A Review of Predator Avoidance Defenses by Aquatic Insects through the Use of
Chemical Substances
Howard Rhodes - Chemical Ecology - Spring 1994
Colorado State University
Fort Collins, Colorado 80523
Introduction
Chemical ecology pertaining to predator-prey relationships
merit significance in the freshwater aquatic insect community,
yet relatively few articles have appeared recently (1990's to
date) that examine the use of the chemosenses relating to any
particular observed behavior of stream-dwelling insects (Martinez
1986; R.K. Zimmer-Faust 1994, personal communication). Indeed,
interest in the isolation and identification of active compounds
related to predator-prey interactions are practically
nonexistent. Research concerning chemical ecology related to
marine biology abound (Dame et al. 1989, Tamburri et al. 1992,
Zimmer-Faust 1991, Ache, et al. 1978). Similar studies in
chemical ecology investigating freshwater fauna relating to prey
avoidance, predation, food selection, and migration are difficult
to locate, and are not so numerous as their marine counterparts.
Articles describing the construction of instrumentation for
measuring odor plumes indicate that the need for such devices
that will detect minute concentrations of odors in the aquatic
environment has arisen (Moore et al. 1992, Zimmer-Faust et al.
1988), and this need has been addressed.
Numerous papers on predator avoidance have been published on
freshwater aquatic insect behavior and ecology, that do not
address the active chemical substances responsible for the
observed behaviors. Peckarsky's (1980) paper investigated the
avoidance behavior by stream mayflies (Ephemeroptera) in second
order streams where predaceous stoneflies (Plecoptera) were
present. Although the avoidance reaction to chemical odors was
one of three stimuli investigated, identifications of the
responsible component(s) were not make. Instead, she suggested
that documentation of the nature of the chemical substances
produced by the stoneflies would have lent additional support to
the observed behavior.
Other investigations along the same theme of avoidance have
ascertained strategies that were pursued by macroinvertebrates
once a predator was recognized. Ostracods were observed to move
into Chara plants for protection when exposed to low quantities
of water from tanks containing juvenile Cyprinidae (Roca et al.
1993). Other forms of aquatic macroinvertebrate responses to
avoid predation have been suggested. For example, mayflies were
observed entering the drift at night in higher numbers than in
the daytime in order to avoid predators (LeRoy et al. 1991). In
this paper and Roca et al. (1993) no attempt was made to isolate
and identify the chemical stimuli responsible for the avoidance
strategies that the authors observed.
Published information on chemical defenses is so scattered
throughout the literature, and unavailable to the casual reader,
that it gives the impression that chemical defenses are rare and
unusual in aquatic organisms (Scrimshaw and Kerfoot 1987). The
purpose of this paper is twofold: (1) to provide an overview of
the recency of the literature that could be located pertaining to
those aquatic insect orders that produce avoidance chemicals or
react to allelochemicals released by other aquatic organisms, and
(2) determine the level of intensity that research is presently
taking in this arena.
Of those freshwater aquatic insect groups where predator
avoidance has been studied, the order Coleoptera appears to have
been given the most attention. Three families in particular are
included here, the Dytiscidae (predacious diving beetles),
Gyrinidae (whirligig beetles), and Elmidae (riffle beetles). Also
included are the truly aquatic families in the order Hemiptera,
or true bugs. All are subject to predation by fish, as well as
interspecific predation.
Order Coleoptera
The Gyrinidae:
Gyrinidae beetles are common in the United States where they
aggregate on the open surfaces of rivers, ponds, and lakes,
preferring quiet, sheltered areas. During the day they aggregate
in large groups that can exceed hundreds of individuals. When the
water's surface is the least agitated, the gyrinids react with
immediate darting and jumping motions. The question is posed
whether dispersal is their only defense against predators, or
whether there may also be chemical defenses involved. Hendrickson
and Stenson (1993) tested the hypothesis that secretions from the
pygidial gland (Figures 1a and 1d), in addition to repelling or
being toxic to fish and newts, can also serve as alarm
substances. Pure water containing pygidial secretions was
released into tanks of the beetles. They appeared to react
immediately after the contaminated water reached them by zig-
zagging, diving, and vigorous underwater swimming. Air containing
the substances and wafted over the beetles elicited no response.
The conclusion was that chemoreceptors on the gyrinid body seem
to be oriented toward the water, and the probability that
chemical signals are transported by the water are justifiable.
Prior to the above research, studies had shown that gyrinids
produce four varieties of norsesquiterpene (Figure 2) that are
strongly repellent to fish, birds, and small mammals (Scrimshaw
and Kerfoot 1987). Benfield (1972) assayed the glands'
effectiveness on bluegill fish. Although some ingestion took
place without incident, the overwhelming response was immediate
buccal flushing. Repellency to Gyrinidae seems to be more a
contact or taste phenomenon, with high sensitivity at relatively
low concentrations.
