Introduction

Herbal tea, also called “cooling tea”, “herbal beverage” or “liáng chá”, is widely used for preventive and/or therapeutic healthcare in Hong Kong and South China. Herbal tea is a decoction with a range of herbal ingredients prepared according to the ethnomedical experiences of indigenous and local people [1]. The use of herbal tea is also regarded as a popular primary healthcare system in many countries [2].

The history of drinking herbal tea in China can be traced back to the Tang dynasty (AD618-907), proposed in the ancient Chinese literature Beiji Qian Jin Yao Fang [3]. Southern China, especially in Lingnan region, has a rich tradition of drinking herbal tea [4]. The tradition of drinking herbal tea in Hong Kong has a long history, yet the documented evidence pertaining to its origins and development remains scarce. In 1869, a Hong Kong government gazette ordinance mandated that ships transporting Chinese labours overseas must carry Chinese medicinal herbs in proportion to the number of passengers [5]. The requirement of preparing herbal tea for passengers during their voyage, signifies the popular utilization of herbal tea for ailment treatment and illness prevention in the nineteenth-century Hong Kong. In 1949, the government announced the requirement for a license to sell herbal tea [6], officially regulating the herbal tea business. Until now, herbal tea shops have been a fixture of Hong Kong landscape and still very much a part of daily life. Herbal tea has been inscribed in the national list of Intangible Cultural Heritage in 2006 [5].

Hong Kong has a subtropical climate with a wet–dry season [7]. Both written in Qing Dynasty (1644–1911), Lingnan fengwu ji (a record of the customs of Guangdong) and Chronicle of Sun-On County described the unfavourable climate and environmental conditions and frequent epidemic plagues happened in Hong Kong. From the perspective of Chinese Medicine, hot and wet weather are thought to contribute to a series of health problems. Exogenous heat and dampness evils attack the body, resulting in the abnormal consumption of qi (basic element that constitutes the cosmos) in the body and impairment of body fluids. In particular, the interstices and mysterious mansion on the skin remain open in summer and spring time to release the internal heat, and hence people are susceptible to the illness. To prevent the above conditions and alleviate the symptoms, people gradually learnt to use local botanical resources for making herbal tea. They boiled herbs for decoctions to treat a range of health problems from infectious diseases, cardiovascular disease to digestive disorders [1, 8]. The practice of drinking herbal tea builds up the immune system and helps to adopt the harsh environmental conditions.

So far, over 400 plant species have been used in producing herbal teas in southern China [9, 10]. The formulae of herbal teas vary in different regions [4]. The most iconic and representative herbal tea is Five Flowers Tea. This decoction is mostly consumed during the spring and summer months. Five Flowers Tea can be found globally because it has developed a widespread consumer base. The growing Chinese populations in cities like Toronto, London, Los Angeles and Sydney have boosted demand for traditional herbal remedies. These markets provide ingredients for expat communities and locals wanting to explore health traditions. The primary functions are to clear heat and expel dampness which helps to alleviate symptoms including stomach ache, abdominal bloating, indigestion and diarrhoea [1, 11]. There is no official formula of Five Flowers Tea in the Chinese Pharmacopeia. On the other hand, some of the local publications suggested that the mainstream formula of Five Flowers Tea includes the flower of Lonicera japonica (honeysuckle), Plumeria rubra (frangipani), Pueraria lobata (kudzu), B. ceiba (tree cotton) and Sophora japonica (Japanese pagoda tree) in equal weight ratio. Apart from the stated herbs, herbal tea shops and distributors may add some other herbs to reinforce the heat clearing and dampness expulsion function [11, 12]. The lack of standardisation in Five Flowers Tea formula poses problems. Different producers may use substitutes or adulterated ingredients due to the availability issues or to cut cost. This inconsistency in formulation can undermine the therapeutic claims and safety. So far, information and analysis on the composition of Five Flowers Tea and the rationale of using the herbs are lacking. We therefore set forth to analyse the authenticity, diversity and quality of Five Flowers Tea found in the Hong Kong market.

