JOURNAL OF INVERTEBRATE PATHOLOGY 69, 79–83 (1997)

ARTICLE NO. IN964634

Instar Susceptibility of the Monarch Butterfly (Danaus plexippus)

to the Neogregarine Parasite, Ophryocystis elektroscirrha
K. L. H. LEONG, M. A. YOSHIMURA, H. K. KAYA,1 AND H. WILLIAMS
Biological Sciences Department, California Polytechnic State University, San Luis Obispo, California 93407

Received May 16, 1996; accepted October 23, 1996

(1988) recognizes six families within the order, one of

The susceptibility of the monarch butterfly (Danaus which is the family Ophryocystidae. This monogeneric
plexippus) larvae to the neogregarine parasite, Ophryo- family has more than 12 described species that infect
cystis elektroscirrha, was tested in the laboratory. the Malpighian tubules of coleopterous insects (Levine,
Spore loads recovered from infected monarch butter- 1988). One exception is the species, Ophryocystis elek-
flies were directly related to the inoculum level, larval troscirrha, which infects hypodermal tissues of the
stage of the host, and spore age. There was a linear
monarch butterfly, Danaus plexippus, and the Florida
relationship between spores ingested by first instar
queen butterfly, Danaus gilippus berenice (McLaughlin
larvae and spore concentration. Larvae feeding on
leaves treated with 0, 50, 500, 5000, or 50,000 spores
and Myers, 1970).
averaged 0, 0, 193, 457, or 1,255 spores, respectively, on Recent studies showed that a high percentage (50–
the abdomens of the adult butterflies. When first, third, 60%) of monarch butterflies, D. plexippus, overwinter-
and fifth instar larvae were given 14.5 spores/mg of ing in California were infected with O. elektroscirrha
body weight, there was no significant difference in the (Leong et al., 1992). Very little is known about the
spore load of the adults resulting from the first and impact of this neogregarine on these large winter
third instars. However, there were significant differ- aggregations of monarch butterflies in California which
ences in the spore load from adults resulting from the are considered to be a natural treasure and an endan-
first and third instars versus the fifth instar. In addi- gered phenomenon (Wells et al., 1983). Moreover, labo-
tion, 1-year-old spores were not as infectious as fresh ratory rearing of the monarch butterfly for commercial
spores. Our findings indicate that under field condi- use increases the prevalence and degree of infection
tions, the first instar is most likely to become infected (William Jensen and John Ausland, personal communi-
because one spore appears sufficient to produce a cation) resulting in infected adults with heavy parasite
detectable spore load in the adult. Older instars are loads that are weak and unable to expand their wings.
less susceptible and have fewer opportunities to en-
These affected insects die shortly after emergence.
counter sufficient viable spores for infection to occur.
Spores of O. elektroscirrha are ingested by the monarch
Thus, vertical transmission appears to be the primary
mode of parasite maintenance in natural populations butterfly larvae and the hypodermal cells become in-
of monarch butterflies. r 1997 Academic Press fected. The parasite, however, does not sporulate until
KEY WORDS: Danaus plexippus; Ophryocystis elektro- the hosts pupate (McLaughlin and Myers, 1970). Upon
scirrha; monarch butterfly, neogregarine; protozoan pupation, the parasite rapidly completes morphogen-
parasite. esis in the tissue that is destined to become the setae
and scales of the adult butterfly. Thus, the spores of this
parasite are found externally over the entire butterfly’s
INTRODUCTION body, with the heaviest concentration on the abdomen
(Leong et al., 1992). The neogregarine does not appear
Neogregarines in the phylum Apicomplexa, Class to influence the butterfly’s survival or mating activity
Gregarinia, Order Neogregarinorida, are characterized during its 6-month stay at the overwintering site
by one or two merogonic schizogonies and a gameto- (Leong et al., 1992). Under laboratory conditions, butter-
gonic and sporogonic schizogony (Weiser and Briggs, flies with or without detectable spore loads had similar
1971; Brooks, 1988; Tanada and Kaya, 1993). Levine longevity.
Because of the paucity of information on the interac-
tions of O. elektroscirrha and the monarch butterfly, we
1 Present address: Department of Nematology, University of Cali- initiated a series of laboratory experiments to deter-
fornia Davis, Davis, CA 95616. mine parasite virulence and larval susceptibility to

79 0022-2011/97 $25.00
Copyright r 1997 by Academic Press
All rights of reproduction in any form reserved.
80 LEONG ET AL.

