- Email: [email protected]
Transmission and burden and the impact of temperature on two species of vertically transmitted microsporidia
International Journal for Parasitology 36 (2006) 409–414 www.elsevier.com/locate/ijpara
Transmission and burden and the impact of temperature on two species of vertically transmitted microsporidia Alison M. Dunn a,*, Jon C. Hogg a, Melanie J. Hatcher b a b
Faculty of Biological Sciences, University of Leeds, Leeds, UK School of Biological Sciences, University of Bristol, Bristol, UK
Received 28 September 2005; received in revised form 9 November 2005; accepted 15 November 2005
Abstract Microsporidia are unusual amongst eukaryotic parasites in that they utilize both vertical and horizontal transmission and vertically transmitted species can cause sex ratio distortion in their host. Here we study vertical transmission in two species of feminising microsporidia, Nosema granulosis and Dictyocoela duebenum, infecting a single population of the crustacean host Gammarus duebeni and measure the effect of temperature on parasite transmission and replication. N. granulosis was vertically transmitted to 82% of the host embryos and D. duebenum was transmitted to 72% of host embryos. For both parasites, we report relatively low parasite burdens in developing host embryos. However, the parasites differ in their pattern of replication and burden within developing embryos. Whilst N. granulosis undergoes replication during host development, the burden of D. duebenum declines, leading us to propose that parasite dosage and feminisation efficiency underlie the different parasite frequencies in the field. We also examine the effect of temperature on parasite transmission and replication. Temperature does not affect the percentage of young that inherit the infection. However, low temperatures inhibit parasite replication relative to host cell division, resulting in a reduction in parasite burden in infected embryos. The reduced parasite burden at low temperatures may underpin reduced feminization at low temperatures and so limit the spread of sex ratio distorters through the host population. q 2005 Australian Society for Parasitology Inc Published by Elsevier Ltd. All rights reserved. Keywords: Vertical transmission; Feminisation; Sex ratio distortion; Gammarus duebeni; Nosema granulosis; Dictyocoela duebenum; Wolbachia
1. Introduction The Microspora are an unusual phylum of intracellular eukaryotic parasites that may use both horizontal and vertical transmission in their life cycle with many vertically transmitted species causing sex ratio distortion (Dunn et al., 2001; Terry et al., 2004). Vertically transmitted parasites are predicted to exhibit lower virulence than their horizontally transmitted counterparts because they rely on host reproduction for their own transmission (Ewald, 1987; Messenger et al., 2000; Bandi et al., 2001). For example, the microsporidium Edhazardia aedis, which has alternating horizontal and vertical routes of transmission, exhibits lower parasite burden and associated fitness effects during the vertical phase (Agnew and Koella, * Corresponding author.. Address: Institute of Integrative and Comparative Biology, Faculty of Biological Sciences, Miall Building, Clarendon Way, University of Leeds, Leeds LS2 9JT, UK. Tel.: C44 113 3432856; fax: C44 113 3432835. E-mail address: [email protected] (A.M. Dunn).
1999). Parasite burden for vertically transmitted parasites will represent a tradeoff between selection to maximize transmission to the host gametes and so to future generations and selection to minimize any reduction in host fitness (Dunn et al., 1995). In vertically transmitted microsporidia, these selective pressures have lead to life history strategies that are tightly linked to host reproduction (Dunn et al., 2001). Vertically transmitted parasites are transovarially transmitted from female hosts to the offspring but the small size of male gametes precludes transmission by this means. As a result, many vertically transmitted parasites have evolved strategies to manipulate host reproduction towards overproduction of female offspring. Microsporidia have been shown to distort host sex ratio through male killing (Lucarotti and Andreadis, 1995) and feminisation (Terry et al., 2004). The amphipod crustacean Gammarus duebeni is host to several species of vertically transmitted microsporidia (Bulnheim, 1978; Terry et al., 2004) of which Nosema granulosis is most studied. Studies of this feminising microsporidium have shown that the main site of infection is
0020-7519/$30.00 q 2005 Australian Society for Parasitology Inc Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.ijpara.2005.11.005
410
A.M. Dunn et al. / International Journal for Parasitology 36 (2006) 409–414
the gonad of its adult G. duebeni host. Parasite burden is low, sporulation takes place in the follicle cells and the oocytes become infected during vitellogenesis, probably through spore germination (Terry et al., 1997). In the infected embryo, merogony occurs. Parasite burden is low and there is evidence that the meronts segregate towards particular cell lineages during host cell division in order to reach the target tissue (the gonad) during host development (Terry et al., 1997; Dunn et al., 1998). However, whilst transovarial transmission is very efficient (96%), only about 70% of infected offspring go on to transmit the parasite to the F2 generation, indicating that within host replication and transmission is a bottleneck to the parasite life cycle (Dunn et al., 1995). In this study, we investigate parasite transmission and within host replication for two species of microsporidian sex ratio distorter; N. granulosis and a second, recently described species Dictyocoela duebenum (Terry et al., 2004) both of which infect a population of the crustacean G. duebeni on the Isle of Man, UK. Also of interest is the potential impact of environmental conditions on parasite burden and vertical transmission. The G. duebeni host is adapted to environments with both seasonal and daily variation in temperature and reproduction takes place between 5 and 20 8C (Bulnheim, 1978; Gledhill et al., 1993). We would predict microsporidia to be adapted to the same range of environmental conditions experienced by the host and this has been shown for some microsporidia of arthropods and fish (Becnel and Undeen, 1992; Ebert, 1995; Takahashi and Ogawa, 1997). Nonetheless, temperature has been shown to affect replication and development in several microsporidia (e.g. Beaman et al., 1999; Sanchez et al., 2000). The maintenance and spread of a parasitic sex ratio distorter is dependent on both its transmission efficiency and its ability to feminise the host (Hatcher and Dunn, 1995). If temperature affects replication of the feminising microsporidia infecting G. duebeni, this is likely to affect both transmission and feminization and may therefore influence the population dynamics of infection. Here we measure the impact of temperature on transmission and replication of N. granulosis and D. duebenum and consider the implications for parasite maintenance and spread.
parasites using a Zeiss Axioplan fluorescence microscope (Terry et al., 1998). A total of 90 of these mothers were infected (although parasite identity was not determined at this stage) and were mated again in the experiments described below. Finally, parasite identification was carried out at the end of the experiment by PCR of ovaries dissected from the mothers. 2.2. Parasite transmission and burden To measure parasite transmission and burden, 90 infected females were assigned randomly to four temperature treatments (6, 9, 12, 15 8C) under long day photoperiod (16 h light:8 h dark). As temperature affects the length of the gonotrophic cycle, we allowed females to acclimatise to the temperature treatment for 2 months (at least one gonotrophic cycle). One male was then placed with each female and the pair was observed until eggs were laid into the brood pouch where fertilisation takes place. A total of 896 developing embryos were collected from the mothers’ marsupia 1–2 days post fertilisation (1–220 cell stage), DAPI stained and scored for parasite transmission. The efficiency of parasite transmission to eggs was measured as the proportion of infected eggs in the brood of a parasitised mother. Parasite burden was recorded for five embryos randomly selected from each mother. The stage of development (embryo cell number) was recorded and the parasite burden was measured for each cell. 2.3. Parasite identification
2. Materials and methods
Parasite identity was determined at the end of the experiment by PCR of the mother’s gonad. DNA extraction and PCR were carried out as described in Hogg et al. (2001). Microsporidian 16S rDNA was amplified from the extracted genomic DNA using microsporidian 16S primers V1 (forward) 5 0 CACCAGGTTGATTCTGCCTGAC 3 0 (Vossbrinck and Woese, 1986) and 530R (reverse) 5 0 CCGCGGCTGCTGGCAC 3 0 (Baker et al., 1995). Parasite identity was determined by band size. Products of 410 bp represented Nosema granulosis infections and products of 440 bp represented Dictyocoela duebenum infections (Hogg et al., 2001; Terry et al., 2004).
