Ethnobotany and its role in drug development

PHYTOTHERAPY RESEARCH Phytother. Res. 14, 479–488 (2000) REVIEW ARTICLE Ethnobotany and its Role in Drug Development† Michael Heinrich* Centre for Pharmacognosy and Phytotherapy, The School of Pharmacy, University of London, 29–39 Brunswick Square, London WC1N 1AX, UK The botanical collections of early explorers and the later ethnobotany have played important roles in the development of new drugs for many centuries. In the middle of the last century interest in this approach had declined dramatically, but has risen again during its last decade, and new foci have developed. The systematic evaluation of indigenous pharmacopoeias in order to contribute to improved health care in marginalized regions has been placed on the agenda of international and national organizations and of NGOs. In this paper the results of various projects on Mexican Indian ethnobotany and some of the subsequent pharmacological and phytochemical studies are summarized. Medicinal plants are an important element of indigenous medical systems in Mexico. This study uses the medicinal plants in four indigenous groups of Mexican Indians—Maya, Nahua, Zapotec and Mixe—as an example. The relative importance of a medicinal plant within a culture is documented using a quantitative method and the data are compared intra- and interculturally. While the species used by the indigenous groups vary, the data indicate that there exist well-defined criteria specific for each culture, which lead to the selection of a plant as a medicine. For example, a large number of species are used for gastrointestinal illnesses by two or more of the indigenous groups. At least in this case, the multiple transfers of species and their uses within Mexico seems to be an important reason for the widespread use of a species. Some of the data we gathered in order to evaluate the indigenous claims are also discussed, focusing on the transcription factor NF-kB as a molecular target. This led to the identification of sesquiterpene lactones such as parthenolide as potent and relatively specific inhibitors of this transcription factor. Copyright # 2000 John Wiley & Sons, Ltd. Keywords: medicinal plants; ethnobotany; traditional medicine; drug development; natural products; Mexico; Maya; Nahua; Zapotec; Mixe; transcription factors; NF-kB. INTRODUCTION Medicinal plants are an important element of indigenous medical systems in Mexico as well as in other countries. These resources are usually regarded as part of a culture’s ‘traditional’ knowledge. For many years Europe has profited from exchange with other continents and many of the pure natural products and some of the phytotherapeutic preparations used today are derived from plants used in indigenous cultures. Examples of 18th century explorers, who described indigenous plant use in detail, are the British Richard Spruce (Schultes, 1983) and the German Alexander von Humboldt (1997). He studied in detail the preparation of curare used as an arrow poison in * Correspondence to: Prof. M. Heinrich, Centre for Pharmacognosy and Phytotherapy, The School of Pharmacy, University of London, 29–39 Brunswick Square, London WC1N 1AX, UK. E-mail: phyto@cua.ulsop.ac.uk † Plenary lecture given at the meeting of the United Kingdom Academy of Pharmaceutical Scientists ‘Pharmacognosy in the 21st Century’, Bradford, UK, April 2000. Contract/grant sponsor: Swiss Agency for Development and Cooperation, Berne. Contract/grant sponsor: Swiss Academy of Natural Sciences. Contract/grant sponsor: Deutsche Akademische Austauschdienst, Bonn. Contract/grant sponsor: Secretariá de Relaciones Exteriores, México, D.F. Contract/grant sponsor: Freiburger Wissenschaftliche Gesellschaft, Freiburg. Copyright # 2000 John Wiley & Sons, Ltd. South America. The role of the ethnobotanist in the search for new drugs has been of continuous importance until the second half of the 20th century (Schultes 1962), when other approaches became more ‘fashionable’. In recent years the use of such information in medicinal plant research has again received considerable interest in the media and in some segments of the scientific community. Also the ‘Western’ use of such information has come under increasing scrutiny and the national and indigenous rights on these resources have become acknowledged by most academic and industrial researchers [see Programme for Traditional Resource Rights (http://users.ox.ac.uk/wgtrr.htm)]. These developments result in a considerable challenge to and in increasing responsibilities for ethnobotanists and ethnopharmacologists. Simultaneously the need for basic scientific investigations on medicinal plants using indigenous medical systems becomes ever more relevant. The public availability of the results of research into medicinal plants is essential for further developing and ‘upgrading’ indigenous and traditional medicine. In our research we are particularly interested in understanding the use of medicinal plants and in developing rigorous methodologies for ethnobotanical research. Additionally we are concerned with indigenous concepts about medicinal plants and the selection processes for such plants (see Brett and Heinrich, 1998, and the other contributions in Received 11 July 2000 Accepted 13 July 2000 480 M. HEINRICH Table 1. Summary of ethnographic background, medical ethnobotany and medical system for the four cultures Culture (ethnic group) Ethnograpic background Subsistence agriculture with maize as the main staple Commercially important crops coffee citrus fruits vegetables/legumes etc. Other important economic activities handicraft (production of huipiles) Bilingual or monolingual speakers of the indigenous language (in %) Information on medicinal plants Total number of botanical species documented Total number of use-reports obtained Maya Nahua Zapotec Mixe ‡ ‡ ‡ ‡ ÿ (‡) (‡) ‡‡ (‡) ‡ ‡‡ ‡ ‡ ‡ ‡ ÿ ‡ 99 ÿ 70 ÿ 60 ‡ 90 320 1549 203 816 445 3059 213 n.a. n.a., data not available; ‡‡, very important or frequent; ‡, relevant aspect of the culture. (‡), only of minor importance; ÿ, not present in the region at the time of the study. this issue) but this goes beyond the scope of this paper. Here I give an overview of our collaborative research efforts of the past years looking at both ethnobotanical and pharmacological–phytochemical aspects of this research. In the latter part I concentrate on plants studied to evaluate their anti-inflammatory activity. ETHNOBOTANICAL BACKGROUND All studies discussed here were conducted in Mexican Indian communities in the southern parts of Mexico (Table 