Introduction
Herbal plants have long been used as medications in all cultures around the globe [1]. The use of Traditional Medicine (TM) has expanded globally during the last decade and continued to gain popularity. TM has been used not only in developing countries for primary health care of the poor but also in countries where conventional medicine is the predominant practice in the national health care system [2]. According to one study, about two-thirds of the world population relies on medicinal plants for treating a variety of illnesses [3, 4].
Despite being more effective than phytomedicines, synthetic drugs and antibiotics often come along with unavoidable side effects and high prices. Moreover, because of the historical and cultural biases prevailing among people, synthetic drugs still have limited usage in different parts of the world, especially in the rural areas of developing countries. Consequently, the researchers have accelerated their quest to explore new drugs from natural sources in recent years [5]. The study of medicinal plants and their traditional uses has increased during the last few decades in different parts of the world [6]. Plants are extensively being studied to identify the phytochemicals and lead compounds responsible for their pharmacological and therapeutic efficacy. In this regard, ethnobotanical surveys of medicinal plants have made essential contributions to the discovery and conservation of novel biological resources [7, 8].
Bangladesh is a country gifted with a rich plant diversity because of its various environmental conditions such as warm and humid climate and fertile alluvial land. About 6000 species of indigenous and naturalized plants grow in the country [9], among which more than 1000 species contain medicinally active chemical substances [10]. Chittagong Hill Tracts (CHT) is the only extensive hilly region of Bangladesh located in the southeastern part. Known as a land of splendid natural beauty with landscape, lakes, and rivers, the area has hills and cliffs covered with dense jungles of bamboo, shrubs, and creepers harboring an abundance of floral species [9]. The tribal and other native communities are mainly dependent on traditional medicinal healers for treating different ailments.
Local inhabitants hold a strong belief in the healing properties of herbal medicine. A large portion of the population are deprived of modern medical facilities. This condition has also contributed to their dependence on TM [11, 12]. Unfortunately, medicinal plants and the abundant knowledge associated with them are facing the risk of serious loss owing to aberrant climate, deforestation and other human-made hazards, and migrations of traditional medicinal healers to different jobs [13]. Besides, the knowledge of traditional therapeutic practice has been passed only verbally from one generation to the other [14], and the written documents are unavailable in most cases [15, 16].
Skin diseases are general disorders that affect people from all age groups and produce damages in various ways [17]. It is difficult to define skin diseases precisely since they include a wide range of different disorders. Their prevalence rates are influenced by nutrition, habits, genetics, and socioeconomic status of a particular community [18]. The growing proportion of dermatological diseases encountered in general practice causes a significant part of morbidity in children; however, little information is available about the frequency of specific skin diseases. Although the overall incidence rate of all diseases combined has decreased in general practice, the incidence rates of the bacterial, mycotic, and atopic skin diseases have increased [19].
Transmissible skin diseases are major public health problems in many developing countries like Bangladesh. Lack of proper hygiene and basic amenities, and especially the difficulty of traveling to distant health care facilities in hilly terrains, are the major risk factors of dermatological diseases [20]. Herbal plants are considered as the first line of treatment against skin disorders in many rural areas of Bangladesh. According to the literature, some ethnobotanical studies have already been performed on dermatological diseases in the country [21, 22, 23, 24, 25]. However, these studies have not covered the areas, which we have specifically focused on our research. Our research is the first to document the plants used for treating dermatological diseases in some particular hilly regions located in CHT of Bangladesh. Because of the diversity of dermatological disorders, we attempted to focus on those disorders which are found to be treated by the preparations of the documented plants in our survey. The most prominent disorders include boils, eczema, bruise, itching, sore on different areas of the body (such as mouth, throat, tongue, and foot), dandruff, acne, scabies, scurvy, chickenpox, measles, leprosy, and skin ulcer. Apart from documenting the plants, our study aimed to provide relevant information about the plants and their potential applications in novel drug discovery.