infection by this neogregarine. We report herein the (1) milkweed leaves daily within these containers until the
effect of spore concentration fed to larvae on the fifth instar, at which time they were transferred one per
neogregarine parasite load in the resulting adult butter- cup to inverted 355-ml clear plastic tumblers with
flies, (2) viability of spores after passage through the plastic lids. Five holes were made in the bottom of the
larval intestine, (3) infectivity of 1-year-old spores, and plastic cups for aeration. The larvae were fed daily
(4) susceptibility of first, third, and fifth larval instars within these clear plastic cups until they pupated.
to the protozoan parasite. Upon adult eclosion, the butterflies were harvested and
their abdomens examined for spores, using the wash
MATERIALS AND METHODS
method described above.
Effect of larval spore concentration on adult parasite
The butterflies used for the study were laboratory load. In the first experiment, first instar larvae were
reared (19.4°C 6 0.4 SE and 44.9 R.H. 6 1.5 SE) and exposed to milkweek bouquet leaves sprayed with one
free of the neogregarine parasite for two generations. of five spore suspensions: 0, 50, 500, 5000, and 50,000
The larvae were fed leaves of the blood flower milk-
spores/ml. Starting at the lowest concentration, each
weed, Asclepias curassavica, or the crown flower milk-
treatment, replicated three times, was applied to the
weed, Calotropis gigantea. The milkweeds were grown
leaves using a hand atomizer (DeVilbiss atomizer 15,
in the greenhouse to provide year-round production of
The DeVilbiss Co., Somerset, PA) until runoff and
food and to eliminate possible contamination by dis-
eased wild butterflies. allowed to air dry before the larvae were introduced. A
total of 15 larvae (5 larvae/replication) were used for
Spore inocula. Unless otherwise indicated, all spore each of the four spore suspension levels and control.
inocula used for the study were harvested from the The larvae were allowed to feed for 48 hr before being
abdomens of infected butterflies that were stored indi- transferred to clean milkweed bouquets and reared
vidually in envelopes at 5°C for less than a month. The using the protocol described earlier.
spores were recovered from infected butterflies using a In the second experiment, third instar larvae were
modified methodology described by Leong et al. (1992). given five levels of spore inocula. The spore level was
The abdomen was placed in a 15-ml centrifuge tube based on exposing each larva to 14.5 spores/mg of body
containing 2 ml of 0.05% (v/v) wetting agent (Tween 20; weight. Because the mean weight of the third instar
J. T. Baker Chemical, Phillipsburg, NJ) in deionized was 45.4 mg 6 0.35 SE, the maximum inoculum was
water and washed for spores by agitating with a vortex 658 spores/larva. Succeeding spore levels were halved
mixer for 1 min, letting stand for 5 min, and reagitating so that each larva received 329, 165, or 82 spores, with
for another minute. The abdomen was removed and the the control larva receiving 0 spores. Nine larvae were
wash suspension was centrifuged for 10 min at 3000 used for each of the five spore levels. In addition,
rpm with a bench-top centrifuge (Beckman GS-6 Centri-
1-year-old spores stored at 5°C were tested to deter-
fuge; Beckman, Palo Alto, CA). The supernatant was
mine their viability. Nine larvae were each given 658 of
discarded, 0.2 ml of the 0.05% wetting solution was
these spores.
added, and the suspension was agitated for 2 sec on a
Each inoculum was applied in 100 µl to a half leaf of
vortex mixer to resuspend the pellet. The number of
A. curassavica. The suspension was air dried, after
neogregarine spores/ml recovered was determined with
a hemocytometer. In all cases, control treatments con- which each larva, confined in a 50-ml plastic vial, was
tained 0.05% wetting agent. given a half leaf with the appropriate spore inoculum.
The cap of the vial was perforated with holes for
Larval rearing. The first to third instar larvae were aeration. Upon consumption of the leaf, the larvae were
maintained on milkweed bouquets. A bouquet consisted transferred to clean milkweed bouquets (3 larvae/
of one or two blood flower milkweed shoots (15 to 20 cm) bouquet, 9 larvae total per inoculum level) and reared
inserted through the middle of a plastic lid for a 473-ml using the same protocol as that described earlier.
drinking cup and secured to the lid with cotton. The
cutting was inserted into an opening of a parafilm- Viability of spores after passage through the larvae.
covered 150-ml beaker filled with water, and the shoot Infectivity of spores after passage through the larval
was adjusted to an upright position. The 473-ml cup, intestine was determined by collecting the feces from
modified by replacing the bottom with a Kimwipe larvae in each of the five inoculum levels in the second
tissue (secured with tape), was then inverted over the experiment described above. The feces from each inocu-
milkweed bouquet and snapped onto the lid. The larvae lum level were pooled and each pooled sample was
were transferred to fresh bouquets every 2 days. placed into a separate 15-ml centrifuge tube. Three
When larvae reached the fourth instar, the lid and milliliters of 0.05% Tween 20 was added and mixed
parafilm-covered beaker were replaced with a new until a suspension was achieved. An aliquot from each
plastic lid cover. The larvae were fed the crown flower inoculum level was examined at low power (103 objec-
 MONARCH BUTTERFLY SUSCEPTIBILITY TO A NEOGREGARINE 81