Given this, the objectives of this study are as follows: (1) to comprehensively analyse and document the botanical compositions of Five Flower Tea samples from Hong Kong using macroscopic and molecular identification methods, (2) to investigate the common and varying herbal ingredients across samples, and identify any adulterants present, (3) to evaluate compositional differences and variations between samples of different retail sources, (4) to determine whether the variation in compositions could impact the traditional uses and purported health benefits relied upon by the public and (5) to provide standardisation recommendations and contribute novel data to support the use and quality control of herbal tea.

Methods

Study area and sample collection

Hong Kong lies at the southern part of China, bordered by the South China Sea on all sides except the north. The territory consists mostly of Hong Kong Island, Kowloon Peninsula and New Territories. The study area has rugged relief and marked variations, with very limited flatland, comprising mostly woodland, shrubland and grassland [13]. Hong Kong has a subtropical climate in which the wet–dry seasonal change is apparent; greater variation in temperature and humidity can be observed when compared to other subtropical areas [7].

Sample collection was conducted between 2021 and 2023 in Hong Kong. Thirty-three bags of Five Flowers Tea were collected from local grocery stores, herbs retailers, market stalls and Chinese medicine pharmacies from 13 districts. The sampling locations were organised by administrative region, and each sample was assigned a unique voucher number. Table 1 and Fig. 1 detail the district of purchase and corresponding voucher numbers for the Five Flowers Tea samples collected. Claims and instructions given by the suppliers, total weight and size of the parcel and species of herbs were recorded (Additional file 1).

Table 1 Sample collection locations and voucher numbers
Full size table
Fig. 1
figure 1

Map of the study area for the collection of sample in Hong Kong. Investigated locations: 1. Central and Western District; 2. East District; 3. Wan Chai District; 4. South District; 5. Yau Tsim Mong District; 6. Sham Shui Po District; 7. Kwun Tong District; 8. Wong Tai Sin District; 9. Kowloon City District; 10. Tai Po District; 11. Sha Tin District; 12. Tuen Mun District; 13. Tsuen Wan District (photo courtesy: GeoInfo Map, The Government of Hong Kong SAR)

Full size image

Voucher specimen collection and macroscopic examination

The voucher specimen of each botanical sample was collected and deposited in Li Dak Sum Yip Yio Chin R&D Centre for Chinese Medicine at the Chinese University of Hong Kong (Additional file 1). Macroscopic identification of the Five Flowers Tea samples includes shape, size, surface, texture, cross section and type of fracture. Organoleptic examination including colour, odour and taste was carried out by comparing with the herbs listed in the Chinese Pharmacopoeia [14].

Molecular authentication

Total DNA was extracted from 2 g of each dried sample using a broad-spectrum plant rapid genomic DNA extraction kit (Biomed, China) according to the instructions of the kit manufacturer. DNA was visualized by electrophoresis in a 1.5% agarose gel, and images were collected (Gel Doc EZ system, Bio-Rad, Hercules, CA). DNA concentration was measured by Nanodrop spectrophotometer (DeNovix, Wilmington, DE).

