Behavioral evidence of pheromonal signaling in desert grassland scorpions

Paruroctonus utahensis
Author(s): Matthew S. Taylor, Caleb R. Cosper, and Douglas D. Gaffin
Source: Journal of Arachnology, 40(2):240-244. 2012.
Published By: American Arachnological Society
DOI: http://dx.doi.org/10.1636/Hi11-75.1
URL: http://www.bioone.org/doi/full/10.1636/Hi11-75.1

BioOne (www.bioone.org) is a nonprofit, online aggregation of core research in the biological, ecological, and
environmental sciences. BioOne provides a sustainable online platform for over 170 journals and books published
by nonprofit societies, associations, museums, institutions, and presses.
Your use of this PDF, the BioOne Web site, and all posted and associated content indicates your acceptance of
BioOne’s Terms of Use, available at www.bioone.org/page/terms_of_use.
Usage of BioOne content is strictly limited to personal, educational, and non-commercial use. Commercial inquiries
or rights and permissions requests should be directed to the individual publisher as copyright holder.

BioOne sees sustainable scholarly publishing as an inherently collaborative enterprise connecting authors, nonprofit publishers, academic institutions, research
libraries, and research funders in the common goal of maximizing access to critical research.
2012. The Journal of Arachnology 40:240–244

Behavioral evidence of pheromonal signaling in desert grassland scorpions Paruroctonus utahensis

Matthew S. Taylor, Caleb R. Cosper and Douglas D. Gaffin1: Department of Zoology, University of Oklahoma, Norman,
Oklahoma 73019-0235 USA

Abstract. Behavioral evidence suggests that, in some scorpion species, females deposit a pheromone that attracts mates.
To date, however, no pheromone has been identified. The goal of our study was to isolate a pheromone from female desert
grassland scorpions, Paruroctonus utahensis (Williams, 1968) (Scorpiones:Vaejovidae). We took in situ cuticular washes
from female P. utahensis in a chloroform-methanol solution; the extract stratified into aqueous and organic layers. In
controlled laboratory experiments, most males exposed to female extract (aqueous and organic fractions combined)
exhibited pre-courtship behavior, whereas those exposed to the solvent control (2:1 chloroform-methanol) showed no
change in behavior. When extract fractions were separately tested, males initiated pre-courtship behavior when exposed to
the organic fraction but not when exposed to the aqueous fraction. These data are the first experimental evidence of a
female pheromone in this species and are important early steps toward characterizing any scorpion pheromone.
Keywords: Pheromone, ground-directed chemical signaling, pectines, arachnid, arthropod