2.1. Animal collection and culture
2.4. Statistical analysis
Adult Gammarus duebeni were collected from a population at Gansey Point (54805 0 N, 4844 0 W) on the Isle of Man where both Nosema granulosis and Dictyocoela duebenum occur (Hogg et al., 2001; Terry et al., 2004). Collection, transport and laboratory maintenance were performed according to the methods of Kelly et al. (2002). As it is not possible to screen live animals for the parasite, the broods of 200 females were screened for the presence of vertically transmitted microsporidia by fluorescence microscopy. The embryos were collected from the mother’s brood pouch, permeabilised with 5 M HCl, washed, fixed in acetone, squashed, stained with DAPI (4,6diamidino-2-phenyl-indole) and screened for the presence of
Data were analysed using the generalized linear modelling package GLIM (GLIM 3.77, Numerical Algorithms Group, Oxford, 1985). We examined parasite transmission efficiency and burden with respect to parasite species and temperature. Significant differences were assessed by examining the reduction in deviance caused by successive deletion of terms starting from the maximal model. Data for parasite transmission efficiency were proportion data (number of infected eggs/total number of eggs) and were analysed specifying a binomial error structure, taking the number of eggs as the binomial denominator. Overdispersion was corrected for by scaling with a heterogeneity factor (HfZPearson’s X2/d.f.) and
A.M. Dunn et al. / International Journal for Parasitology 36 (2006) 409–414
changes in deviance were assessed using an F-test (Crawley, 1993). Burden data were natural-logarithm transformed and treated as a normal error structure, assessing changes in deviance using an F-test. Means (GSEM) reported throughout are the maximum likelihood estimates from the GLIM minimum adequate model (containing only the significant terms). The aim of the study was to examine parasite replication, however, the temperature treatment could also affect the division rate of host embryonic cells therefore we carried out an a posteori analysis of host developmental stage which revealed that embryo cell number (ln-transformed and treated as a normal variate) differed significantly between temperature treatments and between parasite strains (main effect of parasite: F1,378Z12.78, P!0.01; parasite.temperature interaction: F3,375Z3.088; P!0.05). Host cell number was lower in lowtemperature treatments and was also significantly lower in embryos infected with N. granulosis compared with those infected with D. duebenum (mean host cell count: N. granulosis 36.6G 3.036; D. duebenum 66.04G 11.09). Therefore, host developmental stage was controlled for in all analyses of burden by performing ANCOVA, treating embryo cell number as the covariate. 3. Results 3.1. Transmission efficiency to eggs The efficiency of parasite transmission to eggs was high for both parasites and did not differ between species (F1,71Z1.140; PO0.05). On average, 82%G 3.8 of the eggs of mothers infected with Nosema granulosis inherited the infection from their mothers and 72%G 8.67 of the eggs of Dictyocoela duebenum infected mothers were infected. Transmission efficiency did not differ significantly between temperature treatments (F3,70Z0.837; PO0.05). 3.2. Parasite burden Parasite burden per embryo differed significantly for the two parasites (Table 1; the main effect of parasite species
411
accounting for 38% of the total deviance). N. granulosis was present at considerably higher burden than D. duebenum (mean burden per embryo; N. granulosis Z164.5G 4.056; D. duebenum Z38.8G1.896). Analysis of the data using ANCOVA suggests that these differences in burden result from differential replication or decay within the embryo, rather than different initial burdens in the egg. Initial parasite burden does not differ between parasite species (change in deviance on replacement of separate intercepts with joint value: F1,377Z 0.1887; PO0.05; the joint intercept of 4.55 implies an initial burden of 94 in the zygote host for both parasites). However, parasite replication during host embryogenesis differs between the species with D. duebenum dividing less frequently, or even being lost as host development proceeds (Fig. 1). The two parasite species showed different patterns of burden; whilst N. granulosis burden increased during host development, D. duebenum burden declined (Fig. 1). Mean parasite burden decreased with decreasing temperature for both parasites (controlling for host cell, Fig. 1) and there was a significant interaction between parasite species and temperature (F3,379Z3.17; P!0.05; ANCOVA: Table 1). Model simplification showed that, for N. granulosis, there was no significant difference in burden between 9 and 6 8C treatments (change in deviance on replacement with separate factor labels with combined label for 9 and 6 8C: F2,373Z0.4607; PO0.05). For D. duebenum, there was no significant difference in burden between 12, 9 and 6 8C (model simplification to combine 12, 9 and 6 8C treatments for D. duebenum caused no significant increase in deviance, F1,374Z2.20; PO0.05), indicating that burden was significantly greater at high (15 8C) than at low (12, 9 and 6 8C) treatments. Further simplifications resulted in significantly (P!0.05) increased deviance, indicating that burden differs significantly between high and low temperature rearing conditions for both parasites. Parasite burden per infected cell was investigated as this reflects the relative rates of host and parasite cell division (Table 2). For both parasites, the burden per infected cell decreased with increasing host cell count, suggesting that both parasites replicate less frequently than the rate of host cell division during early embryogenesis (Fig. 2). The rate of decline in burden was significantly faster for D. duebenum
Table 1 ANCOVA table for total parasite burden per infected embryo Source
SS
d.f.
MS
F
P
Host cell count (hcc) Temperature (temp) Parasite species (para) Temp.para Temp.hcc Para.hcc Temp.para.hcc Error Total
0.03 11.43 154.30 3.23 3.76 11.78 5.43 208.28 398.38
1 3 1 3 3 1 3 364 379
0.032 3.81 154.30 1.08 1.25 11.78 1.81 0.57
0.05 6.66 269.66 1.88 2.19 20.58 3.17
O0.05 !0.01a !0.001b O0.05 O0.05 !0.01a !0.05c
Variation in ln (parasite burden) was analysed assuming a normal error structure. SS, sum of squares; d.f., degrees of freedom; MS, mean squares; F, F ratio. a Significant at the 1% level. b Significant at the 0.1% level. c Significant at the 5% level.
412
A.M. Dunn et al. / International Journal for Parasitology 36 (2006) 409–414
(a) 8
ln parasite burden
7
15 C 12 C 9C 6C 15 C 12 C 9,6 C
6 5 4 3 2 0
1
2
3
4
5
per cell as host cell count increased (Fig. 2a). These data suggest that replication of N. granulosis is more strongly reduced at low temperatures than the rate of host cell division. In contrast, there was no significant effect of temperature on D. duebenum burden per cell (Fig. 2b; model simplification replacing separate slopes for each temperature for D. duebenum with a single slope caused no significant change in deviance F3,374Z2.342; PO0.05). There was also a significant interaction between host cell stage, parasite strain and temperature.
