Journal of Invertebrate Pathology 93 (2006) 105–113 www.elsevier.com/locate/yjipa
Microsporidian infections in Lymantria dispar larvae: Interactions and eVects of multiple species infections on pathogen horizontal transmission Daniela K. Pilarska a, Leellen F. Solter b, Manana Kereselidze c, Andreas Linde d, Gernot Hoch e,¤ a
Bulgarian Academy of Sciences, Institute of Zoology, 1 Tsar Osvoboditel Blvd., 1000 SoWa, Bulgaria b Illinois Natural History Survey, 1816 S. Oak St., Champaign, IL 61820, USA c Georgian Academy of Science, V. Gulisashvili Institute of Mountain Forestry, Mindely Str. 9, 0186 Tbilisi, Georgia d University of Applied Sciences, Alfred Moeller Str. 1, D-16225 Eberswalde, Germany e BOKU University of Natural Resources and Applied Life Sciences, Department of Forest and Soil Sciences, Hasenauerstrasse 38, A-1190 Vienna, Austria Received 17 February 2006; accepted 12 May 2006 Available online 30 June 2006
Abstract The interactions in multiple species infections and eVects on the horizontal transmission of three microsporidian species, Vairimorpha disparis, Nosema lymantriae and Endoreticulatus schubergi, infecting Lymantria dispar were evaluated in the laboratory. Simultaneous and sequential inoculations of host larvae were performed and the resulting infections were evaluated. Test larvae were exposed to the inoculated larvae to measure horizontal transmission. Dual species infections demonstrated interspeciWc competition between Nosema and Vairimorpha in the host larvae, but no observable competition occurred between Endoreticulatus and either of the other microsporidian species. Timing of inoculation was an important factor determining the outcome of competition between Nosema and Vairimorpha. The species inoculated Wrst showed a higher rate of successful establishment; a time lag of 7 days between inoculations allowed the Wrst species to essentially exclude the second. The microsporidia diVered in eYciency of horizontal transmission. Nosema and Endoreticulatus were transmitted at very high rates, close to 100%. Horizontal transmission of Vairimorpha was less eYcient, ranging from 25% to a maximum of 75%. The patterns of infection observed in inoculated larvae were reXected in the test larvae that acquired infections in the horizontal transmission experiments. Competition with Vairimorpha suppressed horizontal transmission of Nosema after simultaneous and sequential inoculation. In simultaneous inoculation experiments Endoreticulatus had no eVect on transmission of Nosema and Vairimorpha. © 2006 Elsevier Inc. All rights reserved. Keywords: Vairimorpha disparis; Nosema lymantriae; Endoreticulatus schubergi; Lymantria dispar; Co-infection; Horizontal transmission; InterspeciWc competition
1. Introduction The gypsy moth, Lymantria dispar (L.), is known to host a complex of entomopathogenic microsporidia that includes three genera, Endoreticulatus, Nosema and Vairimorpha. Vairimorpha disparis, a fat body parasite, is
*
Corresponding author. Fax: +43 1 3686352 97. E-mail address:
[email protected] (G. Hoch).
0022-2011/$ - see front matter © 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.jip.2006.05.003
primarily transmitted after release of mature, infective, environmentally resistant spores (“environmental spores”) from host cadavers and is highly virulent relative to other lepidopteran microsporidia (Solter et al., 1997; Vavra et al., in press). Endoreticulatus spp. are midgut parasites and are less virulent than Vairimorpha isolates, but are readily transmitted via release of environmental spores in the feces throughout the host larval period (Weiser, 1966). The Nosema group, which includes at least three species in L. dispar, infects the fat body, silk glands and the
106
D.K. Pilarska et al. / Journal of Invertebrate Pathology 93 (2006) 105–113
Malpighian tubules. Environmental spores are probably released from living larvae with feces and silk, as well as from cadavers (Maddox et al., 1999; Goertz et al., 2004). There are few records of naturally occurring mixedspecies microsporidian infections in Lepidoptera (Weiser et al., 1969; Solter et al., 2002). Most report Endoreticulatus mixed with either Nosema or Vairimorpha species (SmirnoV, 1965; Wilson, 1975; Wilson and Burke, 1978; Pilarska et al., 2001). Solter et al. (2002) hypothesized that microsporidia may compete for hosts and that as a result, one species may eliminate another in a host population or prevent establishment in a biological control program. They studied the impact of mixed species infections in order to address potential competition between species in a host population. Three species of microsporidia isolated from Bulgarian populations of L. dispar, Endoreticulatus schubergi, Nosema lymantriae and Vairimorpha disparis, were evaluated in laboratory bioassays in which all possible pairwise combinations were administered either simultaneously or sequentially. The authors determined that the addition of Endoreticulatus sp. to infections of N. lymantriae or V. disparis increased development time of the host larva, and they also demonstrated antagonism between Nosema and Vairimorpha. They suggested that the L. dispar microsporidia, particularly Nosema and the closely related Vairimorpha, which both utilize fat body tissues, may compete in L. dispar populations. Indeed, despite the large number of L. dispar populations surveyed since 1985 (McManus and Solter, 2003) and the variety of microsporidia in this host insect, the only L. dispar population studied in Europe found to harbor mixed species infections is one population in Hungary parasitized by Nosema and Endoreticulatus (Solter and Pilarska, unpublished data). The three microsporidian species studied by Solter et al. (2002) probably utilize diVerent transmission pathways according to their primary sites of infection in the host (Maddox et al., 1998). The relative success of the diVerent pathways could be aVected by the presence of a second, competing parasite. In this study, we tested the hypothesis that interspeciWc competition among microsporidia utilizing an individual host leads to altered establishment of the pathogens and, consequently, to diVerential horizontal transmission to susceptible hosts. We tested this in experiments with simultaneous and sequential infections. The latter were conducted using two experimental protocols: (1) with administration of the second species during the early dispersal phase of the Wrst species in the host tissues (3 days) and (2) with a suYcient time period between per os inoculation events to allow establishment of the Wrst microsporidian species before administration of the second (7 days). Infections produced in the treated larvae were microscopically examined to evaluate competition after controlled oral inoculation. Susceptible (untreated) larvae were exposed to infected larvae under what can be considered a “maximum challenge” situation (Solter and Maddox, 1998) to demonstrate the eVect of interspeciWc competition among microsporidia on horizontal transmission.