The mechanisms mediating ground-directed chemical sig- pectinal sweeping increases in S. mesaensis (Gaffin & Brownell
naling have received little empirical attention compared with 1992). Additionally, in pectine-ablated S. mesaensis males,
mate tracking via airborne chemical cues. However, some conspecific female chemical washes did not release pre-
animals, such as snakes, insects and arachnids, detect courtship behavior (Gaffin & Brownell 2001). In most
nonvolatile sex pheromones while moving along the ground. scorpions, male pectines are longer and contain more chemo-
Male red-sided garter snakes follow the trail of female skin sensory peg sensilla than do female pectines (Polis & Farley
rubbings, allowing males to locate females over long distances 1979; Swoveland 1978). A sexual dimorphism in peg chemo-
(LeMaster & Mason 2001). Similarly, male parasitoid wasps sensitivity might also be present in some scorpion species: single
(Aphelinus asychis) and male minute pirate bugs (Orius sauteri) peg stimulations of P. utahensis with citric acid evoked a higher
detect and respond to conspecific female deposits on leaves, response in male sensilla than in female sensilla (Knowlton &
but they show no response to volatile female odors (Fauvergue Gaffin 2011). Natural selection might have favored larger
et al. 1995; Nakashima & Hirose 1999). pectines with selective chemical sensitivity in males if these
In addition, both chemical and behavioral evidence suggests chemosensory organs help males track female sex pheromones.
that spiders follow pheromone trails to find mates. Recent In this paper, we present the first steps toward our goal of
studies have isolated pheromones from spider silks and isolating and characterizing a scorpion sex pheromone. We
indicate a wide variety of chemical signals—from nonpolar collected female Paruroctonus utahensis (Williams 1968)
fatty acids in the agenelid spider Tegenaria atrica (C.L. Koch scorpions during the peak of their mating season and
1843) to small, polar compounds like dimethyl citrate in the immediately extracted cuticle-associated chemicals. We then
wandering spider (Papke et al. 2000; Trabalon et al. 2005; exposed conspecific males to the female extract in laboratory
Jerhot et al. 2010). Further, male wandering spiders Cupien- experiments. Initial exposures evoked pre-courtship behaviors
nius salei (Keyserling 1877) display courtship behavior upon similar to those described in S. mesaensis (Gaffin & Brownell
contact with a silk dragline treated with a small concentration 1992). Subsequent tests on aqueous and organic fractions
of female pheromone (Barth 1993; Tichy et al. 2001). revealed strong behavioral responses to the organic fraction.
Several studies suggest chemical cues are important in Our evidence suggests that the pheromone has low polarity
scorpion mate-tracking and courtship behavior. In a Y-maze and is highly stable.
test, male desert hairy scorpions Hadrurus arizonensis (Ewing
1928) preferred substrate previously walked on by females but METHODS
showed no preference for substrate previously walked on by
males (Melville et al. 2003). Additionally, male dune scorpions Research animal.—We collected P. utahensis during early
Smeringurus mesaensis (Stahnke 1957) initiate pre-courtship September from sand dunes near Monahans, Texas (31u299N,
behavior when they encounter chemical washes of conspecific 102u399W) at night using UV. We deposited a voucher
female epicuticles (Gaffin & Brownell 1992). specimen in the Sam Noble Oklahoma Museum of Natural
Elaborate, sexually dimorphic organs provide further evi- History in Norman, Oklahoma. The mating season of P.
dence that chemical communication guides male scorpion mate- utahensis runs from mid-August through early September
searching behavior. All scorpions have movable, ground- (Bradley 1988). In many ways, the natural history and surface
directed organs called pectines on their mid-ventral abdomen activity of P. utahensis is similar to the well-described ecology
(Cloudsley-Thompson 1955; Foelix & Müller-Vorholt 1983; of S. mesaensis (formerly P. mesaensis; Polis & Farley 1979).
Hjelle 1990). Pectines are likely important in mate tracking; Animal care.—The scorpions were kept in the laboratory in
upon contact with substrate previously walked on by females, 3.8 l glass jars partially filled with sand from the animals’
desert environment. We used timed lighting to maintain a
1
Corresponding author. E-mail: [email protected] 14:10 h L:D cycle; temperature and humidity were kept within
240
TAYLOR ET AL.—EVIDENCE OF A SCORPION PHEROMONE 241

