Autotoxicity in Plants
Michelle Marko
Colorado State University
Fort Collins, Colorado 80523
mmarko@lamar.colostate.edu
April 2, 1994
Summary
Plant pathogens and herbivores can have significant impacts on
plant populations. One way plants can protect themselves is through the
use of chemical substances. However, plants are faced with the problem
of poisoning systems that are metabolically similar to their own.
'Autotoxicity is self-destruction of a species through the production of
chemicals that escape into the environment and directly inhibit the
growth of that species' (Kumari and Kohli, 1987).
Plants do synthesize, store, and emit potentially autotoxic
compounds. How plants do this is largely unknown. They may take
advantage of physiological differences between itself and its attacker
by producing compounds such as neurotoxins (Fowden and Lea, 1979).
Plants can also package the chemical defenses in tissues, compartments,
or organelles (such as vacuoles) to remove the toxin from its enzyme
systems and primary metabolic processes. Polymerizing the metabolite or
making it insoluble within its own tissues is another mechanism in which
plants can avoid damage (Fowden and Lea, 1979; Baldwin and Callahan,
1993; Hartmann and Witte, 1994).
Autotoxicity in plants has been studied in a variety of manners.
Methodologies range from analyzing the biosynthesis and cellular
metabolism of compounds to adding leachate and plant exudates to soil
samples or petri dishes and measuring experimental plant performance.
Overall autotoxic effects can be measured by the last approach.
However, it is unknown how a plant specie is metabolically able to avoid
the autotoxic effects it has when added to the external environment of
the same species. In order to fully understand to what extent a
compound is toxic and how the plant is able to produce the toxin yet
avoid the negative effects itself, both external applications, and
subcellular micro experiments must be conducted. Discussion of current
knowledge of where secondary compounds are found, and how they arrived
there is first presented followed by research in which overall autotoxic
effects were studied.
Defense compounds emerge from the internal physiology of the
plant, and the importance of their metabolic behavior within the plant
has not been sufficiently appreciated in existing concepts about their
distribution within plants. The principal shortcoming is the tendency
to look at the toxic product in isolation for the metabolic events
associated with its accumulation. The chemical-defense phenotype that
is being acted on by natural selection is not simply the toxic product
itself, isolated from its connection with plant metabolism and
responsive as a unit to selection. Rather, the phenotype being acted
upon by selection is a very complex set of physiological events: the
biosynthetic pathways leading to the toxic product and to its
catabolism, as well a components associated with he manipulation of the
toxic compound, e.g., organelles involved in its compartmentation,
adjustments in other systems vulnerable to toxic effects of the
compound, and mechanisms for its translocation.'Doyle McKey. 1979. The
Distribution of Secondary Compounds Within Plants p. 56.
INTRODUCTION
SECONDARY METABOLISM
Many potentially autotoxic compounds are produced by what is
termed 'secondary metabolism', which describes the production of any
compound not essential to the growth and basic needs of a plant.
Secondary metabolism includes the biosynthesis, transport, accumulation
and storage of metabolic products (Herms and Mattson, 1992). The role of
many secondary metabolites in plants is still widely hypothesized, but
some are known to be toxic to plant pathogens and herbivores (McKey,
1979). What type and amount of a secondary plant metabolite depends on
many factors including the individual plant, its resources, environment,
and evolutionary constraints. Since plants have a limited amount of
resources to support their growth, all physiological processes can not
be met simultaneously, resulting in trade-offs between defense and
growth, and primary and secondary metabolism (McKey 1979).
Plants do not produce enough secondary compounds to protect all of
their tissues at all times, thus concentrations and types of compounds
vary throughout the plant (McKey, 1979). Some compounds are present at
all times in cellular tissue and some are activated after attack
(Karban, 1993). The apportionment of chemical defense will depend on
the unique biochemical and physiological properties of each compound.
In developing tissue, secondary metabolite production can be constrained
by the lack of enzymatic machinery necessary for its synthesis and the
lack of fully developed cell walls, vacuoles, idioblasts, resin ducts,
lactifers, and other specialized structures required for their
compartmentation (Herms and Mattson, 1992). Leaf development could be
slowed to keep up with the metabolic cost of secondary metabolism.
One hypothesis for the origin of secondary compounds may be that
these toxic compounds may have originally been produced via altered
substrate specificity of existing enzymes. The toxins may have been
produced by a metabolic accident resulting in a different function than
its homolog and a new metabolic pathway. The new toxin's distribution
throughout plants could result in a new factor to be evolutionarily
selected upon (McKey, 1979). Metabolically cheap compounds could be
favored. Along with selection of certain compounds, there would also be
selection to minimize metabolic cost by retrieving metabolites from
senescent tissues, transporting compounds from seed to seedling and from
flower to seed.