2. Materials and methods 2.1. Insects and microsporidia Lymantria dispar were obtained from egg masses provided by the USDA, APHIS laboratory at Otis Air Force Base, MA. Larvae were hatched and reared on meridic wheat germ diet (Bell et al., 1981) in 250-ml plastic cups at 21 § 1 °C, 16 h light/8 h dark. Inoculum of the same isolates of three microsporidian species that were studied by Solter et al. (2002), were obtained from the collection stored in liquid nitrogen at BOKU University, Vienna, Austria: Nosema lymantriae (Isolate No. 1996-A, GenBank Accession No. AF141129), Vairimorpha disparis (Isolate No. 1995-D, GenBank Accesion No. DQ272237) and Endoreticulatus schubergi (Isolate No. 1996-B, GenBank Accession No. AY502945). The original isolates were recovered from L. dispar collected in three diVerent sites in Bulgaria where only one species occurred in each site (Pilarska et al., 1998). V. disparis infects the fat body tissues and produces two types of mature, environmentally resistant spores, diplokaryotic single spores and monokaryotic octospores. The closely related N. lymantriae infects the silk glands and the fat body tissues and produces only diplokaryotic spores. E. schubergi develops only in the midgut tissues of the larval host, produces small monokaryotic spores in parasitophorous vesicles. For the purposes of this paper the three microsporidia will be referred to as Vairimorpha, Nosema and Endoreticulatus. Spores for the experiments were propagated in L. dispar larvae (Hoch et al., 2000; Solter et al., 2002). Mature spores were harvested from infected tissues 20 days post-inoculation (dpi), then cleaned by Wltration through cellulose tissue and centrifugation. Spore suspension in distilled water was mixed 1:1 with glycerol and stored in liquid nitrogen (Maddox and Solter, 1996) until used in the experiments, but no longer than 2 months. Lymantria dispar larvae were initially inoculated on the second day post-molt to third instar following the method of Bauer et al. (1998), as used in previous studies (Hoch et al., 2000; Solter et al., 2002). Microsporidian spore suspensions were removed from liquid nitrogen storage, thawed, counted in a Bürker-Türk hemacytometer and adjusted to the required concentration with distilled water. Blocks of wheat germ diet cut to 4 mm3 were placed individually into the wells of Corning® tissue culture 24-well plates and 1 l of spore suspension was applied to the surface of each diet block. L. dispar larvae that had been starved overnight were placed individually into each well. Each larva that consumed the entire diet block within 16 h was used in the experiments. Controls were treated in the same manner but diet blocks were initially inoculated with distilled water. 2.2. Transmission experiments with simultaneous infection with Nosema, Vairimorpha and Endoreticulatus In order to evaluate the eVects of co-infection on the transmission of microsporidia, experiments utilizing
D.K. Pilarska et al. / Journal of Invertebrate Pathology 93 (2006) 105–113
simultaneous inoculations were conducted. Infections were produced using a total dosage of 1 £ 103 spores (5 £ 102 spores of each isolate to produce mixed infections) as follows: Nosema (N), Vairimorpha (V), Endoreticulatus (E), Nosema + Vairimorpha (N + V), Nosema + Endoreticulatus (N + E), Vairimorpha + Endoreticulatus (V + E). Dosages were based on the results of mixed infection studies conducted by Solter et al. (2002) in which a dosage of 2.5 £ 103 spores produced 100% infection and was above the LD50 for Vairimorpha and Nosema. In this study, we chose the dose in order to maximize infection rates but minimize mortality during the time of the transmission experiment for all three species. At 14 dpi, Wve uninfected second instar larvae (hereafter termed test larvae), were exposed to Wve larvae that were inoculated with microsporidia (hereafter termed inoculated larvae) in 250-ml cups on wheat germ diet. Inoculated and test larvae were held together in the cups for 7 days (i.e., until 21 dpi). After this period of exposure, test larvae were reared individually for another 12 days to allow acquired infections to mature while inoculated larvae were immediately frozen and stored in at ¡20 °C until they were microscopically examined. Three trials were conducted for a total of 15 rearing containers per treatment. 2.3. Transmission experiments with sequential infection with Nosema and Vairimorpha Data from the experiments in which insects were simultaneously inoculated with two microsporidian species suggested that competition between Nosema and Vairimorpha aVected transmission, while Endoreticulatus was always transmitted at a high rate. Moreover, previous studies revealed that Endoreticulatus is not excluded from the host even when it is administered 7 days after inoculation with Nosema or Vairimorpha (Solter et al., 2002). Therefore, we focused on experiments in which Nosema and Vairimorpha were sequentially administered (see Table 1 for details of experimental design). In the Wrst experiment, the time diVerence was 3 days, a period when the Wrst microsporidium has begun to infect target tissues of the infection bud mature spores have not begun to form (Solter and Maddox, 1998). The dosages used in these experiments were 1 £ 103 spores Table 1 Treatments (oral inoculation) in sequential infection experiments Inoculation 1 Second day in L3
Inoculation 2 3 dpi or 7 dpi
Code
Nosema Vairimorpha Water Water Nosema + Vairimorpha Nosema Vairimorpha Water
Water Water Nosema Vairimorpha Water Vairimorpha Nosema Water
N/Wa V/W W/N W/V N + V/W N/V V/N W/W
a Forward slash indicates “followed by”, e.g., N/W, Nosema followed by water.
