Quantitative Chemobiology: A Guide into the Understanding of Plant Bioactivity

Hoje, nada e mais importante para a sobrevivencia humana do que compreender os mecanismos da natureza atraves de uma linguagem quimico-biologica. Essa abordagem multidisciplinar e uma complexa operacao, pois envolve a integracao de varios niveis de organizacao, tais como quimica, morfologia e ecogeografia, expressos respectivamente pela diversificacao de metabolitos, formas e ambientes. A comparacao entre essas diferentes expressoes da vida, apesar de sua importância incontestavel, ainda permanece um arduo topico de pesquisa. Aplicacao dessa abordagem unificada poderia revigorar o estudo de um assunto antigo e controvertido, a bioatividade vegetal. Enfrentar o maior desafio desse proposito: confrontar o conhecimento tradicional com uma metodologia cientifica, requer a determinacao de tendencias entre usos de especies de angiospermas, independentemente de empirismos e regionalismos. Assim, incorporacao de novos codigos, expressando funcoes biologicas, na linguagem quimico-biologica e possivel somente atraves de conceitos e padroes evolutivos.


Introduction
A considerable effort during several decades produced only modest knowledge on the chemical composition of the Brazilian flora. 1 The fact is even less easy to understand, if it is considered that the value of medicinal products derived from tropical plants surpassed in recent times US$ 6 billion per annum. 2 Should it really be impossible to find ways and means to recover this enormous wealth in adequate time, or at least to design suitable process with such objectiveness?To face such problems is rather urgent.Already drug discovery is going down and disease resistance is going up.Today, more than 30,000 diseases are clinically described.Less than one third of these can be treated symptomatically, and only a few can be cured. 3eanwhile some pharmaceutical concerns continue to announce great expectations of "miraculous cures" based on medicinal plants.Others are looking for natural products of novel molecular skeletons from a variety of marine organisms.However, as yet, no compounds from the sea have advanced to commercial use as a chemotherapeutic agent. 4On the other hand, the quantitative structureactivity relationships (QSAR) approach has demonstrated great possibilities in drug-designs and drug-target interactions.Finally, synthetic, specifically projected, DNA-based wonder drugs are expected to alleviate human suffering in the future.
Thus, the first problem leads us to a second reasonable challenge: Is the study of medicinal plants really subject to scientific control?Brazil, the country of contrasts and paradoxes, is ideally suited to meet this biological challenge!The more than 8.5 million km 2 shelter an enormous biological and cultural wealth.Besides big industrial populations there exist small native tribes that remain in their original state without any contact with the rest of the country, such as the Korubos tribe on the margins of Ituí river, Javari Valley, Amazonas, frontiers of Peru and Colombia. 5Out of 210 Brazilian indigenous populations, 55 live in similar social isolation since the time of colonization, 500 years before present.
And furthermore, should it also be possible to integrate traditional knowledge accumulated by generations in a scientific framework?If so, it should be possible not only to evaluate popular information scientifically, and also to predict bioactivity via a closer understanding of regulatory mechanisms of natural products.Without doubt, the first step in this endeavor is to assemble the available information for further analysis and rationalization."At any rate, the role of science is not to establish some kind of factual data-bank about nature, but to help us understand nature". 6