a consistent range (22uC, 55–60% RH). The scorpions were fed quadrant, initiating the trial. Scorpions rested 6 d between
a wax worm once every 2–3 weeks and were provided water trials.
twice a week. Trial viewing and scoring.—Each trial was recorded (De-
Extract preparation.—In these experiments, we tested male fender SN501 DVR) and viewed later. Each trial lasted for
response to a female extract. To produce the female extract, 45 min after we placed the scorpion in the arena. The trial scorer
we anesthetized 35 adult (at least 200 mg) female scorpions on was blind to the treatment and test quadrant for each trial.
ice within an hour of collection and then immediately Since the test quadrant could be inferred from scorpion
submerged them in a 2:1 chloroform-methanol solvent. After placement in the arena, another researcher previewed each trial
8 h, the extract was decanted. Two days later in the and recorded when the scorpion began moving; the scorer
laboratory, we condensed the extract to a total volume of began viewing the trial 10 s after movement began. The reviewer
about 25 ml using a filter flask and a vacuum line. A few ml of viewed and scored trials every couple of days. Since the order of
water were added as the solution was drying. The resulting treatment exposure was randomized, the reviewer never knew
solution stratified to an aqueous layer on top and an organic what extract was being tested. The previewer recorded scorpion
(relatively nonpolar) layer on bottom. The solution was kept position relative to the test quadrant after 10 s of movement;
in an opaque glass container at room temperature. analysis of these data suggest that scorpions had no bias toward
Apparatus and trial preparation.—We used infrared cameras any quadrant during these initial movements (X2 5 0.5, df 5 2,
(two Defender SP301-C cameras and one Sony CCD-TRV16 P 5 0.779), keeping the test quadrant unknown to the scorer.
camera) to record male scorpion behavior in contained glass Further, no scorpion exhibited courtship behaviors during the
arenas (PyrexE 15 cm diameter, 7.5 cm deep). We divided each first 10 s of movement. The previewer also determined trial
arena into four fictive quadrants with small pieces of tape legitimacy; in a legitimate trial, the scorpion had to cross the test
adhered to the arena rim. The bottom of each arena was thinly quadrant at least once.
coated with sand. Infrared light emitted from photodiodes on We based our scoring criteria on pre-courtship behaviors
the cameras was the only light source during these experi- documented in S. mesaensis males (Gaffin & Brownell 1992).
ments; scorpions are not sensitive to infrared wavelengths Before starting experiment 1, we conducted pilot studies on P.
(Blass & Gaffin 2008). utahensis to ascertain species-specific differences in pre-
We conducted two experiments to determine if the female courtship behavior. Trials were assigned a score of 1–5
extract would release pre-courtship behavior in male scorpi- according to the following criteria. 1) No change in behavior.
ons. To determine sample size, we conducted an a priori power 2) Slight change in behavior, such as creeping (change of
analysis using data from unpublished research by Gaffin and normal motion to shorter forward steps and frequent turning)
from pilot studies (dz 5 2.06; a 5 0.05; 12b 5 0.95; paired). and sudden stops throughout arena. 3) One occurrence of
We concluded that we needed at least six legitimate trials per pedipalp-reaching or scrunching (drawing the pedipalps back
experiment to be confident in our results (see criteria for and toward the body midline), back-up (an abrupt stop of
legitimacy below); we used extra scorpions in each experiment forward motion followed by one or two steps backwards), or
to ensure we met this minimum. Each trial started between 1– sidestepping. 4) Two three-level responses in the same
3 h after the beginning of the dark cycle, when scorpions are quadrant. 5) Grasping the substrate with the pedipalps, tail
normally active in the field. The order of treatment exposure wagging, juddering, or three three-level responses in the same
to each scorpion was random (as determined by a computer quadrant.
random number generator). The scorpion was placed in the Statistical analysis.—In both experiments, only scorpions
quadrant opposite the test quadrant, initiating the trial. After with two legitimate trials were included in the statistical
each trial, we removed the sand, cleaned the arena with 70% analysis. All 12 scorpions in Experiment 1 had two legitimate
ethanol, and placed fresh sand into the arena. trials, whereas only six out of 10 scorpions in Experiment 2
Experiment 1.—In the first experiment, conducted in had two legitimate trials. The final score assigned to each trial
October and November, we exposed 12 male scorpions to was the highest score observed in the test quadrant during that
the female extract and to a control solution at separate times. trial. We collected paired, nonparametric data in both
The extract was shaken to create a homogeneous solution experiments and used the Wilcoxon signed-rank test (SPSS
before application to the sand on the arena floor; 2:1 software, release 12.0.0) to determine whether treatments
chloroform-methanol solvent was used as the control. The within each experiment evoked statistically different behav-
trial preparer deposited 100 ml of either the extract or the ioral responses.
control solution near the arena wall of the randomly selected
test quadrant (by computer random number generator). RESULTS
Scorpions rested at least 2 d between trials. In these experiments, we exposed P. utahensis males to
Experiment 2.—We conducted a second experiment a washes from the cuticle of female conspecifics. Of the 22 pairs
month after the first experiment ended. Ten of the 12 of trials conducted, 16 (73%) were legitimate. The behavior of
scorpions used in the first experiment were used for the male P. utahensis changed as they encountered female extracts
second experiment. The other two scorpions died between deposited on the sand substrate of the arena. Back-ups and
Experiments 1 and 2. For the second experiment, extract pedipalp-scrunches were the most common behaviors ob-
fractions were separately tested. The trial preparer deposited served, and males displayed these behaviors to varying
100 ml of either the organic fraction or the aqueous fraction degrees. Males that received a score of 5 often continued to
near the arena wall in the randomly selected test quadrant. We perform strong behaviors (sidesteps, back-ups, pedipalp-
placed the scorpion in the quadrant opposite the test scrunches) in the test quadrant for several minutes.
242 THE JOURNAL OF ARACHNOLOGY