107
of each species except for the V + N/W treatment in which 5 £ 102 spores of each species were fed. The methods for inoculating larvae were the same as in the studies of simultaneous infections. In order to distinguish inoculated larvae from test larvae in sequential treatment experiments, the inoculated larvae were marked at 2 dpi by clipping one proleg. Fourteen days after the Wrst inoculation and 11 days after the second inoculation, Wve uninfected test larvae, staged at 1 day post molt to third instar, were exposed to the Wve inoculated larvae. The larvae were reared together for 7 days (i.e., until 21 dpi with the Wrst species) in a 250-ml plastic cup containing wheat germ diet. The inoculated larvae were then removed and stored at ¡20 °C for microscopic examination. The test larvae were transferred into 40-ml diet cups and reared individually for 11 days. Test larvae were then stored at ¡20 °C until microscopic examination. Three trials were conducted for a total of 24 rearing containers per treatment. In a second experiment using sequentially produced infections, we studied the eVect of a longer time period between inoculations with the two microsporidian species. In this experiment the second species was administered to L. dispar larvae 7 days after the Wrst species (Table 1), a time period suYcient to allow establishment of the Wrst microsporidian species. Other than the extended time between inoculations, the same treatments and the same dosages were used in this experiment. Fourteen days after the Wrst inoculation (7 days after the second inoculation), Wve uninfected test larvae staged at 1 day post-molt to third instar were exposed to Wve inoculated larvae. Treated and test insects were reared together for 7 days (i.e., until 21 dpi with the Wrst species) in 250-ml plastic cups. The inoculated larvae were then removed and stored at ¡20 °C until microscopic examination. The test larvae were transferred to 40-ml diet cups and reared individually for 11 days. All test larvae were frozen and stored at ¡20 °C for microscopic examination. Eight rearing containers per treatment were set up for this experiment. 2.4. Evaluation of infections For microscopic evaluation, fresh tissue smears were prepared on glass slides. The smears were examined under phase contrast microscopy (400£), which allowed detection of spores as well as immature stages of microsporidia (Solter and Maddox, 1998). Larvae that were potentially infected with both Nosema and Vairimorpha were dissected and fresh preparations of silk glands, fat body, midgut and Malpighian tubules were examined. In all other cases, cross sections, which included silk gland, fat body, midgut and Malpighian tubule tissues were examined (Solter et al., 1997). The two microsporidian species were determined as follows: The presence of immature or mature octospores in the fat body was used as the criterion for Vairimorpha infection; the presence of mature diplokaryotic spores in the silk glands or Malpighian tubules indicated a Nosema
108
D.K. Pilarska et al. / Journal of Invertebrate Pathology 93 (2006) 105–113
infection (Solter et al., 2002). Endoreticulatus was easily distinguished by the much smaller spore size and vesicles containing more than eight spores; moreover, it is restricted to the midgut tissue (Weiser, 1961; Cali and El Garhy, 1991). 2.5. Statistical analysis Statistical analyses were carried out with SPSS 12.0 for Windows (SPSS Inc.). Infection rates, based on number of individuals acquiring infections in individual rearing containers (replicates), were compared between treatments using the Kruskal–Wallis H test. Rearing containers in which not all inoculated larvae were infected by at least one pathogen species were excluded. SigniWcant H tests were followed up by pairwise comparison with Mann–Whitney U test (controlled for type I errors by the Bonferroni method). Relative frequencies of infection in inoculated larvae (based on total number of larvae) were compared using 2 analysis. The relationship between mortality and infection was evaluated by computing Spearman’s correlation coeYcient. 3. Results 3.1. Simultaneous infection with Nosema, Vairimorpha and Endoreticulatus Results of the experiment on simultaneous infections are presented in Fig. 1 and Table 2. Dual infections did not prevent establishment of any of the administered microsporidia in inoculated larvae; at least 90% of inoculated larvae were diagnosed with mixed infections in the dual species treatments. Transmission of Nosema and Vairimorpha in single infections (median of 4 test larvae infected with Nosema of 5 test insects per cup; median of 2 test larvae infected with Vairimorpha) was lower than the transmission of Endoreticulatus (5 of 5 test larvae infected). There was signiWcantly lower transmission of Nosema to the test larvae when inoculated larvae were challenged with Vairimorpha + Nosema compared to Nosema alone or Nosema + Endoreticulatus (Table 2). Transmission of Endoreticulatus and Vairimorpha was not inXuenced by the mixed infections.