tive lens) with a compound microscope to verify the TABLE 1

presence of spores. Number of Ophryocystis elektroscirrha Spores Recovered
A 0.5-ml aliquot of the fecal suspension was spread from Abdomens of Monarch Butterflies Treated during Their
onto half a leaf of A. curassavica and air dried. The First Larval Instar with Leaves Sprayed with 0, 50, 500,
inoculated half leaf was placed in a 50-ml capped 5000, and 50,000 Spores
plastic vial containing a third instar larva. After the Abdomen
larva had consumed this leaf, it was transferred to a spore Percentage
clean milkweed bouquet. Five larvae (replicate) were Treatment Percentage load of deformed
used for each of the five treatment levels. The larvae (spores/ml) N with spores (x 6 SE) adults
were reared using the same protocol as described 50 12 0 0 0
earlier. 500 12 42 193 6 83 0
5000 14 100 457 6 90 7.1
Instar susceptibility. The first, third, and fifth in- 50,000 12 100 1255 6 285 91.7
stars were tested to determine their susceptibility to 0 14 0 0 0
the neogregarine. Average body weights were 0.7 mg 6
0.03 SE, 61.9 mg 6 1.56 SE, and 770.8 mg 6 6.47 SE for
the first, third, and fifth instars, respectively. Larvae appeared wet and were covered with spores along with
were given 14.5 spores/mg of body weight (approxi- a tar-like substance that may have been melanin. Their
mately 10 spores per first instar larva). Nine larvae wings did not expand and their abdomens were shriv-
(replicate) were used for each of the three developmen- eled.
tal stages and their respective controls. The inoculum In the second experiment, the parasite load of butter-
for first instar was applied in 2-µl aliquots to a given flies when treated during the third instar was directly
area of C. gigantea and allowed to air dry. This inocu- related to the level of inoculum (y 5 61.6 1 111.1x,
lated leaf area was then enclosed by a 3.2-ml shell vial r 2 5 0.91). The mean number of spores (spore load)
(7 mm inner diameter) containing a single larva. After recovered from the adults at various concentration
the larva had consumed the entire enclosed leaf area levels, however, was not significantly different (Ta-
within the vial, it was transferred to clean milkweed ble 2).
bouquets (3 larvae/bouquet, 9 larvae total). The inocula Adults treated during the first instar with 1-year-old
for the third and the fifth instars (2-µl and 50-µl inoculum had significantly fewer spores (123 6 22.4)
aliquots, respectively) were applied to half leaves of A. than those treated with fresh spores (665 6 147.8;
curassavica and air dried. The inoculated half leaf was t 5 3.6, P , 0.01, df 5 8). Spore loads produced with
placed in a 50-ml plastic vial containing a single larva. the 1-year-old inoculum were one-fifth the load pro-
After consuming the leaf, the larva was transferred to duced with the fresh inoculum.
clean milkweed bouquets (3 larvae/bouquet, 9 larvae/ Viability of spores after passage through the larvae.
instar). The larvae were reared using the same protocol Examination of fecal material of treated larvae showed
described earlier. the presence of few neogregarine spores (1 or 2 spores/5
Statistical analyses. The data were analyzed using random microscopic field using 1003 magnification).
the statistical program of Biostat 1 for ANOVA, linear No spores were recovered from the abdomens of adults
regression, and t test (Pimentel and Smith, 1990). resulting from larvae fed fecal contaminated leaves.
Instar susceptibility. Adults of larvae treated dur-
RESULTS
ing the first and third instars had significantly