Figure 2.—Comparison of male behavioral responses to homog-

enized female extract or control fluid of chloroform/methanol.
Plotted is each male’s difference in behavioral response between trials
(extract minus control). Numbers above each point indicate the score
of the extract trial. Subtracting the y-axis value from the extract trial
value gives the corresponding control trial value.

extract; four males (66%) received the highest behavioral

score. No changed behavior was noted as males walked across
the aqueous fraction of the female extract. Males received
significantly higher behavioral scores in organic fraction trials
than in aqueous fraction trials (Fig. 3: Z 5 2.26, p 5 0.026,
two-tailed).
Figure 1.—Distribution of elapsed time before entering treatment
quadrant. Shown are durations in minutes for extract and control DISCUSSION
trials in Experiment 1. Each bar represents the duration for
one scorpion.
This study represents the first steps toward isolating a
scorpion pheromone. Our first experiment revealed that male
P. utahensis behavior changes when encountering a female
Others abruptly stopped when walking over the deposited
chemical extract, which provides the first evidence for
extract, backed up, and then continued walking around the
substrate-borne chemical signaling in this species. Our second
arena (a score of 3). Juddering and tail wagging were each
observed in one trial. The strongest response observed was
often during a male’s first or second crossing of the test
quadrant, but in some trials, the male’s behavior did not
change until he had encountered the deposited extract several
times. When males detected no stimulant, they typically
walked around the perimeter of the arena and/or attempted
to climb the arena walls.
Volatile chemicals from the female extract apparently did
not attract males to the treatment quadrant. The average
elapsed time before entering the treatment quadrant was
6.7 min in extract trials and 5.9 min in control trials (Fig. 1).
In addition, no change in behavior was noted before males
entered the test quadrant, providing additional evidence that
males are responding to a ground-based chemical.
Experiment 1.—Behavior changed in 10 out of 12 males
exposed to homogenized female extracts. The trial scorer also
noted changed behavior in two of the 12 control trials, which
indicates a low level of experimental error. Males received
Figure 3.—Comparison of male behavioral response to fractioned
significantly higher behavioral scores when exposed to female female extracts. Plotted is each male’s difference in behavioral
extract than when exposed to solvent control (Fig. 2: Z 5 response between trials (organic fraction minus aqueous fraction).
2.68, p , 0.001, two-tailed). Numbers above each point indicate the score of the organic trial.
Experiment 2.—The trial-scorer recorded modified behavior Subtracting the y-axis value from the organic trial value gives the
in all six males exposed to the organic fraction of the female corresponding aqueous trial value.
TAYLOR ET AL.—EVIDENCE OF A SCORPION PHEROMONE 243

