- Email: [email protected]
Infectivity of resting spores of Massospora cicadina (Entomophthorales: Entomophthoraceae), an entomopathogenic fungus of periodical cicadas (Magicicada spp.) (Homoptera: Cicadidae)
Journal of Invertebrate Pathology 80 (2002) 1–6 www.academicpress.com
Infectivity of resting spores of Massospora cicadina (Entomophthorales: Entomophthoraceae), an entomopathogenic fungus of periodical cicadas (Magicicada spp.) (Homoptera: Cicadidae) L. Duke,a D.C. Steinkraus,b,* J.E. English,a and K.G. Smitha b
a Department of Biological Sciences, University of Arkansas, Fayetteville, AR 72701, USA Department of Entomology, 321 AGRI, University of Arkansas, Fayetteville, AR 72701, USA
Received 18 June 2001; accepted 26 May 2002
Abstract Massospora cicadina Peck is a fungal pathogen of 13- and 17-year periodical cicadas (Magicicada spp.). In northwest Arkansas, during the spring 1998 emergence of the 13-year periodical cicada, Magicicada tredecassini (Brood XIX), <1% of emerging cicadas were infected with the conidial stage of M. cicadina, similar to data collected from the same population in 1985. However, in northwest Arkansas plots treated with M. cicadina resting spores collected from infected 17-year Magicicada septendecim cicadas (Brood IV) in 1997 from southern Iowa, 10 months prior to the 1998 emergence in Arkansas, conidial stage infections of M. cicadina in 13-year Arkansas M. tredecassini cicadas increased significantly to 10.6% (7.9% in males and 2.6% in females). These data suggest that M. cicadina resting spores do not require a dormancy of 13 or 17 years between cicada emergences. Instead M. cicadina resting spores appear to be capable of germinating and infecting periodical cicadas after less than 1 year. In addition, M. cicadina resting spores derived from one species (17-year M. septendecim cicadas) were infective for a second species (13-year M. tredecassini cicadas). A mean of 1:4 106 ðSE ¼ 1:8 105 Þ mature resting spores were produced per infected male M. septendecim. Ó 2002 Elsevier Science (USA). All rights reserved. Keywords: Massospora cicadina; Resting spores; Entomophthorales; Entomophthoraceae; Fungal infection; Host specificity; Magicicada spp.; Periodical cicadas
1. Introduction Periodical cicadas are homopterans (Cicadidae: Magicicada spp.) with either a 13- or 17-year life cycle depending on where they occur in the eastern half of the United States (Williams and Simon, 1995). Periodical cicadas have the longest known life cycle of any insect (Lloyd and Dybas, 1966). Their unusually long synchronous life cycle is thought to allow Magicicada spp. to escape control by synchronous predators and parasitoids (White et al., 1979). While the taxonomy of periodical cicadas is somewhat unclear it is generally considered that there are six species (Alexander and *
Corresponding author. Fax: +501-575-3348. E-mail address: [email protected] (D.C. Steinkraus).
Moore, 1962). The three species of 17-year cicadas are Magicicada cassini (Fisher), Magicicada septendecim (L.), and M. septendecula (Alexander and Moore) and the three species of 13-year cicadas are Magicicada tredecassini (Alexander and Moore), Magicicada tredecim (Walsh and Riley), and M. tredecula (Alexander and Moore) (Williams and Simon, 1995). Within an emergence all three of the 17- or 13-year species emerge together (Lloyd and Dybas, 1966). Taxonomically the situation is complicated by the fact that each of the three species of 17-year cicadas have a morphologically and behaviorally identical counterpart in one of the three species of 13-year cicadas (Alexander and Moore, 1962; Lloyd and Dybas, 1966). For further information on the classification of periodical cicadas see Marshall and Cooley (2000) and Simon et al. (2000). Periodical
0022-2011/02/$ - see front matter Ó 2002 Elsevier Science (USA). All rights reserved. PII: S 0 0 2 2 - 2 0 1 1 ( 0 2 ) 0 0 0 4 0 - X
2
L. Duke et al. / Journal of Invertebrate Pathology 80 (2002) 1–6
cicadas spend most of their life cycle underground as nymphs. Approximately 45 days before the end of their life cycle fifth instar nymphs emerge from the ground, undergo their final molt, then reproduce as adults; mating and ovipositing near their nymphal host trees. Several weeks before emergence, nymphs burrow to the soil surface where they may encounter resting spores of Massospora cicadina Peck (Entomophthorales: Entomophthoraceae) (Soper et al., 1976). M. cicadina is thought to be the only synchronized natural enemy of periodical cicadas (White et al., 1979; Williams and Simon, 1995) and its absence has been related to extremely high densities of periodical cicadas. Leidy (1851) first reported M. cicadina from periodical cicadas, but other than taxonomy (Peck, 1879, 1907) and morphology (Spear, 1921; Goldstein, 1929), little experimental information has been documented on the fungus since then. According to Soper (1963, 1974), M. cicadina infects all six species of periodical cicadas. However, ‘‘dog-day’’ cicadas (Tibicen spp.) and other solitary cicadas are not susceptible to M. cicadina, at least there are no records of infected solitary species. Soper (1974) described 10 additional Massospora spp. from gregarious cicadas in Canada, Central America, South America, and other regions. All published reports of Massospora spp. have been from naturally infected specimens. There have been no experimental host range studies with M. cicadina to confirm that isolates from one cicada species are infective to a second species. Massospora resting spores in the soil infect mature emerging cicada nymphs. These infections result in the formation of the asexual conidial stage of the fungus (White and Lloyd, 1983). According to Soper et al. (1976) immature nymphs underground are not infected by M. cicadina. The fungus invades only the abdomen of infected cicadas (Spear, 1921). Conidia are produced
within the abdomen of the cicada, turning it into a white mass of conidia that eventually results in sloughing off of the abdominal segments, starting with the genitalia and posterior segments (Goldstein, 1929; Leidy, 1851; Soper, 1974; Spear, 1921). Infected individuals are able to behave normally, flying and engaging in interactions with other cicadas, even with much of their abdomens missing. Because adult cicadas congregate in large lek mating centers (English, 2001) and because infected individuals demonstrate few behavioral changes (Soper et al., 1976), individuals infected with the conidial stage of M. cicadina can spread the fungus to many healthy individuals. Cicadas infected by conidia develop the resting spore stage of the fungus. Conidia and resting spores never occur within the same individuals. Resting spores, like conidia, are produced only within the abdomen and are ochre to brown in color with a powdery consistency (Goldstein, 1929; Spear, 1921). As the infection progresses, the abdomen becomes swollen and distended, and intersegmental membranes between the abdominal sclerites become fragile, stretched and broken (Fig. 1). Once the sclerites separate, abdominal segments fall off and powdery resting spores are dispersed into the air as infected cicadas fly. The resting spores settle to the ground where they have been thought to remain dormant until the next cicada emergence in either 13 or 17 years (Lloyd et al., 1982; Samson et al., 1988; Soper, 1974). Generally it has been considered that M. cicadina has the longest life cycle of any fungus (Tanada and Kaya, 1993). Little is known about the conditions necessary for survival and germination of resting spores or other aspects of the fungal life cycle, largely because the hosts only emerge every 13 or 17 years in a given area. Because of the long host life cycle, some authors have
Fig. 1. Periodical cicadas infected with the resting spore stage of M. cicadina can be easily recognized in the field by bending a cicada’s abdomen. Infected cicadas break open easily and the cake-like mass of ochre-colored resting spores is obvious.