1, Fig. 1). The Indian groups included belong to the Yucatec Maya, Lowland Mixe, Nahua of the Sierra de Zongolica and Sierra Isthmus Zapotec. Generally speaking, these regions are in the more marginal regions of Mexico. Emigration is frequent, especially with regard to the Zapotec and Nahua. In all four groups subsistence agriculture is the economic basis, with corn being the main crop. Other important crops vary from region to region. Indigenous forms of medical treatment are still important. One of the main reasons is the lack of biomedical facilities in the communities and the indigenous peoples’ distrust towards allopathic medical doctors. Indigenous medical specialists—as the term is used in this paper—includes not only those people who are considered to be a healer by the community, but also people who state that they ‘only’ know herbal remedies and who advise members of the community on how to use them (‘specialists in home remedies’, cf. Heinrich, 1997) and who may perform ritual cleansing ceremonies (limpias). According to the unpublished data of the local health authorities and our surveys, it is apparent that gastrointestinal disorders (frequently diarrhoea and— especially in younger children—as a result thereof, dehydration) and respiratory illnesses are major health problems in all regions. Infected wounds and other inflammatory dermatological diseases are also common. state of Yucatán (for details, see Ankli et al., 1999). The entire municipio of Chikindzonot numbers 2750 inhabitants. 56% of those older than 15 years are literate and one third of those older than 5 years are monolingual speakers of Maya, with the remainder being bilingual. The economy is based on subsistence agriculture (mostly maize, beans and squash), as well as honey, fruit (watermelon and citrus fruits) and cattle breeding. Hunting is still practised regularly, especially by younger men. Handicraft articles (hammocks and huipiles, a women’s blouse) are sold in the market of Valladolid. No detailed anthropological monograph on the Maya of this area is available, but the community of Chan Kom, which was first studied by Redfield and Villa R. (1934), is only 27 km to the north. Health and Healing. Poisonous snakebites (from species such as tsab - cascabel - Crotalus durissus) are feared but only a few cases have been recorded in recent years. Diabetes is considered a critical health problem by local health authorities, and elderly women often claim to suffer from it. The best-known group of healers is the h-men. They are not only healers but also religious specialists who perform ceremonies in order to request rain for the milpa (cornfield) from the rain-god or to pray for other needs of the community. He or she is the owner of a sastun, a stone used for divining. Midwives and hierbateros (specialists in medicinal plants) form another group of healers. The medicinal plant specialists are generally proficient in treating broken bones as well, and thus work as hueseros. The sobadores give massages along with the midwives. Maya The data on the Maya were gathered in the communities of Chikindzonot (pop. 1500), and adjacent ones all south of the city of Valladolid in the southeastern part of the Copyright # 2000 John Wiley & Sons, Ltd. Figure 1. Mexican Indian communities Phytother. Res. 14, 479–488 (2000) ETHNOBOTANY AND DRUGS All these groups of healers extensively use medicinal plants. Some use these plants almost entirely as empirical medications, while others (in particular the h-men) also use them for ritual purposes. An outpatient clinic run by the SSA (Secretarı́a de Salud y Asistencia) and staffed with a pasante (a medical student in the final year of training) and a mestizo nurse provides biomedical health care in Chikindzonot. For most ailments, however, the Maya still prefer to be treated by their own healers. In 1993/1994 the pasante, although a woman, was called only once to assist a delivery. Ethnomedical and ethnobotanical data were collected from February 1994 until May 1995, and in September–October, 1996 (Ankli et al., 1999). Mixe The land of the Mixe extends mostly through the cool and humid mountains of the Sierra de Juarez in the Mexican state Oaxaca. San Juan Guichicovi is the only Mixespeaking community in the subtropical Istmo de Tehuantepec. It is the principal community (cabecera) in a sub district (municipio) of the same name. In 1980, the municipio had 20 000 inhabitants, while the cabecera had 5500 to 6500 (Heinrich et al., 1992). 75% of the population in the cabecera are considered to be bilingual. A minute fraction of the population only speak Spanish. The economy is based on subsistence agriculture (mainly maize) and on the production of coffee and citrus fruit. Another relevant commercial product is huipiles of the Tehuantepec style. They are produced by women and men of the community and are usually sold to Isthmus Zapotec merchant women who resell them in numerous communities of the Oaxacan part of the Isthmus and in many other regions of Mexico. No detailed monograph on the lowland Mixe is available. Two cultural aspects that have been dealt with in detail are the ritual calendar, which is still used in some parts of the municipio and, in some adjoining municipios, the relationship of religious ritual to medical concepts (Heinrich, 1989, 1997). Health and healing. There are at least 15 different types of healers (pa‘am iixyp’) known in the community. The largest groups are ‘specialists in home remedies’. Other important groups are midwives, chupadores (healers that suck out an illness), prayer makers (rezadores), spiritists (espiritistas) and spiritualists (espiritualistas). Considerable differences distinguish the various groups of healers (Heinrich, 1997). Recently, the number of those offering help in case of illness has grown to include travelling salespersons (mostly Mixe-speakers with minimal experience in Western medicine), assistants to the local Roman Catholic priest, various Protestant groups and trained nurses. The differences between these health care providers are enormous. The travelling salespersons and other individuals with minimal experience in Western medicine are best considered as charlatans, while others have a solid background in medical therapy. Over the past 10 years, there have always been one to four practising medical doctors present. Some are sent by Mexican government agencies such as IMSS (Instituto Mexicano del Seguro Social) and INI (Instituto Nacional Indigenista). Some of the ‘specialists in home remedies’ and travelling salespersons sell