Effect of larval spore concentration on adult parasite TABLE 2

load. In the first experiment, a linear relationship Number of Ophryocystis elektroscirrha Spores Recovered
was exhibited between the concentration of spores from Abdomens of Adult Butterflies Treated during Third
sprayed on the leaves and the number of spores recov- Instar Based on 14.5 Spores/mg of Body Weight
ered from the adult butterflies (y 5 2545.1 1 404.3x, No. of
r 2 5 0.88). The average number of spores recovered spores Abdomen
from the abdomens of adults ranged from 1255 spores fed per spore load
for those treated with 50,000 spores/ml to 0 spores at 50 individual N % with spores (x 6 SE)
spores/ml. Inocula levels of 0, 50, and 500 spores/ml 658 9 100 496 6 127
resulted in adults that appeared normal. Inocula levels 329 9 100 438 6 123
of 5000 and 50,000 spores/ml resulted in 7.1 and 91.7% 165 9 100 229 6 84
82 9 89 176 6 70
of the emerging adults (Table 1) that were heavily
0 9 0 0
infected and weak and could not separate from their
chrysalis or fell to the container’s floor. These adults Note. The average body weight of third instar was 45.4 mg 6 0.35.
82 LEONG ET AL.

TABLE 3 cause the probability of older larvae encountering a

Instar Susceptibility Based on Number of Ophryocystis high enough spore level for infection (based on spores
elektroscirrha Spores Recovered from Abdomens of Adults recovered by the wash method) to occur is remote.
Treated with 14.5 Spores/mg of Body Weight during the First, Moreover, infected larvae are not the source of infection
Third, and Fifth Larval Instars because sporulation occurs only during the pupal stage
Body Abdomen (McLaughlin and Myers, 1970), and our study has
weight (mg) Percentage spore load b shown that spores passed through the larval digestive
Instar a (x 6 SE) N with spores (x 6 SE) system appear to be noninfectious.
First 0.70 6 0.03 9 100 665a 6 148 Infected adult males or females could contaminate
Third 61.9 6 1.56 9 100 495a 6 127 the larval food source with spores if mating occurs on
Fifth 770.8 6 6.47 9 89 94b 6 59 the milkweed plant, resulting in infection of older
a None of the controls in each of the three instars were infected. instars. Leong (1995) observed canopy mating among
N 5 9 for each instar. overwintering butterflies. The activity between the
b Means with different letter are significantly different (F 5 6.22,
sexes during mating could cause the spores on the
P 5 ,0.01, df 5 2, 24). scales to detach and contaminate the milkweed leaf
surfaces. Although sexual transmission of spores from
(F 5 6.22, P 5 ,0.01, df 5 2, 24) mores spores (665 infected males to healthy females has not been demon-
and 495, respectively) than those treated during the strated, it may be another possible means of spore
fifth instar (94) (Table 3). Although not significant, transfer resulting in contamination of the egg. Liu et al.
more spores were recovered from butterflies treated (1974) observed spores of Apicystis (5 Mattesia) bombi
during the first than during the third instar. in the spermatheca of mated, overwintering bumblebee
queens. They hypothesized that this may be a means of
DISCUSSION
contaminating the bumblebee brood.
The passive transmission of spores from infected to
The number of spores recovered from adults was healthy butterflies in clusters is not likely, since the
directly related to the inoculum levels fed to the larvae proportion of individuals with detectable spore loads
and to the age of the larvae. Larvae treated during the did not change significantly during a 6-month overwin-
first or third instars produced adults that showed a tering period (Leong et al., 1992).
strong linear relationship between the inoculum level Severely infected individuals as observed by
and spore load. Hence, the higher inoculum levels McLaughlin and Myers (1970) and monarch breeders
consumed by the larvae resulted in higher spore loads (William Jenson and John Ausland, personal communi-
on the adults. A high percentage (91.7%) of adults with cation) probably occur at a low frequency under field
a mean abdominal spore load of 1255 were weak and conditions. A female generally deposits eggs singly on
deformed and died shortly after emergence. When the underside of a leaf. In Paso Robles, San Luis Obispo
equivalent inoculum levels (based on spores/mg) were County, California, the number of eggs on a single
given to the first, third, and fifth instars, the larvae in milkweed plant in the field seldom exceeded six eggs
each instar produced infected adults. The degree of (Leong, unpublished data). The low larval density
infection, based on spore loads on the abdomen, was minimizes competition for food and, perhaps, ensures
directly associated with the duration within the host; maintenance of the neogregarine parasite in the popu-
larvae infected during the earlier instars produced lation. We have shown that a first instar larva consum-
adults with greater numbers of spores than those ing approximately 10 spores produces adults with a
infected at a later instar. mean abdominal spore load of 665. Females with this
Although all larval stages tested were susceptible to spore load may be capable of depositing eggs with
the neogregarine, the most probable stage that would sufficient inoculum to produce severely infected indi-
be infected in the field would be the first instar. viduals. Under field conditions, however, some of these
Experimentally, first instars weigh an average of 0.7 spores may be inactivated by environmental factors
mg and an estimated one spore/first instar larva was (UV light or desiccation) or may not adhere to the leaf
sufficient to cause detectable infection (spores recov- surface. Even if heavily infected larvae survive and
ered by the wash method). Consumption of a single reach adulthood, the chances of the adult surviving will
spore by the first instar is highly possible because we not be very good based on laboratory observations.
have frequently observed eggs from infected adults Monarch breeders have often encountered severely
with one or more spores adhering to the chorion. Upon infected individuals after just three generations of
eclosion from the egg, the monarch larva often con- rearing the butterflies. Conditions that produce se-
sumes the entire chorion. Such behavior ensures the verely infected butterflies may be attributed to the
perpetuation of the parasite in the monarch butterfly cultural practices of these breeders. The butterflies are
population. Infection of later instars is unlikely be- forced to oviposit on a common milkweed host in
 MONARCH BUTTERFLY SUSCEPTIBILITY TO A NEOGREGARINE 83