L. Duke et al. / Journal of Invertebrate Pathology 80 (2002) 1–6
stated that mechanisms which trigger germination of M. cicadina resting spores will never be determined (Evans, 1989). Laboratory attempts to germinate resting spores have been unsuccessful (White and Lloyd, 1983). We present evidence that the hypothesis that M. cicadina resting spores remain dormant for 13 or 17 years and are closely synchronized with periodical cicada emergences may be incorrect. Our evidence suggests that M. cicadina resting spores can germinate and infect periodical cicadas whenever they contact an eclosing cicada independent of species or whether the host is from a 13- or 17-year emergence. To further understand the life cycle of M. cicadina, data were collected on prevalence of the fungus among an emerging 13-year periodical cicada population in Arkansas in 1988 that had been exposed to resting spores <1 year old. In addition, the M. cicadina resting spores used for this study were collected in 1997 from a 17-year M. septendecim population in Iowa, which is a different species from the 13-year M. tredecassini in Arkansas.
2. Materials and methods 2.1. Collection of resting spore stage M. cicadina in Iowa During the 1997 emergence of 17-year M. septendecim (Brood IV) in southern Iowa, live adult cicadas were collected on 4 and 5 July 1997, during the late phase of the adult life cycle (approximately day 30 of the cicada breeding season). The prevalence of resting spore stage M. cicadina infections in the cicada populations in three sites in Iowa was determined. The three sites were: Stephens State Forest and Red Haw State Park in Lucas County, and Nine Eagles State Park in Decatur County. Areas sampled were selected because of high cicada densities and accessibility. Cicadas were collected early in the morning by hand or net from low shrubs and herbaceous plants bordering woods and placed in groups of 50 in paper bags and kept in coolers. An effort was made to collect every live cicada possible whether active or moribund. In the laboratory each cicada was sexed then bisected with a razor blade to determine if the resting spore stage of M. cicadina was present (Fig. 2). 2.2. Number of M. cicadina resting spores per cicada The number of resting spores produced per infected male M. septendecim cicada was determined by counting the number of resting spores contained within five infected male cicadas with fully intact abdomens collected from the Iowa emergence. The infected individuals chosen for analysis were ‘‘ripe,’’ i.e., their abdomens were packed with mature powdery resting spores and the abdomen was fragile and about to rupture (Fig. 1).
3
Abdomens were removed and separately placed in 20 ml of deionized water containing 0.01% Tween 80 (FisherScientific, Fair Lawn, NJ 07410) in a 125 ml flask then vortexed for 2 min on the high setting. Four subsamples of each spore suspension were counted to determine the number of resting spores using a phase microscope at 200 with a hemacytometer, following the method of Cantwell (1970). 2.3. Infectivity of Iowa resting spores for Arkansas cicadas A field site on the wooded edge of a cattle pasture in Durham, Washington County, Arkansas, where 13-year periodical cicadas had emerged in 1985 (Williams et al., 1993) was chosen for the experiment. Experimental plots were laid out beneath areas in which the tree and shrub limbs showed intense cicada oviposition damage from the 1985 emergence. Within a total area of 15 m2 , vegetation was removed from the soil surface of 10 plots, each measuring 0:5 m2 . Five of the plots were randomly assigned as treatment plots and each was sprayed with 1 liter of fungal slurry prepared using 1,000,000 resting spores/liter of distilled water. The resting spores were collected from infected male 17-year M. septendecim in 1997 in Iowa as described above and stored in a refrigerator 11 days before use. The resting spores were harvested and quantified as above. The five control plots were sprayed with 1 liter of distilled water only. Each plot was evenly sprayed on 16 July 1997 with a CO2 powered backpack sprayer with a single ConeJet TSS 6 SX nozzle (Spraying Systems, Wheaton, IL 60189) and marked for future identification. The following spring, on April 28, 1998, prior to cicada emergence, traps made of fine mesh (described in Williams et al., 1993) were placed over each individual study plot enabling collection of all emerging cicadas within those plots. Emergence of Brood XIX M. tredecassini began in Durham, Washington, AR, on 14 May 1998 with peak emergence on 21 May 1998 (unpublished data). Although all three species of 13-year periodical cicadas occur on the study site, only M. tredecassini were used in this study, because it was the most abundant of the three species of periodical cicadas at this emergence site (Williams and Smith, 1991). Collection of cicadas began on the first day of emergence, 14 May, and emergence traps were also checked on 16, 17, 20, 21, 22, 24, and 26 May. Wings of all emerged cicadas within each study plot were marked using a permanent marker pen, indicating trap number, treatment or control, and date. The cicadas were placed in a mesh tent (2 m 2 m 2:5 m) containing woody vegetation. Each cicada within the tent was removed and checked daily for visual signs of the conidial stage of M. cicadina. In addition, the natural cicada population at Durham was sampled by randomly collecting 100 periodical
4
L. Duke et al. / Journal of Invertebrate Pathology 80 (2002) 1–6
Fig. 2. Comparison of bisected M. septendecim periodical cicadas collected in Iowa in 1997. (a) Uninfected female, ov, ovaries; (b) uninfected male, note that a healthy male abdomen is almost empty and serves as a resonating chamber for its mating song; (c) infected male with abdomen full of M. cicadina resting spores (rs).
cicadas by hand from vegetation on 14, 16, 17, 20, 21, 22, 24, and 26 May from an area within 60 m of the experimental area. Collected cicadas were marked as above and placed in an identical but separate holding tent. As above, each periodical cicada was checked daily for visual signs of M. cicadina.
determining whether there was a difference between the incubation period of M. cicadina with regards to gender.