pharmaceuticals and give advice on how to administer them. No quantitative data Copyright # 2000 John Wiley & Sons, Ltd. 481 are available on the importance of these forms of medication as opposed to indigenous phytotherapy (Heinrich, 1997). Data in this area were collected in 1985 and 1986, and later in several short trips (1–2 months). Nahua The Sierra de Zongolica in the Mexican state of Veracruz is part of the Sierra Madre Oriental. Its area of 1900 km2 lies to the south of 19 ° N latitude, and is bordered by the states of Puebla and Oaxaca. The area is subdivided according to altitude into three major regions: the cold highlands (tierra frı́a), temperate intermediate zone (tierra templada) and hot lowlands (tierra caliente). Oak and conifer forests dominate the vegetation in the cold highlands, whereas in the hot lowlands the principal ecosystem is the tropical evergreen forest. There are about 200 000 inhabitants in the Sierra and 34 000 in the sub district (municipio) of Zongolica (Weimann and Heinrich, 1997). Since the 19th century the commercial production of coffee has played the dominant economic role. Many different fruits and vegetables are cultivated and occasionally the forest trees are cut for wood. In past decades, tobacco was an important crop. The people of the hot and cold zones are mutually dependent on each other; for example, workers from the cold zone come to the lowlands to earn an annual subsistence. Nahuatl, belonging to the Uto-Aztecan language families, is still spoken by a large portion of the population. More than 90% of the population in the highlands are mono- or bilingual speakers of Nahuatl while in the lowlands it is approximately 70%. Health and healing. Gastrointestinal and respiratory disorders are the most frequent ones. Tuberculosis is still prevalent. Mal aire, which is said to be caused by an evil spirit or wind, and susto/espanto (sudden fright) are culturally important medical problems. Symptoms of the latter may be sleeplessness (despite being tired) and/or shivering fits. We often encountered reports of unexplained gastrointestinal disorders. The Nahuatl have various groups of specialists for curing illness. Ritual healers (curanderos) treat culturebound syndromes. Midwives accompany women throughout the pregnancy, birth and childbed. The hierberos are specialists in medicinal plants. The hueseros are experts on the skeleto-muscular system and deal with sprains, fractures and bruises. Many healers have had experience in several forms of treatment. Ethnomedical and ethnobotanical data were collected in selected regions of the Sierra Nahua de Zongolica from September 1993 until February 1995. In this area, biomedical forms of treatment were more readily available than in the other three regions. Zapotec The area of the Zapotec is adjacent to the one of the Mixe. Forced in the 14th century by Aztec and Mixtec invasions to leave the highland Valley of Oaxaca, the Sierra Zapotec settled in their present area. The communities we worked with, especially Santo Domingo Petapa and Phytother. Res. 14, 479–488 (2000) 482 M. HEINRICH Santa Marı́a Petapa, are linguistically and culturally isolated from the other groups Sierra Zapotec groups. Between 1% and 5% of the inhabitants older than 5 years are monolingual speakers of Zapotec, and 50% –70% are bilingual. As with the Mixe, coffee and citrus fruits are important commercial products. Health and Healing. The spectrum of illnesses known to the Zapotec is similar to that of the Mixe (see above). In addition, many of the groups of healers are similar to those described for the other groups: specialists in home remedies, midwives and herbalists (hierberos). Bonesetters are still active in the communities and in the treatment of sprains, fractures and bruises. Many healers have experience in several forms of treatment. Gastrointestinal and dermatological conditions are the two groups of illnesses treated most frequently with herbal preparations. The treatment is largely based on the principle of binary opposition using a ‘hot–cold’ classification (Frei et al., 1998). ETHNOBOTANICAL METHODS AND DOCUMENTATION By interviewing specialists in medicinal plants and other healers from the different regions we obtained information on the use, preparation, application and properties of the plants as well as descriptions of illnesses and treatments. We used unstructured interviews and discussions on medicinal plants and the modes of treatment at meetings of groups of indigenous healers. Sometimes other community members were present. The reports from the healers for each species were summarized. Further data are based on participant observation, especially on the observation of the healers’ healing sessions and of self-treatment. Quantitative ethnobotany The study design of our ethnobotanical research is based on the assumption that the more often a plant is reported to be useful the more often it is going to be used. Quantifying the data by evaluating each use-report (ur) of a species thus allows one to estimate the relative importance of a plant in a socioculture. Thus, culturally important plants are those that are used by a large number of healers preferably for the same category of indigenous use, while plants that are cited as useful by only one or two informants are considered to be of low cultural importance. In order to further study the use of these indigenous medicinal plants, we first assigned each use report to one of eight to eleven groups of indigenous uses. The main groups are identical in all studies. These include: * * * * Gastrointestinal disorders Dermatological problems Respiratory illnesses and Gynaecological and andrological problems Others are specific to one culture. For example, in the case of the Zapotec skeleto-muscular and febrile diseases form one group each, while in case of the Nahua and Copyright # 2000 John Wiley & Sons, Ltd. Maya one category for pain/febrile diseases and fever each was used. Trotter and Logan (1986) developed a method based on the concept of ‘informant consensus’ for identifying potentially effective medicinal plants. They compared the total case-number for each ailment (number of informants that reported a certain illness) with the number of separate remedies for this ailment. Compared with this, Fic gives the relationship between the ‘number of use-reports in each category (nur) minus the number of taxa used (nt)’ and the ‘number of use-reports in each category minus 1’. Fic is thus calculated using the following formula: nur ÿ nt Fic ˆ nur ÿ 1: The product of this