experiment showed that male behavior only changes upon (2008) made circular tracks composed of a small Petri dish
encountering nonpolar chemicals from the female chemical glued inside a larger Petri dish. In such a setup, the pectines
extract. Aqueous chemicals from the female extract released can be filmed from below through a clear Plexiglas stage.
no pre-courtship behavior in any of the males tested. Extract fractions could be blotted directly on the Petri dish
Males are not attracted to the female extract at a distance. surface or dried onto grains of sand sprinkled in the arena,
In none of our studies did males initiate pre-courtship and the number of pectinal sweeps could be tracked as the
behavior outside of the test quadrant of the arena. In animal moved across various stimuli. It might also be possible
addition, males were not drawn to the test quadrant more to use thin-layer liquid chromatography (TLC) to separate the
quickly in trials with the female extract than in control trials in extract into bands and test the animal directly in a rectangular
Experiment 1. We therefore conclude that the chemical (or arena atop the TLC plate. In this case, the monitoring would
mixture of chemicals) that males responded to in our studies be from the top, but an angled mirror could provide visual
was not volatile. access to pectinal activity.
Although the pheromonal chemical(s) remain unknown, it is Identifying scorpion mate-tracking pheromones has several
possible that male scorpions are reacting to a lipid in the possible implications, from controlling populations of dan-
female cuticle. Nonpolar chemicals, such as cuticular lipids, gerous scorpion species (thousands of humans die of scorpion
might be more stable in a hot, windy habitat than polar stings annually: Warrell 2007) to understanding the physiol-
chemicals. In general, longer carbon chains have lower ogy and evolutionary significance of pectines. Given our
polarity and a higher boiling point than shorter carbon ability to electrophysiologically record directly from scorpi-
chains. Although we cannot be certain that the female P. ons’ pectinal peg sensilla (Knowlton & Gaffin 2010), the
utahensis chemical signal has long hydrocarbon chains, identification of pheromone compounds presents a unique
cuticular extracts from another scorpion, Hadrurus arizonen- opportunity to relate proximate sensory mechanisms with the
sis, show a high proportion of compounds with chains evolution of a ground-based sexual signaling system.
exceeding 18 carbons (Trabalon & Bagnères 2010). ACKNOWLEDGMENTS
Males responded vigorously to the extract in Experiment 2,
four months after initial extraction. In these scorpions’ We thank Elise Knowlton for assisting with animal
habitat, it is likely that heat, wind, and exposure to sunlight collection and extraction, statistical consultation, and main-
would break down chemicals faster than we observed in our taining the animals in the laboratory. We thank Dr. Mariëlle
laboratory. Still, winds may blow sand grains covered with Hoefnagels for critically reviewing and editing this manu-
female-deposited pheromone across the dunes. If the phero- script. We also thank Drs. Robert Cichewicz and Jon
mone remained stable for even a few days, the pheromone Henrikson for help with pilot studies on extract fractionation.
could spread long distances from the female. During our We thank Dr. Richard Bradley for his help planning our
collections in the mating season, we observed several males collection trip. Finally, we thank the Life Fund of the
within 20 m of each female, an unusual grouping for P. University of Oklahoma Foundation for supporting this work.
utahensis at this field site. With the dispersing sands, it seems LITERATURE CITED
unlikely that males follow a female trail directly to her.
Barth, F.G. 1993. Sensory guidance in spider pre-copulatory
Instead, it is more likely that males compare concentrations of behavior. Comparative Biochemistry and Physiology 104:717–733.
female pheromone and move toward increasing concentration. Blass, G.R.C. & D.D. Gaffin. 2008. Light wavelength biases of
The stability of the female pheromone is noteworthy and scorpions. Animal Behaviour 76:365–373.
may lead to some interesting follow-up studies. For example, Bradley, R.A. 1988. The influence of weather and biotic factors on the
we hope to test male response to fractions within the organic behavior of the scorpion (Paruroctonus utahensis). Journal of
female extract. We have conducted some preliminary trials Animal Ecology 57:533–551.
(two trials each on three different fractions) and did not see Cloudsley-Thompson, J.L. 1955. On the function of the pectines of
the vigorous male responses we observed with the entire scorpions. Annals & Magazine of Natural History 8:556–560.
organic extract. One scorpion exposed to the most nonpolar Fauvergue, X., K.R. Hopper & M.F. Antolin. 1995. Mate finding via
a trail sex pheromone by a parasitoid wasp. Proceedings of the
fraction stopped in the test quadrant during his first crossing,
National Academy of Sciences USA 92:900–904.
backed up, and moved forward four times, but no other male Foelix, R. & G. Müller-Vorholt. 1983. The fine structure of scorpion
responded. The low response rate in these preliminary trials sensory organs. II. Pecten sensilla. Bulletin of the British
might suggest that male scorpions respond to combinations of Arachnological Society 6:68–74.
female chemicals. However, these observations must be Gaffin, D.D. & P.H. Brownell. 1992. Evidence of chemical signaling
approached with caution, because of the small sample size. in the sand scorpion, Paruroctonus mesaensis (Scorpionida:
In addition, because reproduction is seasonal, male sensitivity Vaejovidae). Ethology 91:59–69.
might fluctuate throughout the year, influencing their Gaffin, D.D. & P.H. Brownell. 2001. Chemosensory behavior and
responses to these fractions. It is also possible that the active physiology. Pp. 184–203. In Scorpion Biology and Research. (P.H.
compound or compounds had begun to break down before Brownell & G.A. Polis, eds.). Oxford University Press, New York.
Hjelle, J.T. 1990. Anatomy and morphology. Pp. 9–63. In The
these trials, which were conducted eight months after the
Biology of Scorpions. (G.A. Polis, ed.). Stanford University Press,
extraction was made. Stanford, California.
Since it is likely that males use their pectines to detect female Jerhot, E., J.A. Stoltz, M.C.B. Andrade & S. Schulz. 2010. Acylated
pheromones (Gaffin & Brownell 2001), future studies might serine deriviatives: a unique class of arthropod pheromones of the
focus on pectinal sweeping as animals move across extract- Australian redback spider, Latrodectus hasselti. Angewandte
contaminated substrates. For example, Blass and Gaffin Chemie International Edition 49:2037–2040.
244 THE JOURNAL OF ARACHNOLOGY

