Pyrrolizidine Alkaloids as a Fulcrum for Co-Evolution

The theory of evolutionary radiation of flowering plants has been linked to interaction with pollinators and herbivores (Ehrlich and Raven, 1964: Herrera et al., 2002). One class of phytochemicals that researchers investigated to understand the mechanisms for a co-evolutionary impact is pyrrolizidine alkaloids (PAs). Much of the past research has been on the toxicity of these compounds in various structural iterations. When they contain 1,2 double-bond (unsaturated) in their base, necine moiety (Figure 1), PAs can be activated to become hepatotoxic, carcinogenic, genotoxic and teratogenic to humans (IPCS 1988).

Symphytum officinale

This toxicity impacted the dietary supplement market in 2001, when the USA FDA required withdrawal of PA-containing Comfrey (Symphytum officinalis) preparations from the market. Nevertheless, other organisms have benefited from ingesting the compounds.

Biological data revealed how insects interactions with PA containing plants can be protective to the insect (Berenbaum and Aangerl, 1993; Nelson and Kursar, 1999). More recently,  Liu, Vrieling and Klinkhamer (2018) explored these synergistic effects with herbivores. Their study evaluated the anti-herbivory effect of PAs on the study insects, Western Flower Trips (WFT). They found that plant secondary metabolite fractions underestimates the effect on herbivory. All fractions contained N-oxides and free bases. Most of the anti-herbivory effect was restored when fractions were recombined.

The insects were more susceptible to the more polar fraction of butanol than the chloroform fraction. Both N-oxides and free bases were present in higher concentration in the butanol as opposed to the chloroform, but total PA amounts were 3x

Necine structure of Pyrrolizidine Alkaloids
Figure 1: Necine structure of Pyrrolizidine Alkaloids (Sahzly and Wink, 2014) 

greater in later. They also noted that when combined with chlorogenic acid (CGA), the N-oxide was more active than when combined with free base, which is the opposite of what was expect based on toxicity profiles.

One of the more important findings from their study was that aside from the N-oxide and freebase forms of the PA, other metabolites can act as antagonists or synergists between CGA and the PA retrorsine. The complexity of the phytochemical background altered the interactions between plant metabolites and their potential bioactivity. What we don’t know is if the use of laboratory spiking studies will work in the field, and will it have population level effects necessary for long term alterations to behavior?

Another piece of interesting research focused on where these compounds get sequestered in the plant. Stegemann et al. (2018) investigated the accumulation and role of PAs’ in comfrey (Symphytum officinale) flowers and fruits. It was originally believed that PA synthesis occurred in the plant roots, which was then distributed to the rest of the structure. They reported secondary sites of synthesis in young leaves subtending developing inflorescences, with transport from leaf to flower to protect reproductive structures.

The authors found variability in accumulation patterns in different tissue, strongly suggesting that the synthesis is developmentally driven. Generally, the level of N-oxides to tertiary PAs was ~95% in all tissue , and the fresh weight PA concentration (ppm) present in different plant parts are listed as follows (from least to most):

  • sepals (tr)
  • petals (6)
  • pollen (14)
  • ovaries (98)
  • complete flower (141)
  • fruits (183)

The patterns of the alkaloids themselves were different in various tissue. Application of methyl jasmonate didn’t appear to alter the levels of expression, thus they appear not to be constitutively produced. The highest levels appeared at peak inflorescence, dropping as the flowers withered. And additional PAs are stored in young leaves and reproductive structures.

In table below, even though only trace amounts of all PAs are located in petals and pollen, almost all is the PA is myosorpine. It’s not clear why that pattern of individual PAs is unique to specific tissue. The presence of specific PAs in different tissues may be “leftover” from the original metabolic pathway, with the plant shuttling those particular metabolites to specific tissue.

7-acetylintermedine ~65%

  • ovaries
  • fruits
  • whole flower

~45%

  • leaves
  • roots
7-acetyllycopsamine ~30%

  • leaves
  • roots
intermedine ~15% all
lycopsamine ~15%

  • ovaries
  • fruits
  • whole flowers

5%

  • leaves

12%

  • roots
myosorpine 99%

  • petals
  • pollen
3-acetylmyoscorpine largely trace

We can compare these unique distribution and accumulation patterns to plants that normally don’t produce PAs. Now et al. (2016) tested horizontal transfer of natural products from leachate of rotting plants containing PAs into plants that do not produce those compounds. The accumulation of PA’s occurred in the leaves of guest plant, not flowers, suggesting xylem transfer driven via transpiration. Accumulation in flowers via transport (as opposed to being a production site) is typically source-sink-translocation via phloem. The researchers found that the PAs accumulated as salt like N-oxides.  They had expected the presence of free base PAs since they are able to cross biomembranes via simple diffusion. Either the guest plant has a transporter able to translocate N-oxides, or the free base PAs are taken up and oxidized to N-oxide. It made me wonder if this root uptake of exogenous PA’s might happen in all plant species, or was it limited specific plant species, genus or families?