6
ln host cell number
4. Discussion
(b) 7 ln parasite burden
6 5
15 C
4
12 C
3
9C 6C
2
15 C
1
12,9,6 C
0 0
1
2
3
4
5
6
ln host cell number Fig. 1. Parasite burden in relation to embryo cell stage and rearing temperature. The graphs plot ln (total parasite count per embryo) against host cell count (the number of host cells per embryo) for the four temperature rearing conditions. Lines indicate statistically significant (P!0.05) lines of best fit from GLIM analysis (ANCOVA with normal errors). Regression lines combine data for temperatures for which no significant difference was found between treatments. (a) Nosema granulosis burden increased significantly with host cell stage (ln burdenZ4.55C0.153*ln(host cell stage)) for all temperatures. (b) Dictyocoela duebenum burden decreased significantly with host cell stage for all temperatures (ln burdenZ4.55–0.2157*ln(host cell stage)). Inset box gives temperature treatments in 8C.
compared with N. granulosis (F2,377Z257.5, P!0.001), suggesting that D. duebenum replicates less frequently, which supports the analysis of total burden above. For N. granulosis, the rate of decline was also dependent on temperature: lower temperature led to a more rapid reduction in parasite numbers
Parasite transmission was high for both N. granulosis and D. duebenum and did not differ between species or temperatures. High rates of vertical transmission have also been reported for Octosporea bayeri, a microsporidium which is both horizontally and vertically transmitted to its Daphnia magna host (Vizoso et al., 2005). The feminisers N. granulosis and D. duebenum rely solely or mainly on vertical transmission and cause low pathogenicity in the host (Terry et al., 1998; Ironside et al., 2003) and theoretical models predict maintenance in the host population at the levels of transmission observed in the current study (Hatcher and Dunn, 1995). Transmission efficiency for N. granulosis was in accord with previous studies (Terry et al., 1998; Kelly et al., 2002). In addition, the initial parasite dose inherited by the zygote did not differ between species nor was it affected by temperature. However, parasite replication differed between the two species and was influenced by temperature. N. granulosis parasite burden increased as the host developed, whereas in D. duebenum decreased during host development. Low temperatures led to a reduced parasite burden at any given host developmental stage for both species indicating that replication was influenced by temperature. Temperature may affect different stages of the microsporidian life cycle. Elevated temperatures have been shown to inhibit sporogony and xenoma formation by Loma salmonae infecting rainbow trout (Beaman et al., 1999; Sanchez et al., 2000). Similarly, heat shock treatments were shown to reduce
Table 2 ANCOVA table for parasite burden per infected cell (infected embryos only) Source
SS
d.f.
MS
F
P
Host cell count (hcc) Temperature (temp) Parasite strain (para) temp.para temp.hcc para.hcc temp.para.hcc Error Total
236.70 4.38 63.92 6.78 5.26 2.49 5.05 204.40 529.00
1 3 1 3 3 1 3 364 379
236.70 1.46 63.92 2.26 1.75 2.49 1.68 0.56
421.55 2.60 113.84 4.03 3.12 4.45 3.00
!0.001a O0.05 !0.001a !0.01b !0.05c !0.01b !0.05c
Variation in ln (mean parasite burden per infected cell per infected embryo) was analysed assuming a normal error structure. SS, sum of squares; d.f., degrees of freedom; MS, mean squares; F, F ratio. a significant at the 0.1% level. b Significant at the 1% level. c Significant at the 5% level.
A.M. Dunn et al. / International Journal for Parasitology 36 (2006) 409–414
ln burden per infected cell
(a) 7 15 C
6
12 C
5
9C
4 3
6C 15 C
2
12 C 9C
1
6C
0 0
1
2
3
4
5
6
ln host cell number
ln burden per infected cell
(b) 7 6 15 C 12 C
5 4
9C
3
6C
2
all temps
1 0 0
1
2
3
4
5
6
ln host cell number Fig. 2. Parasite burdens within infected host cells. Graphs plot ln (parasite burden per infected cell) for the four temperature rearing conditions. Each data point represents the mean burden per cell for an individual embryo. Lines plot statistically significant (P!0.05) lines of best fit from GLIM analysis (ANCOVA with normal errors). (a) Nosema granulosis; (b) Dictyocoela duebenum. For D. duebenum, there was no significant difference in linear parameter estimate for the different temperatures, so a single line is fitted. The rate of decline in burden was significantly faster for D. duebenum. (ln(burden)Z4.2–0.70*ln(host cell count)) compared with N. granulosis (combined for all temperatures ln(burden)Z4.2–0.47*ln(host cell count) (F2,377Z257.5, P!0.001). Inset box gives temperature treatments in 8C.
spore formation and parasite load of Nosema muscidifuracis infecting Musidifurax raptor (Boohene et al., 2003) and elevated temperatures were found to impede replication and the onset of sporogony in Brachiola algerae in mammalian cell culture (Lowman et al., 2000). Studies of E. aedis showed that spore germination was reduced at high and low temperature extremes (Undeen et al., 1993). The primary site of infection for N. granulosis is the host follicle cells and infection of the gametes takes place when spores germinate during host vitellogenesis (Terry et al., 1997). In the current study, we maintained females in the different temperature treatments for at least one gonotrophic cycles before measuring parasite transmission. We found no effect of temperature on transmission efficiency (percentage of eggs infected) or on initial parasite dose, leading us to conclude that sporulation and spore germination are not affected by temperature. However, in embryos that inherited the infection, we observed reduced merogonic replication at the lower temperatures for both N. granulosis and D. duebenum. The effects of temperature reported in this study reflect constraints on both parasites and hosts and our results suggest that, in general, reduced temperature has a greater impact on parasite replication than on host cell division as total burden
413
and burden per cell are lower at any given host cell stage at low temperatures. In addition, we observed lower host replication in N. granulosis infected embryos than in D. duebenum infected embryos. Since, we endeavored to control for host development by screening embryos within a limited period after fertilization, these results may reflect an effect of the infection on host cell division similar to that reported for mammalian cells infected with Encephalitozoon (Scanlon et al., 2000). However, this hypothesis would require further verification and is not pursued further in this paper. The optimum burden for a vertically transmitted sex ratio distorter parasite will represent a trade off between the benefits of enhanced transmission and feminisation and the metabolic costs imposed on host fitness with increasing burden. Nosema granulosis has been shown to induce feminisation in G. duebeni by preventing differentiation of the androgenic gland and the production of androgenic gland hormone which controls male sexual differentiation (Rodgers-Gray et al., 2004). The current observations suggest that reduced parasite replication underpins the previously observed reduction in N. granulosis feminsation efficiency at low temperatures (Kelly et al., 2002). We suggest that successful feminisation is dependent on parasite density in the host. Similarly, studies of the sex ratio distorting bacterium Wolbachia have found that expression of the parasite may be density dependent and that dosage is sensitive to environmental temperatures. Wolbachia dosage is positively correlated with parthenogenesis induction in Muscidifurax uniraptor (Zchori-Fein et al., 2000). Elevated temperatures have been shown to reduce the density and male killing efficiency of Wolbachia infecting Drosophila (Hurst et al., 2000) as well as the density and feminisation efficiency of Wolbachia that infect woodlice (Porcellionides pruinosus, Rigaud et al., 1997). The maintenance and spread of a parasitic sex ratio distorter in the host population is dependent on its transmission efficiency and its ability to feminise the host (Hatcher and Dunn, 1995). Prevalence of D. duebenum in the field (3.6%, Terry et al., 2004) is much lower than prevalence of N. granulosis (30–39%; Terry et al., 1998, 2004). As transmission efficiency and initial dose do not differ for the two parasites, we suggest that the difference in parasite frequency reflects the lower replication and hence a lower ability of D. duebenum to feminise its host. Acknowledgements JCH was funded by NERC; MJH acknowledges funding from the Royal Society (Dorothy Hodgkin Fellowship). We thank Andy Kelly, Joe Ironside, Chris Tofts and Judith Smith for stimulating discussions and Andy Kelly for technical assistance. References Agnew, P., Koella, J.C., 1999. Constraints on the reproductive value of vertical transmission for a microsporidian parasite and its female-killing behaviour. J. Anim. Ecol. 68, 1010–1019.