SOME SECONDARY METABOLITES (McKey 1979):
-Phenolic compounds- often weak toxicity to plants, microorganisms, and
animals. They can damage membranes, uncouple oxidative phosphorylation,
and perform other acts of disruption of general biochemical processes.
Tannins-, some phenolic resins, and quinones age general protein
poisons. Some phenols are acutely cytotoxic, e.g. podophyllotoxic.
-Terpenoids- mono- and sesquiterpenes are known allelopathic agents.
Sesquiterpene lactones are able to alkylate thiol groups of proteins;
some of which inhibit plant growth. Saponins alter the permeability of
cell membranes and may interact nonspecificity with proteins, thus
exerting general toxicity. Cardiac glycosides are a classic example of
a type of compound specifically toxic to animals. However, plants have
many similar ATPases which are similar to those in animals. Digitalis
purpurea aqueous extracts have been shown to cause necrosis and
chlorosis of leave tissue in itself as well as other plants.
-Nonprotein amino acids- analogs of common amino acids can be toxic by
replacing the necessary protein.
-Enzymes, in ribosomes, cyanogenic glycosides, and glucosinolates- all
have the possibility of being toxic to plants.
-Mixed-Function-Oxidase (MFO)- detoxifying enzymes of animals can be
disrupted as seen in Senecio -produced pyrrolizidine alkaloids and
methylenedioxyphenyl synergists.
-Alkaloids-nicotine toxic to plants when applied exogenously.
MECHANISMS FOR AVOIDING AUTOTOXICITY
The ability of a plant to prevent autotoxicity is essential to the
well being of the plant. In one case, (Wattenbarger et al., 1968),
found cyanogenic and noncyanogenic forms of clover. Since HCN is formed
in freezing tissues, the cost of producing cyanogenic forms in cold
climates is more costly than herbivory. However, clover can safely
produce cyanogenic forms in warmer climates to deter herbivory. Below
are listed some potential mechanisms (McKey, 1979; Luckner, 1980;
Luckner, 1990):
1) Segregation of toxic compounds in cell vacuoles and other organelles
or invention of specialized cells or tissues. Tannins, alkaloids, and
anthocyanidins are found mostly in vacuoles (McKey 1979). Protein and
cellular aggregates can work to keep enzymes and substances in localized
cellular regions where they can protect unstable intermediates (Luckner,
1980).
2) Secondary compounds as they occur in the plant may be inactive
precursors rendered toxic by some event during cell disturbance.
Intermediates either bind covalently to the enzyme complexes or are kept
in catalytic centers by other mechanisms (Luckner, 1980). Cyanogenic
glycosides release HCN only when disturbed. Glucosinolates are normally
accompanied by hydrolyzing enzyme myrosinase contained in idioblast
cells and liberated on disintegration of plant tissue (McKey, 1979).
3) Backup detoxification systems within the plant protect the plant
when toxins come in contact with sensitive tissue. Rhodanese and §-
cyanoalanine synthase are found in mitochondrial fractions of the plant
Manihot esculenta (Euphorbiaceae), which is involved in cyanide
metabolism McKey, 1979).
4) Plants which produce a particular secondary chemical may have more
specificity than nonproducer plants to a protein and its toxic analog
which causes the producer to be less sensitive to the metabolite. This
metabolic insensitivity is seen in co-factor binding of compounds and
compounds which are nontoxic in plants, but become toxic due to the
animals detoxification systems (Baldwin,1993; McKey, 1979).
Dichapetalum spp. (Dichapetalaceae) contain fluoroacetate that is
metabolized by animals to fluorocitrate, a potent inhibitor of Krebs
cycle reactions. In lily -of-the-valley (Convallaria majalis) the
aminoacyl-tRNA synthetase in a plant that produces nonprotein amino acid
will discriminate between the analogue and the protein amino acid and
thereby activate only the latter. In this way, the analogue of proline,
azetidine-20-carboxylic acid, is stored in shoots of lily -of-the-valley
without adverse effects, but causes toxicity when applied to other
plants.
COMPARTMENTATION
Three types of compartments are used for secondary products;
microcompartments, organelles surrounded by a single membrane, and
organelles surrounded by double membranes (Luckner, 1980). Lipophilic
secondary products like rubber particles, essential oil droplets, and
carotenoids, may accumulate in membranes or in the cytoplasm.