Universal primers (COI-F: 5’-GGT CAA CAA ATC ATA AAG ATA TTG G-3’ and COI-R: 5’-TAA ACT TCA GGG TGA CCA AAA AAT CA-3’) were used for the amplification of the COI (Cytochrome Oxidase I) region. Another pair of universal primers (psbAF: 5’-GTT ATG CAT GAA CGT AAT GCT C-3’ and trnHR: 5’- CGC GCA TGG TGG ATT CAC AAT CC-3’) were used for amplification of the psbA-trnH region. A pair of universal primers (ITS2F: 5’- ATG CGA TAC TTG GTG TGA AT-3’ and ITS3R: 5’-GAC GCT TCT CCA GAC TAC AAT-3’) were used for amplification of the ITS2 region. The 30μl PCR reaction mixes included 21.4μl of ultrapure water, 3μl of DNA template, 1.2U of DNA polymerase, 3μl of 10X PCR buffer, 0.6μl of deoxynucleoside triphosphates (10nM), 0.3μl of each primer (10μM) and 1.2μl of 50mM MgSO4. Amplifications were carried out using a thermal cycler (Veritipro 96W). The thermal regime consisted of an initial denaturation at 94°C for 5 min; 30 cycles of 94 °C for 1 min, 56 °C for 1 min and 72 °C for 1.5 min; and a final extension at 72 °C for 7 min. PCR products were visualised on 1.5% agarose gel. Amplicons of the correct sizes were selected and purified for subsequent sequencing by BGI, China. The obtained sequences were aligned by ClustalW multiple alignment in MEGA11. The identities of species were identified via BLASTN, NCBI in GenBank database at their highest similarity.

Data analysis

Data analysis was carried out to evaluate the composition of different ingredients of Five Flowers Tea samples using Microsoft Excel. Voucher number, name of herbal materials, voucher code, weight, total weight of the package, percentage of herbal materials, indication and administration, condition and remarks are listed in Additional file 1. Excel was also used for statistical analyses. Sequences generated from samples in this study were assembled and aligned using CodonCode aligner for comparison and checking.

Results

Five Flowers Tea samples

We collected 33 Five Flowers Tea samples from different locations in Hong Kong. Most were packed by transparent plastic bags, with some of them having logos of the shops. A sample was wrapped by paper in the traditional way. Heat seal was the major way to close the plastic bag. Other ways included tying by string and a knot directly on the bag (Fig. 2).

Fig. 2
figure 2

Packaging styles of Five Flowers Tea samples. A HK07: Transparent plastic bag with a piece of note stating their brand, name of the tea, ingredients and functions. B KL03: Transparent plastic bag closing with heat seal. C KL07: Transparent plastic bag tying by nylon string. D HK08: Wrapped by paper printed with company name and address

Full size image

Among the 33 samples, 14 were labelled with description including the names of the herbal tea, herbs included or functions. Four were labelled with the brand name and 14 did not have label. One sample was labelled as Heat Dampness Clearing Tea instead of Five Flowers Tea.

Profiles of Five Flowers Tea samples

A total of 23 herbs in 18 families were identified by macroscopic identification (Additional file 2). Components of the herbs in each sample are listed in Additional file 1. Figure 3 shows that the most prevalent herb is B. ceiba, followed by C. morifolium. The third place is Artemisia capillaris, a herb collected in non-flowering period. It is popular for the shop owners to change some ingredients in the Five Flowers Tea formula to lower the cost or increase the potency.

Fig. 3
figure 3

Herbs used in the Five Flowers Tea. Number denotes the frequency of herbs used

Full size image

Composition of parcel samples

Additional file 1 and Table 2 illustrates the composition of herbs in the different parcel samples. HK07, HK12, KL04, KL05 and NT04 contain B. ceiba, P. rubra, C. morifolium, L. japonica and S. japonica, which are herbs in the classical formula. However, the weights of L. japonica and S. japonica in HK07 were exceptionally low compared to other samples. Two extra herbs, A. capillaris and P. vulgaris, were found. For HK12, the weight of B. ceiba was around 45 g, triple to that other four ingredients P. rubra, C. morifolium, L. japonica and S. japonica, which were around 15 g. For KL04, the weight of B. ceiba was over 40 g. For sample KL05, B. ceiba outweighed all the herbs by 2 to 10 times, followed by an extra herb A. capillaris. For NT04, the quantity of B. ceiba and P. rubra was higher than the rest of the herbs.

Table 2 Composition of Five Flowers Tea samples
Full size table

Non-traditional herbs prevailed in both HK06 and NT06. For HK06, 41.3% of the weight of the whole package was dried whole plant of Polygonum chinense L. For NT06, 55.9% was Morus alba L.

