stable and can be transported and stored at ambient temperature. Nevertheless, they should be protected against moisture

This is the amount of food sample used in conducting one or more tests. The sample unit must be greater than that required for analysis, with sufficient quantities for the counter-sample.

In Brazil, tests for the quantitation of microorganisms in honey comprise mould and yeast counts, the count of total and faecal coliforms and S*almonella* analysis, the trials of which are usually done with an analytical unit of 25 g of honey (in special cases at least 10 g of honey can be used). Analytical units of 25 g meet the requirements of ISO 6887-1 [20], and those of the Compendium, for all tests. Two analytical units are required for analysis of a honey sample one for mould and yeast quantification, total and thermotolerant coliform count and the other

Disinfect the area outside the packaging with 70% ethanol and remove the jar lid aseptically. Observe and note the presence of abnormalities in the packaging or in the internal content such as bloating, leakage, the presence of foreign bodies, odour and/or strange appearance.

Before the withdrawal of analytical units, the content of the sample should be homogenized to ensure that the removed portion is representative of all the material. In the case of honey in a jar with enough room for agitation, the package should be inverted 25 times. If there is no free space for agitation, use a second sterile vial and transfer the sample from one vial to another three times. Remove the analytical unit with a sterile spatula (ISO 6887-5: 2010) [21].

To increase the reliability of the results obtained, all tests should be performed in triplicate,

The bacteria are in Gram-negative bacilli form, are facultative, not sporogenic anaerobes, capable of fermenting lactose with gas production, and are temperature dependent. Total coliforms, also known as coliforms at 35°C, are a sub-group of the *Enterobacteriaceae* family.

**1.4. Homogenization of the honey sample and withdrawal of the analytical unit**

from an intermediate lot [10, 11].

test for the absence or presence of *Salmonella*.

**2. Description of methods**

following the methods described below.

**2.1. Total and thermotolerant coliforms**

**1.2. Transport of samples**

262 Honey Analysis

and excessive heat [20].

**1.3. Analytical unit**

Lactose fermentation capacity is analysed through the formation of gas and/or acid in the lactose-containing culture media. These characteristics are used in traditional methods of total coliform counting.

With modern methods, it is possible to directly detect the activity of the β-galactosidase enzyme involved in the fermentative metabolism of lactose, incorporating the substrates for the enzyme in culture media. One of these substrates is ONPG (ortho-nitrophenyl-β-Dgalactopyranoside) which when degraded by β-galactosidase results in a product that is yellow in colour. It also possesses the X-GAL (5-bromo-4-chloro-3-indolyl-β-D-galactopyranoside) substrate, which results in a product with an intense blue staining, and Salmon-Gal (6-chloro-3-indolyl-β- D-galactopyranoside), whose degradation product is a salmon red colour [23, 24].

The thermotolerant coliform group, also known as coliforms at 45°C but usually called faecal coliforms, is a subgroup of total coliforms which are restricted to members capable of fermenting lactose in 24 hours at 44.5–45.5°C with gas production [23–25]. While this definition aims in principle to select only enterobacteria that originate from the gastrointestinal tract (*E. coli*), it is currently known that the group includes members of non-faecal origin (various strains of *Klebsiella pneumoniae, Pantoea agglomerans, Enterobacter aerogenes, Enterobacter cloacae* and *Citrobacter freundii*) [23].

*Escherichia coli* have as their natural habitat the intestinal tract of warm-blooded animals, but can be introduced into food from non-faecal sources. They can grow in eosin methylene blue agar where their growth characteristics allow them to be distinguished from other coliforms [23, 25–27].

These bacteria by themselves do not generally represent a major risk, but can indicate poor quality food that may contain harmful agents. According to the International Commission on Microbiological Specifications for Foods [7], total coliforms, thermotolerant coliforms and *Escherichia coli* are microorganisms with a low or indirect risk to health. Their presence may indicate inadequate hygiene and sanitation, demonstrating failures during post-processing, as they are easily inactivated by sanitizers and heat treatment.

**Method:** This is based on the most probable number (MPN) technique involving inoculation in tubes with lauryl sulphate tryptose broth (LST). This technique is the most used for coliform-bacteria counting. The most probable number in a sample is determined by using a confidence interval table at 95% probability for the various positive tube combinations in three or five tube series [27, 28].

This method enables the density of the viable organisms present in a sample under analysis to be estimated and is based on the principle that the bacteria present in a sample can be separated by agitation, resulting in a suspension of bacterial cells, evenly distributed in the sample. It is based on the inoculation of an increasing sample volume in a suitable culture medium for the growth of microorganisms, with each volume being inoculated into a series of tubes. Inoculum is obtained by sampling successive dilutions, the streaking of which provides positive and/or negative results allowing the calculation of the density of bacteria investigated by the application of probability calculations.