Fig. 1. Establishment of infections with Nosema, Vairimorpha and Endoreticulatus in L. dispar larvae after simultaneous oral inoculation (above) and infections after transmission to test larvae (below). Columns represent percent of larvae combined from 9 to 15 rearing containers (total of 45–70 larvae) per treatment. Larvae were inoculated with 1 £ 103 spores on second day in third instar. Test larvae were exposed to inoculated larvae from 14 to 21 dpi. Treatments: N, Nosema; V, Vairimorpha; E, Endoreticulatus; N + V D Nosema + Vairimorpha; N + E D Nosema + Endoreticulatus; V+ E D Vairimorpha + Endoreticulatus.
Table 2 Comparison of prevalence of infection in test larvae after horizontal transmission from inoculated larvae (experiments with simultaneous inoculations) Species
Prevalencea
Kruskal–Wallis H testb
Nosema
N N+V N+E
4a 2b 4a
2 D 11.020 df D 2 P D 0.004
Vairimorpha
V N+V V+E
2 2 1
2 D 2.727 df D 2 P D 0.256
Endoreticulatus
E N+E V+E
5 5 5
2 D 2.205 df D 2 P D 0.332
3.2. Sequential infections with Nosema and Vairimorpha For studies of sequential infections, we evaluated in detail the establishment of infection in inoculated larvae. The results from the Wrst experiment evaluating sequential infections with Nosema and Vairimorpha are presented in Fig. 2. Timing of inoculations was very important, however, the 3-day time period between inoculation events did not lead to complete exclusion of the species that was administered second. When Vairimorpha was administered 3 days before Nosema, the percentage of Vairimorpha-only infection in inoculated insects was 85.5%. When Nosema was administered Wrst, mixed infections dominated and reached
Treatments
c
a
Median number of infected test larvae per rearing container. Based on individual rearing containers (n D 9, ƒ , 15). c DiVerent letters indicate signiWcant diVerences between treatments based on pariwise Mann–Whitney U tests (corrected for type I errors by Bonferroni method) following up signiWcant H test. b
91.6%. Nosema infection rates in inoculated larvae diVered between treatments (2 D 82.0, df D 4, P 6 0.001) while no such diVerences were determined for Vairimorpha (2 D 0.35, df D 4, P > 0.05).
D.K. Pilarska et al. / Journal of Invertebrate Pathology 93 (2006) 105–113
Fig. 2. Establishment of infections with Nosema and Vairimorpha in L. dispar larvae after sequential oral inoculation with a 3 day time period between inoculations (above) and infections after transmission to test larvae (below). Columns represent percent of larvae combined from 19 to 24 rearing containers (total of 95–120 larvae) per treatment. Larvae were inoculated with 1 £ 103 spores on second day in third instar and again 3 dpi. Test larvae were exposed to inoculated larvae from 14 to 21 days after inoculation with the Wrst species. Refer to Table 1 for experimental design and codes of treatments.
In the 3-day sequential inoculation experiment, Nosema infections were transmitted to nearly 100% of test larvae. Transmission of Vairimorpha infections was lower; a median of 4 of 5 test larvae were infected in the Vairimorpha/water treatments and 2.5 of Wve were infected in water/ Vairimorpha treatments. This diVerence was not statistically signiWcant (Table 3).
109
Transmission of Vairimorpha to test larvae did not diVer signiWcantly between single-species treatments and mixedspecies treatments (Table 3). Transmission of Nosema, however, was negatively aVected by competition with Vairimorpha. Test larvae showed signiWcantly lower prevalence of infection with Nosema when inoculated larvae had mixed infections compared to Nosema-only infections. The lowest transmission for Nosema (median of 0 of 5 test larvae) occurred in test larvae exposed to larvae inoculated with Vairimorpha Wrst. Test insects that were exposed to larvae that were inoculated with Nosema three days before Vairimorpha showed a high rate of mixed infections (56.8%). A signiWcant positive correlation was determined between mortality of inoculated hosts and transmission of Vairimorpha to test larvae (Spearman’s D 0.495, P < 0.001); a signiWcant but weaker negative correlation was determined for Nosema (Spearman’s D ¡0.335, P < 0.001) (Fig. 3). A 7-day time period between the two successive inoculations led to nearly complete exclusion of the second species administered (Fig. 4). In contrast, simultaneous infection with Nosema and Vairimorpha followed by a water treatment (Nosema + Vairimorpha/water) did not lead to signiWcant reduction of establishment of either species (Nosema: 2 D 0.076, df D 3, P > 0.05; Vairimorpha: 2 D 6.06, df D 3, P > 0.05). Mixed infections occurred in 55% of inoculated larvae. Even when established, however, the infection level of Nosema was clearly reduced in 77% of mixed infections as indicated by incomplete infestation of the silk glands. When administered 7 days prior to Vairimorpha, Nosema was transmitted to all test larvae. Transmission of Vairimorpha was overall lower; the median was one test larva infected in the Vairimorpha/water treatment and 1.5 in Vairimorpha/ Nosema treatment. For Nosema + Vairimorpha/water, the transmission of Vairimorpha was even lower, although not signiWcantly diVerent from Vairimorpha-only treatment (Table 3). The transmission of Nosema was not signiWcantly lower in Nosema + Vairimorpha/water treatment compared
Table 3 Comparison of prevalence of infection in test larvae after horizontal transmission from inoculated larvae (experiments with sequential inoculations) 3 days diVerence Species
7 days diVerence Treatments
Prevalencea
Kruskal–Wallis H testb
Species
Treatments
Prevalencea
Kruskal–Wallis H testc
Nosema
N/W W/N N/V V/N V + N/W
5a 5a 4b 0d 2c
D 81.445 df D 4 P D 0.000
Nosema
N/W W/N N/V V/N V + N/W
5a 0b 5a 0b 5a
2 D 36.840 df D 4 P D 0.000
Vairimorpha
V/W W/V N/V V/N V + N/W
4 2.5 4 3 4
2 D 6.582 df D 4 P D 0.160
Vairimorpha
V/W W/V N/V V/N V + N/W
1 ab 0b 0b 1.5 a 0 ab
2 D 19.562 df D 4 P D 0.001
a
d
2
d
Median number of infected test larvae per rearing container. Based on individual rearing containers (n D 19, ƒ , 24). c Based on individual rearing containers (n D 7, ƒ , 8). d DiVerent letters indicate signiWcant diVerences between treatments based on pariwise Mann–Whitney U tests (corrected for type I errors by Bonferroni method) following up signiWcant H test. b
110
D.K. Pilarska et al. / Journal of Invertebrate Pathology 93 (2006) 105–113
Fig. 3. Correlations between number of inoculated larvae that died during the exposure period and number of infected test larvae ( D transmission) per rearing container for Vairimorpha and Nosema. Results of all relevant treatments from the experiment with sequential infections with 3 days diVerence were combined. Larger dots represent more cases.