factor ranges from 0 to 1. A high value (close to 1) indicates that relatively few taxa (usually species) are used by a large proportion of the healers, while a low value indicates that the informants disagree on the taxa to be used in the treatment within a category of illness. Botanical documentation Voucher specimens for all our studies are deposited at the Herbarium of the Universidad Nacional Autonoma de México (MEXU) and the Centre of Pharmacognosy and Phytotherapy, School of Pharmacy, University of London (UK). Collections from individual studies are at the Herbarium of the Instituto de Ecologı́a (XAL - Xalapa, Veracruz, Mexico) (Nahua collection), the Centro de Investigación Cientı́fica de Yucatán (CICY - Mérida, Yucatán, Mexico) (Maya collection), the Herbario Medicinal del Instituto Mexicano del Seguro Social (IMSS-M - México, D.F.) (Nahua and Mixe collections), the ETH Zürich (ZT - Zurich, Switzerland) (Maya and Zapotec collections). The collection numbers are A. Ankli 1-450, C. Weimann 1-324, B. Frei 1-550, M. Heinrich and N. Antonio B. 1-350. Plants were identified by comparison with authentic specimens and in some cases with the assistance of several specialists at the National Herbarium of Mexico. ETHNOBOTANICAL RESULTS Drug development Drug development in this context refers less to the design of new drugs for the European and North American market, but to the development of improved therapeutic options for the regions where these plants were originally used. It thus encompasses: (i) Preservation of indigenous knowledge; (ii) The selection of particularly useful species and the exclusion of toxic or ineffective ones; (iii) The dissemination of the ethnobotanical and phytochemical pharmacological results. While the first and the third point are clearly interrelated, they are distinct with respect to the methodologies used. In the first one the scientific study of the ethnobotanical Phytother. Res. 14, 479–488 (2000) ETHNOBOTANY AND DRUGS 483 Table 2. Comparison of Maya, Nahua and Zapotec medicinal plant use Number of taxa Number of use-reports Informants' consensus factor (Fic) Category of indigenous uses (group of illness) Maya Nahua Zapotec Maya Nahua Zapotec Maya Nahua Zapotec Gastrointestinal Dermatological Respiratory Gynaecological/andrological Culture-bound syndromes Pain/febrile diseases Fever (incl. malaria) Skeleto-muscular Ophthalmological Urological Poisonous animal bites Cardiovascular Other/unclassi®ed TOTAL 140 138 77 67 ± 105 ± ± 26 40 42 ± 60 320a 72 86 39 40 26 56 ± ± ± 34 ± ± 11 203a 176 205 88 122 144 ± 76 105 20 ± ± 20 ca. 100 445a 476 287 174 129 ± 204 ± ± 39 66 76 ± 101 1549 222 159 99 78 80 126 ± ± ± 67 ± ± 11b 816 518 605 303 364 563 ± 285 321 48 ± ± 52 n.a. 3611 0.71 0.52 0.56 0.48 ± 0.49 ± ± 0.34 0.40 0.45 ± 0.41 0.68 0.46 0.61 0.49 0.68 0.56 ± ± ± 0.50 ± ± 0.0 0.66 0.66 0.71 0.67 0.75 ± 0.73 0.68 0.60 ± ± 0.59 ± Fic ˆ nuseÿreports ÿntaxa (A higher value indicates a high rate of agreement between the informants, a low one a low degree of nuseÿreportsÿ1 agreement). n.a., not analysed. ± category absent (for details see text). a A taxon may be listed in several of the categories of indigenous uses. b Only diabetes. resources is at the centre, in the third interactive, developmental projects have to be initiated. In this contribution I will focus on the first two aspects. Medicinal plants in the four cultures The Maya, Mixe, Nahua and Zapotec frequently use plants in treatment. All four cultures heavily rely on their surrounding flora for this purpose (as well as for others such as food, nutrition, as toys for children and many others). While many of the plants are endemic or native to Southern Mexico, introduced plants also play an important role. One of the most popular plants is, for example, Ruta chalapensis (rue), which was brought to Mexico in colonial times and has come to be one of the highly appreciated medicinal plants. In the four regions plants are used by healers as well as by the general population. Contrary to other regions [e.g. Highland Chiapas and the Chorti region of Guatemala (Kufer et al., unpublished)] healers are essential in the four sociocultures in providing medicinal plants to the general public. Healers often have a series of plants they routinely use. Often these plants are those well known to have strong pharmacological effects (e.g. Datura stramonium in the case of the Mixe). No detailed description of the ethnobotanical results is feasible in this contribution (cf. Ankli et al., 1999 on the Maya, Heinrich, 1989 on the Mixe, Weimann and Heinrich, 1997 on the Nahua, Frei et al., 1998 on the Zapotec). Quantitative ethnobotanical evaluation Generally Fic is higher among the Zapotec (Table 2, last three columns). This indicates a more consistent use of the medical resources. A category of use that yielded a high factor of informant consensus was the culture-bound syndromes. For the Zapotec Fic is highest for this category (0.75), and among the Nahua it ranks first Copyright # 2000 John Wiley & Sons, Ltd. (0.68) together with gastrointestinal illnesses. These data indicate that a well-defined selection of species is culturally important for treating these illnesses. Among the Nahua the most important plant in the category ‘culture-bound syndromes’ is pennyroyal† (Satureja brownei), an aromatic species that accounts for 20% of all use-reports or 12 out of 60 (Weimann and Heinrich, 1997). Among the Zapotec nine species had ten or more use-reports, which accounts for 18.1% of all use-reports. All culturally important species of the Nahua are aromatic plants and rich in essential oils pointing to an important selection criterion of this ethnic group (Weimann and Heinrich, 1998). Mayan culture-bound syndromes do not form a separate category, since these illnesses are not clearly distinguishable from gastrointestinal (mal de ojo; cirro, cf. see Table 3) or skeletomuscular disorders or fever (mal viento). Also the factor is generally high in the category ‘gastrointestinal disorders’. The core taxa in this category are shown in Table 3 (see also below). Once again, aromatic plants and also astringent plants are employed by all three groups (cf. Ankli et al., 1999; Frei et al., 1998; Weimann and Heinrich, 1997) Fic is relatively low for dermatological illnesses among the Maya and Nahua indicating that there is a low consensus on the treatment of dermatological problems. This variability in use, although not quantified, was observed for the Mixe, too, and is due to experimentation (Heinrich, 1989). First, new and potential medical