414
A.M. Dunn et al. / International Journal for Parasitology 36 (2006) 409–414
Baker, M.D., Vossbrinck, C.R., Didier, E.S., Maddox, J.V., Shadduck, J.A., 1995. Small-subunit ribosomal DNA phylogeny of various microsporidia with emphasis on aids-related forms. J. Euk. Microbiol. 42, 564–570. Bandi, C., Dunn, A.M., Hurst, G.D.D., Rigaud, T., 2001. Inherited microorganisms, sex specific virulence and reproductive parasitism. Trends Parasitol. 17, 88–94. Beaman, H.J., Speare, D.J., Brimacombe, M., 1999. Regulatory effects of water temperature on Loma salmonae (Microspora) development in rainbow trout. J. Aqua. Anim. Health 11, 237–245. Becnel, J.J., Undeen, A.H., 1992. Influence of temperature on developmental parameters for the parasite/host system Edhzardia aedis (Microsporia: Amblyosporidae) and Aedes aegypti (Diptera: Culicidae). J. Invert. Pathol. 60, 299–303. Boohene, C.K., Geden, C.J., Becnel, J.J., 2003. Evaluation of remediation methods for Nosema disease in Muscidifurax raptor (hymernoptera: Pteromalidae). Environ. Entomol. 32, 1146–1153. Bulnheim, H.P., 1978. Interactions between genetic, external and parasitic factors in sex determination of the crustacean amphipod Gammarus duebeni. Helgoland. wiss. Meer. 31, 1–33. Crawley, M.J., 1993. GLIM for Ecologists. Blackwell Science, Oxford. Dunn, A.M., Hatcher, M.J., Terry, R., Tofts, C., 1995. Evolutionary ecology of vertically transmitted parasites: strategies of transovarial transmission of a microsporidian sex ratio distorter in Gammarus duebeni. Parasitology 111, S91–S109. Dunn, A.M., Terry, R.S., Taneyhill, D.E., 1998. Within-host transmission strategies of transovarial, feminising parasites of Gammarus duebeni. Parasitology 117, 21–30. Dunn, A.M., Terry, R.S., Smith, J.E., 2001. Transovarial transmission in the microsporidia. Adv. Parasitol. 48, 57–100. Ebert, D., 1995. The ecological interactions between a microsporidian parasite and its host Daphnia magna. J. Anim. Ecol. 64, 361–369. Ewald, P.W., 1987. Transmission modes and evolution of the parasitism– mutulism continuum. Ann. NY Acad. Sci. 503, 295–306. Gledhill, T., Sutcliffe, D.W., Williams, W.D., 1993. British Freshwater Crustacea Malacostraca, Freshwater Biological Association publication No. 52. Titus Wilson and Son, Kendal. Hatcher, M.J., Dunn, A.M., 1995. Evolutionary consequences of cytoplasmically inherited feminising factors. Phil. Trans. R. Soc. Lond., B 348, 445–456. Hogg, J.C., Ironside, J.E., Sharpe, R.G., Hatcher, M.J., Smith, J.E., Dunn, A.M., 2001. Infection of Gammarus duebeni populations by two vertically transmitted microsporidia; parasite detection and discrimination by PCR-RFLP. Parasitology 125, 59–64. Hurst, G.D., Johnson, A.P., Schulenburg, J.H., Fuyama, Y., 2000. Male-killing Wolbachia in Drosophila: a temperature-sensitive trait with a threshold bacterial density. Genetics 156, 69–709. Ironside J.E., Smith J.E., Hatcher M.J., Rollinson, D.R. Smith, J.E., Dunn, A.M., 2003. Two species of feminising microsporidian parasites in Gammarus duebeni. J. Evol. Biol. 16, 467–473. Kelly, A., Dunn, A.M., Hatcher, M.J., 2002. Incomplete feminisation by the microsporidian sex ratio distorter, Nosema granulosis, and reduced transmission and feminisation efficiency at low temperatures. Int. J. Parasitol. 32, 825–831.
Lowman, P.M., Takvorian, P.M., Cali, A., 2000. The effects of elevated temperatures and various time-temperature combinations on the development of Brachiola (Nosema) algerae n. comb. in mammalian cultures. J. Euk. Microbiol. 47, 221–234. Lucarotti, C.J., Andreadis, T.G., 1995. Reproductive strategies and adaptations for survival among obligatory microsporidian and fungal parasites of mosquitoes: a comparative analysis of Amblyospora and Coelomomyces. J. Am. Mosquito Control Assoc. 11, 111–121. Messenger, S.L., Molineux, I.J., Bull, J.J., 2000. Bull Virulence evolution in a virus obeys a trade-off. Proc. R. Soc. Lond. B 266, 397–404. Rigaud, T., Antoine, D., Marcade, I., Juchault, P., 1997. The effect of temperature on sex ratio in the isopod Porcellionides pruinosus: environmental sex determination or a by-product of cytoplasmic sex determination? Evol. Ecol. 11, 205–215. Rodgers-Gray, T.P., Smith, J.E., Ashcroft, A.E., Isaac, R.E., Dunn, A.M., 2004. Mechanisms of parasite induced sex reversal in Gammarus duebeni. Int. J. Parasitol. 34, 747–753. Sanchez, J.G., Speare, D.J., Markham, R.J.F., 2000. Normal and aberrant tissue distribution of Loma salmonae (Microspora) within rainbow trout, Oncorhynchus mykiss (Walbaum), following experimental infection at water temperatures within and outside of the xenoma-expression temperature boundaries. J. Fish Dis. 23, 235–242. Scanlon, M., Shaw, A.P., Zhou, C.J., Visvesvara, G.S., Leitch, G.J., 2000. Infection by microsporidia disruption to host cell cycle. J. Eukaryot. Microbiol. 47, 525–531. Takahashi, S., Ogawa, K., 1997. Efficacy of elevated water temperature treatment of ayu infected with the fish microsporidian Glugea plecoglossi. Fish Pathol. 32, 193–198. Terry, R.S., Dunn, A.M., Smith, J.E., 1997. Cellular distribution of a feminising microsporidian: a strategy for vertical parasite transmission. Parasitology 115, 157–163. Terry, R.S., Smith, J.E., Dunn, A.M., 1998. Impact of a novel feminising microsporidian parasite on its crustacean host. J. Eukaryot Microbiol. 45, 497–501. Terry, R.S., Smith, J.E., Sharpe, R.G., Rigaud, T., Littlewood, T.J., Ironside, J.E., Rollinson, D., Bouchon, D., MacNeil, C., Dick, J.T.A., Dunn, A.M., 2004. Widespread horizontal transmission and associated host sex-ratio distortion within the eukaryotic phylum Microspora. Proc. R. Soc. B 271, 1783–1789. Undeen, A.H., Johnson, M.A., Becnel, J.J., 1993. The effects of temperature on the survival of Edhazardia aedis (Microspora, Amblyosporidae), a pathogen of Aedes aegypti. J. Invertebr. Pathol. 61, 303–307. Vizoso, D.B., Lass, S., Ebert, D., 2005. Different mechanisms of transmission of the microsporidium Octosporea bayeri: a cocktail of solutions for the problem of parasite permanence. Parasitology 130, 51–509. Vossbrinck, C.R., Woese, C.R., 1986. Eukaryotic ribosomes that lack a 5.8S RNA. Nature 320, 287–288. Zchori-Fein, E., Gottlieb, Y., Coll, M., 2000. Wolbachia density and host fitness components in Muscidifurax uniraptor (Hymenoptera: Pteromalidae). J. Invertebr. Pathol. 75, 267–272.