4. Discussion
Fig. 4. Establishment of infections with Nosema and Vairimorpha in L. dispar larvae after sequential oral inoculation with a 7-day time period between inoculations (above) and infections after transmission to test larvae (below). Columns represent percent of larvae combined from 7 to 8 rearing containers (total of 35–40 larvae) per treatment. Larvae were inoculated with 1 £ 103 spores on second day in third instar and again 7 dpi. Test larvae were exposed to inoculated larvae from 14 to 21 days after inoculation with the Wrst species. Refer to Table 1 for experimental design and codes for treatments.
to one-species infection (Nosema/water). No transmission was recorded in test larvae of either microsporidian species when administered 7 days after another species or after water treatment.
Our study on dual species infections clearly indicated that competition can occur between two microsporidian parasites, Nosema and Vairimorpha, in L. dispar larvae, but competition was not evident between Endoreticulatus and either of the aforementioned species. Competition between Nosema and Vairimorpha was evident in the establishment of infections in inoculated larvae as well as in horizontal transmission to test larvae. When inoculated with two microsporidian species simultaneously, the majority of the larvae developed dual infections. However, when Nosema and Vairimorpha were administered separately, one 7 days after the Wrst, the Wrst species inoculated clearly out-competed the second. At 7 dpi, the Wrst parasite inoculated had invaded and produced mature spores in the target tissues of the host. When the second species was administered 3 days after the Wrst, at a time when the Wrst microsporidium inoculated had just begun to invade the target tissues, the outcome was more ambiguous. This experiment indicated dominance by Vairimorpha based on successful invasion of host tissues, however, the other experiments did not conWrm this dominance. Apparently the two species are rather codominant in inoculated larvae when fed simultaneously with a certain level of variability between the experiments. Competition among multiple pathogens infecting the same host has been described for diVerent entomopathogen interactions including microsporidia vs. virus (e.g., Fuxa, 1979; Cossentine and Lewis, 1984; Bauer et al., 1998), virus vs. virus (Ishii et al., 2002), virus vs. fungus (Malakar et al., 1999), and fungus vs. fungus (Thomas et al., 2003). Fuxa (1979) hypothesized that interference upon entry into the hemocoel, metabolic interference, and competition for host
D.K. Pilarska et al. / Journal of Invertebrate Pathology 93 (2006) 105–113
cells are three possible causes for antagonism between a microsporidium and a nuclear polyhedrosis virus. Such interference and exploitation competition appear to be typical of competitive interactions in multiple infections (Thomas et al., 2003). Hypothetically, Nosema and Vairimorpha could each interfere with the other during the early stage of infection since they both utilize cells of the midgut epithelia and muscularis for the Wrst reproductive cycle (Maddox et al., 1999; Vavra et al., in press). Unless spore dosages are extremely high, however, suYcient potential sites (cells) for initial reproduction of both species should be available. Metabolic interference cannot be excluded, but we believe that the observed competition between Nosema and Vairimorpha is likely due to exploitation competition because the two obligatory intracellular parasites must compete for limited resources in the host. Both species utilize the host’s fat body cells for the secondary reproductive cycle; Vairimorpha proliferation in the fat body tissues is extreme by 5 dpi (Henn and Solter, 2000). Nosema also infects the silk glands and Malpighian tubules. Vairimorpha may invade the silk glands but the developmental cycle is not completed in silk gland tissues. This marginal invasion is, however, apparently suYcient to prevent spore development of Nosema in parts of infected silk glands (Solter et al., 2002). Vairimorpha infection causes depletion of carbohydrates and lipids in the L. dispar host larva (Hoch et al., 2002), which was shown to be a reason for reduced larval development of a braconid endoparasitoid in infected hosts. The earlier L. dispar larvae are inoculated with the microsporidium relative to parasitization by the braconid, the more pronounced is the negative eVect on the parasitoid larvae (Hoch et al., 2000). A severe lack of nutrients may likewise contribute to the exclusion of a microsporidian species that invades a host after another species is established. Endoreticulatus remains in the midgut epithelium throughout its developmental cycle and could potentially interfere with the initial developmental cycle of Nosema or Vairimorpha. There is, however, no exclusion by Endoreticulatus, even when larvae are inoculated with Nosema and Vairimorpha after inoculation with Endoreticulatus (Solter et al., 2002). We do not know whether the midgut infection with Endoreticulatus aVects the availability of nutrients in the host larva. Timing of inoculation was an important factor determining the outcome of competition between Nosema and Vairimorpha. The Wrst species inoculated established successfully; a time lag of 7 days before the second species was inoculated allowed the Wrst species to essentially exclude the second. Timing can aVect both interference and exploitation competition and was shown to be critical to outcomes of a number of very diVerent examples of multiparasitism including a microsporidium and virus (Bauer et al., 1998), fungus and virus (Malakar et al., 1999), virus and parasitoid (Beegle and Oatman, 1975), microsporidium and parasitoid (Hoch et al., 2000), or two species of parasitoids (Marktl et al., 2002).
