Revisiting the linkage between ethnomedical use and development of new medicines : A novel plant collection strategy towards the discovery of anticancer agents

1 Department of Medicinal Chemistry and Pharmacognosy, University of Illinois at Chicago, 833 S. Wood St., Chicago, Illinois 60612, USA. 2 Institute of Traditional Medicine, Ministry of Health, Vientiane Capital, Lao People's Democratic Republic. 3 Division of Medicinal Chemistry and Pharmacognosy, College of Pharmacy, Ohio State University, 500 W. 12th Ave., Columbus, OH 43210, USA. 4 Science and Education, Field Museum, 1400 S. Lake Shore Drive, Chicago, Illinois 60605, USA.


INTRODUCTION
In trusting human agency and the intentionality of traditional ecological knowledge, one may be liable to conclude that botanical ethnopharmacopeias consist of plants that: (a) bear useful pharmacological activities and can improve health status and (b) can be administered to patients in ways that render them clinically safe within reasons (Fadeyi et al., 2013;Pan et al., 2013;Getasetegn and Teferi, 2016).Given these two likelihoods, these ethnopharmacopeias can be studied in their entirety through biological screening efforts (Mazzio and Soliman, 2009;Fadeyi et al., 2013;Leonti and Weckerle, 2015;Odonne et al., 2017), setting the stage in the drug discovery pipeline towards human health applications not necessarily anticipated by indigenous knowledge and folk clinical application of the particular plants evaluated in this way.It might even be expected that this strategy could yield results superior to "random" collection efforts constrained to geographic areas with similar levels of species richness and biodiversity (Bletter, 2007;Saslis-Lagoudakis et al., 2012).Prior metaanalysis suggests that on a per sample basis, depending on ethnobotanical use and screening assays performed for samples evaluated from Laos and Vietnam, plants employed in traditional medicine can have a higher hit rate for bioactivity in empirical studies (Gyllenhaal et al., 2012).Although there are caveats and nuances to this finding, which will be discussed subsequently, this insight is worth bearing in mind in the context of the present paper.The hypothesis of this study, accordingly, states that the agentive, purposive nature of botanical ethnopharmacopeias biases sample sets derived from them for useful bioactivity; this selection criterion therefore lends itself to success in initial biological screening and drug discovery studies.
Towards the end of further, and unambiguous exploration of the possibility for ethnobotanically driven screening efforts, in this instance for the preliminary stages of anticancer drug discovery, two of the most extensively inventoried preserves in the Lao PDR, the Xiengkhouang MBP and the Bolikhamxay MPP were selected as promising expedition sites for the extramurally funded Program Project (P01) from among the ten preserves that are extant (Sydara et al., 2014;Soejarto et al., 2015) (Figure 1).These reservoirs for local traditional medicine plants served as the premiere P01 project sites for the exploration of the ethnopharmacopeias of Laos through the lens of this pragmatic, serendipitous-activity-through-human-utility paradigm.This paper presents the results of this endeavor.

Memorandum of agreement
A Memorandum of Agreement (MOA) for the conduct of collaborative research targeted to the flowering plants, between the University of Illinois at Chicago and the Institute of Traditional Medicine (ITM, Vientiane, Lao PDR), covering issues on intellectual property and the sharing of benefits in the event of the discovery and development of a pharmaceutical product was established.This MOA allowed for the collection of plant materials (plant samples and their voucher herbarium specimens) in Laos and their transfer to and biological evaluation in the USA.

Plant collection
Following the signing of this agreement, a joint plant collection expedition between the University of Illinois at Chicago (UIC) and the Institute of Traditional Medicine (ITM) was undertaken in the Lao PDR.One expedition site was the Xiengkhouang Medicinal Biodiversity Preserve (MBP) of the Kham District, Xiengkhouang Province, and the other was Bolikhamxay Medicinal Plant Preserve (MPP) of the Paksan District, Bolikhamxay Province (Figure 1).