DNA barcoding for the authentication of herbs

We used DNA barcoding technique to identify the suspected adulterants. Barcoding region psbA-trnH was used followed by BLASTN search from GenBank database for the authentication of L. japonica and A. capillaris. The result suggested that four samples had Lonicera macranthoides instead of L. japonica. For the suspected adulterant of A. capillaris, BLASTN research indicated that one sample was Potentilla chinensis, while another was Potentilla discolor. ITS2 region was sequenced to examine four doubtful samples of Lysimachia christinae, and it was found that all were Desmodium styracifolium. Table 3 illustrates the ten found adulterants.

Table 3 Summary of identification of suspected adulterants using BLASTN search from GenBank
Full size table

Procedures for decoction generation

A total of 16 parcels were advised to bring up a boil, and all the parcels were advised to brew with different volumes of water, time length and level of heat. It was suggested that other ingredients can be added to increase the potency or flavour. Rock sugar and Siraitia grosvenorii (Lo Han Kuo) were the most prevalent optional additives to Five Flowers Tea, recommended by five store owners. For KL02, slab sugar but not rock sugar was suggested. Store owner of KL07 suggested to add L. japonica. Additional file 1 summarizes the recommended procedures for preparing the decoctions.

Storage conditions and identity of insects

All 12 samples that contained insects were discovered to be moist. We found that all collected insect samples were L. serricorne  (Fig. 4), as revealed by sequencing the COI region and compared with the data in BOLD Identification System (BOLD-IDS) or BLAST search in GenBank.

Fig. 4
figure 4

Image of insects collected from sample HK08 as an example

Full size image

Discussion

Variation of five flowers tea formula

Etymologically, the term Wu Hua Cha can be directly translated into Tea with Five Flowers. The most used formula is based on the combination of five herbs, B. ceiba, P. rubra, C. morifolium, L. japonica and S. japonica. These five herbs possessed the function of heat clearing, dampness draining and detoxifying. The herbs would act on intestinal area and liver. In terms of pharmacological effect, most of the five herbs possess the function of anti-inflammatory, antibacterial, antiviral, antidiarrhoeal and diuretic effect, aligning with the ethnomedical claims.

In this study, we only found eight samples comprised of the five floral herbs. Some herbal shop owners mentioned that some herbs are cut to reduce the cost. Herbal tea is a grassroot remedy in respond to the expensive fee for consulting a clinician. Therefore, there is often a trade-off between the cost and potency of the herbal tea. Prunella vulgari, for example, is a cheaper alternative to other expensive herbs like Pueraria lobata and S. japonica. Over half of the parcels (63.6%) contain it. Prunella vulgari is renowned in the heat clearing function and as a major ingredient in another popular herbal tea Xiasangju [15].

A. capillaris was found in 28 out of 33 parcels (84.8%) (Fig. 3). This herb is not a component in the classic formula. Also, the herb is not in the flowering stage. The harvesting season of A. capillaris is between March and April [16]. It is believed that during this period, the medicinal properties of A. capillaris are at their strongest for clearing heat and draining dampness, surpassing its effectiveness when collected during the flowering period.

A. capillaris was used to substitute S. japonica. According to the principle of Chinese medicine, A. capillaris has a greater dampness draining ability, but S. japonica is better in heat clearing. Some studies suggested that A. capillaris may have hepatoprotective, anticancer and anti-inflammatory properties [17,18,19]. Therefore, herbalists may preferentially select A. capillaris for its superior dampness dispelling capacity.

DNA authentication of herbs

Herbal remedies often face issues of adulteration due to the difficulty of visually identifying authentic herbs from fraudulent substitutes. The visual appearance of closely related herbal species can be highly similar, making organoleptic identification difficult for those without specialized knowledge of the subtle morphological differences. This confusion has led to many instances of adulteration, such as D. styracifolium being used as L. christinae, Potenilla species substituting for A. capillaris and Lonicera confuse being used in place of L. japonica.