3. Results 3.1. Collection of resting spore stage M. cicadina in Iowa
2.4. Statistical analyses A v2 goodness-of-fit test was used to determine if differences existed between numbers of cicadas infected with M. cicadina in treatment plots compared to control plots. v2 was also used to determine if M. cicadina was gender biased in conidial stage during the 1998 emergence of Brood XIX, and resting spore stage in the 1997 Brood IV emergence. A t test (two sample for variance) was used in
A total of 431 M. septendecim (83.1% males and 16.9% females) were collected from Stephens State Forest (Lucas County). Of these 7.3% of the males were infected with the resting spore stage of M. cicadina and no females were infected. A total of 249 M. septendecim (76.7% males and 23.3% females) were collected at Red Haw State Park (Lucas County). Of these 2.6% of the males and 3.4% of the females were infected with the
L. Duke et al. / Journal of Invertebrate Pathology 80 (2002) 1–6
resting spore stage of M. cicadina. A total of 372 (82.8% males, 17.2% females) cicadas were collected from Nine Eagles State Park (Decatur County). Of these 16.9% of the males were infected with the resting spore stage of M. cicadina while no females were found infected. Overall, 1052 M. septendecim cicadas (81.5% male and 18.5% female) were captured in southern Iowa of which 9.7% of the males and 1.0% of the females were infected with the resting spore stage of M. cicadina. Brood IV M. septendecim, collected from southeastern Iowa 4–5 July 1997 showed a significant difference between males and females infected (v2 ¼ 94.87, df ¼ 1, P ¼ 0:0001) with males constituting 98% of total infected. A mean of 1:4 106 ðSE ¼ 1:8 105 Þ resting spores were produced per infected male M. septendecim. 3.2. Infectivity of Iowa resting spores for Arkansas cicadas A total of 346 cicadas were collected from the 10 study plots. The number of emerged cicadas was not significantly different between treatment (189 cicadas) and control (157 cicadas) plots (v2 ¼ 2.96, df ¼ 1, P ¼ 0:085). A total of 20 individuals were infected with the conidial stage of M. cicadina from the resting spore treated plots with equal distribution between all treatment plots (v2 ¼ 2.6, df ¼ 4, P ¼ 0:6271). Only one infected cicada emerged from control plots resulting in a significant difference between control and treatment plots (v2 ¼ 21.1, df ¼ 1, P ¼ 0:0001). These numbers indicate a 10.6% prevalence of conidial stage M. cicadina infections in emerging populations of periodical cicadas in plots treated with resting spores versus a 10.6% prevalence from cicadas in the control plots. Of the 21 infected cicadas, 15 (71%) were males and 6 (29%) were females, resulting in a significant difference with regard to gender (v2 ¼ 3.98, df ¼ 1, P ¼ 0:046). No cicadas were found infected with conidial stage M. cicadina infections in the natural population based on a sample of 800 individuals collected over a period of 13 days. At emergence, four cicadas showed visible signs of M. cicadina conidial stage infection. The mean period of time before which adult cicadas showed visual signs of conidial stage M. cicadina infections was 1.25 days after emergence with no significant difference with regard to gender (t test ¼ )0.243, df ¼ 19, P ¼ 0:81). Visible signs of conidial stage M. cicadina infections were evident in all infected cicadas by the 14th day into the breeding period, with peak numbers occurring on the 8th and 9th days.
4. Discussion Results of our study demonstrate that: (1) resting spores of M. cicadina collected from a 17-year cicada
5
species in Iowa were infective to a 13-year cicada species in Arkansas, (2) resting spores collected from M. septendecim were infective to a separate species, M. tredecassini, (3) resting spores are capable of infecting cicadas after only one year of dormancy, (4) each infected male M. septendecim produced about 1:4 106 resting spores, and (5) significantly more male than female cicadas were infected with the resting spore stage of M. cicadina. We believe that this is the first experimental study on the infectivity of M. cicadina resting spores and the first quantification of the number of resting spores produced by individual cicadas. We found no evidence that M. cicadina resting spores have a 13- or 17-year dormant period in synchrony with periodical cicadas and we hypothesize that M. cicadina resting spores may be capable of infecting periodical cicadas whenever they contact eclosing periodical cicadas. The prevalence of conidial stage M. cicadina infections varies among populations of periodical cicadas, but rarely exceeds 10% (White and Lloyd, 1983). Conidial-stage M. cicadina infections were very low in the Durham, Arkansas, cicada population in 1985 (Williams et al., 1993) and 1998 (this study), so the infection rate of 10.6% from cicadas collected from our five resting sporetreated plots appears to be a meaningful increase in prevalence. White and Lloyd (1983) suggested that emerging cicadas encounter resting spores randomly, and therefore, an equal sex ratio of cicadas infected with the conidial stage of M. cicadina conidia might be expected. However, we found significantly more males infected with the resting spore stage of M. cicadina at the end of the emergence in Iowa in 1997 and the conidial stage at the beginning of the emergence in 1998 in Arkansas. Emergence of periodical cicadas is protandrous, with most males emerging quickly in a matter of days, while female emergence occurs somewhat later and is more prolonged (Williams et al., 1993). Given that males emerge first, typically in large numbers (e.g., over 1,000,000/ha, [Dybas and Davis, 1962]), males would encounter resting spores before females, perhaps leading to a higher infection rate among males. Also, males form large chorus centers during mating (Williams and Smith, 1991), so that males are constantly coming into contact with other males (Alexander and Moore, 1962), including infected males shedding conidia. This may account for the higher percentage of males infected with the resting spore stage. Females tend to be more solitary therefore the possibility of females becoming infected through direct contact is reduced. Females are thought to mate only once and English (2001) has recently suggested that the higher percentages of males with resting spores may be due to avoidance of infected males by females. Females use an acoustic assessment for large size while courting males during the early part of the breeding period (English, 2001). Conidia or resting
6
L. Duke et al. / Journal of Invertebrate Pathology 80 (2002) 1–6
spores fill the abdomens of infected males (Fig. 2C), altering the pitch of their song, resulting in infected males sounding smaller than they actually are. Thus, the higher prevalence in males may be a consequence of both greater male aggregation and female avoidance of infected males. Little has been known about the ecological requirements necessary for survival and germination of M. cicadina resting spores. Fungal spores often need specific environmental conditions of relative humidity and temperature to survive and germinate (Steinkraus and Slaymaker, 1994). Resting spores may require a dormant period before they can germinate. For instance, Perry and Latge (1982) found that Conidiobolus obscurus resting spores required a 3-month cold period before they could germinate. We found that M. cicadina resting spores collected in July 1997, then applied to soil, were able to infect cicadas 10 months later in May 1998. Apparently 10 months and one winter period were a sufficient dormant period for some M. cicadina resting spores to infect emerging periodical cicadas in 1998. Therefore, while it is evident that M. cicadina resting spores are able to survive for up to 17 years in the soil, they do not require a 13 or 17 year dormant period, and are not as closely synchronized with periodical cicada life cycles as previously thought.