Result and Discussion
Ethnobotany: Evolutionary patterns for useful plants Regional ethnobotanical inventories.In a first essay we introduced as database per regional ethnobotanical inventories of plants, utilized by three indigenous societies, living in different parts of Amazonia: Chácobo from Bolívia, 7 Kayapó from Xingú 8 and Ka'apor from Maranhão. 9The correlation among the frequency of useful plants (in percentages) per superorders (sensu Dahlgren 10 ) with an evolutionary parameter based on herbaceousness indices (HI) reveals a common evolutionary trend: while more primitive plant species (lower HI) are used as foods, more recently evolved plant species (higher HI) are selected as medicines. 11,12he same conclusions arise when we analyze other inventories of useful plants, confirming the existence of general patterns.This discovery is surprising, if we consider that in this case non-human primates, represented by cebuella from Amazonia, 13 spider monkey from Pará and Guyana 14 and muriqui from Atlantic forest, 15,16 possess "knowledge" about useful plants similar to the human populations. 17Thus, who learned what from whom?
Extensive ethnobotanical inventory.In a further attempt to confirm the universality of the previous results, we analyzed a vast ethnobotanical survey elaborated during the first years of the 20th century, chiefly in Brazil. 18To unify different databases, as the three regional indigenous surveys and this extensive Brazilian dictionary, it was necessary initially to develop an appropriate methodology.In short, the quantitative method for the determination of chemo-biological patterns implied the following stages: i) selection of dicotyledon species to which useful (edible and medicinal) properties had been assigned in these reports; ii) classification of these species at the level of families according to a system of classification (e.g.Dahlgren's system 10 ); iii) characterization of each listed dicotyledon family by an ethnobotanical parameter (according to frequency of their useful species) and by an evolutionary parameter [according to Sporne indices (SI) 19 ]; iv) arrangement of these families according to their evolutionary status (i.e. by SI); v) Determination of the "evolutionary spectrum" for the ethnobotanical survey, considering the total frequency of useful species (in percentages) for each SI (cumulative frequencies).
This procedure allowed the characterization of each dicotyledon family by three types of information: systematic (according to Dahlgren's system 10 ), evolutionary (according to Sporne indices, SI 19 ) and ethnobotanical (according to percentage of useful species).Juxtaposing of these informations showed extraordinary consistencies.Frequently, the identical preferential (edible and/or medicinal) use of orders included in the same superorder was observed.This regularity of ethnobotanical indication was more common in medicinal orders of higher evolutionary status (Asteridae sensu Cronquist 20 ) than in edible orders of intermediate evolutionary status (Hamamelidade-Dilleniidae-Rosidae complex sensu Cronquist 20 ). 21Practically the same trends of use for a vast survey 18 and for three regionally and ethnically restricted Amazonian inventories (Chácobos, 7 Kayapós 8 and Ka'apors 9 ) were observed. 22ther ethnobotanical inventories.Additionally to these four ethnobotanical surveys, eight inventories of different regions and continents were selected: three of food plants (one from Africa, 23 one from North America 24 and one from Brazilian Amazonia 25 ); and five of medicinal plants (one from Africa, 26 one from North America, 27 two from Brazilian Amazonia, 28,29 and one from Peruvian Amazonia 30 ).Many features distinguish each one of these human groups, from the environmental point of view, as available resources, type of vegetation, climate and soil, to the sociocultural point of view, as management strategies, nutritional customs, inheritance of the traditional use of plants, illness characterizations and even mystic beliefs.
In order to compare these different databases, we selected in each inventory the dicotyledon groups, arranged according to their evolutionary status (SI), with more than 5% of the total food species (Table 1) and medicinal species (Table 2).This procedure allowed the elimination of the individual characteristics, i.e. the noises.While food species predominated in dicotyledon groups characterized by Sporne indices of 45, 48 and 57 (Table 1), medicinal species predominated in groups with Sporne indices of 37, 48, 57 and 72 (Table 2) in practically all inventories.Confirming the previous results, food species are limited by intermediate evolutionary status (SI 45 -57), only rarely attaining higher or lower SI values.Exceptions (SI 72) for Africa may indicate lack of foods; and for North America, where species utilized as beverage and other stimulants are considered additional food plants.In opposition, medicinal species occupy a large SI range (SI 37 -72), often present at high evolutionary status (SI 72, 80).Thus, similar systematic and evolutionary patterns should be observed for all inventories analysed, independently of their dimensions and/or geographic localization. 31

Phytochemistry: Regulatory mechanisms of plant bioactivity
The next, and very important, challenge consisted to look for chemical mechanism responsible for plant bioactivity.However, this task is very difficult, and until now without significative results, due to the many factors involved in the production, expression and regulation of natural products.An introspection in this subject, required the application of the same methodology above described, to phytochemical data (represented by number of occurrences of natural compounds).This procedure allowed the determination of evolutionary chemical patterns comparable to the evolutionary ethnobotanical patterns obtained.][34] The "spectral" features of the gallic-model were similar to the "spectral" features of the ethnobotany-guided food plants.In contradistinction, the analogous features of the caffeic-model were similar to the features of the ethnobotany-guided medicine plants.Indeed, the maxima of food species for practically all inventories corresponding to families with SI = 45, 48 and 57 (Table 1) corresponded to the maxima for gallic acid.Analogously, the maximum at SI = 72 was relative to medicine plants (Table 2) and caffeic acids. 31

Conclusion
Now we are apt to face the challenge question announced in the introduction: How should we proceed to integrate traditional knowledge and scientific endeavour into a single framework?Trends among the uses of angiosperm species must possess systematic and evolutionary relevance in order to promote ethnobotanical descriptions to valid quantitative evolutionary codes.Thus, incorporation of new codes, expressing biological functions, in the chemo-biological language becomes possible only through evolutionary concepts and patterns.Indeed, remember Dobzhansky: "Nothing in biology makes sense", and from our point of view not even integration of ethnobotany and science makes sense, "except in the light of evolution". 35inally, the complementary nature of food and medicine species can be rationalized by the involvement of plant metabolic cycles regulated via gallate/caffeate feedback loops. 31Hopefully this contribution should help to clarify the chemical and pharmacological potentialities of this enormous wealth in adequate time.This original procedure reveals the flexibility and importance of a dynamic and holistic approach into a crucial matter, an introspection of nature's functioning.
"We are the generation for whom the only message for a tropical biologist is: Set aside your random research and devote your life to activities that will bring the world to understand that tropical nature is an integral part of human life.If our generation does not do it, it won't be there for the next."(D. H. Janzen, 1986)

Table 1 .
Values of Sporne indices (SI) for groups of dicotyledon families with more than 5% of food species cited in different ethnobotanical inventories