Knowlton, E.D. & D.D. Gaffin. 2010. A new tip-recording method to Polis, G.A. & R.D. Farley. 1979. Behavior and ecology of mating in
test scorpion pecten chemoresponses to water-soluble stimulants. the cannibalistic scorpion, Paruroctonus mesaensis Stahnke (Scor-
Journal of Neruroscience Methods 193:264–270. pionida: Vaejovidae). Journal of Arachnology 7:33–46.
Knowlton, E.D. & D.D. Gaffin. 2011. Functionally redundant peg Swoveland, M.C. 1978. External morphology of scorpion pectines.
sensilla on the scorpion pecten. Journal of Comparative Masters Thesis, California State University, San Francisco, USA.
Physiology A 197:895–902. Tichy, H., E. Gingl, R. Ehn, M. Papke & S. Schulz. 2001. Female sex
LeMaster, M.P. & R.T. Mason. 2001. Evidence for a female sex pheromone of a wandering spider (Cupiennius salei): identification
pheromone mediating male trailing behavior in the red-sided garter and sensory reception. Journal of Comparative Physiology A
snake, Thamnophis sirtalis parietalis. Chemoecology 11:149–152. 187:75–78.
Melville, J.M., S.K. Tallarovic & P.H. Brownell. 2003. Evidence of Trabalon, M. & A.-G. Bagnères. 2010. Contact recognition pheromones
mate trailing in the giant hairy desert scorpion, Hadrurus in spiders and scorpions. Pp. 344–375. In Insect Hydrocarbons:
arizonensis (Scorpionida, Iuridae). Journal of Insect Behavior Biology, Biochemistry, and Chemical Ecology. (G.J. Blomquist &
16:97–115. A.-G. Bagnères, eds.). Cambridge University Press, New York.
Nakashima, Y. & Y. Hirose. 1999. Trail sex pheromone as a cue for Trabalon, M., J. Niogret & C. Legrand-Frossi. 2005. Effect of 20-
searching mates in an insect predator Orius sauteri. Ecological hydroxyecdysone on cannibalism, sexual behavior, and contact sex
Entomology 24:115–117. pheromone in the solitary female spider, Tegenaria atrica. General
Papke, M., S. Schulz, H. Tichy, E. Gingl & R. Ehn. 2000. and Comparative Endocrinology 144:60–66.
Identification of a new sex pheromone from the silk dragline of Warrell, D.A. 2007. Venomous animals. Medicine 35:659–662.
the tropical wandering spider Cupiennius salei. Angewandte
Chemie International Edition 39:4339–4341. Manuscript received 6 September 2011, revised 14 April 2012.