Microbial Evaluation of Turkish Herbal Teas Before and After Infusion

Background: Herbal teas are produced and sold in packaged or unpackaged forms all over the world. The aim of this study was microbial evaluation of Turkish herbal teas before and after infusion in boiled water. Methods: A total of 20 packaged and unpackaged samples of Turkish herbal teas, including chamomile, salvia, green, mix, apple, mate, ginger, linden, fennel, and senna tea were collected. All of the samples were analyzed before and after infusion in hot water (~100 °C). Microbiological analyses were performed with tenfold serial dilution for yeast-mold, Salmonella, and Cronobacter sakazakii after enrichment by using the spread-on-plate method on selective agar. Data were statistically analyzed by using Microsoft Office Professional Plus 2013 Excel software. Results: Three out of the 20 tea samples (15%) were contaminated with C. sakazakii. None of the samples were contaminated with Salmonella spp. No significant difference was found in occurrence of C. sakazakii before and after infusion of the samples (p>0.05). Mold and yeast contamination were found in 12 out of 20 teas samples (60%). No statistical significance (p>0.05) was found between the mold-yeast contents of the unpackaged and packaged herbal teas. After infusion, neither mold nor yeast was observed in any of the samples. Conclusion: All of the herbal teas in this research were found to be within the microbiological limits for consumption according to Turkish Food Codex. However, the microbiological results highlighted the need for quality control of senna tea during production to eliminate the risk of C. sakazakii contamination. © 2020, Shahid Sadoughi University of Medical Sciences. This is an open access article under the Creative Commons Attribution 4.0 International License.

protects the beneficial effects more and is mostly preferred because of its simplicity (Jäger et al., 2011). However, during their growth, production, and storage, teas might be microbiologically contaminated through different routes, making them harmful rather than beneficial (Székács et al., 2018). Hence, heat treatment may not be sufficient to eliminate the microbial pathogens.
A variety of herbal teas are available on the market, providing various benefits for health, such as sedative and antioxidant effects in addition to supporting cardiovascular physiology, protecting against cold, strengthening the immune system, and providing defense against pathogens via their antimicrobial properties (Poswal et al., 2019). Phenolic compounds in herbal teas have been particularly investigated for their potentials against some gastrointestinal pathogens, including Salmonella, Helicobacter pylori, as well as Campylobacter jejuni (Jarriyawattanachaikul et al., 2016). Despite the various health benefits of herbal teas, some microbiological assessments indicate high microbial contamination in numerous herbal tea products worldwide. Only some of these microorganisms can be inactivated with hot water, and importantly, most of the bacterial spores are not affected by hot water (Nas, 2004). Besides, antimicrobial efficacies of herbal phytochemicals might change with the pH level of tea (Oh et al., 2013).
Analysis of the consumer profile indicates that women prefer herbal teas more than men in Turkey, and this tendency decreases with age (Ulusoy and Seker, 2013). Even infants within their first years of life can encounter with herbal teas worldwide (Zhang et al., 2011). Relative to adults, infants are more vulnerable to pathogens due to their weak physiology and metabolism, and dose amounts per body weight they receive. Thus, although infants have a low risk of pathogenic exposure related to herbal tea consumption, they are more susceptible to develop serious infectious diseases (Bianco et al., 2008).
Another risk factor considered is that unpackaged herbal teas are assumed to be at higher risk of food-borne pathogens than packaged and controlled products. Nevertheless, Salmonella spp. contamination in packaged herbal tea products has previously caused multiple recalls (Gurtler et al., 2014). Cronobacter sakazakii is another pathogenic agent sometimes reported in herbal teas. It is a heat-resistant bacterium and presents a health risk particularly for infants (Stojanović et al., 2011). Manufacturers can reduce the microbial load in herbs by food processing practices, such as drying. However, several bacterial species such as C. sakazakii which are resistant to drying may endure the drying process (Beuchat, 2009). Besides, inappropriate drying or storage conditions can cause mold contamination and/or mycotoxin production, which can not be eliminated by boiling (Kosalec et al., 2009;Monbaliu et al., 2010).
In accordance with the above-mentioned reasons, the main aim of this study was microbial evaluation of Turkish herbal teas before and after infusion in boiled water.

Infusion process
Infusion was performed by mixing 10 g of dry herb with 90 ml of boiling water (~100 °C). Samples with boiled water were left for 30 min to mimic teabrewing. Microbiological analyses were conducted with pre-infusion (dry herbal tea) and post-infusion samples.

Microbiological analysis
Infused (10 ml) and dry herb (10 g) samples were separately mixed with 90 ml Enterobacteriaceae Enrichment Broth (Oxoid CM0317, ThermoFisher, United Kingdom) for the analyses of C. sakazakii, and with 90 ml Buffered Peptone Water (Oxoid CM0509, ThermoFisher, UK) for the analysis of Salmonella spp.