Transmission and burden and the impact of temperature on two species of vertically transmitted microsporidia Alison M. Dunn a,*, Jon C. Hogg a, Melanie J. Hatcher b a b
Faculty of Biological Sciences, University of Leeds, Leeds, UK School of Biological Sciences, University of Bristol, Bristol, UK
Received 28 September 2005; received in revised form 9 November 2005; accepted 15 November 2005
Abstract Microsporidia are unusual amongst eukaryotic parasites in that they utilize both vertical and horizontal transmission and vertically transmitted species can cause sex ratio distortion in their host. Here we study vertical transmission in two species of feminising microsporidia, Nosema granulosis and Dictyocoela duebenum, infecting a single population of the crustacean host Gammarus duebeni and measure the effect of temperature on parasite transmission and replication. N. granulosis was vertically transmitted to 82% of the host embryos and D. duebenum was transmitted to 72% of host embryos. For both parasites, we report relatively low parasite burdens in developing host embryos. However, the parasites differ in their pattern of replication and burden within developing embryos. Whilst N. granulosis undergoes replication during host development, the burden of D. duebenum declines, leading us to propose that parasite dosage and feminisation efficiency underlie the different parasite frequencies in the field. We also examine the effect of temperature on parasite transmission and replication. Temperature does not affect the percentage of young that inherit the infection. However, low temperatures inhibit parasite replication relative to host cell division, resulting in a reduction in parasite burden in infected embryos. The reduced parasite burden at low temperatures may underpin reduced feminization at low temperatures and so limit the spread of sex ratio distorters through the host population. q 2005 Australian Society for Parasitology Inc Published by Elsevier Ltd. All rights reserved. Keywords: Vertical transmission; Feminisation; Sex ratio distortion; Gammarus duebeni; Nosema granulosis; Dictyocoela duebenum; Wolbachia
1. Introduction The Microspora are an unusual phylum of intracellular eukaryotic parasites that may use both horizontal and vertical transmission in their life cycle with many vertically transmitted species causing sex ratio distortion (Dunn et al., 2001; Terry et al., 2004). Vertically transmitted parasites are predicted to exhibit lower virulence than their horizontally transmitted counterparts because they rely on host reproduction for their own transmission (Ewald, 1987; Messenger et al., 2000; Bandi et al., 2001). For example, the microsporidium Edhazardia aedis, which has alternating horizontal and vertical routes of transmission, exhibits lower parasite burden and associated fitness effects during the vertical phase (Agnew and Koella, * Corresponding author.. Address: Institute of Integrative and Comparative Biology, Faculty of Biological Sciences, Miall Building, Clarendon Way, University of Leeds, Leeds LS2 9JT, UK. Tel.: C44 113 3432856; fax: C44 113 3432835. E-mail address: [email protected] (A.M. Dunn).
1999). Parasite burden for vertically transmitted parasites will represent a tradeoff between selection to maximize transmission to the host gametes and so to future generations and selection to minimize any reduction in host fitness (Dunn et al., 1995). In vertically transmitted microsporidia, these selective pressures have lead to life history strategies that are tightly linked to host reproduction (Dunn et al., 2001). Vertically transmitted parasites are transovarially transmitted from female hosts to the offspring but the small size of male gametes precludes transmission by this means. As a result, many vertically transmitted parasites have evolved strategies to manipulate host reproduction towards overproduction of female offspring. Microsporidia have been shown to distort host sex ratio through male killing (Lucarotti and Andreadis, 1995) and feminisation (Terry et al., 2004). The amphipod crustacean Gammarus duebeni is host to several species of vertically transmitted microsporidia (Bulnheim, 1978; Terry et al., 2004) of which Nosema granulosis is most studied. Studies of this feminising microsporidium have shown that the main site of infection is
0020-7519/$30.00 q 2005 Australian Society for Parasitology Inc Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.ijpara.2005.11.005
410
A.M. Dunn et al. / International Journal for Parasitology 36 (2006) 409–414
the gonad of its adult G. duebeni host. Parasite burden is low, sporulation takes place in the follicle cells and the oocytes become infected during vitellogenesis, probably through spore germination (Terry et al., 1997). In the infected embryo, merogony occurs. Parasite burden is low and there is evidence that the meronts segregate towards particular cell lineages during host cell division in order to reach the target tissue (the gonad) during host development (Terry et al., 1997; Dunn et al., 1998). However, whilst transovarial transmission is very efficient (96%), only about 70% of infected offspring go on to transmit the parasite to the F2 generation, indicating that within host replication and transmission is a bottleneck to the parasite life cycle (Dunn et al., 1995). In this study, we investigate parasite transmission and within host replication for two species of microsporidian sex ratio distorter; N. granulosis and a second, recently described species Dictyocoela duebenum (Terry et al., 2004) both of which infect a population of the crustacean G. duebeni on the Isle of Man, UK. Also of interest is the potential impact of environmental conditions on parasite burden and vertical transmission. The G. duebeni host is adapted to environments with both seasonal and daily variation in temperature and reproduction takes place between 5 and 20 8C (Bulnheim, 1978; Gledhill et al., 1993). We would predict microsporidia to be adapted to the same range of environmental conditions experienced by the host and this has been shown for some microsporidia of arthropods and fish (Becnel and Undeen, 1992; Ebert, 1995; Takahashi and Ogawa, 1997). Nonetheless, temperature has been shown to affect replication and development in several microsporidia (e.g. Beaman et al., 1999; Sanchez et al., 2000). The maintenance and spread of a parasitic sex ratio distorter is dependent on both its transmission efficiency and its ability to feminise the host (Hatcher and Dunn, 1995). If temperature affects replication of the feminising microsporidia infecting G. duebeni, this is likely to affect both transmission and feminization and may therefore influence the population dynamics of infection. Here we measure the impact of temperature on transmission and replication of N. granulosis and D. duebenum and consider the implications for parasite maintenance and spread.
parasites using a Zeiss Axioplan fluorescence microscope (Terry et al., 1998). A total of 90 of these mothers were infected (although parasite identity was not determined at this stage) and were mated again in the experiments described below. Finally, parasite identification was carried out at the end of the experiment by PCR of ovaries dissected from the mothers. 2.2. Parasite transmission and burden To measure parasite transmission and burden, 90 infected females were assigned randomly to four temperature treatments (6, 9, 12, 15 8C) under long day photoperiod (16 h light:8 h dark). As temperature affects the length of the gonotrophic cycle, we allowed females to acclimatise to the temperature treatment for 2 months (at least one gonotrophic cycle). One male was then placed with each female and the pair was observed until eggs were laid into the brood pouch where fertilisation takes place. A total of 896 developing embryos were collected from the mothers’ marsupia 1–2 days post fertilisation (1–220 cell stage), DAPI stained and scored for parasite transmission. The efficiency of parasite transmission to eggs was measured as the proportion of infected eggs in the brood of a parasitised mother. Parasite burden was recorded for five embryos randomly selected from each mother. The stage of development (embryo cell number) was recorded and the parasite burden was measured for each cell. 2.3. Parasite identification
2. Materials and methods
Parasite identity was determined at the end of the experiment by PCR of the mother’s gonad. DNA extraction and PCR were carried out as described in Hogg et al. (2001). Microsporidian 16S rDNA was amplified from the extracted genomic DNA using microsporidian 16S primers V1 (forward) 5 0 CACCAGGTTGATTCTGCCTGAC 3 0 (Vossbrinck and Woese, 1986) and 530R (reverse) 5 0 CCGCGGCTGCTGGCAC 3 0 (Baker et al., 1995). Parasite identity was determined by band size. Products of 410 bp represented Nosema granulosis infections and products of 440 bp represented Dictyocoela duebenum infections (Hogg et al., 2001; Terry et al., 2004).