111
The patterns of infection in inoculated larvae were reXected in the horizontal transmission experiments. The simultaneous infection experiments revealed no eVect of dual species infections with either Nosema and Endoreticulatus or Vairimorpha and Endoreticulatus; the microsporidia were each transmitted at the same rate as for single-pathogen infections when inoculated simultaneously. For Nosema–Vairimorpha dual species infections, the competition observed in inoculated larvae was also evident in transmission to test larvae; mixed infections occurred only in 25% of test larvae. Transmission of Nosema was signiWcantly lower in dual species infections than when this species occurred alone. In one sequential infection experiment, Vairimorpha was dominant over Nosema in inoculated larvae (Fig. 2). Consequently, also Nosema transmission was negatively aVected by this competition. In the other sequential infection experiment, Nosema had an advantage in inoculated larvae (Fig. 4). In this case, Vairimorpha transmission was slightly but not signiWcantly lower. Reduced transmission due to competitive interactions could be a result of reduced parasite reproduction as has been shown for nuclear polyhedrosis and entomopox viruses (Bauer et al., 1998; Malakar et al., 1999; Ishii et al., 2002) or altered release of infective parasite stages from hosts with multiple pathogen infections. We ascertained signiWcant correlations between mortality in inoculated larvae and transmission to test larvae; the correlation was positive for Vairimorpha and negative for Nosema. Altered host mortality due to multiple pathogen infection could aVect transmission of either species. A higher mortality would hypothetically favor Vairimorpha, which appears to be released predominantly after host death. But there were no consistent indications of such increased mortality in hosts infected with two microsporidian species in our study, nor in that of Solter et al. (2002). Several routes of horizontal transmission exist for microsporidia of lepidopteran hosts. Transmission via feces is typical for gastric microsporidia (Weiser, 1961); after maturation, infective spores are released with the feces throughout the larval stage. This is reXected in the extremely high transmission (nearly 100%) of Endoreticulatus in our experiments. Production of spores in silk glands and dissemination of spores in extruded silk are thought to be important for transmission of Nosema portugal in L. dispar (JeVords et al., 1989; Maddox et al., 1999). For a diVerent Nosema sp. from L. dispar, silk was less important for transmission and feces were the predominant source of spores (Goertz, 2004). Spore release via feces is also important for the N. lymantriae isolate used in the present study (Solter and Hoch, unpublished data). Since this species also infects the fat body tissues, release of mature, infective spores from cadavers might be an additional route of transmission. The weak negative correlation between mortality in inoculated larvae and transmission of Nosema to test larvae, however, suggests that spore release via the feces is more important for transmission of Nosema. The decomposition
112
D.K. Pilarska et al. / Journal of Invertebrate Pathology 93 (2006) 105–113
of cadavers is probably necessary for release of Vairimorpha spores from host fat body tissues. The importance of this pathway is supported by a positive correlation between mortality in inoculated larvae and transmission of Vairimorpha to test larvae. We also observed transmission when larvae were still alive and determined that, in later stages of infection, mature spores are present in some cells of the midgut and the Malpighian tubules (Vavra et al., in press), enabling release of spores via the feces. No horizontal transmission occurred in our experiments for the microsporidian species that was administered 7 days after administration of the Wrst species or water treatment. Apparently the exposure time was too short after the latent period of the second species; insuYcient spores were released by the inoculated larvae for transmission. Thus, transmission data for the species that infected second were not interpreted for this experiment (Fig. 4). Overall, the microsporidia diVered in eYciency of horizontal transmission. Nosema and Endoreticulatus were transmitted at the highest rate due to release of infective microsporidian spores from living larvae via feces. Horizontal transmission of Vairimorpha was clearly less eYcient, ranging from 25% to a maximum of 75%. Hochberg and Holt (1990) proposed that a host–parasite–parasite system could be stable when one parasite was superior in intra-host competition and the other in extra-host competition (i.e., transmission). Because Nosema appeared to be superior in extra-host competition, Vairimorpha would need to be a more aggressive invader of the hosts in order to persist in the multiple parasite system. Vairimorpha is a more virulent microsporidium (Maddox et al., 1999); however, there is no clear evidence that it