Materials and Methods

Study area
CHT, with an area of about 13184 km2, is bordered by Myanmar to the southeast, the Tripura state of India to the north, Mizoram state of India to the east, and Chittagong district of Bangladesh to the west. It is situated between 21°25'N to 23°45'N and 91°54'E to 92°50'E [9, 26]. Tropical monsoon climate prevails in the region with an average annual rainfall of 2540 mm in the north and east, and 2540 mm to 3810 mm in the south and west. The hills soils, characterized by strongly acidic nature, are chiefly yellowish-brown to reddish-brown loams graded into broken shales, mottled sands or sandstones at varying depth. The vegetation type of the area falls under semi-evergreen (deciduous) and tropical evergreen forests. Along with natural vegetation, Jhum cultivation is practiced on the slopes of the hills. The main rivers are Karnafuli, Sangu, Feni, and Matamuhuri, which drain into the Bay of Bengal. According to the 1991 census, the area population is about 1.042 million dominated by Mongolian, Chakma, Tripura, Murong, and Magh tribes. The inhabitants mainly depend on the resources coming from the hilly areas [9]. The present ethnobotanical study was conducted in three districts of CHT: Rangamati, Bandarban, and Khagrachari. Figure 1 shows the different areas of data collection.

Sampling of informants
The ethnomedicinal survey was conducted between January 2016 to December 2017, and the aim was to cover at least five Kabiraj/Hakim/Unani/Ayurvedic practitioners in each area. In the districts where tribes live, we emphasized the fieldwork. Reputed Hakims and Ayurvedic drug manufacturers such as Hamdard, Ayurvedia Pharmacy, Shakti, Sadhana, and Kundeshwari, along with the experts in Unani and Ayurvedic Board were consulted, too. We also interviewed local people with practical or empirical knowledge on medicinal plants, and a total of 387 people were chosen for this purpose. The inhabitants with enough knowledge of local medicinal plants or involved in medical practice with plants for a long time were the ones selected for interviews. Besides, the age and gender of interviewees along with their educational background and experience on the use of traditional medicinal plants, were also taken into consideration. Ayurvedic and Unani medical practitioners usually have their formularies available in printed form. However, in this survey, the formularies were not enough as their authenticity could not always be confirmed. There could have been biased information, either intentionally put on by the practitioners, or based on local beliefs. This was another aspect where interviewing people with practical knowledge of medicinal plants and a long history of practice provided us with more reliable information.

Ethnomedicinal data collection
Verbal consent was obtained by the interviewer from each informant, ensuring that the objectives of the study were clearly explained to them. Most interviews were arranged by local people who were familiar with traditional healers and could communicate with native communities at the same time. Bengali, the official language of the country, was used for conducting interviews. Local bilingual translators helped during the communication with indigenous populations with different mother tongues.
The survey employed open-ended and semi-structured questionnaires [13], which included the following information: a. the local name; b. plants part/s used; c. source of plant material; d. the method of preparation; e. solvent/adjuvant used; f. mode of application; g. dermatological and other medicinal uses; h. voucher specimen number; and j. dose and dosage forms. The scientific names, family names, habit, habitat, nature, relative abundance, and conservation status of plants were documented either upon consultation with Botanist Mr. Md. Manzur-ul-Kadir Mia, former Principal Scientific Officer and Curator of Bangladesh National Herbarium, Dhaka (DACB), or by the literature search. The voucher specimen of each plant was deposited in Bangladesh National Herbarium, Dhaka, and after consulting with Botanist Mr. Md. Manzur-ul-Kadir Mia, the accession numbers of the plants, were documented. The accession number of each plant is mentioned in its respective voucher specimen, and all voucher specimens are provided in the supplementary section (Supplementary Tables, Table S.1).
Books, research articles, and relevant web pages were also studied during the survey to collect data on phytochemical compounds of the plants as well as any reported toxicity studies. We also documented the compounds commonly found in the reported plant species.





Data analysis
The species of plants were listed in alphabetical order by their scientific name, family, local name, generic name, habit, habitat, geographical distribution, relative abundance, nature, plant parts used, mode of preparation, the solvent used and Frequency of Citation (FC) [21, 27]. The FC of the species of plants in this survey was evaluated using the following formula:
FC=Number of times a particular species was mentioned/The total number of times that all species was mentioned× 100. Frequency distribution was calculated using the SPSS v. 19 [27].
The taxonomic identification of each plant was performed following the guidelines on the website http://www.tropicos.org/NameSearch.aspx and upon consultation with Botanist Mr. Md. Manzur-ul-Kadir Mia.