Finally, Wink (2019) discussed how PAs are often stored by specialized insects and that their chemical ecology is intricate. As an example Wink noted how they function as nuptial gift for defense of eggs in the caterpillar larvae of Creatonotus spp. after metamorphosis into adult insects. The presence of the PAs secreted on to the eggs protects them from potential predators. Because of that benefit, it appears that increased PA ingestion by male caterpillars enhances their attractiveness.

This last puzzle in the world of PA chemical ecology had me wondering what examples might exist in the human world that mimic this type of behavior. Although beer consumption has long been thought to enhance the appearance of human male virility, my own experience has disproved that particular absurdity. Do you have any examples?

References:

  1. Berenbaum MR, Zangerl AR (1993) Furanocoumarin metabolism in Papilio polyxenes: biochemistry, genetic variability, and ecological significance. Oecologia 95:370–37. https://doi.org/10.1007/BF00320991.
  2. Ehrlich P & Raven P (1964Butterflies and plants: a study in coevolutionEvolution 18586– 608. DOI: 10.2307/2406212
  3. El Sahzly, A. and Wink, M. (2014) Diversity of Pyrrolizidine Alkaloids in the Boraginaceae Structures, Distribution, and Biological Properties. Diversity 6(2):188 – 282. https://doi: 10.3390/d6020188.
  4. International Programme on Chemical Safety (IPCS) (1988) Pyrrolizidine alkaloids. Environmental health criteria 80. WHO, Geneva.
  5. Herrera, C.M., Medrano, M., Rey, P.J., Sánchez-Lafuente, A.M., García, M.B., Guitián, J., and Manzaneda, A.J. (2002) Interaction of pollinators and herbivores on plant fitness suggests a pathway for correlated evolution of mutualism- and antagonism-related traits. PNAS. 99 (26): 16823-16828. DOI: 10.1073/pnas.252362799.
  6. Liu, X., Vrieling, K., and Klinkhamer, PGL. (2018) Phytochemical background medicates effects of pyrrolizidine alkaloid in Western Flower Trips. J. Chem. Eco. 45(2): 116–127.
  7. https://doi.org/10.1007/s10886-018-1009-2.
  8. Nelson AC, Kursar TA (1999) Interactions among plant defense compounds: a method for analysis. Chemoecology 9:81–92. https://doi.org/10.1007/s000490050037.
  9. Now, M., Wittke, C., Lederer, I., Klier, B., Kleinwächter, M., and Selma, D. (2016) Interspecific transfer of pyrrolizidine alkaloids: An unconsidered source of contaminations of phytopharmaceuticals and plant derived commodities. Food Chem. 213: 163–168. https://doi.org/10.1016/j.foodchem.2016.06.069.
  10. Sahzly, A. and Wink, M. (2014) Diversity of Pyrrolizidine Alkaloids in the Boraginaceae Structures, Distribution, and Biological Properties. Diversity 6(2):188 – 282. https://doi: 10.3390/d6020188.
  11. Stegemann T, Kruse LH, Brütt M, and Ober D. (2018) Specific Distribution of Pyrrolizidine Alkaloids in Floral Parts of Comfrey (Symphytum officinale) and its Implications for Flower Ecology.  J Chem Ecol. doi: https://10.1007/s10886-018-0990-9.
  12. Wink, M. (2019) Quinolizidine and Pyrrolizidine Alkaloid Chemical Ecology – a Mini-Review on Their Similarities and Differences. J Chem Ecol. 45(2):109-115. https://doi.org/10.1007/s10886-018-1005-6.

Case Study to Develop NDIs for Cannabis Oil

In their final trimester, Masters students in the Herbal Product Design and Manufacture Program at Maryland University of Integrative Health (MUIH) take a case study course where companies and various stakeholders present real world problems for our students to solve. The current iteration will investigate the significant public interest in cannabis (hemp)-based and CBD-containing food and herbal products. Image of Marijuana leaf with CBD molecule superimposed over it.

Many questions were generated about the science, safety, and quality of these products at the May 31, 2019 public hearing with the Food and Drug Administration.  Key take-away messages from this meeting are the following:

  • Data are needed to determine safety thresholds for CBD.
  • Datasets/information should be objective, of adequate quality and available for transparent review.
  • Lab testing and data analyses need to be replicable.
  • Consumers need consistent information and labeling.
  • State/government entities need support in knowing what to do.
  • We need to understand the implications for children when they take CBD-containing products at different dosage ranges.

Because public has expressed considerable interest in marijuana products (see state legal status map: https://disa.com/map-of-marijuana-legality-by-state,)the FDA is considering how to issue a regulation creating an exception, such as allowing the marketing of CBD in conventional foods or as a dietary supplement.  Regardless, the former FDA commissioner noted that such foods must safe based standards for food additives and related food safety assessment procedures, such as those presented in the NDI guidance (https://www.fda.gov/news-events/press-announcements/statement-fda-commissioner-scott-gottlieb-md-signing-agriculture-improvement-act-and-agencys).