is overall superior to Nosema in intra-host competition. Available Weld data led to the hypothesis that microsporidia may compete in L. dispar populations; no mixed infections caused by a Nosema and Vairimorpha were detected in any of the surveyed populations in Bulgaria, Czech Republic, Slovakia, and Hungary where both species are known to occur in L. dispar populations (Pilarska et al., 1998; Novotny et al., 2000; Solter et al., 2000). Vairimorpha has not been reported from L. dispar populations in Austria, Poland, Germany and Portugal where Nosema was recovered. The laboratory data (Solter et al., 2002 and this paper) show that Vairimorpha and the closely related Nosema do, indeed, compete when both infect individual hosts, resulting in suppressed establishment in inoculated hosts and suppressed horizontal transmission. This information can help to better understand the interactions in pathogen–host complexes and is valuable for developing methods for release of the microsporidia into naïve host populations in Europe and the US. Acknowledgments We thank Mr. Thomas Kolling and Dr. Almaz Orozumbekov for technical assistance and Dr. Dörte Goertz for the valuable discussion of our experiments and comments on earlier drafts of this manuscript. This work was
made possible by scholarships no. 436 BUL 17/8/04 from the German Research Foundation (DFG) and no. A/03/ 10431 from the German Academic Exchange Service (DAAD) to D.K.P. and M.K., scholarships from the Catholic Institute Janineum, Vienna, to D.K.P., USDA Forest Service Cooperative Agreement No. AG 01-CA-11242343107, and OYce of Research/Illinois Agricultural Experiment Station Project No. 65-309S265. References Bauer, L.S., Miller, D.L., Maddox, J.V., McManus, M.L., 1998. Interactions between a Nosema sp. (Microspora: Nosematidae) and nuclear polyhedrosis virus infecting the gypsy moth, Lymantria dispar (Lepidoptera: Lymantriidae). J. Invertebr. Pathol. 74, 147–153. Beegle, C.C., Oatman, E.R., 1975. EVect of a nuclear polyhedrosis virus on the relationship between Tricholusia ni (Lepidoptera: Noctuidae) and the parasite, Hyposoter exiguae (Hymenoptera: Ichneumonidae). J. Invertebr. Pathol. 25, 59–71. Bell, R.A., Owens, C.D., Shapiro, M., TardiV, J.R., 1981. Mass rearing and virus production. In: Doane, C.C., McManus, M.L. (Eds.), The Gypsy Moth: Research Toward Integrated Pest Management, US Dept. Agric. For. Serv. Tech. Bull., vol. 1584, USDA. Washington, DC, pp. 599–600. Cali, A., El Garhy, M., 1991. Ultrastructural study of the development of Pleistophora schubergi Zwolfer, 1927 (Protozoa, Microsporida) in larvae of the spruce budworm, Choristoneura fumiferana and its subsequent taxonomic change to the genus Endoreticulatus. J. Protozool. 38, 271–278. Cossentine, J.E., Lewis, L.C., 1984. Interactions between Vairimorpha necatrix, Vairimorpha sp., and a nuclear polyhedrosis virus from Rachiplusia ou in Agrotis ipsilon larvae. J. Invertebr. Pathol. 44, 28–35. Fuxa, J.R., 1979. Interactions of the microsporidium Vairimorpha necatrix with a bacterium, virus, and fungus in Heliothis zea. J. Invertebr. Pathol. 33, 316–323. Goertz, D., 2004. Der EinXuss eines Parasiten auf seinen Wirt: das Beispiel Lymantria dispar L. (Lepidoptera, Lymantriidae) und Nosema sp. (Microsporidia). PhD Thesis, Freie Universität Berlin. Goertz, D., Pilarska, D., Kereselidze, M., Solter, L., Linde, A., 2004. Studies on the impact of two Nosema isolates from Bulgaria in the gypsy moth (Lymantria dispar L.). J. Invertebr. Pathol. 87, 105–113. Henn, M.W., Solter, L.F., 2000. Food utilization values of gypsy moth Lymantria dispar (Lepidoptera: Lymantriidae) larvae infected with a microsporidium Vairimorpha sp. (Microsporidia: Burenellidae). J. Invertebr. Pathol. 76, 263–269. Hoch, G., Schopf, A., Maddox, J.V., 2000. Interactions between an entomopathogenic microsporidium and the endoparasitoid Glyptapanteles liparidis within their host, the gypsy moth larva. J. Invertebr. Pathol. 75, 59–68. Hoch, G., Schafellner, C., Henn, M.W., Schopf, A., 2002. Alterations in carbohydrate and fatty acid levels of Lymantria dispar larvae caused by a microsporidian infection and potential adverse eVects on a co-occurring endoparasitoid, Glyptapanteles liparidis. Arch. Insect Biochem. Physiol. 50, 109–120. Hochberg, M.E., Holt, R.D., 1990. The coexistence of competing parasites. I. The role of cross-species infection. Am. Nat. 136, 517–541. Ishii, T., Takatsuka, J., Nakai, M., Kunimi, Y., 2002. Growth characteristics and competitive abilities of a nucleopolyhedrovirus and an entomopoxvirus in larvae of the smaller tea tortrix, Adoxophyes honmai (Lepidoptera: Tortricidae). Biol. Contr. 23, 96–105. JeVords, M.R., Maddox, J.V., McManus, M.L., Webb, R.E., Wieber, A., 1989. Evaluation of the overwintering success of two European microsporidia inoculatively released into gypsy moth populations in Maryland. J. Invertebr. Pathol. 53, 240–253. Maddox, J.V., Solter, L.F., 1996. Long-term storage of viable microsporidian spores in liquid nitrogen. J. Euk. Microbiol. 43, 221–225.