The Dytiscidae:
The Dytiscids are the most specialized of the aquatic
beetles, and have by far received the most attention with respect
to the isolation and identification of chemicals and their effect
on target organisms. As of 1987 (Scrimshaw and Kerfoot 1987) some
41 compounds had been identified from 58 species in four
subfamilies. Much of the interest in this family derives from:
(1) evidence that some species synthesize "vertebrate" hormones
as defenses against fishes, and (2) the opportunity for
comparisons between the Dytiscidae and the closely-related
Carabidae.
Prothoracic glands of dytiscids (Figures 1a and 1b) have
been well described elsewhere (Forsyth 1968). Their secreted
compounds that have been isolated and identified includes
steroids, a diterpene, and a nucleoprotein (Figures 3 and 4,
Table 1). Secretory ducts open individually into a chitinous
reservoir. There are no muscles around the reservoir. Instead
secretions are forced out by turgor pressure and indirect action
of adjacent muscles.
The most common compounds isolated from dytiscid pygidial
glands are aromatic aldehydes, esters, and acids. In total, 16
compounds have been identified from 54 species in four
subfamilies (Table 2, Figure 5). These compounds differ from the
dytiscids' closely related terrestrial carabids (Table 3). For
example, most carabid beetles possess pygidial secretions of
either formic acid or a mixture of 95% methacrylic acid and 5%
tiglic acid (Schildknect 1970). In contrast, two dytiscid
subfamilies Dytiscinae and Colymbetinae possess only one major
carboxylic acid, benzoic acid. The remainder of the secretions
are phenols.
When fish are force-fed either the secretions from the
thorax or the pygidial bladders of dytiscids, immediate rejection
and buccal flushing result. Thus there is no doubt about the
defensive nature of these secretions, nor of their effectiveness
on fish. Finally, one further possible use pygidial products may
serve is in the conditioning of plant tissues associated with the
deposition of eggs.
The Elmidae:
Riffle beetles or Elmidae are unique because they spend
their entire life in the same aquatic areas, submerged among the
rocks and algae of fast-flowing streams and rivers. Since they
are very poor swimmers, and are slow-moving, seemingly lethargic
insects, they depend on crawling around deliberately using their
tarsal claws for attachment. They are not predators, but live on
wood and vegetation attached to the bottom or substrate.
In feeding tests (Scrimshaw and Kerfoot 1987) seven species
of adult elmid beetles were offered four species of fish. In 98%
of the offerings (n=105 presentations) the beetles were rejected.
Closely related families (Dryopidae, Limnichidae, Eubriidae,
Psephenidae) were accepted. There are no known descriptions of
secretory glands, so the precise mechanism causing rejection is
unclear. Yet the combination of a general lack of predation and
distinct color patterns on the elytra and thorax suggest that
elmids might possess yet undisclosed chemical defense mechanisms.
To test this White (1989) offered adult and larval elmid beetles
to a variety of fish from locations where the beetles were
unknown. Prior to introducing the beetles, food of the beetles'
same proportional size were fed to the fish as a conditioning
feature. During testing none of the fish ignored or rejected the
larval elmids, but 97% of all adult elmids were rejected. All
prey were expelled unharmed but none of the fish attacked a
second time. When offered the beetles at three and six days
later, rejection responses appeared identical to the first
offering, thus appearing that the fish had not learned to
recognize and avoid adult elmids. Sculpin fish which coexist with
riffle beetles ignored them as a prey item.
Order Hemiptera
Relatively little has been published about chemical defenses
within the aquatic Hemiptera, considering the research done on
their life histories and taxonomy. Eight of the 10 families that
are considered aquatic (Merritt and Cummins 1978) have
specialized glands in the metasternal region of the prothorax
that emit pungent, protective secretions against potential
predators (Scrimshaw and Kerfoot 1987). These paired glandular
lobes are attached to a saclike reservoir through a connecting
duct. Other ducts lead from the reservoir to lateral orifices.
Secretory substances are not forcibly dislodged, but ooze onto
the cuticular surface, and then along groves or into patches of
bristles.
Hemiptera nymphs also possess glands, but they are located
dorsally in the abdominal region. Each gland opens through two
pores near intersegmental membranes. The so-called scent glands
are located between segments 3 and 4, 4 and 5, and/or 5 and 6.
Each gland opens through two pores near the intersegmental
membranes.