3. Results
Informants
Of 387 informants interviewed, the majority were male. Most of them were 50-60 years old, followed by informants aged 40-50 years. Nearly every interviewee was an independent healer and completed at least 10 years of education. Also, a substantial number of professional alternative medicine practitioners were interviewed. The majority of interviewees had 5-10 years of practical experience or empirical knowledge who were followed by people with 10-20 years and 2-5 years of experience (Table 1).

Plants used in the treatment of dermatological disorders and other relevant information
Among the 189 plants collected from 3 different districts, 56 species were used by the traditional healers to treat dermatological diseases in Bangladesh, and they belong to 32 different families. The most significant number of species belonged to the Fabaceae family (8 species), followed by Rubiaceae (5 species), Lamiaceae (4 species) and Asteraceae (Compositae) (3 species) (Figure 2).

Different plant parts were reported to be used, among which leaves (33.9%) were the most frequently used ones. They were followed by leaves and root (10.7%), leaves and stems (7.1%), bark and leaves (5.4%) and the whole plant (5.4%). The dominant mode of preparation was paste (35.7%) followed by juice (19.6%) and extract (14.3%) (Figure 3).

The solvents used for the preparations were water (53.6%), milk (19.6%), honey (16.1%), and wine (10.7%). The mode of administration was found to be topical (67.9%), oral (21.4%), and both topical and oral (10.7%). Among the reported species, 32.1% was shrub followed by herb (28.6%), and climber and tree (10.7% both). Of the plant species, 19.6%, 14.3% and 12.5% grow in the forest, homestead, hill forest and homestead, respectively. Significant portions also grow in the garden (8.9%) and marshy places (3.6%). Based on availability, the species are categorized as common (66.1%), less frequent (23.2%) and rare (10.7%). The majority of the plants are wild (66.1%); some are both cultivated and wild (17.9%) while others are only grown (16.1%). Figure S.1, Figure S.2, and Figure S.3 elaborate on the use of different solvents, mode of administration and nature of plants, respectively (supplementary figures).

The doses of the available plants varied widely (Table 2).



Frequency of citation of the plants
The most cited species of plants were Aloe barbadensis Mill., Azadirachta indica A. Juss., Commelina erecta L., Flemingia congesta Roxb. ex W.T. Aiton, Ichnocarpus frutescens (L.) R. Br., Lawsonia inermis L., Melia azedarach L., Sarchochlamys pulcherrima, and Sida acuta Burm. f. (Supplementary Tables, Table S.1).

Discussion
In the present investigation, Fabaceae was the most dominant family of plants. Being the second-largest family of medicinal plants, Fabaceae covers more than 490 plant species, the majority of which are applied in TM. The worldwide prevalence of a high number (approximately 20000) of species of trees, vines, shrubs, and herbs can also relate to the predominance of this family in our study [28]. Leaves were the major plant parts used against dermatological disorders either alone or mixed with other plant parts. Other previously conducted studies produced similar results [1, 13, 29, 30, 31, 32]. The medicinal value of leaves is attributed to the presence of photosynthates, which make them the primary photosynthetic organs [32]. The ease of collection of leaves, compared to other parts of plants such as roots, fruits, and flowers, is another reason behind their widespread use [29, 33]. Fresh leaves are prepared as paste and applied to skin infections as reported in previous studies [34]. That is why paste is the main mode of preparation in our investigation. Shrubs, herbs, and trees were the most common habit of plants. This state could be attributed to the natural abundance of shrubs or herbaceous plants in this geographical area and their accessibility to the communities of local traditional healers [35, 36].
A comparative study between the cited plants of our research and those of other investigations revealed some noteworthy similarities. Of 56 species, 36 plants have been reported in the treatment of many dermatological problems in other countries such as India, Nepal, Pakistan, South America, etc. (Table 3). The use of the same plants in different communities of the world for similar purposes support the pharmacological efficacy of these plants. However, we did not categorize the medicinal uses of the plants in other communities. Few plants were found in different surveys that were used to treat a particular ailment, while for most of the other plants, the available information was minimal. Therefore, Table 3 lacks the informant consensus of the plants to determine the most widely used plants for a particular disease in that study region.