The pH analysis
The pH of all infused samples were measured with a pH meter (MILWAUKEE MW 102, United States).

Data analysis
All data of this study were statistically analyzed using Microsoft Office Professional Plus 2013 Excel software.

Results
The results of the microbiological analyses are given in Table 2. Three out of 20 tea samples (15%) were contaminated with C. sakazakii. This bacterium survived after infusion in both of the packaged and unpackaged senna tea samples. None of the samples were contaminated with Salmonella spp. No significant difference was found in occurrence of C. sakazakii before and after infusion of the samples (p>0.05).
Mold and yeast contaminations were found in 12 out of 20 teas samples (60%). No statistical significance (p>0.05) was found between the mold-yeast contents of the unpackaged and packaged herbal teas. After infusion, neither mold nor yeast was observed in any of the samples.
The pH values were measured after infusion. The lowest and highest pH values were found in the unpackaged apple and packaged linden tea samples as pH 3.17 and pH 6.63, respectively (Table 2). There was no statistical significance (p>0.05) between the pH values of the unpackaged and packaged teas. The overall median pH of the samples was 5.82.

Discussion
In this research, C. sakazakii in the senna tea (Cassia angustifolia) still existed after infusion in boiled water. In the study of Stojanović et al. (2011), C. sakazakii was found in 32% of the 14 different types of tea from Belgrade, Serbia, including senna, sage, and several fruit teas. This bacterium can remain viable after infusion due to its resistance to drying and heating. Also, in our study, both of the packaged and unpackaged senna tea samples were contaminated with this bacterium, raising the possibility that senna plant may intrinsically have a high risk of contamination with this bacterium, and it needs to be investigated at what stage this contamination is likely to occur. In Pakistan, 27% of the herbal teas have been found contaminated with Salmonella within the range of 2×10 1 -8.4×10 3 Colony Forming Unit (CFU)/g (Khattak, 2012). Salmonella was not found in any of the samples analyzed in the present study, complying with the microbiological criteria of Turkish Food Codex (TFC, 2011). However, for the presence of C. sakazakii in herbal teas, there is no criterion in the microbiological criteria of Turkish Food Codex (TFC, 2011).
In the United States, Tournas and Katsoudas (2008) detected mold and yeast contamination in 88% of the chamomile teas. European Herbal Infusions Association (EHIA, 2018) restricted the yeast and mold levels to 10 4 /g and 10 5 /g, respectively. According to the Directive on Microbiological Criteria of Turkish Food Codex, presence of yeast and mold in herbal teas is limited to 10 5 CFU/g-ml (TFC, 2011) and no sample in the present study exceeded these limits.
Generally, the antimicrobial properties of plant leaves are higher than that of roots (Oh et al., 2013). A protective effect of green tea against Salmonella, an inhibitory effect of mate tea against Gram-positive bacteria, and antimicrobial properties of sage tea were also investigated against many bacterial species, such as Bacillus cereus, Klebsiella pneumoniae, Morganella morganii, Pseudomonas aeruginosa, and Staphylococcus aureus (Albayrak et al., 2012;Oh et al., 2013). Additionally, senna tea has been found to have antimicrobial activity against B. cereus after infusion process (Albayrak et al., 2012).
The pH levels of herbal teas may vary according to their growth and processing conditions. Consumption preferences may also vary among individuals in part due to differences in the pH values (Phelan and Rees, 2003). The mean pH of herbal teas are recorded to be 3.15 by Brunton and Hussain (2001), and ranges from 2.73 to 5.85 according to the study of Malik et al. (2013), which are slightly lower than the median result (5.82) of the current work. Phelan and Rees (2003) detected a higher pH value in chamomile tea from Thailand (7.08) than that of the present study (6.40). The pH levels may affect the active ingredients of phytochemical compounds and antimicrobial activities of herbal teas. Allicin, a chemical metabolite extracted from sulphide/thiol plants, such as garlic, onion, and shallot, is most stable at the pH 5-6 (Jarriyawattanachaikul et al., 2016). In our study, 6 out of 10 unpackaged and 5 out of 10 packaged herbal teas were within this pH range. The apple tea had the lowest pH value (3.31±0.19) among our samples.

Conclusion
All of the herbal teas in this research were found to be within the microbiological limits for consumption according to Turkish Food Codex. However, the microbiological results highlighted the need for quality control of senna tea during production to eliminate the risk of C. sakazakii contamination. Due to this sensitivity of contamination, necessary precautions should be taken for the storage conditions, such as temperature and humidity, as well as for the prevention of contamination by pests and microbes at every stage of production. Nevertheless, increasing the number of samples in future studies is expected to be useful in obtaining a wider data and making the necessary risk assessment.