Hydrophilic secondary compounds and some enzymes involved in secondary
product formation and degradation (glucosidases, phenol oxidases,
peroxidases, peroxidases, hemicellulases, polyuronases, glucanases, and
chitinases) may accumulate in nonplasmic compartments separated by
membranes from cytoplasm-lumen of vacuoles, endoplasmic reticulum, and
golgi vesicles, or extraplasmic space. Compounds can also be
transported to extraplasmic space. Although removed from the cell, they
can still be acted upon by secondary metabolic enzymes secreted into
nonplasmic compartments by the mechanisms known for primary metabolic
enzymes.
Storage sites (Luckner, 1990, p. 35):
The controlled environment and regulation of metabolism makes it
possible for cells to (Luckner, 1990):
-orderly synthesize secondary products from intermediates and
cosubstrates of primary metabolism;
-create high local concentrations of precursors and intermediates
(reactants ready for inducible defenses)
-channel precursors and intermediates to favor certain reactions
sequences;
-facilitate and control metabolic sequences;
-accumulate large amounts of toxic secondary products within or near the
cell (ready for defense);
-control the release of stored secondary products;
-control the transformation and degradation of secondary compounds.
TESTING FOR AUTOTOXICITY
Autotoxicity has experimentally been studied either by determining
the secondary metabolite's biosynthetic pathway or storage site, or
through larger-scale experiments such as leachate-application
experiments. Cellular-based experiments frequently have looked at only
part of the picture; where particular compounds are found or what makes
up the biosynthetic pathway. Studying the biosynthetic pathway focuses
on determining mechanisms of autotoxicity avoidance, but few of these
experiments have indicated the exact location precursors can be found,
in what concentrations, and what would happen if the pathway was
altered. Techniques which can be used to study these cellular processes
are staining, microscopic examination, radioisotope labelling, and
genetic manipulations (Fowden and Lea 1979).
CELLULAR EXPERIMENTS
Pyrrolizidine Alkaloids (PA) are one set of metabolites where some
metabolic mechanisms are known (Hartmann and Witte, in prep). The exact
tissues of each stage of biosynthesis are still unclear. It is known,
though, that PAs are synthesized, translocated, and stored as N-oxides.
Pyrrolizidine Alkaloids are synthesized in roots or shoots and in some
plants occur in leaves. All plant parts contain some PAs with the
greatest concentrations of PAs in highly apparent areas, such as
epidermal and subepidermal tissues, roots, inflorescences, and leaves.
Celluarly, PAs are found in protoplast and vacuole storage.
Translocation in vacuoles by specific carriers is by mediated transport.
While the biosynthetic process is know for PAs, no work has been done on
potential autotoxic effects.
In poppies, another mechanism has been elucidated; vesicles
derived from endoplasmic reticulum carry alkaloids. In some cyanide-
producing plants, cyanide detoxifying enzymes, such as rhodanese and §-
cyanoalanine synthetase are localized in the mitochondria to protect the
cytochrome oxidase from inhibition by HCN (Luckner, 1990). Tissue
senescence and necrosis can be seen as signs of autotoxicity when HCN is
released in cells.
Non-protein amino acids may be produced as analogues of common,
protein amino acids. They are toxic to many organisms, including
plants, although no autotoxic effects have yet been studied. The
aminoacyl-tRNA synthetase activates the amino acids, before the
messenger RNA and chain initiation and termination factors determine the
nature of the protein. Thus it may activate the analogue instead of the
common protein amino acid.
These studies are important to understanding how metabolism of
secondary compounds occurs, but it needs to go one step further. What
happens if one enzyme is modified? Does the cell have mechanisms to
counteract potential damage by metabolites and precursors? Can the
metabolic pathway be setup in plants that do not naturally produce a
certain metabolite and what happens to manipulations of this system?
Parasitic plants could be used to study these types of questions.
Castilleja sp.incorporate pyrrolizidine alkaloids from their host
plants, yet do not produce them on their own (Stermitz and Harris,
1987). Are they able to and if so how do the toxic effects of these
compounds?
WHOLE PLANT
Environmental instances of autotoxicity have been identified by
using larger-scale types of experiments. These experiments determine,
on a plant to plant basis, whether exudates from a parent or same
species plant affect the growth of new plants. The plant exudates can
be applied directly on the plant, to the plants soil base, and by
planting offspring in its parent's bed. Another type of macro
experiment involves exogenous application of a secondary compound, or
its precursor to the plant's nutrient base, and measuring the impact on
the plant. Three cases of ecologically oriented experiments are
presented.