2.1. Animal collection and culture
2.4. Statistical analysis
Adult Gammarus duebeni were collected from a population at Gansey Point (54805 0 N, 4844 0 W) on the Isle of Man where both Nosema granulosis and Dictyocoela duebenum occur (Hogg et al., 2001; Terry et al., 2004). Collection, transport and laboratory maintenance were performed according to the methods of Kelly et al. (2002). As it is not possible to screen live animals for the parasite, the broods of 200 females were screened for the presence of vertically transmitted microsporidia by fluorescence microscopy. The embryos were collected from the mother’s brood pouch, permeabilised with 5 M HCl, washed, fixed in acetone, squashed, stained with DAPI (4,6diamidino-2-phenyl-indole) and screened for the presence of
Data were analysed using the generalized linear modelling package GLIM (GLIM 3.77, Numerical Algorithms Group, Oxford, 1985). We examined parasite transmission efficiency and burden with respect to parasite species and temperature. Significant differences were assessed by examining the reduction in deviance caused by successive deletion of terms starting from the maximal model. Data for parasite transmission efficiency were proportion data (number of infected eggs/total number of eggs) and were analysed specifying a binomial error structure, taking the number of eggs as the binomial denominator. Overdispersion was corrected for by scaling with a heterogeneity factor (HfZPearson’s X2/d.f.) and
A.M. Dunn et al. / International Journal for Parasitology 36 (2006) 409–414
changes in deviance were assessed using an F-test (Crawley, 1993). Burden data were natural-logarithm transformed and treated as a normal error structure, assessing changes in deviance using an F-test. Means (GSEM) reported throughout are the maximum likelihood estimates from the GLIM minimum adequate model (containing only the significant terms). The aim of the study was to examine parasite replication, however, the temperature treatment could also affect the division rate of host embryonic cells therefore we carried out an a posteori analysis of host developmental stage which revealed that embryo cell number (ln-transformed and treated as a normal variate) differed significantly between temperature treatments and between parasite strains (main effect of parasite: F1,378Z12.78, P!0.01; parasite.temperature interaction: F3,375Z3.088; P!0.05). Host cell number was lower in lowtemperature treatments and was also significantly lower in embryos infected with N. granulosis compared with those infected with D. duebenum (mean host cell count: N. granulosis 36.6G 3.036; D. duebenum 66.04G 11.09). Therefore, host developmental stage was controlled for in all analyses of burden by performing ANCOVA, treating embryo cell number as the covariate. 3. Results 3.1. Transmission efficiency to eggs The efficiency of parasite transmission to eggs was high for both parasites and did not differ between species (F1,71Z1.140; PO0.05). On average, 82%G 3.8 of the eggs of mothers infected with Nosema granulosis inherited the infection from their mothers and 72%G 8.67 of the eggs of Dictyocoela duebenum infected mothers were infected. Transmission efficiency did not differ significantly between temperature treatments (F3,70Z0.837; PO0.05). 3.2. Parasite burden Parasite burden per embryo differed significantly for the two parasites (Table 1; the main effect of parasite species
411
accounting for 38% of the total deviance). N. granulosis was present at considerably higher burden than D. duebenum (mean burden per embryo; N. granulosis Z164.5G 4.056; D. duebenum Z38.8G1.896). Analysis of the data using ANCOVA suggests that these differences in burden result from differential replication or decay within the embryo, rather than different initial burdens in the egg. Initial parasite burden does not differ between parasite species (change in deviance on replacement of separate intercepts with joint value: F1,377Z 0.1887; PO0.05; the joint intercept of 4.55 implies an initial burden of 94 in the zygote host for both parasites). However, parasite replication during host embryogenesis differs between the species with D. duebenum dividing less frequently, or even being lost as host development proceeds (Fig. 1). The two parasite species showed different patterns of burden; whilst N. granulosis burden increased during host development, D. duebenum burden declined (Fig. 1). Mean parasite burden decreased with decreasing temperature for both parasites (controlling for host cell, Fig. 1) and there was a significant interaction between parasite species and temperature (F3,379Z3.17; P!0.05; ANCOVA: Table 1). Model simplification showed that, for N. granulosis, there was no significant difference in burden between 9 and 6 8C treatments (change in deviance on replacement with separate factor labels with combined label for 9 and 6 8C: F2,373Z0.4607; PO0.05). For D. duebenum, there was no significant difference in burden between 12, 9 and 6 8C (model simplification to combine 12, 9 and 6 8C treatments for D. duebenum caused no significant increase in deviance, F1,374Z2.20; PO0.05), indicating that burden was significantly greater at high (15 8C) than at low (12, 9 and 6 8C) treatments. Further simplifications resulted in significantly (P!0.05) increased deviance, indicating that burden differs significantly between high and low temperature rearing conditions for both parasites. Parasite burden per infected cell was investigated as this reflects the relative rates of host and parasite cell division (Table 2). For both parasites, the burden per infected cell decreased with increasing host cell count, suggesting that both parasites replicate less frequently than the rate of host cell division during early embryogenesis (Fig. 2). The rate of decline in burden was significantly faster for D. duebenum
Table 1 ANCOVA table for total parasite burden per infected embryo Source
SS
d.f.
MS
F
P
Host cell count (hcc) Temperature (temp) Parasite species (para) Temp.para Temp.hcc Para.hcc Temp.para.hcc Error Total
0.03 11.43 154.30 3.23 3.76 11.78 5.43 208.28 398.38
1 3 1 3 3 1 3 364 379
0.032 3.81 154.30 1.08 1.25 11.78 1.81 0.57
0.05 6.66 269.66 1.88 2.19 20.58 3.17
O0.05 !0.01a !0.001b O0.05 O0.05 !0.01a !0.05c
Variation in ln (parasite burden) was analysed assuming a normal error structure. SS, sum of squares; d.f., degrees of freedom; MS, mean squares; F, F ratio. a Significant at the 1% level. b Significant at the 0.1% level. c Significant at the 5% level.
412
A.M. Dunn et al. / International Journal for Parasitology 36 (2006) 409–414
(a) 8
ln parasite burden
7
15 C 12 C 9C 6C 15 C 12 C 9,6 C
6 5 4 3 2 0
1
2
3
4
5
per cell as host cell count increased (Fig. 2a). These data suggest that replication of N. granulosis is more strongly reduced at low temperatures than the rate of host cell division. In contrast, there was no significant effect of temperature on D. duebenum burden per cell (Fig. 2b; model simplification replacing separate slopes for each temperature for D. duebenum with a single slope caused no significant change in deviance F3,374Z2.342; PO0.05). There was also a significant interaction between host cell stage, parasite strain and temperature.