resources are applied topically. Then, depending on the result further use might be explored. This means that many plants initially used to treat dermatological problems are later no longer used because the plant is considered ineffective or toxic. The Zapotec, however, have a well-defined category of plants used in the treatment of dermatological conditions. The most popular † English names are provided for the convenience of the reader only. The native names can be found in our previous publications and on the voucher specimens. Phytother. Res. 14, 479–488 (2000) 484 M. HEINRICH species are Aloe (Aloe barbadensis) and Tournefortia densiflora (Frei et al., 1998). There also are instances of under-representing the consensus of the healers. Fic is, for example, low for the Maya in case of gynaecological/andrological disorders (Ankli et al., 1999). The most popular species in this category is bay cedar or caulote (Guazuma ulmifolia). It is mentioned 12 times, but is normally given in combination with other species. These species vary from informant to informant, and since these taxa are listed individually, the total number increases. Another problem are the non-identified species, since these are listed separately and thus the total number of taxa increases significantly. In the group gynaecological/andrological disorders of the collection ‘Maya’ 29 of 67 taxa are included for which information on uses is available but which have not yet been identified. Generally speaking, the factor Fic is higher among the Maya and Nahua for the larger categories of use, indicating a higher informant consensus. An increased number of use-reports does not result in significantly more species being added to a category. Thus the factor attests that the number of taxa used medicinally is limited and that only a certain percentage of the total flora is used (cf. Moerman, 1996). Since no information on the total number of species in the regions is available a direct comparison is not possible. The relatively high value of Fic in many groups of uses indicates that the ethnobotanical sample is large enough to identify plants that are culturally important and that may be of relevance for detailed phytochemical and pharmacological studies (see Bork et al., 1997). In particular, the category ‘gastrointestinal illnesses’, which has a high Fic value, contains a number of potentially interesting plants. Some of these were studied by our group (e.g. Hör et al., 1995). On the other hand, due to the characteristics of culture-bound syndromes and the lack of adequate pharmacological models no attempts have been made to evaluate the claims for plants in this category using phytochemical or pharmacological methods. Compared with Trotter and Logan’s analysis (1986), it is noteworthy that several of their disease categories which have a high informant consensus value are gastrointestinal illnesses and problems, such as earache, eye irritation, insect bites, and burns treated with topical pharmaceutical preparations. Our data thus support the earlier findings of these authors and point to particularly relevant categories of uses. chalepensis) are employed by four of five ethnic groups. Species of six genera are used by three groups: locust berry tree or nanche (Byrsonima crassifolia), velvetleaf (Cissampelos pareira), sweet orange and related Citrus species (Citrus sinensis and spp.), cat mint (Lippia alba) as well as peppermint (Mentha X piperita and spp.). Eight closely related plants are utilised by any two indigenous groups (see Table 3). The parallel use of plant taxa may be due to: Principal species used to treat gastrointestinal and dermatological conditions As discussed above the focus of this paper is on the development of improved therapeutic options in the regions of ethnobotanical study. The following examples highlight some research, which—as we hope—has contributed to this process. Most of our own research has focused on gathering data, which will allow the evaluation of indigenous medicinal plant uses. After documenting the indigenous uses we selected species, which seemed to be of particular interest, for further ethnopharmacological evaluation and phytochemical study. Plants were selected taking into consideration the following other criteria: In this section three different categories of indigenous use will be discussed: gastrointestinal, respiratory and dermatological. The data concerning those plants that were principally used are presented in Tables 3 and 4, which also contain comparative data for the Mixe and (in Table 3) for the Tzeltal and Tzotzil (Berlin and Berlin 1996). A large number of medicinal plant species are used by two or more cultures. For example, guava (Psidium guayava and other species of this genus) and American wormseed [(Teloxys ambrosioides (syn.: Chenopodium ambrosioides)] are shared by five groups; black sage (Artemisia ludoviciana ssp. mexicana) and ruda (Ruta Copyright # 2000 John Wiley & Sons, Ltd. * * * Coincidence (a random selection of similar species), Similar criteria for selecting plants (see Heinrich, 1998), Shared information on the potential usefulness of a plant (i.e. information on the use of a plant is diffused in various regions). Sharing information is probably responsible for the parallel usage of both Psidium guajava and Teloxys ambrosioides, which are known to mestizo groups in Mexico and are widely distributed as fruit tree and common weed, respectively. Whatever the reason for parallel usage, these plants are of particular interest for phytomedical and health care research. Regarding the category of dermatological diseases (Table 4), only one species (Mexican ‘Arnica’ or Mexican ‘sunflower’ - Tithonia diversifolia) is commonly used by three of the four groups and is a rarely used medicinal plant with the fourth group—the Maya. No species was common to all groups (Table 4). T. diversifolia is also known from various regions of Veracruz for treating dermatological conditions. It is native to the lowlands of southeastern Mexico and Central America but it is not well known in the ethnobotanical literature (Heinrich et al. 1998a). Although a tall shrub (2–3 m), the plant is often referred to as ‘arnica’ because the conspicuous yellow flower heads resemble the European Arnica montana. Bork et al. (1997) have shown that the leaf extract acts as a potent inhibitor of an inflammatory transcription factor. The widespread use of this plant in Mexico is presumably due to its pharmacological effect and its superficial similarity to European arnica, which European settlers might