densities not observed in nature. This allows eggs of REFERENCES

infected females to be oviposited with those of the Brooks, W. M. 1988. In ‘‘CRC Handbook of Natural Pesticides,’’ Vol. 5,
uninfected ones (McLaughlin and Myers, 1970). After a ‘‘Microbial Insecticides,’’ Part A, ‘‘Entomogenous Protozoa and
few generations, the majority of the females would be Fungi’’ (C. M. Ignoffo, Ed.), pp. 1–149. CRC Press, Boca Raton, FL.
infected and the inoculum level on the milkweed used Leong, K. H. L. 1995. Initiation of mating activity at the tree canopy
level among overwintering monarch butterflies in California. Pan.
for oviposition would be sufficient to produce severely Pac. Entomol. 71, 68–71.
infected individuals. Leong, K. L. H., Kaya, H. K., Yoshimura, M. A., and Frey, D. 1992. The
Our study has demonstrated that vertical transmis- occurrence and effect of a protozoan parasite, Ophryocystis elektro-
sion of O. elektroscirrha is the most probable because of scirrha sp. n. McLaughlin and Myers (Neogregarinida: Ophryocys-
tidae) on overwintering monarch butterflies (Danaus plexippus
the high susceptibility of the first instar larva to the (L.)) from two California winter sites. Ecol. Entomol. 17, 338–342.
neogregarine spore. Although infected adults could Levine, N. D. 1988. ‘‘The Protozoan Phylum Apicomplexa,’’ Vol. 1.
contaminate the food supply, older instars need to CRC Press, Boca Raton, FL.
consume a fairly high number of spores before infection Liu, H. J., Macfarlane, R. P., and Pengelly, D. H. 1974. Mattesia
occurs. The lack of larval to larval (horizontal) transmis- bombi n sp. (Neogregarinida: Ophrocystidae), a parasite of Bombus
sion further suggests that vertical transmission is the (Hymenoptera: Apidae). J. Invertebr. Pathol. 23, 225–231.
McLaughlin, R. E., and Myers, J. 1970. Ophryocystis elektroscirrha
primary means of parasite maintenance in monarch
sp. n., a neogregarine pathogen of the monarch butterfly Danaus
butterfly populations. Under laboratory breeding condi- plexippus (L.) and the Florida queen butterfly D. gilippus berenice
tions, however, spores on the scales of the butterfly are Cramer. J. Protozool. 17, 300–305.
the probable source of contamination resulting in the Pimentel, R., and Smith, J. 1990. ‘‘Biostat I: A Univariate Statistical
rapid spread of the parasite throughout the laboratory Tool Box,’’ 2nd ed. Sigma Soft. Placentia, CA.
population. Tanada, Y., and Kaya, H. K. 1993. ‘‘Insect Pathology.’’Academic Press,
San Diego, CA.
Weiser, J., and Briggs, J. D. 1971. Identifications of pathogens. In
‘‘Microbial Control of Insects and Mites’’ (H. D. Burges and N. W.
Hussey, Eds.), pp. 13–66. Academic Press, New York.
ACKNOWLEDGMENT
Wells, S. M., Pyle, R. M., and Collins, N. M. 1983. ‘‘The IUCN
We thank Mark Kubinski for his critical review of and comments on Invertebrate Red Book.’’ International Union for Conservation of
the manuscript. Nature and Natural Resources, Gland, Switzerland.