Acknowledgments We acknowledge Chris Simon for advice, and Gabriele Boys, Ping Li, and Shauna Ginger for field assistance and Mike Cassidy for access to his property. The David Causey Grant-in-Aid awarded from the Department of Biological Sciences, University of Arkansas provided partial support.
References Alexander, R.D., Moore, T.E., 1962. The evolutionary relationships of 17-year and 13-year cicadas, and three new species (Homoptera, Cicadidae, Magicicada). Misc. Publ. Mus. of Zool., Univ. Michigan. Tej eikle, vol. 121, pp. 1–59. Cantwell, G.E., 1970. Standard methods for counting Nosema spores. Am. Bee J. 110, 222–223. Dybas, H.S., Davis, D.D., 1962. A population census of 17-year periodical cicadas (Homoptera: Cicadidae: Magicicada spp.). Ecology 43, 432–444. English, J.E., 2001. Sexual selection in the lek-mating system of periodical cicadas Magicicada cassini and M. decim; selection for male body size, dominant courtship call and pathogen avoidance. Doctoral dissertation, University of Arkansas, Fayetteville.
Evans, H.C., 1989. Mycopathogens of insects of epigeal and aerial habitats. In: Wilding, N., Collins, N.M., Hammond, P.M., Webber, J.F. (Eds.), Insect–Fungus Interactions. Academic Press, London, pp. 205–238. Goldstein, B., 1929. A cytological study of the fungus Massospora cicadina, parasitic on the 17-year cicada, Magicicada septendecim. Am. J. Bot. 16, 394–401. Leidy, J., 1851. Fungus disease of Cicada septendecim. Proc. Acad. Nat. Sci. Phila. 5, 235. Lloyd, M., Dybas, H.S., 1966. The periodical cicada problem. II Evolution. Evolution 20, 466–505. Lloyd, M., White, J., Stanton, N., 1982. Dispersal of fungus-infected periodical cicadas to new habitat. Environ. Entomol. 11, 852–858. Marshall, D.D., Cooley, J.R., 2000. Reproductive character displacement and speciation in periodical cicadas, with a description of a new species, 13-year Magicicada neotredecim. Evolution 54, 1313– 1325. Peck, C., 1879. Massospora gen. Nov. 31st. Ann. Rept. New York State Museum Nat. Hist., 44. Peck, C., 1907. Massospora cicadina n.g. et sp. Ann. Rept. New York State Botanist 31, 44–48. Perry, D.F., Latge, J.-P., 1982. Dormancy and germination of Conidiobolus obscurus azygospores. Trans. Br. Mycol. Soc. 78, 221–225. Samson, R.A., Evans, H.C., Latge, J.P., 1988. Atlas of Entomopathogenic Fungi. Springer, Berlin. Simon, C., Tang, J., Dalwadi, S., Staley, G., Deniega, J., Unnasch, T.R., 2000. Genetic evidence for assortative mating between 13year cicadas and sympatric 17-year cicadas with 13-year life cycles provides support for allochronic speciation. Evolution 54, 1326– 1336. Soper, R., 1963. Massospora levispora, a new species of fungus pathogenic to the cicada, Okanagana rimosa. Can. J. Bot. 41, 875– 878. Soper, R., 1974. The Genus Massospora, entomopathogenic for cicadas, Part I, Taxonomy of the Genus. Mycotaxon 1, 13–40. Soper, R.S., Smith, L.F.R., Delyzer, A.J., 1976. Epizootiology of Massospora levispora in an isolated population of Okanagana rimosa. Ann. Entom. Soc. Am. 69, 275–283. Spear, A.T., 1921. Massospora cicadina Peck, a fungus parasite of the periodical cicada. Mycologia 13, 72–82. Steinkraus, D.C., Slaymaker, P.H., 1994. Effect of temperature and humidity on formation, germination, and infectivity of conidia of Neozygites fresenii (Zygomycetes: Neozygitaceae) from Aphis gossypii (Homoptera: Aphididae). J. Invertebr. Pathol. 64, 130–137. Tanada, Y., Kaya, H.K., 1993. Insect Pathology. Academic Press, San Diego. White, J., Lloyd, M., Zar, J.H., 1979. Faulty eclosion in crowded suburban periodical cicadas: populations out of control. Ecology 60, 305–315. White, J., Lloyd, M., 1983. A pathogenic fungus, Massospora cicadina Peck (Entomophthorales), in emerging nymphs of periodical cicadas. (Homoptera: Cicadidae). Environ. Entomol. 12, 1245– 1252. Williams, K.S., Simon, C., 1995. The ecology, behavior, and evolution of periodical cicadas. Ann. Rev. Entomol. 40, 269–295. Williams, K.S., Smith, K.G., 1991. Dynamics of periodical cicada chorus centers (Homoptera: Cicadidae: Magicicada). J. Insect Behav. 4, 275–291. Williams, K.S., Smith, K.G., Stephen, F.M., 1993. Emergence of 13year periodical cicadas (Cicadidae: Magicicada): phenology, mortality, and predator satiation. Ecology 74, 1143–1152.