6
ln host cell number
4. Discussion
(b) 7 ln parasite burden
6 5
15 C
4
12 C
3
9C 6C
2
15 C
1
12,9,6 C
0 0
1
2
3
4
5
6
ln host cell number Fig. 1. Parasite burden in relation to embryo cell stage and rearing temperature. The graphs plot ln (total parasite count per embryo) against host cell count (the number of host cells per embryo) for the four temperature rearing conditions. Lines indicate statistically significant (P!0.05) lines of best fit from GLIM analysis (ANCOVA with normal errors). Regression lines combine data for temperatures for which no significant difference was found between treatments. (a) Nosema granulosis burden increased significantly with host cell stage (ln burdenZ4.55C0.153*ln(host cell stage)) for all temperatures. (b) Dictyocoela duebenum burden decreased significantly with host cell stage for all temperatures (ln burdenZ4.55–0.2157*ln(host cell stage)). Inset box gives temperature treatments in 8C.
compared with N. granulosis (F2,377Z257.5, P!0.001), suggesting that D. duebenum replicates less frequently, which supports the analysis of total burden above. For N. granulosis, the rate of decline was also dependent on temperature: lower temperature led to a more rapid reduction in parasite numbers
Parasite transmission was high for both N. granulosis and D. duebenum and did not differ between species or temperatures. High rates of vertical transmission have also been reported for Octosporea bayeri, a microsporidium which is both horizontally and vertically transmitted to its Daphnia magna host (Vizoso et al., 2005). The feminisers N. granulosis and D. duebenum rely solely or mainly on vertical transmission and cause low pathogenicity in the host (Terry et al., 1998; Ironside et al., 2003) and theoretical models predict maintenance in the host population at the levels of transmission observed in the current study (Hatcher and Dunn, 1995). Transmission efficiency for N. granulosis was in accord with previous studies (Terry et al., 1998; Kelly et al., 2002). In addition, the initial parasite dose inherited by the zygote did not differ between species nor was it affected by temperature. However, parasite replication differed between the two species and was influenced by temperature. N. granulosis parasite burden increased as the host developed, whereas in D. duebenum decreased during host development. Low temperatures led to a reduced parasite burden at any given host developmental stage for both species indicating that replication was influenced by temperature. Temperature may affect different stages of the microsporidian life cycle. Elevated temperatures have been shown to inhibit sporogony and xenoma formation by Loma salmonae infecting rainbow trout (Beaman et al., 1999; Sanchez et al., 2000). Similarly, heat shock treatments were shown to reduce
Table 2 ANCOVA table for parasite burden per infected cell (infected embryos only) Source
SS
d.f.
MS
F
P
Host cell count (hcc) Temperature (temp) Parasite strain (para) temp.para temp.hcc para.hcc temp.para.hcc Error Total
236.70 4.38 63.92 6.78 5.26 2.49 5.05 204.40 529.00
1 3 1 3 3 1 3 364 379
236.70 1.46 63.92 2.26 1.75 2.49 1.68 0.56
421.55 2.60 113.84 4.03 3.12 4.45 3.00
!0.001a O0.05 !0.001a !0.01b !0.05c !0.01b !0.05c
Variation in ln (mean parasite burden per infected cell per infected embryo) was analysed assuming a normal error structure. SS, sum of squares; d.f., degrees of freedom; MS, mean squares; F, F ratio. a significant at the 0.1% level. b Significant at the 1% level. c Significant at the 5% level.
A.M. Dunn et al. / International Journal for Parasitology 36 (2006) 409–414
ln burden per infected cell
(a) 7 15 C
6
12 C
5
9C
4 3
6C 15 C
2
12 C 9C
1
6C
0 0
1
2
3
4
5
6
ln host cell number
ln burden per infected cell
(b) 7 6 15 C 12 C
5 4
9C
3
6C
2
all temps
1 0 0
1
2
3
4
5
6
ln host cell number Fig. 2. Parasite burdens within infected host cells. Graphs plot ln (parasite burden per infected cell) for the four temperature rearing conditions. Each data point represents the mean burden per cell for an individual embryo. Lines plot statistically significant (P!0.05) lines of best fit from GLIM analysis (ANCOVA with normal errors). (a) Nosema granulosis; (b) Dictyocoela duebenum. For D. duebenum, there was no significant difference in linear parameter estimate for the different temperatures, so a single line is fitted. The rate of decline in burden was significantly faster for D. duebenum. (ln(burden)Z4.2–0.70*ln(host cell count)) compared with N. granulosis (combined for all temperatures ln(burden)Z4.2–0.47*ln(host cell count) (F2,377Z257.5, P!0.001). Inset box gives temperature treatments in 8C.
spore formation and parasite load of Nosema muscidifuracis infecting Musidifurax raptor (Boohene et al., 2003) and elevated temperatures were found to impede replication and the onset of sporogony in Brachiola algerae in mammalian cell culture (Lowman et al., 2000). Studies of E. aedis showed that spore germination was reduced at high and low temperature extremes (Undeen et al., 1993). The primary site of infection for N. granulosis is the host follicle cells and infection of the gametes takes place when spores germinate during host vitellogenesis (Terry et al., 1997). In the current study, we maintained females in the different temperature treatments for at least one gonotrophic cycles before measuring parasite transmission. We found no effect of temperature on transmission efficiency (percentage of eggs infected) or on initial parasite dose, leading us to conclude that sporulation and spore germination are not affected by temperature. However, in embryos that inherited the infection, we observed reduced merogonic replication at the lower temperatures for both N. granulosis and D. duebenum. The effects of temperature reported in this study reflect constraints on both parasites and hosts and our results suggest that, in general, reduced temperature has a greater impact on parasite replication than on host cell division as total burden
413
and burden per cell are lower at any given host cell stage at low temperatures. In addition, we observed lower host replication in N. granulosis infected embryos than in D. duebenum infected embryos. Since, we endeavored to control for host development by screening embryos within a limited period after fertilization, these results may reflect an effect of the infection on host cell division similar to that reported for mammalian cells infected with Encephalitozoon (Scanlon et al., 2000). However, this hypothesis would require further verification and is not pursued further in this paper. The optimum burden for a vertically transmitted sex ratio distorter parasite will represent a trade off between the benefits of enhanced transmission and feminisation and the metabolic costs imposed on host fitness with increasing burden. Nosema granulosis has been shown to induce feminisation in G. duebeni by preventing differentiation of the androgenic gland and the production of androgenic gland hormone which controls male sexual differentiation (Rodgers-Gray et al., 2004). The current observations suggest that reduced parasite replication underpins the previously observed reduction in N. granulosis feminsation efficiency at low temperatures (Kelly et al., 2002). We suggest that successful feminisation is dependent on parasite density in the host. Similarly, studies of the sex ratio distorting bacterium Wolbachia have found that expression of the parasite may be density dependent and that dosage is sensitive to environmental temperatures. Wolbachia dosage is positively correlated with parthenogenesis induction in Muscidifurax uniraptor (Zchori-Fein et al., 2000). Elevated temperatures have been shown to reduce the density and male killing efficiency of Wolbachia infecting Drosophila (Hurst et al., 2000) as well as the density and feminisation efficiency of Wolbachia that infect woodlice (Porcellionides pruinosus, Rigaud et al., 1997). The maintenance and spread of a parasitic sex ratio distorter in the host population is dependent on its transmission efficiency and its ability to feminise the host (Hatcher and Dunn, 1995). Prevalence of D. duebenum in the field (3.6%, Terry et al., 2004) is much lower than prevalence of N. granulosis (30–39%; Terry et al., 1998, 2004). As transmission efficiency and initial dose do not differ for the two parasites, we suggest that the difference in parasite frequency reflects the lower replication and hence a lower ability of D. duebenum to feminise its host. Acknowledgements JCH was funded by NERC; MJH acknowledges funding from the Royal Society (Dorothy Hodgkin Fellowship). We thank Andy Kelly, Joe Ironside, Chris Tofts and Judith Smith for stimulating discussions and Andy Kelly for technical assistance. References Agnew, P., Koella, J.C., 1999. Constraints on the reproductive value of vertical transmission for a microsporidian parasite and its female-killing behaviour. J. Anim. Ecol. 68, 1010–1019.