have used as an explanatory model for using this plant. ETHNOPHARMACOLOGICAL EVALUATION OF INDIGENOUS USES * * The species is native to the area The species or closely related ones have been little studied phytochemically and/or pharmacologically Phytother. Res. 14, 479–488 (2000) Copyright # 2000 John Wiley & Sons, Ltd. Table 3. Principal species used to treat gastrointestinal illnesses among the Maya, Nahua and Zapoteca as well as comparative data from Tzeltal/Tzotzil and Mixe Maya (YucataÂn) Nahua Zapotec Tzeltal/Tzotzilb Mixeb Tagetes erectac (11) Marrubium vulgare (10) Ruta chalepensise (10) Psidium guajavaf (10) Ruta chalepensise (10) Teloxys ambrosioidesf (10) Baccharis serraefoliac Baccharis trinerveisc Baccaris vaccinoidesc Aristolochia maxima (11) Lippia stoechadifoliaa (11) Lippia albad (9) Matricaria recutitac (8) Byrsonima crassifoliad Teloxys ambrosioidesf Teloxys ambrosioidesf (10) Artemisia ludoviciana spp. mexicanae (7) Psidium guajavaf (7) Verbena menthaefoliac (7) Artemisia absinthiumc (6) Baccharis confertac (6) Psidium salutaref (8) Artemisia ludovicianae ssp. mexicanae (7) Pluchea symphytifoliae (7) Anethum graveolensc Artemisia absinthiumc Artemisia ludoviciana ssp. mexicanae Byrsonima crassifoliad Teloxys ambrosioidesf Cissampelos pareirad Cissampelos pareirad Citrus limond (7) Anethum graveolensc (6) Annona reticulata (6) Byrsonima crassifolidd (6) Psidium guajavaf Psidium guineensef Tagetes ®lifoliac Tagetes lucidac Citrus limond Guazuma ulmifoliac Lippia albad Matricaria recutitae Cocos nucifera (6) Guazuma ulmifoliac (6) Equisetum sp. (6) Juliana adstringens (6) Mentha x piperitad (6) Psidium x hypoglaucumf (6) Zea mays (6) Verbena carolinac Verbena litoralisc Mentha x piperitad Pluchea symphytifoliac Psidium guajavaf Psidium guineensef Ruta chalepensise Erigeron karvinskianus (6) Teloxys ambrosioidese (6) Rosa chinensis (6) Phyla scarberrima (6) a Only the species mentioned six or more times within one culture are listed. Data based on Berlin und Berlin (1996:83), total number of principal species used in the treatment of gastrointestinal illnesses:38 (Tzeltal/Tzotzil) and from Heinrich (1989:36±38) (Mixe); only taxa, which were documented as important medicinal plants in at least one of the studies with the Maya, Nahua or Zapotecs (28, 12 and 15, respectively) are listed. c Parallel reports of the species (or the genus) with two indigenous groups. d Parallel reports of the species (or the genus) with three indigenous groups. e Parallel reports of the species (or the genus) with four indigenous groups. f Parallel reports of the species (or the genus) with ®ve indigenous groups. b 485 Phytother. Res. 14, 479–488 (2000) Dorstenia contrajerva (10) Mentha aff. citratad (10) Psidium guajavaf (10) Artemisia ludoviciana ssp. mexicanae (9) Callicarpa acuminata (9) Cissampelos pareirad (9) Citrus aurantiumd (9) Lippia alba (9) Malvaviscus arboreus (9) Ruta chalepensise (9) Bidens squarrosa (8) Cissus trifoliata (8) Citrus sinensisd (8) Triumfetta semitrilobata (8) Zingiber of®cinale (8) Mentha aff. arvensisd (7) Ocinum micranthum (7) Punica granatum (7) Citrus aff. aurantifolidd (6) Hylocereus undatus (6) Microgramma nitida (6) Piscidia piscipula (6) ETHNOBOTANY AND DRUGS Mentha aff. piperitad (18) Abrus precatorius (12) Manilkara zapota (12) 486 M. HEINRICH Table 4. Principal species used to treat dermatological problems among the Maya, Nahua and Zapotec and comparative data from Mixe* Maya Samolus ebrecteatus (7) Anredera vesicaria (6) Calea urticifolia (6) Diospyros anisandra (6) Kalanchoe integra (6) Ocimum micranthum (6) Psidium sartorianum (6) Salvia micrantha (6) Nahua Phyllanthus niruri (6) Lobelia laxi¯ora (5) Sida rhombifoliab (5) Heterotheca inuloides (5) Anagallis arvensis (4) Bryophyllum calycinum (4) Buddleja cordata (4) Mecardonia procumbens (4) Stachys sp. (4) Stellaria nemorum (4) Tithonia diversifoliac (4) Zapoteken Aloe barbadensisb (13) Tounefortia densi¯ora (12) Piper auritumb (9) Piper tuberculatum (7) Tithonia diversifoliac (7) Capraria bi¯orab (6) Comoclea engleriana (6) Hamelia patens (6) Hyptis verticillatab (6) Jatropha curcasb (6) Pinus oocarpa (6) Solanum torvum (6) Swietina humilis (6) Thevetia thevetioides (6) Zebrina pendula (6) Mixe Aloe barbadensisb Capraria bi¯orab Hyptis verticillatab Jatropha curcasb Piper auritumb Sida rhombifoliab Tithonia diversifoliac * Species mentioned six or more times (Nahua 4four times). Principal species for the Mixe based on Heinrich (1989:46±49); and includes only taxa documented as important medicinal plants for at least one of the other studies. b Parallel reports of the species (or the genus) with two indigenous groups. c Parallel reports of the species (or the genus) with three indigenous groups. a * * * It is used systemically (i.e. orally, rectally) or in case of dermatological conditions topically It is not an endangered species and It is possible to collect sufficient plant material for study. As a result of these studies we were able to evaluate a total of approximately 150 species in various pharmacological test systems. Only a handful was studied phytochemically in detail. NF-kB as a molecular target One target used in these studies is the inducible transcription factor NF-kB. It is an important mediator of immune and inflammatory response and is induced by many pro-inflammatory stimuli including tumour necrosis factor (TNF), the phorbol ester phorbol 12-myristate 13 acetate (PMA), and oxidants such as H2O2. These inducers trigger signalling cascades that lead to the induced degradation of the inhibitory IkB molecule, resulting in the translocation of NF-kB into the nucleus and the induced expression of numerous target genes coding for inflammatory cytokines such as IL-6, cell adhesion molecules, immunoreceptors, haematopoietic growth factors, acute phase proteins and further transcription factors. Therefore, activation pathways of NFkB are frequent targets for antiinflammatory substances, many of them being antioxidants. Also established antiinflammatory agents such as glucocorticoids and acetyl salicylic acid are known to be inhibitors of NF-kB (Schmitz et al., 1998; Dumont et al., 1998; Renard and Raes, 1999). Electrophoretic mobility shift assays (EMSA) were used for detecting the inhibitory activity of plant extracts and pure compounds on NF-kB and are described in detail in Bork et al. (1997). HeLa (ATCC CCL2) or TC10 (a mouse endothelial cell line) cells were grown Copyright # 2000 John Wiley & Sons, Ltd. overnight, exposed to the pure compounds or the plant extracts and then stimulated