Infectivity of resting spores of Massospora cicadina (Entomophthorales: Entomophthoraceae), an entomopathogenic fungus of periodical cicadas (Magicicada spp.) (Homoptera: Cicadidae) L. Duke,a D.C. Steinkraus,b,* J.E. English,a and K.G. Smitha b
a Department of Biological Sciences, University of Arkansas, Fayetteville, AR 72701, USA Department of Entomology, 321 AGRI, University of Arkansas, Fayetteville, AR 72701, USA
Received 18 June 2001; accepted 26 May 2002
Abstract Massospora cicadina Peck is a fungal pathogen of 13- and 17-year periodical cicadas (Magicicada spp.). In northwest Arkansas, during the spring 1998 emergence of the 13-year periodical cicada, Magicicada tredecassini (Brood XIX), <1% of emerging cicadas were infected with the conidial stage of M. cicadina, similar to data collected from the same population in 1985. However, in northwest Arkansas plots treated with M. cicadina resting spores collected from infected 17-year Magicicada septendecim cicadas (Brood IV) in 1997 from southern Iowa, 10 months prior to the 1998 emergence in Arkansas, conidial stage infections of M. cicadina in 13-year Arkansas M. tredecassini cicadas increased significantly to 10.6% (7.9% in males and 2.6% in females). These data suggest that M. cicadina resting spores do not require a dormancy of 13 or 17 years between cicada emergences. Instead M. cicadina resting spores appear to be capable of germinating and infecting periodical cicadas after less than 1 year. In addition, M. cicadina resting spores derived from one species (17-year M. septendecim cicadas) were infective for a second species (13-year M. tredecassini cicadas). A mean of 1:4 106 ðSE ¼ 1:8 105 Þ mature resting spores were produced per infected male M. septendecim. Ó 2002 Elsevier Science (USA). All rights reserved. Keywords: Massospora cicadina; Resting spores; Entomophthorales; Entomophthoraceae; Fungal infection; Host specificity; Magicicada spp.; Periodical cicadas
1. Introduction Periodical cicadas are homopterans (Cicadidae: Magicicada spp.) with either a 13- or 17-year life cycle depending on where they occur in the eastern half of the United States (Williams and Simon, 1995). Periodical cicadas have the longest known life cycle of any insect (Lloyd and Dybas, 1966). Their unusually long synchronous life cycle is thought to allow Magicicada spp. to escape control by synchronous predators and parasitoids (White et al., 1979). While the taxonomy of periodical cicadas is somewhat unclear it is generally considered that there are six species (Alexander and *
Corresponding author. Fax: +501-575-3348. E-mail address: [email protected] (D.C. Steinkraus).
Moore, 1962). The three species of 17-year cicadas are Magicicada cassini (Fisher), Magicicada septendecim (L.), and M. septendecula (Alexander and Moore) and the three species of 13-year cicadas are Magicicada tredecassini (Alexander and Moore), Magicicada tredecim (Walsh and Riley), and M. tredecula (Alexander and Moore) (Williams and Simon, 1995). Within an emergence all three of the 17- or 13-year species emerge together (Lloyd and Dybas, 1966). Taxonomically the situation is complicated by the fact that each of the three species of 17-year cicadas have a morphologically and behaviorally identical counterpart in one of the three species of 13-year cicadas (Alexander and Moore, 1962; Lloyd and Dybas, 1966). For further information on the classification of periodical cicadas see Marshall and Cooley (2000) and Simon et al. (2000). Periodical
0022-2011/02/$ - see front matter Ó 2002 Elsevier Science (USA). All rights reserved. PII: S 0 0 2 2 - 2 0 1 1 ( 0 2 ) 0 0 0 4 0 - X
2
L. Duke et al. / Journal of Invertebrate Pathology 80 (2002) 1–6
cicadas spend most of their life cycle underground as nymphs. Approximately 45 days before the end of their life cycle fifth instar nymphs emerge from the ground, undergo their final molt, then reproduce as adults; mating and ovipositing near their nymphal host trees. Several weeks before emergence, nymphs burrow to the soil surface where they may encounter resting spores of Massospora cicadina Peck (Entomophthorales: Entomophthoraceae) (Soper et al., 1976). M. cicadina is thought to be the only synchronized natural enemy of periodical cicadas (White et al., 1979; Williams and Simon, 1995) and its absence has been related to extremely high densities of periodical cicadas. Leidy (1851) first reported M. cicadina from periodical cicadas, but other than taxonomy (Peck, 1879, 1907) and morphology (Spear, 1921; Goldstein, 1929), little experimental information has been documented on the fungus since then. According to Soper (1963, 1974), M. cicadina infects all six species of periodical cicadas. However, ‘‘dog-day’’ cicadas (Tibicen spp.) and other solitary cicadas are not susceptible to M. cicadina, at least there are no records of infected solitary species. Soper (1974) described 10 additional Massospora spp. from gregarious cicadas in Canada, Central America, South America, and other regions. All published reports of Massospora spp. have been from naturally infected specimens. There have been no experimental host range studies with M. cicadina to confirm that isolates from one cicada species are infective to a second species. Massospora resting spores in the soil infect mature emerging cicada nymphs. These infections result in the formation of the asexual conidial stage of the fungus (White and Lloyd, 1983). According to Soper et al. (1976) immature nymphs underground are not infected by M. cicadina. The fungus invades only the abdomen of infected cicadas (Spear, 1921). Conidia are produced
within the abdomen of the cicada, turning it into a white mass of conidia that eventually results in sloughing off of the abdominal segments, starting with the genitalia and posterior segments (Goldstein, 1929; Leidy, 1851; Soper, 1974; Spear, 1921). Infected individuals are able to behave normally, flying and engaging in interactions with other cicadas, even with much of their abdomens missing. Because adult cicadas congregate in large lek mating centers (English, 2001) and because infected individuals demonstrate few behavioral changes (Soper et al., 1976), individuals infected with the conidial stage of M. cicadina can spread the fungus to many healthy individuals. Cicadas infected by conidia develop the resting spore stage of the fungus. Conidia and resting spores never occur within the same individuals. Resting spores, like conidia, are produced only within the abdomen and are ochre to brown in color with a powdery consistency (Goldstein, 1929; Spear, 1921). As the infection progresses, the abdomen becomes swollen and distended, and intersegmental membranes between the abdominal sclerites become fragile, stretched and broken (Fig. 1). Once the sclerites separate, abdominal segments fall off and powdery resting spores are dispersed into the air as infected cicadas fly. The resting spores settle to the ground where they have been thought to remain dormant until the next cicada emergence in either 13 or 17 years (Lloyd et al., 1982; Samson et al., 1988; Soper, 1974). Generally it has been considered that M. cicadina has the longest life cycle of any fungus (Tanada and Kaya, 1993). Little is known about the conditions necessary for survival and germination of resting spores or other aspects of the fungal life cycle, largely because the hosts only emerge every 13 or 17 years in a given area. Because of the long host life cycle, some authors have
Fig. 1. Periodical cicadas infected with the resting spore stage of M. cicadina can be easily recognized in the field by bending a cicada’s abdomen. Infected cicadas break open easily and the cake-like mass of ochre-colored resting spores is obvious.