Xiengkhouang medicinal biodiversity preserve expedition
The Xiengkhouang Medicinal Biodiversity Preserve (MBP) (as depicted in Figure 1) is located at about 50 km northeast of the capital city (Phonsavanh) of Xiengkhouang Province, near Ban Tha, a rural Lao Lum (Lowland Lao) village in the Kham District, and about 15 km from both Muang Kham and Ban Tha.The elevation is approximately 1,140 m above sea level, at 19°43' N; 103°35' E (GPS reading).This preserve comprises approximately 500 hectares of high quality, secondary, montane tropical rainforest (Figure 2).
Two of the ITM's 5-passenger Toyota Hilux pickup trucks, with collapsible soft tops for covering the rear cargo area, provided excellent mobility throughout the expedition period (December 8 to 13, 2015).These vehicles allowed for the transportation of 6 to 10 passengers (including drivers) in addition to the loads of collected plant materials and field supplies.Two workers and several other locals were employed from the village of Ban Tha for the purposes of harvesting and processing plant material.
Near the Xiengkhouang MBP, quarters at the Seng Deuane guest house in Meuang Kham were rented as the expedition base and as equipment and supplies storage.The concrete patio of the guest house served as the base for drying plant samples and working quarters.The space and facilities of this guest house permitted the efficient performance and completion of the expedition within a short time period.The warm and mostly dry conditions in the winter season during the expedition eased the initial stages of processing of the voucher specimens and the plant samples collected.
Part of the supplies and equipment for use in the expedition, such as tarpaulins, rice sacks, nylon mesh bags, cardboard, plant presses, strings, branch cutters, twig clippers, knives (pointed knives known locally as mid [/miːd/]), a GPS, digital cameras, binoculars, used newspapers, and other plant collecting and processing supplies, was brought from Vientiane through the ITM's supplies or as part of the ITM and UIC personnel's belongings.Forays into the field and the search for the plants collected were guided by a list and photographs of plants previously collected (Soejarto et al., 2015) from this MBP.Plant samples and voucher herbarium specimens were collected, carefully numbered and documented by field notes and photographic images.Numbered herbarium specimens were pressed between newspapers in the Towards the end of further, and unambiguously, exploring the possibility for ethnobotanicallydriven screening efforts, in this instance for the preliminary stages of anticancer drug discovery, two of the most extensively inventoried preserves in the Lao PDR, the Xiengkhouang MBP and the Bolikhamxay MPP, were selected as promising expedition sites for our Program Project (P01) from among the ten preserves that are extant (Sydara et al., 2014;Soejarto et al., 2015) (Fig. 1).These reservoirs for local traditional medicine plants served as the premiere P01 project sites for the exploration of the ethnopharmacopeias of Laos through the lens of this pragmatic, serendipitousactivity-through-human-utility paradigm.This paper presents the results of this endeavor.(Soejarto et al., 2015), showing the location of expedition sites described in this paper.field.Screening samples, 1 to 3 kg fresh weight, depending on nature or fleshiness of the plant material, were packed in sample collection bags made of nylon mesh and were placed in the trucks.On return from the field, these were arranged in rows to dry on the concrete patio of the Seng Deuane guest house, only being removed to the interior of the guest house or a nearby wooden, roofed platform when the conditions were overcast.At the end of the field operation (December 13, 2015), semi-dry samples in the nylon mesh bags were loaded into the trucks, while voucher herbarium specimens in newspapers were cinched in straps and cardboard sheets and also loaded into the trucks.All field equipment and plant materials collected were transported by the pickup trucks to the ITM, where the drying of the samples and voucher herbarium specimens was completed.