High FC values indicate the various and numerous medicinal properties of the plants, and thus, they demand further phytochemical, pharmacological, and toxicological analysis for the discovery of potential novel drugs. For instance, Aloe barbadensis Mill., and Azadirachta indica A. Juss., are popular market preparation of reputed pharmaceuticals mainly for their emollient and wound healing properties.
The majority of the plants reported in our study, apart from dermatological disorders, are used in many other diseases, which are listed in the supplementary section (Table S. 2). This trend is an indication of the tradition, as mentioned elsewhere, to develop a local healing system through trials and errors for ideal treatment practices [37, 38].





Several studies have been conducted to find the specific compounds responsible for the use of the reported plants to cure dermatological disorders. However, not all plant species in our study have been subjected to extensive phytochemical investigations. Carrying out a detailed phytochemical investigation in all plant species is a gigantic task and somewhat beyond the scope of the study. Nevertheless, based on the previous investigations, Table 4 has identified and listed the compounds behind the dermatological potential of several plants.


To explain the mode of action of all 56 reported plants individually is beyond our scope. Generalizing the mechanism of actions of all plants would not be rational either since it would result in the exclusion of lots of valuable information. Nonetheless, several studies have been carried out to underpin the role of these active compounds, which would validate their contribution to treating dermatological disorders. For instance, anthraquinone derivatives are known as antibacterial agents that, in particular, are highly effective against the gram-positive bacterium S. aureus. They can penetrate the bacterium by interacting with the cell membrane which ultimately results in the inhibition of the cell growth at its logarithmic phase [39]. Mannose-6-phosphate is responsible for the wound healing property of certain medicinal plants by inhibiting transforming growth factor-beta [40]. Besides, bergenin is one of the compounds accountable for antifungal action, which works by inhibition of spore germination [41]. However, alkaloids, tannins, and flavonoids are well-known antioxidants that work by inhibiting either the formation or the proliferation of free radicals. The increase of free radicals is blocked either by scavenging species that initiate peroxidation or chelating metal ions so that they are incapable of producing reactive species or decompose lipid peroxides [42, 43]. The anti-inflammatory property of hydroxycoumarins and terpenoids is due to the inhibition of interleukin-6 production [44] and modification of critical cell signaling pathways, respectively [45].
Through the literature study, 13 plant species have been found to possess potential toxic compounds (Table 5). Apart from picrotoxin and azadirachtin found respectively in the berries of Anamirta cocculus [46, 47] and seeds of Azadirachta indica [48, 49], the rest are toxic mainly due to high doses of ingestion.

Conclusion
The present inventory sheds light on the contribution and importance of the natural flora of Bangladesh in treating different dermatological disorders. The knowledge in this survey supports the potential development of novel plant-based medications. Preliminary studies in these medicinal plants have provided the basis of their pharmacological efficacy against dermatological disorders; however, further investigations are necessary for ensuring safe therapy concerning traditional medicinal plants.

Ethical Considerations
Compliance with ethical guidelines
The authors took verbal consent from all the traditional medicine practitioners and indigenous people for participating and providing the relevant information in this survey.

Funding
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Authors' contributions
Conceptualization: M.M.K. Mia; Methodology: M.M.K. Mia, Mohammad Fahim Kadir; Investigation: M.M.K. Mia, Md. Rafi Anwar, Nurul Islam Setu, Raunak Jahan; Writing-original draft: Shejuti Rahman Brishty, Mohammad Fahim Kadir; Writing-review & editing: Shejuti Rahman Brishty and Rabiul Islam; Resources: Shejuti Rahman Brishty, Rafi Anwar, Rabiul Islam, Nurul Islam Setu; Supervision: M.M.K. Mia, Mohammad Fahim Kadir, and Rabiul Islam.

Conflict of interest
The authors declared no conflict of interest.

Acknowledgments
The authors are utterly thankful to all the traditional medicine practitioners and Ayurvedic drug manufacturers for providing information about their practice. Appreciations are also expressed to the local authorities as they offered help and administrative facilities during the survey. The authors are grateful to Muhammad Shahdaat Bin Sayeed, Department of Clinical Pharmacy and Pharmacology, University of Dhaka, Bangladesh for his critical comments on this paper.


References