414
A.M. Dunn et al. / International Journal for Parasitology 36 (2006) 409–414
Baker, M.D., Vossbrinck, C.R., Didier, E.S., Maddox, J.V., Shadduck, J.A., 1995. Small-subunit ribosomal DNA phylogeny of various microsporidia with emphasis on aids-related forms. J. Euk. Microbiol. 42, 564–570. Bandi, C., Dunn, A.M., Hurst, G.D.D., Rigaud, T., 2001. Inherited microorganisms, sex specific virulence and reproductive parasitism. Trends Parasitol. 17, 88–94. Beaman, H.J., Speare, D.J., Brimacombe, M., 1999. Regulatory effects of water temperature on Loma salmonae (Microspora) development in rainbow trout. J. Aqua. Anim. Health 11, 237–245. Becnel, J.J., Undeen, A.H., 1992. Influence of temperature on developmental parameters for the parasite/host system Edhzardia aedis (Microsporia: Amblyosporidae) and Aedes aegypti (Diptera: Culicidae). J. Invert. Pathol. 60, 299–303. Boohene, C.K., Geden, C.J., Becnel, J.J., 2003. Evaluation of remediation methods for Nosema disease in Muscidifurax raptor (hymernoptera: Pteromalidae). Environ. Entomol. 32, 1146–1153. Bulnheim, H.P., 1978. Interactions between genetic, external and parasitic factors in sex determination of the crustacean amphipod Gammarus duebeni. Helgoland. wiss. Meer. 31, 1–33. Crawley, M.J., 1993. GLIM for Ecologists. Blackwell Science, Oxford. Dunn, A.M., Hatcher, M.J., Terry, R., Tofts, C., 1995. Evolutionary ecology of vertically transmitted parasites: strategies of transovarial transmission of a microsporidian sex ratio distorter in Gammarus duebeni. Parasitology 111, S91–S109. Dunn, A.M., Terry, R.S., Taneyhill, D.E., 1998. Within-host transmission strategies of transovarial, feminising parasites of Gammarus duebeni. Parasitology 117, 21–30. Dunn, A.M., Terry, R.S., Smith, J.E., 2001. Transovarial transmission in the microsporidia. Adv. Parasitol. 48, 57–100. Ebert, D., 1995. The ecological interactions between a microsporidian parasite and its host Daphnia magna. J. Anim. Ecol. 64, 361–369. Ewald, P.W., 1987. Transmission modes and evolution of the parasitism– mutulism continuum. Ann. NY Acad. Sci. 503, 295–306. Gledhill, T., Sutcliffe, D.W., Williams, W.D., 1993. British Freshwater Crustacea Malacostraca, Freshwater Biological Association publication No. 52. Titus Wilson and Son, Kendal. Hatcher, M.J., Dunn, A.M., 1995. Evolutionary consequences of cytoplasmically inherited feminising factors. Phil. Trans. R. Soc. Lond., B 348, 445–456. Hogg, J.C., Ironside, J.E., Sharpe, R.G., Hatcher, M.J., Smith, J.E., Dunn, A.M., 2001. Infection of Gammarus duebeni populations by two vertically transmitted microsporidia; parasite detection and discrimination by PCR-RFLP. Parasitology 125, 59–64. Hurst, G.D., Johnson, A.P., Schulenburg, J.H., Fuyama, Y., 2000. Male-killing Wolbachia in Drosophila: a temperature-sensitive trait with a threshold bacterial density. Genetics 156, 69–709. Ironside J.E., Smith J.E., Hatcher M.J., Rollinson, D.R. Smith, J.E., Dunn, A.M., 2003. Two species of feminising microsporidian parasites in Gammarus duebeni. J. Evol. Biol. 16, 467–473. Kelly, A., Dunn, A.M., Hatcher, M.J., 2002. Incomplete feminisation by the microsporidian sex ratio distorter, Nosema granulosis, and reduced transmission and feminisation efficiency at low temperatures. Int. J. Parasitol. 32, 825–831.
Lowman, P.M., Takvorian, P.M., Cali, A., 2000. The effects of elevated temperatures and various time-temperature combinations on the development of Brachiola (Nosema) algerae n. comb. in mammalian cultures. J. Euk. Microbiol. 47, 221–234. Lucarotti, C.J., Andreadis, T.G., 1995. Reproductive strategies and adaptations for survival among obligatory microsporidian and fungal parasites of mosquitoes: a comparative analysis of Amblyospora and Coelomomyces. J. Am. Mosquito Control Assoc. 11, 111–121. Messenger, S.L., Molineux, I.J., Bull, J.J., 2000. Bull Virulence evolution in a virus obeys a trade-off. Proc. R. Soc. Lond. B 266, 397–404. Rigaud, T., Antoine, D., Marcade, I., Juchault, P., 1997. The effect of temperature on sex ratio in the isopod Porcellionides pruinosus: environmental sex determination or a by-product of cytoplasmic sex determination? Evol. Ecol. 11, 205–215. Rodgers-Gray, T.P., Smith, J.E., Ashcroft, A.E., Isaac, R.E., Dunn, A.M., 2004. Mechanisms of parasite induced sex reversal in Gammarus duebeni. Int. J. Parasitol. 34, 747–753. Sanchez, J.G., Speare, D.J., Markham, R.J.F., 2000. Normal and aberrant tissue distribution of Loma salmonae (Microspora) within rainbow trout, Oncorhynchus mykiss (Walbaum), following experimental infection at water temperatures within and outside of the xenoma-expression temperature boundaries. J. Fish Dis. 23, 235–242. Scanlon, M., Shaw, A.P., Zhou, C.J., Visvesvara, G.S., Leitch, G.J., 2000. Infection by microsporidia disruption to host cell cycle. J. Eukaryot. Microbiol. 47, 525–531. Takahashi, S., Ogawa, K., 1997. Efficacy of elevated water temperature treatment of ayu infected with the fish microsporidian Glugea plecoglossi. Fish Pathol. 32, 193–198. Terry, R.S., Dunn, A.M., Smith, J.E., 1997. Cellular distribution of a feminising microsporidian: a strategy for vertical parasite transmission. Parasitology 115, 157–163. Terry, R.S., Smith, J.E., Dunn, A.M., 1998. Impact of a novel feminising microsporidian parasite on its crustacean host. J. Eukaryot Microbiol. 45, 497–501. Terry, R.S., Smith, J.E., Sharpe, R.G., Rigaud, T., Littlewood, T.J., Ironside, J.E., Rollinson, D., Bouchon, D., MacNeil, C., Dick, J.T.A., Dunn, A.M., 2004. Widespread horizontal transmission and associated host sex-ratio distortion within the eukaryotic phylum Microspora. Proc. R. Soc. B 271, 1783–1789. Undeen, A.H., Johnson, M.A., Becnel, J.J., 1993. The effects of temperature on the survival of Edhazardia aedis (Microspora, Amblyosporidae), a pathogen of Aedes aegypti. J. Invertebr. Pathol. 61, 303–307. Vizoso, D.B., Lass, S., Ebert, D., 2005. Different mechanisms of transmission of the microsporidium Octosporea bayeri: a cocktail of solutions for the problem of parasite permanence. Parasitology 130, 51–509. Vossbrinck, C.R., Woese, C.R., 1986. Eukaryotic ribosomes that lack a 5.8S RNA. Nature 320, 287–288. Zchori-Fein, E., Gottlieb, Y., Coll, M., 2000. Wolbachia density and host fitness components in Muscidifurax uniraptor (Hymenoptera: Pteromalidae). J. Invertebr. Pathol. 75, 267–272.