Bolikhamxay medicinal plant preserve expedition
The Bolikhamxay Medicinal Plant Preserve (MPP), also known as the Somsavath MPP due to its propinquity, usefulness and political connection to Somsavath Village, is about 27 km south of Pakxan, the capital city of Bolikhamxay Province and is 163 m above sea level, at 18°27' N; 103°48' E. The preserve comprises approximately 13 hectares of high quality, secondary, lowland broad-leaved tropical rainforests recovering from past fires and logging (Figure 3), with adjoining land cleared for agriculture and plantations, primarily, economic botanicals (Hevea rubber, agarwood, etc.).
One of the ITM's 5-passenger Toyota Hilux pickup trucks, with a collapsible soft top for covering the rear cargo area, provided excellent mobility throughout the expedition period (December 14-17, 2015).This vehicle allowed for the transportation of 3 to 5 passengers (including drivers) in addition to the loads of collected plant materials and field supplies.Guest house quarters in Pakxan were rented as the expedition base and as equipment and supplies storage.The concrete patio and blacktop of the guesthouse served as a base for drying plant samples and working quarters.The space and facilities of this guesthouse permitted the efficient performance and completion of the expedition within a short time period.
As in the Xiengkhouang expedition, part of the supplies and equipment for use in the expedition was brought from Vientiane through the ITM's supplies or as part of the ITM and UIC personnel's belongings.In the collection area, one worker and several other locals were employed from the village of Ban Khampai for the purposes of harvesting and processing plant material.Forays into the field and the search for plants to be collected were guided by a list and by photographs of the plants previously collected (Soejarto et al., 2015) from this MPP.The warm and mostly dry conditions during the winter season of the expedition eased the initial stages of processing for voucher specimens and screening samples in the field.The roofed cement floor of the visitor center, near where the automobiles were usually parked, served as a temporary processing and drying area before the plant materials were loaded into the cargo bay of the Toyota Hilux on returning to the Pakxan guesthouse.
Plant samples and voucher herbarium specimens were carefully numbered and documented by field notes and photographic images.Numbered herbarium specimens were pressed between newspapers in the field.Screening samples, 1 to 3 kg fresh weight, depending on nature or fleshiness of the plant material, were packed in sample collection bags made of nylon mesh and were placed in the truck during forays.Later these were arranged in rows on the concrete patio of the guesthouse.At the end of the field operation (December 17, 2015), semi-dry samples in the nylon mesh bags were loaded into the truck, while voucher herbarium specimens in newspapers were cinched in straps and cardboard sheets.All were transported by the pickup truck to the ITM, where the drying of the samples and voucher herbarium specimens was completed.

Plant identification
One set of voucher herbarium specimens was shipped, processed and accessioned at the John G. Searle Herbarium of the Field Museum of Natural History (F), Chicago.Two additional sets of vouchers were deposited at the herbaria of the ITM.Taxonomic identification of the plants collected was initially performed (by JMH, KS, OS, MX, DDS) at the ITM Herbarium and was completed at the herbarium of the Field Museum.

Plant extraction
Azeotropic ethanolic extracts of the plant samples were generated from 30 g aliquots of plant samples at the laboratories of the ITM.Hence, 30 g aliquots of the 201 unground plant samples were milled and subsequently macerated twice overnight in 250 ml, followed by 200 ml of solvent, successively.The pooled extracts from each sample were desiccated by rotary evaporation at 40°C and transferred to small vials for shipment.All extracts were dispatched to and received safely at the University of Illinois at Chicago, where they were submitted to the MTS assay to determine their effect on the viability of HT-29 colon adenocarcinoma cells.

Colorimetric MTS assay for cell viability
Human colon cancer cells HT-29 were purchased from the American Type Culture Collection (Manassas, VA).The cell line was propagated at 37°C in 5% CO2 in RPMI 1640 medium, supplemented with fetal bovine serum (10%), penicillin (100 units/ml), and streptomycin (100 µg/ml).Cells in log phase growth were harvested by trypsinization followed by two washings to remove all traces of the enzyme.A total of 5,000 cells were seeded per well of a 96-well clear, flat-bottom plate (Microtest 96®, Falcon) and incubated overnight (37°C in 5% CO2).Samples dissolved in DMSO were then diluted and added to the appropriate wells (concentrations: 20 and 2 µg/ml; total volume: 100 µl; DMSO: 0.5%).The cells were incubated in the presence of test substance for 72 h at 37°C and evaluated for viability with a commercial absorbance assay (CellTiter 96® AQueous One Solution Cell Proliferation Assay, Promega Corp Promega) measured viable cells.Survival percentage, based on microplate reader (Synergy Mx, BioTek) readings of absorbance at 490 nm, was expressed in percentage relative to the solvent (DMSO) control.One of the authors (Dr.Wei-Lun Chen) performed the bioassay and interpreted the results (Ren et al., 2017).

Fieldwork
A total of 201 plant samples for preliminary biological screening, comprising 96 species of seed plants, documented by 96 sets of voucher herbarium specimens, were gathered during the two expeditions (Table 1).