Leachate experiments have been used to find alfalfa to be
autotoxic. One study by Hall and Henderlong (1989) tested the effects
of alfalfa extracts and solid materials on growth and germination of
alfalfa seeds. Two types of experiments were used in this case,
greenhouse and laboratory. The greenhouse study involved the use of
mixing alfalfa plant material into soil in half of the plant beds in a
1:10 (alfalfa:soil) ratio. The ecological significance of this ratio
was not mentioned. Some soil flats were autoclaved to take care of
microbial and fungal influences. Experiment found alfalfa emergence was
reduced by 90% in alfalfa containing soil. However, there was no effect
when the soil was autoclaved. This indicated that microbes or fungi may
be the cause of the deterrence. Laboratory experiments were conducted
to try to distinguish between microbial and fungal influence and
autotoxic effects. Four lab experiments were conducted. In one
experiment, alfalfa was ground, extracted with distilled water, and
supernatant, sediment and solid fractions were added to alfalfa-seed
containing petri dishes. Seed germination and radicle length were
decreased by supernatant and solid fractions. In another series of
experiments, a microbial filter was used to remove fungi and microbes.
Again, reduced radicle length was noted in experimental versus control
dishes. Seed germination however, was no different from controls. The
complete effect of microbes and fungi still seems somewhat unclear.
Although not looking for metabolic pathways, this study demonstrates
ecological factors, such as fungi and microbes, which are generally not
considered when looking autotoxicity.
The coffee tree provides another example of autotoxicity tested on
a non- cellular scale (Friedman and Waller, 1985). The scarcity of
weeds around coffee trees is related at least in part to a slow rate of
leaching of caffeine from the tree canopy as well as from litter and
lost coffee beans. Leaching of metabolites from above ground plant
parts occur with rain mist, fog, and dew, and irrigation sprinklers.
Friedman and Waller hypothesize that autotoxicity may act to regulate
seed germination with autoinhibitors and defend against phytopathogenic
agents. Germination may be inhibited under conditions which would
endanger seedling establishment. The role of caffeine in seed
germination is indicated by two factors. The first stages of cell
division in the root tip start only after the root tip has moved away
from the caffeine-endosperm. Growth occurs until the 13th week, when
germination stops as caffeine is shuttled form the endosperm to the
cotyledons and hypoctyl. Rainfall or microbial leaching would ensure
germination occurs only when conditions also favor seedling
establishment--the inhibitors have been washed away. Excessive caffeine
surrounding coffee trees can be detrimental to surface roots which are
highly susceptible to caffeine. Long term effects such as shortened
tree life caused by caffeine concentrations may evolutionarily be
overridden by the benefits of defense and seedling delay.
This last case, demonstrates how exogenous applications of
secondary metabolites have been combined with cellular metabolism to
understand the utility of localized regions of secondary metabolism.
Sorghum bicolor transforms L-tyrosine, to 4-hydroxymandelonitrile in
microsomal areas. However, there is limited activity when L-tyrosine is
applied exogenously. The presence of intermediates at the site of
metabolism can result in increased speed of production of the final
product and allow for more control than occurs when adding intermediates
externally. The same increased activity is seen with blue-green algae,
chloroplast and for catalyzing the transformation of L-phenylalanine to
0- and p-coumaric acid or benzoic acid (Luckner, 1980; Luckner, 1990).
The combination of cellular and whole plant experiments is important in
understanding the whole picture of autotoxicity.
Other autotoxic effects have been demonstrated in: Corsoi palustre
(L.) Scop. using ethanol extracts (Ballegaard and Warncke), Ragweed
Parthenium (Kumari and Kohli), Anastatica hierochuntica (L.)
(Cruciferae) (Edwards et al., 1988; Hegazy et al. 1990) and with
american pokeweed. What is important to note about these studies is
that ecological relevance is not always tested. Water was not always
the solvent used to obtain the extracts, whereas water is the most like
natural compound which will create the extract. The concentration of
extract in diluted water was also not compared to field observations.
In order to get a true understanding of autotoxicity, these factors must
be taken into consideration.
CONCLUSIONS
Autotoxicity is a complex and difficult topic to address.
Microbial and fungal interactions with plants may cause or prevent
autotoxicity. Cellular and whole plant experiments must be combined in
order to understand both the basic biology of autotoxicity and its
implications to ecological communities. Autotoxicity can be an
important factor in crop management. Autotoxic effects of plants could
be used to suppress weed populations, timing of chemical applications
and crop rotation, yet accurate information on how autotoxicity affects
crops is still largely unstudied (Friedman and Waller, 1985; Tesar,
1993). Studies which have looked at autotoxicity frequently have not
considered ecological relevance of metabolite concentrations in
applications. The combination of well designed microbiological and
ecological experiments are much needed to further understanding the role
of and autotoxicity in secondary chemistry.