L. Duke et al. / Journal of Invertebrate Pathology 80 (2002) 1–6
stated that mechanisms which trigger germination of M. cicadina resting spores will never be determined (Evans, 1989). Laboratory attempts to germinate resting spores have been unsuccessful (White and Lloyd, 1983). We present evidence that the hypothesis that M. cicadina resting spores remain dormant for 13 or 17 years and are closely synchronized with periodical cicada emergences may be incorrect. Our evidence suggests that M. cicadina resting spores can germinate and infect periodical cicadas whenever they contact an eclosing cicada independent of species or whether the host is from a 13- or 17-year emergence. To further understand the life cycle of M. cicadina, data were collected on prevalence of the fungus among an emerging 13-year periodical cicada population in Arkansas in 1988 that had been exposed to resting spores <1 year old. In addition, the M. cicadina resting spores used for this study were collected in 1997 from a 17-year M. septendecim population in Iowa, which is a different species from the 13-year M. tredecassini in Arkansas.
2. Materials and methods 2.1. Collection of resting spore stage M. cicadina in Iowa During the 1997 emergence of 17-year M. septendecim (Brood IV) in southern Iowa, live adult cicadas were collected on 4 and 5 July 1997, during the late phase of the adult life cycle (approximately day 30 of the cicada breeding season). The prevalence of resting spore stage M. cicadina infections in the cicada populations in three sites in Iowa was determined. The three sites were: Stephens State Forest and Red Haw State Park in Lucas County, and Nine Eagles State Park in Decatur County. Areas sampled were selected because of high cicada densities and accessibility. Cicadas were collected early in the morning by hand or net from low shrubs and herbaceous plants bordering woods and placed in groups of 50 in paper bags and kept in coolers. An effort was made to collect every live cicada possible whether active or moribund. In the laboratory each cicada was sexed then bisected with a razor blade to determine if the resting spore stage of M. cicadina was present (Fig. 2). 2.2. Number of M. cicadina resting spores per cicada The number of resting spores produced per infected male M. septendecim cicada was determined by counting the number of resting spores contained within five infected male cicadas with fully intact abdomens collected from the Iowa emergence. The infected individuals chosen for analysis were ‘‘ripe,’’ i.e., their abdomens were packed with mature powdery resting spores and the abdomen was fragile and about to rupture (Fig. 1).
3
Abdomens were removed and separately placed in 20 ml of deionized water containing 0.01% Tween 80 (FisherScientific, Fair Lawn, NJ 07410) in a 125 ml flask then vortexed for 2 min on the high setting. Four subsamples of each spore suspension were counted to determine the number of resting spores using a phase microscope at 200 with a hemacytometer, following the method of Cantwell (1970). 2.3. Infectivity of Iowa resting spores for Arkansas cicadas A field site on the wooded edge of a cattle pasture in Durham, Washington County, Arkansas, where 13-year periodical cicadas had emerged in 1985 (Williams et al., 1993) was chosen for the experiment. Experimental plots were laid out beneath areas in which the tree and shrub limbs showed intense cicada oviposition damage from the 1985 emergence. Within a total area of 15 m2 , vegetation was removed from the soil surface of 10 plots, each measuring 0:5 m2 . Five of the plots were randomly assigned as treatment plots and each was sprayed with 1 liter of fungal slurry prepared using 1,000,000 resting spores/liter of distilled water. The resting spores were collected from infected male 17-year M. septendecim in 1997 in Iowa as described above and stored in a refrigerator 11 days before use. The resting spores were harvested and quantified as above. The five control plots were sprayed with 1 liter of distilled water only. Each plot was evenly sprayed on 16 July 1997 with a CO2 powered backpack sprayer with a single ConeJet TSS 6 SX nozzle (Spraying Systems, Wheaton, IL 60189) and marked for future identification. The following spring, on April 28, 1998, prior to cicada emergence, traps made of fine mesh (described in Williams et al., 1993) were placed over each individual study plot enabling collection of all emerging cicadas within those plots. Emergence of Brood XIX M. tredecassini began in Durham, Washington, AR, on 14 May 1998 with peak emergence on 21 May 1998 (unpublished data). Although all three species of 13-year periodical cicadas occur on the study site, only M. tredecassini were used in this study, because it was the most abundant of the three species of periodical cicadas at this emergence site (Williams and Smith, 1991). Collection of cicadas began on the first day of emergence, 14 May, and emergence traps were also checked on 16, 17, 20, 21, 22, 24, and 26 May. Wings of all emerged cicadas within each study plot were marked using a permanent marker pen, indicating trap number, treatment or control, and date. The cicadas were placed in a mesh tent (2 m 2 m 2:5 m) containing woody vegetation. Each cicada within the tent was removed and checked daily for visual signs of the conidial stage of M. cicadina. In addition, the natural cicada population at Durham was sampled by randomly collecting 100 periodical
4
L. Duke et al. / Journal of Invertebrate Pathology 80 (2002) 1–6
Fig. 2. Comparison of bisected M. septendecim periodical cicadas collected in Iowa in 1997. (a) Uninfected female, ov, ovaries; (b) uninfected male, note that a healthy male abdomen is almost empty and serves as a resonating chamber for its mating song; (c) infected male with abdomen full of M. cicadina resting spores (rs).
cicadas by hand from vegetation on 14, 16, 17, 20, 21, 22, 24, and 26 May from an area within 60 m of the experimental area. Collected cicadas were marked as above and placed in an identical but separate holding tent. As above, each periodical cicada was checked daily for visual signs of M. cicadina.
determining whether there was a difference between the incubation period of M. cicadina with regards to gender.