MTS assay
At least six samples out of the 201 collected presented with activity bearing enough interest for further testing with HT-29 colon adenocarcinoma cells, using the colorimetric MTS assay for cell viability, that is, cytotoxicity.At the 20 μg/ml incubation condition, less than 60% of the HT-29 cells in these samples survived, and in five out of six of these samples, less than 50% of the HT-29 cells survived, indicating that the IC 50 of these five extracts should be under 20 μg/ml.Given that all six samples were from different taxa, this means that 2.99% of the collected samples and 6.25% of taxa with samples that were screened in this way were sufficiently cytotoxic to merit initial recollection for further studies.
The six samples of interest (Table 2 and Figure

Usage of HT-29 and other human tumor cell lines in anticancer screening of plants
The prevalence of cytotoxic taxa in this sample set derived from two expeditions in the Lao PDR is generally consistent with the results anticipated by the laboratory personnel of the P01CA125066 grant (Kinghorn et al., 2009(Kinghorn et al., , 2016)).The HT-29 cell line was selected as the sole gatekeeper for initial cytotoxicity testing, from amongst a number of human tumor cell lines accessible to Ohio State University scientists involved with the program project.HT-29 was the only human tumor cell line on hand that displayed high selectivity for the most cytotoxic extracts through its low susceptibility, with unpublished, internal project data suggesting that a signature of only ~5% (or less) of sample extracts evaluated bore significant activity against HT-29 cells.In terms of the P01 project itself, the consequent focus on HT-29 cells and human colon adenocarcinoma by association has led to screening for inhibitory activity in the K-ras pathway, becoming one of the major mechanistic assays emphasized (Peruchot et al., 1987;Ren et al., 2014).HT-29 cells are also significant in that the cell line has long been and continues to be included in the NCI-60 panel utilized for the COMPARE algorithm to predict mechanism of action for cytotoxic compounds (Paull et al., 1989;Shoemaker, 2006).More broadly HT-29 cells are derived from a human colon adenocarcinoma, and this subtype of colon cancer, derived from the epithelial lining surrounding the lumen of the large intestine, is responsible for over 90% of colon cancer cases (Kumar et al., 2010).According to the most current data from the Center for Disease Control, furnished with the assistance of the National Cancer Institute, as of 2013 colorectal cancer is the third most prevalent cancer shared by both sexes in the United States of America and is the second deadliest (Center for Disease Control, 2014).
For the cell lines utilized in the ICBG project, the cytotoxicity hit rate for Lao ethnomedical samples from plants ranged between 5 and 9%, with the exceptions of the non-epithelial leukemia cell line HL-60 (1.9%) and the breast cancer cell line MCF-7, as no bioassays were performed with this latter cell line for these samples (Gyllenhaal et al., 2012).No comparable meta-analysis in relation to the cell lines utilized was produced by the NCDDG, although a synthesis communicated significant insight into potential chemotaxonomic and plant anatomy-based relationships to activity hit rates and levels of cytotoxicity (Balunas et al., 2006).
comparing literature review and ethnopharmacological field work in tandem with biological screening suggests in general that: (a) greater toxicity categories exhibit higher hit rates than all medicinal plants taken as one category and (b) certain categories had three (anthelminthics) to four (arrow poisons and homicidal agents) times the hit rates of plants screened from "random" collections (Spjut, 2005).Still it is not necessarily straightforward that maximizing cytotoxicity in prioritizing bioactive leads always catalyzes advancement in the US pharmaceutical pipeline, as evidenced by the development of antihepatitis C nucleoside analogues for instance, which was among the first preclinical pharmacological research to suggest that it may behoove scientists to emphasize leads that retain moderate activity while minimizing toxicity (Sluis-Cremer et al., 2009;Coats et al., 2014).Plants and other sources of bioactive metabolites may be ideally positioned to take advantage of the uncertain interplay between bioactivity and toxicity in the transition from preclinical to clinical evaluation of leads, given that they often produce a suite of closely related analogues.This is particularly salient for plants collected on the basis of use in local medicine, by which their safety for human administration is comparatively and more likely assured in contrast to other sources of bioactive compounds (combinatorial chemistry, microbes, etc.).While as a result of his meta-analysis Spjut furthermore advocates selectively mining medicinal plants for categorical, chemotaxonomic, and/or novelty-related reasons in ongoing screening efforts, he also admits that the faster pace and greater sample collection rate for "random" (biodiversity-based) collection expeditions is attractive (Spjut, 2005).This is interesting because his analysis includes a further admission of the hybridity of collection strategies (Spjut, 2005).This hybridity is bound up in multiple explanations as to why the "random" collection strategy should be punctuated in quotations since, for instance, phytogeography always and chemotaxonomy still often constrain the hit rate results of plants selected for sampling from a given area.
It should be noted that our Lao collection strategy for P01 anticancer screening resembles the pace and sample collection rate of the "random" strategy that Spjut outlines far more than the typical ethnobotany-driven or active sample re-collection expedition in his experience, with an average of ~40/day obtained between the two expeditions, and within the 1 to 3 active samples (out of 60 to 100 acquired) per day projected for a "random" collection effort (Spjut, 2005).This also implies that at the very maximum, a "random" collection effort might be expected to have twice the hit rate of these Lao P01 collection expeditions, while at the low end, such "random" screening could have only a third of the hit rate observed.As with "random" collection expeditions for anticancer screening, chemotaxonomic considerations had an impact on the samples that were collected (Spjut, 2005;Balunas et al., 2006).This principle along with related inherited wisdom from the National Cancer Institute and the NCDDG/P01 project experience informed our exclusion of certain taxa for sample collection throughout the expeditions.These built-in considerations help demonstrate that this ethnobotanically driven screening strategy implemented for these two Lao expeditions is not an example of ideological purity but rather of hybridity.It would be argued that this eclectic pragmatism contributes to the effectiveness of the innovative strategy as a viable screening paradigm, which is supported perfectly by the fact that twice the number of active hits was generated by broad collection of plant parts as would have been through collection solely of plant parts employed in local ethnomedicine.