REFERENCES
BALDWIN, I.T. and CALLAHAN P. 1993. Autotoxicity and chemical defense:
nicotine accumulation and carbon gain in solanaceous plants. Oecologia
94:534-541.
BALLEGAARD, T.K. and WARNCKE, E. 1985. Observations on autotoxic
effects on seed germination and seedling growth in Cirsium palustre from
a spring area in Jutland, Denmark. Holarctic Ecology 8:63-65.
BARBIERL, L., BOLOGNESI, A., CENINE, P., FALASCA, A.I., MINGHETTI, A.,
GAROFANO, L., GUICCIARDI, A., LAPPI, D., MILLER, S.P., and STIRPE, F.
1989. Ribosome-inactivating proteins from plant cells in culture.
Biochemical Journal 257:801-807.
CHEW, F.S. and RODMAN, J.E. 1979. Plant Resources for Chemical
Defense. In: Rosenthal, G.A., Janzen, D.H. (eds) HERBIVORES: Their
Interaction with Secondary Plant Metabolites. Academic Press San Diego.
pp 271-307.
EDWARDS, M.E., HARRIS, E.M., WAGNER, F.H., CROSS, M.C., and MILLER,G.S.
1988. Seed germination of american pokeweed (Phytolacca americana). I.
Laboratory techniques and autotoxicity. American Journal of Botany
75:1794-1802.
FRIEDMAN, J. and WALLER,G.R. 1985. Allelopathy and autotoxicity.
Trends in Biochemical Sciences 10:47-50.
FOWDEN, L.and LEA, P.J. 1979. Mechanism of Plant Avoidance of
Autotoxicity by Secondary Metabolites, Especially by Nonprotein Amino
Acids. In: Rosenthal G.A., Janzen D.H. (eds) HERBIVORES: Their
Interaction with Secondary Plant Metabolites. Academic Press, San
Diego. pp 135-160.
HALL, M.H., and HENDERLONG, P.R. 1989. Alfalfa Autotoxic Fraction
Characterization and Initial Separation. Crop Science 29:425-428.
HARTMANN, T. and WITTE, L. (in prep.) Chemistry, Biology and
Chemoecology of the Pyrrolizidine Alkaloids.
HEGAZY, K., MANSOUR, K.S., and ABDEL-HADY, N.F. 1990 Allelopathic and
autotoxic effects of Anastatica hierochuntica L. Journal of Chemical
Ecology 16:2183-2193.
HERMS, D.A., and MATTSON., W.J. 1992. The dilemma of plants: to grow
or defend. The Quarterly Review of Biology 67:283-335.
KARBAN, R. 1993. Costs and benefits of induced resistance and plant
density for a native shrub, Gossypium thurberi. Ecology 74:9-19.
KUMARI, A. and KOHLI, R.K. 1987. Autotoxicity of ragweed parthenium
(Parthenium, Hysterophorus). Weed Science 35:629-632.
LUCKNER, M. 1990. Secondary Metabolism in Microorganisms, Plants, and
Animals. 3rd edition. Springer-Verlag, Berlin pp 15-61.
LUCKNER, M. 1980. Expression and Control of Secondary Metabolism. In:
Bell, E.A. and B.V. Charlwood (eds) Secondary Plant Products, Volume 8.
Springer-Verlag, Berlin-Heidelberg. pp. 23-50.
MCKEY, D. 1979. The Distribution of Secondary Compounds Within Plants.
In: Rosenthal G.A., Janzen D.H. (eds) HERBIVORES: Their Interaction
with Secondary Plant Metabolites. Academic Press, San Diego. pp 56-
133.
STERMITZ, F.R., and HARRIS, G.H. 1987. Transfer of pyrrolizidine and
quinolizidine alkaloids to Castilleja (Scrophulariaceae) hemiparasites
from composite and legume host plants. Journal of Chemical Ecology
13:1917-1925.
TESAR, M.B. 1993. Delayed Seeding of Alfalfa Avoids Autotoxicity after
Plowing Glyphosate Treatment of Established Stands. Agronomy Journal
85:256-263.
WATTENBARGER, D.W., GRAY, E., RICE, J.S., and REYNOLDS, J.H. 1968.
Effects of frost and freezing on hydrocyanic acid potential of sorghum
plants. Crop Science 8:526-528. As cited in: BALDWIN, I.T. and
CALLAHAN P. 1993. Autotoxicity and chemical defense: nicotine
accumulation and carbon gain in solanaceous plants. Oecologia 94:534-
541.