This has its own challenges.  The FDA has identified several safety based-questions about CBD products that need answers as noted above.  For example:

  • How much CBD is safe to consume in a day?
  • How does it vary depending on what form it’s taken?
  • Are there drug interactions that need to be monitored?
  • What are the impacts to special populations, like children, the elderly, and pregnant or lactating women?
  • What are the risks of long-term exposure?

Many industry insiders believe that hemp-based cannabis products, such as cannabis oil or tinctures will be addressed using the DSHEA statutory framework, but that these products will require New Dietary Ingredient (NDI) filings. A description of the 2016 NDI guidance document can be found at https://www.fda.gov/media/99538/download.

To that end, our case study this trimester students will collaborate with stakeholder Dr. Roger Clemens.  He is internationally known scientist, and Professor of Pharmacology and Pharmaceutical Sciences and Assistant Professor of Regulatory Science within USC’s School of Pharmacy, International Center for Regulatory Science. Dr. Clemens is a past president of the Institute of Food Technologists (IFT) and is the current presiding officer of the International Academy of Food Science and Technology.  He co-founded, established and contributes to a Food, Medicine and Health column published monthly in Food Technology for the past 15 years.  He continues to serve on several editorial boards (e.g., Food Chemical Toxicology, Journal of Food Science, Toxicology Research and Applications, Journal of Dietary Supplements) and continues to serve as a reviewer for many high-impact journals.  Dr. Clemens has published > 50 original manuscripts and commentaries. In addition, he has presented more than 250 invited lectures at domestic and international scientific conferences.  He is an elected Fellow in four scientific organizations.

They will be tasked with designing an effective model for a hemp oil NDI (New Dietary Ingredient), which will serve them in two important ways: 1) acquire a deeper understanding of hemp constituents and their safety (classic ADME and toxicology), public health implications, and potential medicinal application; and 2) learn to write an NDI, which is vital in the herbal supplement industry.

Wish them the best of luck.

Plant Defense Signaling: How is this related to good medicine?

Botrytis cinerea growing on a PDA
Botrytis cinerea growing on a PDA (Wikipedia)

Plants make numerous small molecules (metabolites) that are either directly toxic to insects, grazing animals, fungi and bacteria, or that stimulate the production of other toxic metabolites. The production of various types of metabolites such as alkaloids, terpenes, and phenolics can be turned on (induced) after damage to the plant occurs. The control mechanisms involve  jasmonate, salicylate, phytohormone (plant hormone) ethylene, the volatile (gaseous) methyljasmonate and methylsalicylate signaling pathways.

The signal process is complex. It works on a local level or systemically by traveling throughout the entire plant. It behaves like a network, allowing chemical communication between parts of a plant or within a population of plants in a given locale. With recent developments in metabolic profiling, highly sensitive separation and detection systems are used to create metabolite profiles, high-throughput gene expression analysis is used to detect genes transcripts and rigorous statistical mining resulted in some interesting data that reveals more about how plant defense signaling is controlled.

Researchers used metabolic profiling of overall patterns rather than relying on targeted metabolites. This is important, since previous work focused on a few major metabolites and provided conflicting data. They found differences in profiles from plants exposed to generalist insect feeders versus plants treated with phytohormones. Although both jasmonate and salicylate pathways were activated in each treatment (co-induced), the metabolite patterns were distinct; both treatments lead to a stronger localized rather than systemic response; and there appeared to be a great deal of cross-talk between both pathways influencing pools of precursor metabolites.

Botrytis cinerea growing on tomato leaf
Botrytis cinerea growing on tomato leaf

Another study followed the spatial accumulation of hydroxycinnamates (phenylpropanoids) and lignins in cell walls of Arabidopsis (a model organism) in response to changes in the ethylene signaling induced by the necrotrophic fungal pathogen, Botrytis cinerea. They correlated metabolic profiles with cytological (cell based) changes to provide biological validation of the analytical data.

When a fungus like Botrytis attacks a plant, it generally destroys cells and eventually the entire plant. In the presence of the fungus, plant genes for the biosynthesis of phenylpropanoids and lignins are expressed (turned on) to modify and reinforce the plant cell wall against fungal penetration. Botrytis induces over 30 ethylene regulated transcription factors – cellular molecules that target and induce underlying genes to become active. So these researchers used a metabolite profile of 3 ethylene mutants, plants that had gene mutations at different DNA sequence points of the ethylene signaling pathway.

It turned out that the mutant plants were less resistant to the fungus and that ethylene resistance, when present, appeared after the fungus had made contact with the plant cell wall and had begun to build the structures necessary to penetrate the cell wall barrier. One phenylpropanoid metabolite in particular, ferulate, seemed to be highly influenced by ethylene signaling. Ferulate cross-links the polymer strands of cell wall polysaccharides, enhancing their structural integrity as a barrier.

So what does this have to do with good medicine? Don’t focus on one or two metabolites to make an efficacious extract. Despite what you hear, we are not trying to mimic “magic bullet” medicine. Expose plants to a full set of ecological challenges to produce a metabolite profile of greater diversity. Mono-crop farming doesn’t cut it. Damn if those hippies had it right after all.