Prior results and integrative programing in Laos and Vietnam as a predictor of future success and strategic considerations for medicinal plant bioprospecting
The expansive and human-centric scope of the Vietnam-Laos ICBG has established and funded infrastructure bolstering traditional medicine through the creation of rural preserves and traditional medicine stations (TMS) in the Lao PDR (Riley, 2001;Sydara et al., 2014;Soejarto et al., 2015); permitted scholarly pursuit of botanical medicines described in Lao palm leaf manuscripts (Elkington et al., 2009); and thoroughly achieved the inventory and preliminary drug discovery lead investigation of the medicinal plants, and the floristic diversity as a whole, in these two countries (Soejarto et al., 2002(Soejarto et al., , 2006(Soejarto et al., , 2012;;Sydara et al., 2014).This project's legacy is an exemplar of the fact that medical ethnobotany, alongside such interdisciplinary subfields as historical ecology, landscape archaeology, and political economy (Balée and Erickson, 2006;Campbell, 2007;Scott, 2009), holds human agency intentionality in high esteem, being that they are crucial to value-added applications resulting from examination, discovery, and rediscovery of traditional knowledge.Indeed, humanenvironment interactions and the cultural transmission of natural and physiological observations are the bedrock underlying the endeavors of medical ethnobotany and ethnopharmacology, ensuring their value in perpetuity (Ott, 1998;Shepard, 2004).
Consideration of past plant collection practices and biological screening results in Vietnam and Laos to date should be able to guide the principles by which future expeditions are performed in the collection of plant samples for the P01 project in the Lao PDR.For instance, in the Vietnam-Laos ICBG, whereas cancer cell cytotoxicity hit rates for "random" samples from Vietnam and ethnomedical samples from the Lao PDR were relatively high, the ethnomedical samples from Vietnam had yielded a low hit rate (Gyllenhaal et al., 2012).It has been suggested that the differences in these hit rates in ethnobotanical collections between the two countries is attributable to the following factors: (a) a greater proportion of accessions in Laos that were purported correctly by healers to treat cancer in particular and (b) a more specialized knowledge of medicinal properties of flora owing to the Lao healers' greater prominence and mastery (Gyllenhaal et al., 2012).While it has been demonstrated that this study strategy can lead to new bioactive leads for anticancer drug discovery, categoryfocused ethnobotanical collection for biological screening seemingly is a tractable and rewarding strategy for evaluating the medicinal potential of Lao plant biodiversity as well.
Although no bio-prospecting and isolation work from the Lao PDR has thus far proceeded to the stage of patent filing thereby meriting additional studies (unlike for Vietnam; Zhang et al., 2014Zhang et al., , 2017)), through the Vietnam-Laos ICBG, six medicinal plant taxa from around the country have been evaluated through the compound isolation stage and have been found to contain mostly novel phytochemicals bearing anti-tuberculosis (Elkington et al., 2014), anti-malarial (He et al., 2005(He et al., , 2006;;Libman et al., 2008;Ma et al., 2008); and cancer cell cytotoxicity activities (Zhang et al., 2004).With respect to anticancer bio-prospecting from the ICBG field work conducted from 1998 to 2012 in the Lao PDR, at least 34 species yielded 50 extracts with significant activity (IC 50 < 20 μg/ml) in one or more cancer cell lines of a six cell line panel, which notably led to the isolated compounds asparacoside and 3''-methoxy-nyasol (IC 50 > 4 μg/ml, < 10 μg/ml) from Asparagus cochinchinensis (Lour.)Merr.(Soejarto et al., 2012).Integrative research with plants sourced from the network of ten preserves now established in Laos, to search for unanticipated bioactivity, is only in the most incipient of stages.Given adequate resources, biological screening and new active compound discovery from higher plants in these diverse habitats of the Lao PDR could continue indefinitely for decades to come.The link between bio-prospecting results and the biodiversity of these preserves and the country overall is as yet somewhat confounded by the underexplored relationship of abiotic and biotic factors to the phytochemistry of higher plant taxa throughout the monsoon cycle and under local management.These dynamic preserves are liable to yield additional surprises for as long as interdisciplinary programs such as the P01 project are able to support additional expeditions and follow-up research on taxa of interest (Kinghorn et al., 2009(Kinghorn et al., , 2016)).