with phorbol 12-myristate 13-acetate (PMA) or TNF-a. The cells were washed, harvested, lysed, and the supernatant protein was tested for DNA binding activity of NF-kB by EMSAs. Plants as inhibitors of NF-kB In the course of our studies we were able to screen all taxa collected in the four regions. Extracts of a total of 150 species from the four ethnobotanical collections were screened for inhibitory activity on NF-kB using EMSA experiments. Particularly noteworthy was the activity of four species from the Asteraceae: Artemisa ludoviciana ssp. mexicana; Calea zacatechichi; Polymnia maculata and Tithonia diversifolia. All four species are used for illnesses associated with chronic or acute forms of inflammation. Tithonia diversifolia is particularly interesting. It is used by all four of the indigenous groups we worked with and is a prominent species with the Mixe, Nahua and Zapotec. Particularly prominent are uses for inflammatory skin conditions like infected wounds, topical inflammations and bruises. All four species are well known to contain sesquiterpene lactones and it was also noteworthy that members of the Asteraceae, which are known not to contain large amounts of sesquiterpene lactones, were inactive. Consequently we tested several commercially available pure sesquiterpene lactones like parthenolide (Fig. 1) and isohelenin (Bork et al., 1997). These data were later corroborated by Rungeler et al. (1998) who isolated a series of sesquiterpene lactones from a Costa Rican collection of T. diversifolia and showed that they inhibit NF-kB activation. The physiological relevance of the inhibition was demonstrated by our group in a reporter gene assay using a construct consisting of a luciferase gene cloned downstream from an IL-6 Phytother. Res. 14, 479–488 (2000) ETHNOBOTANY AND DRUGS 487 Mixe) and of various kidney diseases (Caribbean). The ethanol extract showed an in vitro inhibitory activity on cyclooxygenase (COX 1) and in vivo activity in the HETCAM assay. Using a bioassay-guided fractionation procedure with COX 1 activity as a lead resulted in the isolation of linoleic acid and a-linolenic acid. Additionally the cation content of the water fraction was determined. The contents of Na, K, Ca and Mg represents roughly 25% of the water fraction or 13% of the crude ethanol extract. The high content of K (7.7% of the crude extract) and the low content of Na (0.5 %) are notable. It is well known that medicinal plants, which are used as diuretics often have a high content of K‡ (e.g. Betula spp. or Orthosiphon aristatus with 3% K‡) and are known as saluretics. Obviously pharmacological and/or clinical studies and studies using an aqueous infusion will be necessary in order to evaluate the indigenous claims, but the phytochemical data corroborate the indigenous uses in Mexico and the Caribbean (Heinrich et al., 1998c). Figure 2. Examples of plant natural products with inhibitory activity on the transcription factor NF-kB: parthenolide and phaeophorbide A. promotor. Artemisia ludoviciana ssp. mexicana as well as the pure sesquiterpene lactones isohelenin and parthenolide were shown to down regulate the transcription of the reporter gene (Bork et al., 1997). Solanum diflorum Vell. (Solanaceae) is used in the treatment of erysipelas, local swellings, oedema and fever by the Istmo Sierra Zapotec Indians of Oaxaca, Mexico. The ethanolic crude extract of the leaves showed strong inhibitory activity on NF-kB activation in EMSA experiments at 100 mg/mL using the protocol described previously (see above, Bork et al., 1997). Similarly, another member of this genus, Solanum lancoelatum Cav., used to treat local infections, wounds and ‘espinilla’ showed weak inhibitory activity at the same concentration. The crude extract (100 mg/mL) was not cytotoxic even after 24 h of incubation. It became obvious very early on that the active substance seems to be a degradation product of chlorophyll. Pheophorbide A (Fig. 2) was identified as one of the key compounds responsible for the NF-kB inhibitory activity. The compound interferes with NF-kB activation, was cytotoxic if exposed to light, but devoid of any cytotoxic activity in the dark. These data provide some confirmation for the topical use of the leaves of Solanum diflorum in the treatment of inflammatory skin conditions as is reported in the ethnobotanical literature. Phototherapy using photosensitizers is currently discussed as a therapeutic option, for example, in the treatment of cancers (Heinrich et al., 2000; Sharman et al., 1999). The options for the development of such therapies will have to be investigated further. CONCLUSION This paper gives an overview of some of our research in the last years. It clearly demonstrates the value of integrating ethnobotanical with phytochemical and pharmacological–phytochemical studies. While we did not develop new therapeutic agents for use in biomedicine, we were able to demonstrate the value of some of the plants. Other plants were shown to have too many side effects or are highly toxic (see Heinrich, 1989). In an example from the Highlands of Mexico, Bah et al. (1994) showed that a species popularly used there contains hepatotoxic pyrrolizidine alkaloids. Such plants are thus potential health risks. While this information is often available to the scientific community, the general public mostly is not aware of these risks. Such data have to be summarized in an appropriate way and have to be made available to the people of the respective regions, where the plants are used. It will now be essential to develop partnerships with institutions, which can translate these results into an effective strategy. Our studies also lead to the identification of a new class of NF-kB inhibitors, simultaneously pointing to a biochemical mechanism of this group of compounds, which corroborates the in vivo experiments on sesquiterpene lactones. The studies thus contribute to a better understanding of the mode of action of an important group of natural products. The studies on Solanum diflorum and Peperomia pellucida are typical examples of a phytochemical study, which led to the identification of compounds, which are therapeutically of little relevance. This example should caution us in our expectations on the use of such data in the development of new medications. Acknowledgements Peperomia pellucida Another example is Peperomia pellucida Kunth (Piperaceae). The crushed aerial parts of Tsuxk Huaj Aptx (the herb of the green cow) are popularly used in the treatment of inflammatory dermatological problems (Lowland Copyright # 