NA barcoding method was performed to identify the herbs used in Five Flowers Tea parcel. DNA barcoding is a molecular technique used to identify species based on species-specific differences in a short, standardised fragment of nucleus and organelle DNA from a specific gene or genome region. That DNA fragment is then compared to a reference DNA database to determine the identity [20].

This method can precisely and rapidly identify species from various raw materials. In plant identification, the common barcode regions used are ITS2 and psbA-trnH. The region ITS2 (Internal Transcribed Spacer 2) is located in the nuclear genome, between 5.8S and 28S rRNA genes [21]. The region psbA-trnH is in the non-coding region of the chloroplast genome. Due to their high variability, the region can serve as a distinctive marker for a particular species [22].

The COI region, which is located in the mitochondrial genome of eukaryotic organisms, was used to identify insects due to its significant level of intra-specifically similarity and inter-specifically variation [23].

Using DNA barcoding technique, we detected the substitution of L. japonica, A. capillaris and L. christinae by L. macranthoides, Potentilla species and D. styracifolium, respectively. This may pose risks of having a reduced therapeutic efficacy or unintended side effects.

The pros and cons of using the substitutes and adulterants

The first herb in question was L. japonica. While L. japonica and L. macranthoides show similar pharmacological efficacy as evidenced by their medicinal usage history, the Chinese Pharmacopeia (2005 Edition) lists them separately. Both L. japonica and L. macranthoides exhibit low toxicity [24]. There have been many clinical trials conducted on L. japonica. In contrast, clinical investigations on L. macranthoides are limited. Both L. japonica and L. macranthoides share similar properties, including being categorized as cold in nature, sweet in flavour and having comparable therapeutic functions and applications. L. macranthoides is commonly cultivated and used as a substitute for L. japonica in China. Despite having similar chemical profiles and pharmacological effects, L. macranthoides offers a higher yield and lower production cost. This makes L. macranthoides an economical alternative to L. japonica [25].

The second herb in question was A. capillaris. Studies have shown that A. capillaris has greatly benefited the treatment of many diseases such as liver inflammation, cirrhosis and liver cancer, contributing to the wider development and advancement of traditional Chinese medicine [26]. A. capillaris and its active compounds have antioxidant, anti-inflammatory, antifibrotic, antiviral and anticancer effects. This also supports its traditional uses for treating liver and viral disorders as well as cancer [27,28,29].

We found that Potentilla species was used to replace A. capillaris in Five Flowers Tea. Potentilla species have been noted for their medicinal functions that are akin to those of A. capillaris, although there are variations between the herbs in terms of chemical profiles and effectiveness in treating certain conditions. Potentilla species have been used in traditional medicine, particularly in Chinese medicine, for a long time to treat ailments such as diarrhoea, hepatitis, rheuma, scabies and for detoxification purposes [30]. A. capillaris was found to be adulterated by Potentilla species, in particular P. discolor and P. chinensis. The aerial and underground parts of P. discolor have been used to treat inflammation, wound, cancer, infection caused by bacteria, fungi, and viruses, diarrhoea and diabetes mellitus [31]. There have been no reports of toxic effects during the long history of using Potentilla species and their extracts in traditional medicine [30].

Substituting A. capillaris by P. discolor and P. chinensis may be appropriate due to the pharmacological properties of the two latter herbs. A. capillaris is indicated for dispelling dampness due to its bitter and pungent flavour. P. discolor and P. chinensis, with their stronger cooling properties, are more suitable for resolving excessive heat conditions compared to A. capillaris.

The third herb in question was L. christinae. L. christinae is a traditional herb that possesses heat-clearing, diuretic, detumescent and detoxifying properties [32]. The adulterant D. styracifolium is widely used in traditional Chinese medicine for its diuretic and heat-clearing properties and for the management of urinary calculi, cardiovascular and cerebrovascular disorder and hepatitis [33].