D.K. Pilarska et al. / Journal of Invertebrate Pathology 93 (2006) 105–113 Maddox, J.M., McManus, M.L., Solter, L.F., 1998. Microsporidia aVecting forest Lepidoptera. In: McManus, M.L., Liebhold, A.M. (Eds.), Proceedings: Population Dynamics, Impacts, and Integrated Management of Forest Defoliating Insects. USDA Forest Service General Technical Report, NE-247, pp. 198–205. Maddox, J.V., Baker, M.D., JeVords, M.R., Kuras, M., Linde, A., Solter, L.F., McManus, M.L., Vavra, J., Vossbrinck, C.R., 1999. Nosema portugal, n. sp., isolated from gypsy moth (Lymantria dispar L.) collected in Portugal. J. Invertebr. Pathol. 73, 1–14. Malakar, R., Elkinton, J.S., Hajek, A.E., Burand, J.P., 1999. Within-host interaction of Lymantria dispar (Lepidoptera:Lymantriidae) nucleopolyhedrosis virus and Entomophaga maimaiga (Zygomycetes: Entomophtorales). J. Invertebr. Pathol. 73, 91–100. Marktl, R.C., StauVer, C., Schopf, A., 2002. InterspeciWc competition between the braconid endoparasitoids Glyptapanteles porthetriae and Glyptapanteles liparidis in Lymantria dispar larvae. Entomol. Exp. Appl. 105, 97–109. McManus, M.L., Solter, L., 2003. Microsporidian pathogens in European gypsy moth populations. In: McManus, M.L., Liebhold, A.M. (Eds.), Proceedings: Ecology, Survey, and Management of Forest Insects. USDA Forest Service, Northeast Research Station Gen. Tech. Rep. NE-311. USDA Forest Service, Newtown Square, pp. 44–51. Novotny, J., Pilarska, D., Maddox, J., McManus, M., 2000. Studies of frequency of entomopathogens in a population of gypsy moth (Lymantria dispar L.) during a latency period in Slovakia. Acta Zool. Bulg. 52, 45–49. Pilarska, D., Solter, L., Maddox, J., McManus, M., 1998. Microsporidia from gypsy moth (Lymantria dispar L.) populations in Central and Western Bulgaria. Acta Zool. Bulg. 50, 109–113. Pilarska, D., Linde, A., Goertz, D., McManus, M., Solter, L., Bochev, N., Rajkova, M., 2001. First, report on the distribution of microsporidian infections of browntail moth (Euproctis chrysorrhoea L.) populations in Bulgaria. J. Pest Sci. 74, 33–37. SmirnoV, W.A., 1965. The occurrence of Nosema and Plistophora microsporidians on Archips cerasivoranus (Fitch) in Quebec. Ann. Soc. Entomol. Que. 10, 121–124.
113
Solter, L.F., Maddox, J.V., 1998. Timing of an early sporulation sequence of microsporidia in the genus Vairimorpha (Microsporidia: Burenellidae). J. Invertebr. Pathol. 72, 323–329. Solter, L.F., Maddox, J.V., McManus, M.L., 1997. Host speciWcity of microsporidia (Protista: Microspora) from European populations of Lymantria dispar (Lepidoptera: Lymantriidae) to indigenous North American Lepidoptera. J. Invertebr. Pathol. 69, 135–150. Solter, L.F., Pilarska, D.K., Vossbrinck, C.R., 2000. The host speciWcity of microsporidia pathogenic to forest Lepidoptera in Bulgaria. Biol. Contr. 19, 48–56. Solter, L.F., Siegel, J.P., Pilarska, D.K., Higgs, M.C., 2002. The impact of mixed infection of three species of microsporidia isolated from the gypsy moth, Lymantria dispar L. (Lepidoptera: Lymantriidae). J. Invertebr. Pathol. 81, 103–113. Thomas, B.M., Watson, L.E., Valverde-Garcia, P., 2003. Mixed infections and insect–pathogen interactions. Ecol. Lett. 6, 183–188. Vavra, J., Maddox, J.V., Hylis, M., Vossbrinck, C.R., Pilarska, D., Linde, A., Weiser, J., McManus, M.L., Hoch, G., Solter, L.F., 2006. Vairimorpha disparis n.comb. (Microsporidia: Burenellidae): A redescription and taxonomic revision of Thelohania disparis Timofejeva 1956, a microsporidian pathogen of the gypsy moth Lymantria dispar (L.) (Lepidoptera:Lymantriidae). J. Euk. Microbiol. Weiser, J., 1961. Die Mikrosporidien als Parasiten der Insekten. Monogr. Angew. Entomol. 17, 1–149. Weiser, J., 1966. Nemoci hmyzu. Academia, Praha. 554 pp. Weiser, J., 1969. Immunity of insects to protozoa. In: Jackson, Herman, Singer (Eds.), Immunity to Parasitic Animals, vol. I. AppletonCentury-Crofts Publ., New York. Wilson, G.G., 1975. Occurrence of Thelohania sp. and Pleistophora sp. (Microsporida: Nosematidae) in Choristoneura fumiferana (Lepidoptera: Tortricidae). Can. J. Zool. 53, 1799–1802. Wilson, G.G., Burke, J.M., 1978. Microsporidian parasites of Archips cerasivoranus (Fitch) in the district of Algoma, Ontario. Proc. Entomol. Soc. Ont. 109, 84–85.