The Veliidae and Gerridae:
Being strict surface dwellers, the Veliidae (water striders)
and Gerridae (skaters) must be alert for attacks from above and
below the water's surface. Their survival is enhanced through
agility of movement, sensitivity to surface disturbances,
schooling, and countershading. In addition to this suite of
survival capabilities, both groups possess chemical defenses,
though not as effective as those of the Coleoptera. Although
metasternal glands exist for veliids, no chemicals have been
isolated, and this despite predator-prey tests indicating the
presence of distasteful compounds (Br_nmark et al 1984). Glands
are present in adult Gerridae that release secretions observed to
evaporate almost immediately. To this author's knowledge, no
isolation and identification of these substances have been
successful in these two families.
The Notonectidae and Corixidae:
Glands exist for both the Notonectidae (backswimmers) and
the Corixidae (water boatmen). Secretions from notonectids are
odorless and brownish in color. Characterization by Pattenden and
Staddon (1968) revealed two major components: p-hydroxy-
benzaldehyde and methyl p-hydroxybenzoate. It is significant to
note that these compounds also occur in secretions from the
pygidial glands of dytiscid beetles.
Corixidae also have glands, but with external openings
located on the metacoxa. The pale yellow oily liquid has a
pleasant aldehyde odor and has been characterized as trans-4-oxo-
hex-2-enal in Corixa dentipes and Sigari falleni (Pinder and
Straddon 1965). Whether these compounds are truly defensive for
the corixids and notonectids is not fully answered. Observations
indicate that different species may be preyed upon more readily
than others, but are not outright rejected.
The Pleidae:
Currently the metasternal glands of the Pleidae (pygmy
backswimmers) are unidentified, but their reservoirs open
laterally through a pair of pores where a 15 to 17% solution of
hydrogen peroxide is released, at least in Plea leachi
(Maschiwitz 1971). The hydrogen peroxide appears effective
against predation by Gambusia and Neoheterandria (Kerfoot 1982).
When disturbed, pleids release hydrogen peroxide that forms a
frothy layer around the body. In a fish's mouth this foamy
formation may be more of an explosive event. Whether the hydrogen
peroxide is the active protective substance is unknown. It may
act as a carrier substance that disperses other, as yet
unidentified, compounds.
The Naucoridae and Belostomatidae:
The Naucoridae (creeping water bugs) possess metathoracic
glands that produce, in at least Ilycoris cimicoides, a viscous,
colorless, and odorless liquid. Dominant substances are p-
hydroxybenzyaldehyde and methyl p-hydroxybenzoate (Staddon and
Weatherston 1967). Note that these two compounds were found
previously in the Dytiscidae and Notonectidae, and are repellents
against Tilapia fish.
In the family Belostomatidae (giant water bugs) there is a
general lack of metathoracic glands among members of the
subfamily Belostomatinae, including Belostoma (Staddon 1971). By
contrast, Lethocerus in the subfamily Lethocerinae is known to
possess metasternal scent glands. Males of the species produce
about 25 times the secretion of females, which is clear and
highly odoriferous, and serves to excite the females. The
dominant substance (98%) is trans-hex 2 -enyl-acetate. Based on
the low number of compounds that have been identified from the
giant water bugs, speculation has suggested that this group has
increased to a body size where adults no longer need the same
degree of protection required by smaller water bugs. Or is it
their sit-and-wait feeding mode that reduces their
conspicuousness and subsequent need for reliance on chemical
defenses? As with Lethocerus, the scent glands evidently fulfill
no more than a sexual role.
Discussion
Numerous species within the aquatic Coleoptera and Hemiptera
possess prothoracic and pygidial glands that are linked to the
production of noxious defensive substances. Through various
combinations of chemicals varying in concentration, those
different species are afforded various degrees of protection.
Some of these substances appear to have secondary functions,
i.e., sexual attraction, rather than protection or repulsion.
Some species have been studied extensively, both from the
standpoint of their defensive physiologies, as well as the
isolation and identification of the more important compounds.
While researching this paper, the time period when the
majority of the work was performed appeared to apparently be in
the 1980's. The 1970's were the years of buildup in looking for
defensive chemical compounds in the aquatic insects, although the
1960's saw some research in this area. To date in the 1990's
little followup has been done.
Certainly much room is left for numerous research thrusts
into the chemical ecology of defensive compounds created and
secreted by freshwater aquatic insects. Modes of action for
isolating and identifying substances immediately come to mind.
Concentration of the stimuli that elicit the response might be
investigated among the Gerridae, since a low concentration
appears to be present in the water within the community without
eliciting defensive responses. Where there are several components
for a particular secretion, investigations of the primary active
chemicals need to be isolated and identified. The arena appears
open and ripe for exploitation.
REFERENCES
Ache, B.W., B.J. Johnson, E. Clark. 1978. Chemical attractants of
the Florida spiny lobster Panuliris argus. Florida Sea Grant
Technical Paper No. 10. 28 pp.