According to the methods studied, total and thermotolerant coliform and *E. coli* counting by the most probable number method is conducted in four steps [5, 7, 9, 23, 26–28]:


one series for each dilution. The tubes should be incubated at 35°C for 24–48 hours. If after this time there is turbidity of the medium and the formation of gas in the Durham tube, the presumptive test is positive for the presence of coliforms and should be subjected to confirmatory tests. If there is no turbidity in the medium or gas formation during the incubation period, the analysis ends at this stage and the test result is negative.

**Figure 1.** Decimal dilutions prepared (10−1; 10−2 and 10−3) from honey sample.

It is based on the inoculation of an increasing sample volume in a suitable culture medium for the growth of microorganisms, with each volume being inoculated into a series of tubes. Inoculum is obtained by sampling successive dilutions, the streaking of which provides positive and/or negative results allowing the calculation of the density of bacteria investigated by

According to the methods studied, total and thermotolerant coliform and *E. coli* counting by

(1) Presumptive test for total coliforms: using lauryl sulphate tryptose broth the observation of growth with gas production is considered suspect (presumptive) for the presence of coliforms. The presence of the surfactant in the lauryl sulphate tryptose broth inhibits the growth of the cytoplasmic membrane of Gram-positive bacteria and enables the presence of lactose fermentation, which releases carbon dioxide. The presence of this gas is evident

(2) Confirmation of total coliform test: using brilliant green bile broth (BGBB) there is notable development of bacteria of the coliform group, which is again confirmed by the formation of gas. This occurs because this broth is selective due to the presence of bovine bile and a triphenylmethane dye derivative which inhibits Gram-positive bacteria and sporulated lactose fermenting bacteria. This step of the examination reduces the possibility of false positive results arising from the activity of sporulated bacteria and Gram-positive lactose fermenting bacteria. Observation of growth through gas production in brilliant green bile

(3) Confirmation test for thermotolerant coliforms: this method uses *Escherichia coli* broth (EC) containing lactose, a selective medium containing a mixture of phosphate which maintains the pH of the medium at an appropriate amount. This selectivity is due to bile salts, which inhibit the growth of the Gram-positive microorganism. If there is gas formation in these conditions the thermotolerant coliform is confirmed [29]. The positive *Escherichia coli* tubes for thermotolerant coliforms are suspect for the presence of *E. coli*.

(4) Confirmation testing for *E. coli*: this method uses eosin methylene blue agar, which is a selective differential medium that distinguishes *E. coli* from other thermotolerant coliforms. If there is development of typical colonies of *E. coli* in this agar, these colonies are isolated for the biochemical proof of indole, methyl red, Voges-Proskauer and citrate

 - *Sample preparation*: weigh 25 g of honey and add to 225 mL of peptone water 0.1% and homogenize the sample. This provides a 10−1 dilution; where 1 mL of the same corresponds to 0.1 g of the sample. A quantity of 1.0 mL of this solution (10−1) is transferred using a new sterile pipette to a 9.0 mL of dilution water, thus obtaining a second decimal dilution (10−2), where 1 mL corresponds to 0.01 g of the sample. In the same way, a 10−2 dilution provides

 - *Presumptive Test:* for presumptive evidence, 1 mL of the three subsequent dilutions should be inoculated in a series of three test tubes containing broth lauryl sulphate tryptose, with

the most probable number method is conducted in four steps [5, 7, 9, 23, 26–28]:

tubes is considered confirmatory for the presence of total coliforms.

the application of probability calculations.

in the Durham tube.

264 Honey Analysis

(IMViC).

a 10−3 solution (**Figure 1**).


*Reading of test using most probable number (MPN):* the most probable number technique is based on the statistical probability related to the frequency and occurrence of the most probable positive results in terms of the real number of microorganisms present. Three sets of three tubes are inoculated, employing dilutions 0.1; 0.01 and 0.001 mL/g of honey. Thus, the number of tubes per series of three consecutive dilutions is three, giving a total of nine tubes. The number of microorganisms in the original sample is determined using the most probable number tables (**Tables 1** and **2**), according to the Brazilian Association of Technical Standards [30].

**Figure 2.** Presumptive and confirmatory tests of coliforms at 35 and 45°C.



**Table 1.** Most probable number (MPN) with 95% confidence limits for various combinations of positive results, using three tubes per series to inoculate 1 mL of dilutions 0.1; 0.01 and 0.001 g of honey/ml.


**Table 2.** Examples using dilution (g) combining 0.1; 0.01; 0.001 g/mL.

**Figure 2.** Presumptive and confirmatory tests of coliforms at 35 and 45°C.

Honey Analysis

**Combination of positive tubes Combination of positive tubes**

**0.1g 0.01g 0.001g MPN 0.1g 0.01g 0.001g MPN** 0 0 <3 2 0 0 9.1 0 1 3.0 2 0 1 14.0 0 2 6.0 2 0 2 20.0 0 3 9.0 2 0 3 26.0 1 0 3.0 2 1 0 15.0 1 1 6.1 2 1 1 20.0 1 2 9.2 2 1 2 27.0 1 3 12.0 2 1 3 34.0 2 0 6.2 2 2 0 21.0 2 1 9.3 2 2 1 28.0