From the Chinese medicine perspective, L. christinae and D. styracifolium have some similarities but also distinct differences in their indications and usage. Both herbs are valued for their diuretic and heat-clearing properties, which help resolve accumulated dampness heat and dissipate pathogenic factors. However, L. christinae focuses on dispelling dampness heat in liver and gallbladder. It is indicated for conditions that involve pathogenic factors affecting the hepatobiliary system. In contrast, D. styracifolium focuses more on dispelling dampness heat in the urinary bladder and kidney for urological conditions.

L. christinae and D. styracifolium therefore should be used interchangeably as they are for treating different types of conditions.

Herbal tea poisoning incidents have occasionally been reported, typically resulting from the use of toxic adulterants [34, 35]. In 2003, a local individual sought help after experiencing symptoms of poisoning. Datura metel, a plant known to contain toxic tropane alkaloids, was discovered in the brewing residue of the Five Flowers Tea [36]. The presence of adulterants in herbal tea can pose significant dangers to public health, ensuring the safety and quality of herbal tea products is therefore crucial.

Instructions on brewing and precautions

Instructions for brewing the tea differed among the different shops. The time, quantity of water, procedure of brewing and heat level vary (Additional file 1).

Furthermore, indications on the contradictions are simple and not mentioned in every parcel. For instance, sample HK10 stated that it is suitable for everyone regardless of their health condition. Sample NT02 suggested that people who are of “cold” constitution can only drink half of a bowl. G6PD deficient patients were suggested to avoid taking L. japonica [37, 38]. Of the 33 samples surveyed, only one parcel provided notice that people with G6PD deficiency should avoid using the product.

Apart from the potential danger of G6PD, a retailer has mentioned that pregnant women should avoid taking the Five Flowers Tea. However, a survey has indicated that more than half of Chinese pregnant women living in Hong Kong have taken Chinese herbal medicine, including Five Flowers Tea and L. japonica, for improving their gastrointestinal and digestive health. The use of Chinese herbal medicine during pregnancy did not appear to affect the condition, incidence or severity of jaundice in both normal and G6PD-deficient infants at birth [39]. The herbs used in Five Flowers Tea are not included in the List of Restricted Chinese Herbal Medicines for Pregnant Women [40].

Lasioderma serricorne infestation and quality control

L. serricorne was first reported and identified on tobacco in France in 1848 [41]. It is currently globally distributed and exceptionally prevalent in tropical and subtropical areas and is the most destructive pest affecting a wide variety of materials derived from both plants and animals [42]. It has long been recognised as the dominant pest species that infests stored Chinese medicinal materials, resulting in significant economic loss [43]. In our study, 12 out of 33 Five Flowers Tea samples were contaminated by L. serricorne.

The larvae of L. serricorne can cause significant damage by feeding on herbal tea leaves, resulting in reduced quality and value. Some conditions give rise to the growth of L. serricorne. Firstly, since herbal tea parcels are stored in compacted form, they provide an ample food source for the beetle to feed on. Secondly, the beetles like to hide away from bright light and low humidity. The best temperature range and relative humidity for rapid development of L. serricorne are 29–35 °C at 75% [44]. All our samples found with insects were moist. Hong Kong is warm and humid, not to mention the hot and humid storage conditions in warehouses and wet markets.

Thirdly, packaging material and styles would be another issue. The adult stage of the beetle is capable of biting through the packaging material [45]. In our study, different packaging styles are found (Fig. 2). Insect infestation is higher when the herbs are wrapped in a piece of paper or put in a transparent plastic bag with opening tied with a nylon. L. serricorne can easily penetrate the paper wrap, and paper wrapping cannot protect the herbs from the humidity. Nylon string tied on the bag opening is easily loosened, creating an entry point for the insect and moisture to enter.