Br_nmark, C., Malmqvist, B., Otto, C. 1984. Anti-predator
adaptations in a nuestonic insect Velia capria. Oecologia.
61:189-191.
Dame, R.F., J.D. Spurrier, T.G. Wolaver. 1989. Carbon, nitrogen
and phosphorus processing by an oyster reef. Mar. Ecol.
Prog. Ser. 54:249-256.
Forsyth, D.J. 1968. The structures of the defense glands in the
Dytiscidae, Noteridae, Haliplidae, and Gyrinidae
(Coleoptera). Transactions of the Royal Entomology Society
of London. 120:159-182.
Hendrickson, B.-I. 1993. Alarm substances in Gyrinus aeratus
(Coleoptera:Gyrinidae). Oecologia. 93:191-194.
Kerfoot, W.C. 1982. A question of taste: Crypsis and warning
coloration in freshwater zooplankton communities. Ecology.
63:538-553.
LeRoy-Poff, N., De Cino, R.D., Ward, J.V. 1991. Size-dependent
drift responses of mayflies to experimental hydrologic
variation: Acute predator avoidance or passive hydrodynamics
displacement. Oecologia. 88(4):155-586.
Martinez, L.A. 1986. Morphological and behavioral evidence for
chemoreception by predaceous stonefly nymphs and their
mayfly prey. Chem. Senses. 11(4):633.
Maschiwitz, U. 1967. Eine neuartige Form der Abwer von
Mikroorganismen bei Insekten. Naturwissensschaften. 54:649.
Merritt, R.W., Cummins, K.W. 1978. An Introduction to the Aquatic
Insects of North America. Kendall/Hunt Publishing. Dubuque,
IA, USA.
Moore, P.A., R.K. Zimmer-Faust, S.L. Bement, M.J. Weissburg, J.M.
Parrish, G.A. Gerhardt. 1992. Measurement of microscale
patchiness in a turbulent aquatic odor plume using a
semiconductor-based microprobe. Biol. Bull. 183:000-000.
Paddenden, G., Staddon, B.W. 1968. Secretions of the metathoracic
glands of the water bug Notonecta glauca (Heteroptera:
Notonectidae). Experientia (Basel). 24:1209.
Peckarsky, B.L. 1980. Predator-prey interaction between
stoneflies and mayflies: Behavioral observations. Ecology.
61(4):932-943.
Pinder, A.R., Staddon, B.W. 1965. Trans-4-oxohexen-2-al in the
odoriferrous secretion of Sigara falleoi
(Hemiptera:Heteroptera). Nature. 205:106.
Roca, J.R., Baltanas, A., Uiblein, F. 1993. Adaptive responses in
Cypridopsis vidua (Crustacea:Ostracoda) to food and shelter
offered by a macrophyte (Chara fragilis). Hydrobiologia.
262(2):127-131.
Schildknect, H. 1970. The defensive chemistry of land and water
beetles. Angewandte Chemie International Edition. 9:1-8.
Scrimshaw, S., Kerfoot, W.C., 1987. Chemical Defenses of Fresh-
water Organisms: Beetles and Bugs, Chapter 16 in: Predation:
Direct and Indirect impacts on Aquatic Communities. W.C.
Kerfoot, A. Sih, eds. University Press of New England.
Hanover, New Hampshire.
Staddon, B.W. 1971. Metasternal scent glands in Belostomatidae
(Heteroptera). J. Entomol. (A) 46:69-71.
Staddon, B.W. and Weatherston, J. 1967. Constituents of the stink
glands of Ilyocoris cimicoides (Heteroptera:Naucoridae).
Tetrahedron Letters. 46:4567-4571.
Tamburri, M.N., R.K. Zimmer-Faust, M.L. Tamplin. 1992. Natural
sources and properties of chemical inducers mediating
settlement of oyster larvae: A re-examination. Biol. Bull.
183:327-338.
Weatherston, J., Percy, J.E. 1978. Venoms of Coleoptera. In:
Arthropod Venoms. S. Bettini (ed.) pp. 511-554. Springer-
Verlag. New York.
White, D.S. 1989. Defense mechanisms in riffle beetles
(Coleoptera:Dryopidea). Ann. Entomol. Soc. Amer. 82(2):237-
241.
Zimmer-Faust, R.K., J.M. Stanfill, S.B. Collard, III. 1988. A
fast, multichannel fluorometer for investigating aquatic
chemoreception and odor trails. Limnol. Oceanogr. 33(6, part
2):1586-1595.
Zimmer-Faust, R.K. 1991. Chemical signal-to-noise detection by
spiny lobsters. Biol. Bull. 181:419-426.