3. Results 3.1. Collection of resting spore stage M. cicadina in Iowa
2.4. Statistical analyses A v2 goodness-of-fit test was used to determine if differences existed between numbers of cicadas infected with M. cicadina in treatment plots compared to control plots. v2 was also used to determine if M. cicadina was gender biased in conidial stage during the 1998 emergence of Brood XIX, and resting spore stage in the 1997 Brood IV emergence. A t test (two sample for variance) was used in
A total of 431 M. septendecim (83.1% males and 16.9% females) were collected from Stephens State Forest (Lucas County). Of these 7.3% of the males were infected with the resting spore stage of M. cicadina and no females were infected. A total of 249 M. septendecim (76.7% males and 23.3% females) were collected at Red Haw State Park (Lucas County). Of these 2.6% of the males and 3.4% of the females were infected with the
L. Duke et al. / Journal of Invertebrate Pathology 80 (2002) 1–6
resting spore stage of M. cicadina. A total of 372 (82.8% males, 17.2% females) cicadas were collected from Nine Eagles State Park (Decatur County). Of these 16.9% of the males were infected with the resting spore stage of M. cicadina while no females were found infected. Overall, 1052 M. septendecim cicadas (81.5% male and 18.5% female) were captured in southern Iowa of which 9.7% of the males and 1.0% of the females were infected with the resting spore stage of M. cicadina. Brood IV M. septendecim, collected from southeastern Iowa 4–5 July 1997 showed a significant difference between males and females infected (v2 ¼ 94.87, df ¼ 1, P ¼ 0:0001) with males constituting 98% of total infected. A mean of 1:4 106 ðSE ¼ 1:8 105 Þ resting spores were produced per infected male M. septendecim. 3.2. Infectivity of Iowa resting spores for Arkansas cicadas A total of 346 cicadas were collected from the 10 study plots. The number of emerged cicadas was not significantly different between treatment (189 cicadas) and control (157 cicadas) plots (v2 ¼ 2.96, df ¼ 1, P ¼ 0:085). A total of 20 individuals were infected with the conidial stage of M. cicadina from the resting spore treated plots with equal distribution between all treatment plots (v2 ¼ 2.6, df ¼ 4, P ¼ 0:6271). Only one infected cicada emerged from control plots resulting in a significant difference between control and treatment plots (v2 ¼ 21.1, df ¼ 1, P ¼ 0:0001). These numbers indicate a 10.6% prevalence of conidial stage M. cicadina infections in emerging populations of periodical cicadas in plots treated with resting spores versus a 10.6% prevalence from cicadas in the control plots. Of the 21 infected cicadas, 15 (71%) were males and 6 (29%) were females, resulting in a significant difference with regard to gender (v2 ¼ 3.98, df ¼ 1, P ¼ 0:046). No cicadas were found infected with conidial stage M. cicadina infections in the natural population based on a sample of 800 individuals collected over a period of 13 days. At emergence, four cicadas showed visible signs of M. cicadina conidial stage infection. The mean period of time before which adult cicadas showed visual signs of conidial stage M. cicadina infections was 1.25 days after emergence with no significant difference with regard to gender (t test ¼ )0.243, df ¼ 19, P ¼ 0:81). Visible signs of conidial stage M. cicadina infections were evident in all infected cicadas by the 14th day into the breeding period, with peak numbers occurring on the 8th and 9th days.
4. Discussion Results of our study demonstrate that: (1) resting spores of M. cicadina collected from a 17-year cicada
5
species in Iowa were infective to a 13-year cicada species in Arkansas, (2) resting spores collected from M. septendecim were infective to a separate species, M. tredecassini, (3) resting spores are capable of infecting cicadas after only one year of dormancy, (4) each infected male M. septendecim produced about 1:4 106 resting spores, and (5) significantly more male than female cicadas were infected with the resting spore stage of M. cicadina. We believe that this is the first experimental study on the infectivity of M. cicadina resting spores and the first quantification of the number of resting spores produced by individual cicadas. We found no evidence that M. cicadina resting spores have a 13- or 17-year dormant period in synchrony with periodical cicadas and we hypothesize that M. cicadina resting spores may be capable of infecting periodical cicadas whenever they contact eclosing periodical cicadas. The prevalence of conidial stage M. cicadina infections varies among populations of periodical cicadas, but rarely exceeds 10% (White and Lloyd, 1983). Conidial-stage M. cicadina infections were very low in the Durham, Arkansas, cicada population in 1985 (Williams et al., 1993) and 1998 (this study), so the infection rate of 10.6% from cicadas collected from our five resting sporetreated plots appears to be a meaningful increase in prevalence. White and Lloyd (1983) suggested that emerging cicadas encounter resting spores randomly, and therefore, an equal sex ratio of cicadas infected with the conidial stage of M. cicadina conidia might be expected. However, we found significantly more males infected with the resting spore stage of M. cicadina at the end of the emergence in Iowa in 1997 and the conidial stage at the beginning of the emergence in 1998 in Arkansas. Emergence of periodical cicadas is protandrous, with most males emerging quickly in a matter of days, while female emergence occurs somewhat later and is more prolonged (Williams et al., 1993). Given that males emerge first, typically in large numbers (e.g., over 1,000,000/ha, [Dybas and Davis, 1962]), males would encounter resting spores before females, perhaps leading to a higher infection rate among males. Also, males form large chorus centers during mating (Williams and Smith, 1991), so that males are constantly coming into contact with other males (Alexander and Moore, 1962), including infected males shedding conidia. This may account for the higher percentage of males infected with the resting spore stage. Females tend to be more solitary therefore the possibility of females becoming infected through direct contact is reduced. Females are thought to mate only once and English (2001) has recently suggested that the higher percentages of males with resting spores may be due to avoidance of infected males by females. Females use an acoustic assessment for large size while courting males during the early part of the breeding period (English, 2001). Conidia or resting
6
L. Duke et al. / Journal of Invertebrate Pathology 80 (2002) 1–6
spores fill the abdomens of infected males (Fig. 2C), altering the pitch of their song, resulting in infected males sounding smaller than they actually are. Thus, the higher prevalence in males may be a consequence of both greater male aggregation and female avoidance of infected males. Little has been known about the ecological requirements necessary for survival and germination of M. cicadina resting spores. Fungal spores often need specific environmental conditions of relative humidity and temperature to survive and germinate (Steinkraus and Slaymaker, 1994). Resting spores may require a dormant period before they can germinate. For instance, Perry and Latge (1982) found that Conidiobolus obscurus resting spores required a 3-month cold period before they could germinate. We found that M. cicadina resting spores collected in July 1997, then applied to soil, were able to infect cicadas 10 months later in May 1998. Apparently 10 months and one winter period were a sufficient dormant period for some M. cicadina resting spores to infect emerging periodical cicadas in 1998. Therefore, while it is evident that M. cicadina resting spores are able to survive for up to 17 years in the soil, they do not require a 13 or 17 year dormant period, and are not as closely synchronized with periodical cicada life cycles as previously thought.