The presence of contaminated and gnawed herbs, as well as the excrement, cast-off skin and other products, can pose a menace to the health of consumers. L. serricorne, in particular, has been linked to canthariasis, a human-enteric infection [46]. Cases of intestinal canthariasis have been reported in China and Malaysia [46, 47].

Effective pest management methods include phosphine fumigation, CO2 treatment and radiation treatments and high temperature. In recent years, pest management has put more emphasis on friendly to the environment and cost-effectiveness [48]. For instance, the package can be treated with natural repellents that prevent L. serricorne from consuming the contents [49]. Examples of such repellents include crude leaf extracts of Tecoma stans and Datura metel, as well as various plant essential oils such as shiso oil and savory oil [50]. For storage, multilayer packaging is an alternative to applying pesticide or temperature control warehouse. L. serricorne was found to get through polyethylene, polypropylene and polyester [51]. Thicker packaging material and multilayer packaging better deterred the spreading of the insect [45].

The interplay of cultural and environmental factors in the composition and evolution of Five Flowers Tea

In our study, we observed a diverse range of ingredients and proportions, with no unique recipe found. Several factors contribute to the variation in herbal tea recipes, including medicinal effects, Chinese philosophy, culture and environments.

Chinese philosophy emphasizes the interaction between human being and outside world and so is the traditional Chinese medicine. Factors such as individual person, time and place should be taken into account. In the holistic approach, each factor influences the others, and any adjustments should be considered in the overall context. We previously mentioned the impact of local climate and environment on the choice of herbs. For instance, locals brew “cold medicine” to get rid of heat-related ailments in Hong Kong’s hot and wet weather. Locals would consume specific herbal teas or Chinese medicine during different solar terms or seasons. For example, illness prevention teas are favoured in spring, Five Flowers Tea is for the hot summer, while soothing herbal teas are popular in autumn [5]. The proportions of ingredients in Five Flower Tea can also be adjusted for year–round consumption, not limited to the summer season.

Individual constitution and nature are factors leading to vast variations in herbal tea recipes. The proportions and the selection of herbs differ based on individual needs. When buying samples, vendors suggested different brewing methods and additives to the herbal tea in order to suit individuals need as well (Additional file 1).

The ingredients of Five Flowers Tea in Hong Kong are influenced and limited by the resources, local culture and habit. Recent studies indicated that not all the five flowers are readily available, leading to the brewing of herbal tea with fewer ingredients, and once the number of the ingredients down to B. ceiba alone [52]. The flexibility in Five Flowers Tea composition reflects its evolution from a specific herbal tea recipe to a cultural symbol, expanding the definition of Five Flowers Tea to a kind of drink which provides similar medicinal effects using various herbs with flowers.

The emerging needs of society, such as the occurrence of pandemics like COVID-19, also shaped the development of herbal tea. While standardisation of Five Flowers Tea can instil the confidence of consumers, certain variations may be beneficial for the medicinal efficacy.

In this study, the Five Flowers Tea samples were from vendors in Hong Kong. As herbal tea is also popular in South China, it will be interesting to expand the area of sample collection, to compare the quality and use of this medicinal product in nearby regions. The identification of herbs in this study is by organoleptic and DNA sequences. These sequences are either from our previous work [53] or GenBank. The matching of these two independent approaches can reveal the herb identity to high accuracy. We are constructing a DNA database of herbal tea material from authenticated samples, which will further enhance the confidence of authentication in the future.

Conclusion

Five Flowers Tea has been consumed for prevention and therapeutic purposes in Southern China for centuries. This work provides a systematic analysis on the composition and quality of Five Flowers Tea parcels sold in Hong Kong. We have found the use of variety of herbs, adulterants, irregular instructions and insect infestation, poses significant risk to public health. A better quality control of this popular herbal tea is needed. Also, with a standardised formulation with recommended variations can consumers be confident that this culturally significant herbal remedy delivers its attributed effects.