Conclusions
The two initial P01 expeditions to the Lao PDR demonstrate that biological screening efforts, particularly anticancer bioprospecting, can proceed in areas with ancient and ongoing cultures of medicinal plant use, with the cooperation of local residents who protect, manage, and depend on the various forest habitats of Laos.Given the unique opportunity for direct collaboration with the Institute of Traditional Medicine (ITM) and some of the authors' involvement with healers and protected areas throughout Laos, it is fully expected that further expeditions will yield comparable or improved results.There are clear reasons why ethnobotanically motivated plant collection in areas rich in medicinal plant reliance can pay greater dividends for drug discovery efforts than other strategies.These rationales stem from the requirements of human safety, minimization of toxicity, and mitigation of related side effect profiles, as well as efficacious pharmacological activity (as vetted by longstanding use by local people), and from the pace of sample collection and turnover of expedition results in terms of the substantial timetables for biological screening and compound isolation work.The Bolikhamxay MPP and Xiengkhouang MBP, which were investigated during the winter dry season, are only two of the ten preserves established in the country.It is hoped that more of such preserves will be established in the future and maintained locally to bolster resilience in medicinal plant resource use and management, ensuring long-term protection of these forested regions throughout Laos.
Far more inventory work and sample collection expeditions to these preserves would be needed to truly exhaust the prospective bioactive leads from these ethnopharmacopeial treasure troves of traditional plant medicine, even for anticancer screening alone.Our interdisciplinary methodology and data analysis produce botanical and phytochemical leads not documented by local plant use patterns, but on the whole and verifiably anticipated by their value in ethnomedicine.The results of this research support the hypothesis that investigating plants known to be employed in local ethnopharmacopeias can produce promising starting points for natural product screening programs and pharmaceutical development, particularly as a first stage for anticancer drug discovery.

Figure 1 .
Figure 1.Medicinal plants preserve and medicinal biodiversity preserves network of Lao PDR(Soejarto et al., 2015), showing the location of expedition sites described in this paper.

Figure 2 .
Figure 2. Xiengkhouang Medicinal Biodiversity Preserve, December 9-11, 2015.Top left: Roadside signage downslope of the expedition site.Top right: View of the montane tropical rainforest on a cloudy day with forbs and soil in the foreground.Bottom left: View of the tree cover from the roadside.Bottom right: Forested rivulet within the preserve.

Figure 3 .
Figure 3. Bolikhamxay Medicinal Plant Preserve, December 15-16, 2015.Top left: Roadside signage at the entrance of the preserve.Top right: View from the road of the path into the preserve, on left.Bottom left: Processing samples on the patio of the pavilion near the entrance.Bottom right: Lowland tropical rainforest within the preserve featuring trees, treelets, lianas, and forbs.