2000 John Wiley & Sons, Ltd. The ethnobotanical part of this research would not have been possible without the help of many people in Mexico, especially the midwives, healers and specialists in medicinal plants. These persons are the original keepers of the ethnobotanical knowledge presented here. I am very grateful to all former PhD students in Freiburg and to Professor Rimpler. The data discussed in this paper are part of the PhD theses of C. Weimann and Dr Peter Bork, as well as the MSc Phytother. Res. 14, 479–488 (2000) 488 M. HEINRICH thesis of Inga Koehler. I am also very grateful to Dr A. Ankli, Dr B. Frei, Mr M. Leonti and Professor O. Sticher (all Zurich, CH) for many years of fruitful collaboration on ethnobotany and phytochemistry. The research on plant derived NF-kB inhibitors would not have been possible without Professor M. L. Schmitz (Venceremos), Dr Bacher and their teams in Freiburg and Heidelberg (D). The botanical identification was performed at the Herbario Nacional de México (MEXU) and also at the Herbarium of the Instituto de Ecologı́a (XAL) in Xalapa, Veracruz and the Centro de Investigaciones Cientı́ficas de Yucatán (CICY), Mérida. The new team at the ‘Square’ made the transition to the new environment as easy as possible. Financial support by the SDC (Swiss Agency for Development and Cooperation, Berne, Switzerland) and the SANW (Swiss Academy of Natural Sciences) and the Deutsche Akademische Austauschdienst (Bonn, FRG), the Secretarı́a de Relaciones Exteriores (México, D.F., México), and the Freiburger Wissenschaftliche Gesellschaft (Freiburg, FRG) to myself or to the respective PhD students is gratefully acknowledged. REFERENCES Ankli A, Sticher O, Heinrich M. 1999. Medical ethnobotany of the Yucatec Maya: healers' consensus as a quantitative criterion. Econ Bot 53: 144±160. Bah M, Bye R, Pereda-Miranda R. 1994. Hepatotoxic pyrrolizidine alkaloids in the Mexican medicinal plant Packera candidissima (Asteraceae: Senecioneae). J Ethnopharmacol 43: 19±30. Bork PM, Schmitz ML, Kuhnt M, Escher C, Heinrich M. 1997. Sesquiterpene lactone containing Mexican Indian medicinal plants and pure sesquiterpene lactones as potent inhibitors of transcription factor B (NF-B). FEBS Lett 402: 85±90. Brett J, Heinrich M. 1998. Culture perception and the environment. J Appl Bot 72: 67±69. Dumont A, Hehner SP, Schmitz ML et al. 1998. Cross-talk between steroids and NF-kappa B: what language? Trends Biochem Sci 23: 233±235. Frei B, Sticher O, Heinrich M. 1998. Medical ethnobotany of the Zapotecs of the Isthmus-Sierra (Oaxaca, Mexico): Documentation and assessment of indigenous uses. J Ethnophamacol 62: 149±165. Heinrich M. 1989. Ethnobotanik der Tie¯andmixe (Oaxaca, Mexico) und phytochemische Untersuchung von Capraria bi¯ora L. (Scrophulariaceae). Dissertationes Botanicae No. 144. Berlin und Stuttgart. J. Cramer in Gebr. Borntraeger Verlagsbuchhdlg. Heinrich M. 1997. Herbal and symbolic forms of treatment of the lowland Mixe (Oaxaca, Mexico). In The Anthropology of Medicine. From Culture to Method. Romanucci-Ross L, Moerman DE, Tancredi LR (eds). Bergin and Garvey: Westport; 71±95. Heinrich M. 1998. Plants as antidiarrhoeals in medicine and diet. In Plants for Food and Medicine. Etkin NL, Harris DR, Prendergast HDV, Houghton PJ (eds). Royal Botanic Gardens: Kew; 17±30. Heinrich M, Rimpler H, Antonio B. 1992. Indigenous Phytotherapy of Gastrointestinal Disorders in a Mixe Lowland Community. J Ethnopharmacol 36: 63±80. Heinrich M, Ankli A, Frei B, Weimann C, Sticher O. 1998a. Medicinal plants in Mexico: healers' consensus and cultural importance. Social Sci Med 47: 1863±1875. Heinrich M, Robles West JE, Ortiz de Montellano BR, Rodriguez E. 1998b. Ethnopharmacology of Mexican Asteraceae (Compositae). Ann Rev Pharmaco Toxicol 38: 539±565. Heinrich M, Koehler I, Rimpler H, Bauer R. 1998c. Bioactive compounds from the Mixe Indian medicinal plant Peperomia pellucida. Soc Quimica de MeÂxico Rev 42 (6): 245±248. Heinrich M, Bork PM, Schmitz ML, Rimpler H, Frei B, Sticher Copyright # 2000 John Wiley & Sons, Ltd. O. 2000. Phaeophorbid A from Solanum di¯orum interferes with NF-kB activation. Planta Med in press HoÈr M, Rimpler H, Heinrich M. 1995. Inhibition of intestinal chloride secretion by proanthocyanidins from Guazuma ulmifolia. Planta Med 61: 208±212. Humboldt A von. (Hrsg. H. Beck). 1997. Die Forschungsreise in den Tropen Amerikas. Studienausgabe Bd. 2, Teilband 3. Darmstadt. Wissenschaftliche Buchgesellschaft. Moerman DE. 1996. An analysis of the food plants and drug plants of Native North America. J Ethnopharmacol 52: 1± 22. Red®eld R, Villa Rojas A. 1934. Chan Kom: A Maya Village. Carnegie Institute of Washington: Washington D.C., Publication No 348. Renard P, Raes M. 1999. The proin¯ammatory transcription factor NF-B: a potential target for novel therapeutical strategies. Cell Biol Toxicol 15: 341±344. Rungeler P, Lyss G, Castro V, Mora G, Pahl H, Merfort I. 1998. Study of three sesquiterpene lactones from Tithonia diversifolia on their anti-in¯ammatory activity using the transcription factor NF-B and enzymes of the arachidonic acid pathway as targets. Planta Med 64: 588±593. Schmitz ML, Hehner SP, Bacher S, DroÈge W, Heinrich M. 1998. Hemmung der entzuÈndlichen Immunantwort: Aktivierungswege des Transkriptionsfaktors NF-kB als Target fuÈr biogene und semisynthetische anti-in¯ammatorische Substanzen. Deutsche Apothekerzeitung 138: 4881±4891. Schultes RE. 1962. The role of the ethnobotanist in the search for new medicinal plants. Lloydia 25: 257±266. Schultes RE. 1983. Richard Spruce: an early ethnobotanist and explorer of the Northwest Amazon and Northern Andes. J Ethnobiol 3: 139±147. Schultes RE, Nemry von Thenen de Jaramillo-Arango MJ. 1998. The Journals of HipoÂlito Ruiz: Spanish Botanist in Peru and Chile 1777 ±1788. Timber Press: Portland (OR). Schultes RE, Raffauf RF. 1991. The Healing Forest. Dioscorides Pr: Portland, OR. Sharman W, Allen CM, Lier van JE. 1999. Photodynamic therapeutics. Drug Devel Today 4: 507±517. Trotter R, Logan M. 1986. Informant consensus: A new approach for identifying potentially effective medicinal plants. In Plants in Indigenous Medicine and Diet: Biobehavioural Approaches. Nina L Etkin (ed.). Redgrave Publishers: Bedford Hills, NY; 91±112. Weimann C, Heinrich M. 1997. Indigenous medicinal plants in Mexico: The example of the Nahua (Sierra de Zonglicia). Botan Acta 110: 62±72. Weimann C, Heinrich M. 1998. Concepts of medicinal plants among the Nahua of the Sierra de Zongolica, Veracruz (Mexico). J Appl Bot 72: 87±91. Phytother. Res. 14, 479–488 (2000)