Acknowledgments We acknowledge Chris Simon for advice, and Gabriele Boys, Ping Li, and Shauna Ginger for field assistance and Mike Cassidy for access to his property. The David Causey Grant-in-Aid awarded from the Department of Biological Sciences, University of Arkansas provided partial support.
References Alexander, R.D., Moore, T.E., 1962. The evolutionary relationships of 17-year and 13-year cicadas, and three new species (Homoptera, Cicadidae, Magicicada). Misc. Publ. Mus. of Zool., Univ. Michigan. Tej eikle, vol. 121, pp. 1–59. Cantwell, G.E., 1970. Standard methods for counting Nosema spores. Am. Bee J. 110, 222–223. Dybas, H.S., Davis, D.D., 1962. A population census of 17-year periodical cicadas (Homoptera: Cicadidae: Magicicada spp.). Ecology 43, 432–444. English, J.E., 2001. Sexual selection in the lek-mating system of periodical cicadas Magicicada cassini and M. decim; selection for male body size, dominant courtship call and pathogen avoidance. Doctoral dissertation, University of Arkansas, Fayetteville.
Evans, H.C., 1989. Mycopathogens of insects of epigeal and aerial habitats. In: Wilding, N., Collins, N.M., Hammond, P.M., Webber, J.F. (Eds.), Insect–Fungus Interactions. Academic Press, London, pp. 205–238. Goldstein, B., 1929. A cytological study of the fungus Massospora cicadina, parasitic on the 17-year cicada, Magicicada septendecim. Am. J. Bot. 16, 394–401. Leidy, J., 1851. Fungus disease of Cicada septendecim. Proc. Acad. Nat. Sci. Phila. 5, 235. Lloyd, M., Dybas, H.S., 1966. The periodical cicada problem. II Evolution. Evolution 20, 466–505. Lloyd, M., White, J., Stanton, N., 1982. Dispersal of fungus-infected periodical cicadas to new habitat. Environ. Entomol. 11, 852–858. Marshall, D.D., Cooley, J.R., 2000. Reproductive character displacement and speciation in periodical cicadas, with a description of a new species, 13-year Magicicada neotredecim. Evolution 54, 1313– 1325. Peck, C., 1879. Massospora gen. Nov. 31st. Ann. Rept. New York State Museum Nat. Hist., 44. Peck, C., 1907. Massospora cicadina n.g. et sp. Ann. Rept. New York State Botanist 31, 44–48. Perry, D.F., Latge, J.-P., 1982. Dormancy and germination of Conidiobolus obscurus azygospores. Trans. Br. Mycol. Soc. 78, 221–225. Samson, R.A., Evans, H.C., Latge, J.P., 1988. Atlas of Entomopathogenic Fungi. Springer, Berlin. Simon, C., Tang, J., Dalwadi, S., Staley, G., Deniega, J., Unnasch, T.R., 2000. Genetic evidence for assortative mating between 13year cicadas and sympatric 17-year cicadas with 13-year life cycles provides support for allochronic speciation. Evolution 54, 1326– 1336. Soper, R., 1963. Massospora levispora, a new species of fungus pathogenic to the cicada, Okanagana rimosa. Can. J. Bot. 41, 875– 878. Soper, R., 1974. The Genus Massospora, entomopathogenic for cicadas, Part I, Taxonomy of the Genus. Mycotaxon 1, 13–40. Soper, R.S., Smith, L.F.R., Delyzer, A.J., 1976. Epizootiology of Massospora levispora in an isolated population of Okanagana rimosa. Ann. Entom. Soc. Am. 69, 275–283. Spear, A.T., 1921. Massospora cicadina Peck, a fungus parasite of the periodical cicada. Mycologia 13, 72–82. Steinkraus, D.C., Slaymaker, P.H., 1994. Effect of temperature and humidity on formation, germination, and infectivity of conidia of Neozygites fresenii (Zygomycetes: Neozygitaceae) from Aphis gossypii (Homoptera: Aphididae). J. Invertebr. Pathol. 64, 130–137. Tanada, Y., Kaya, H.K., 1993. Insect Pathology. Academic Press, San Diego. White, J., Lloyd, M., Zar, J.H., 1979. Faulty eclosion in crowded suburban periodical cicadas: populations out of control. Ecology 60, 305–315. White, J., Lloyd, M., 1983. A pathogenic fungus, Massospora cicadina Peck (Entomophthorales), in emerging nymphs of periodical cicadas. (Homoptera: Cicadidae). Environ. Entomol. 12, 1245– 1252. Williams, K.S., Simon, C., 1995. The ecology, behavior, and evolution of periodical cicadas. Ann. Rev. Entomol. 40, 269–295. Williams, K.S., Smith, K.G., 1991. Dynamics of periodical cicada chorus centers (Homoptera: Cicadidae: Magicicada). J. Insect Behav. 4, 275–291. Williams, K.S., Smith, K.G., Stephen, F.M., 1993. Emergence of 13year periodical cicadas (Cicadidae: Magicicada): phenology, mortality, and predator satiation. Ecology 74, 1143–1152.