International Journal for Parasitology 29 (1999) 1011±1016
Research note
Experimental trypanosomiasis of natural hybrids between house mouse subspecies Jean-Marc Derothe, Claude Loubes, Marco Perriat-Sanguinet, Annie Orth, Catherine Moulia * Laboratoire GeÂnome, Populations, Interactions, UPR 9060, cc 105, Universite Montpellier II, France Received 27 January 1999; received in revised form 16 April 1999; accepted 20 April 1999
Abstract This study characterises the extent of the susceptibility to parasites (®rst demonstrated with helminths) of hybrids between Mus musculus domesticus and Mus musculus musculus. Experimental infections with Trypanosoma musculi of M. m. domesticus, M. m. musculus and their natural hybrids have been performed to compare their level of resistance/susceptibility. It appears that contrary to the results with helminths, hybrid mice present the same level of resistance/susceptibility to the trypanosome as M. m. musculus and M. m. domesticus individuals. This result is interpreted in the light of the modalities of host±parasite interactions and leads us to hypothesise on the role of parasitism in the evolution of the house mouse hybrid zone. # 1999 Australian Society for Parasitology Inc. Published by Elsevier Science Ltd. All rights reserved. Keywords: Host±parasite interactions; Mus musculus; Natural hybrids; Resistance/susceptibility; Trypanosoma musculi; Experimental infections
The understanding of ecological and evolutionary implications of the hybridisation process is one of the main concerns of biology. Exploring the way hybridisation either lowers or improves the ®tness of individuals can lead us to understand the phenomenon of speciation and the current biodiversity. However, hybrid ®tness is dicult to assess. Most studies are carried out on few parameters (fertility, mortality, developmental stability...) [1±3] which provide only a partial estimation of hybrid ®tness. Moreover, in the case of animal hybridisation, parasitism is fre* Corresponding author. fax: (33) 4-67-14-46-46: e-mail:
[email protected].
quently a forgotten parameter acting on hybrid ®tness [4]. Unlike most studies, the hybrid zone between the two European house mouse subspecies (Mus musculus domesticus and Mus musculus musculus) has been extensively analysed using several estimators of hybrid ®tness [3, 5±7]. In this situation, natural hybrids were shown to be more infested by helminth parasites than individuals of the two parental taxa [8, 9]. The genetic determinism of this hybrid susceptibility was demonstrated experimentally by Moulia et al. [6], using the mouse pinworm Aspiculuris tetraptera. These results suggested that the co-adapted gene systems controlling the parasite load (probably genes in¯uencing immunity) have diverged in the
0020-7519/99/$20.00 # 1999 Australian Society for Parasitology Inc. Published by Elsevier Science Ltd. All rights reserved. PII: S 0 0 2 0 - 7 5 1 9 ( 9 9 ) 0 0 0 6 9 - 7
1012
J.-M. Derothe et al. / International Journal for Parasitology 29 (1999) 1011±1016
two subspecies genomes during their geographic isolation. Genetic recombination in the hybrid genomes could lead to the break-down of functional associations and to hybrid susceptibility. The parasites could then contribute to counterselect the hybrid mice and to limit the gene ¯ow between the two parental taxa [6, 8, 9]. On the other hand, a study on ¯uctuating asymmetry of this hybrid zone led to a contradictory conclusion: hybrids seem to present greater stability of development than parental taxa [10]. Moreover, the parasitological studies have only concerned helminths of the digestive tract (Cestoda, Nematoda) [6, 8, 9]. One of the main questions to deal with is to estimate the extent of this hybrid susceptibility. Indeed it may either be speci®c to one kind of parasite or concern most of the biotic aggressors of the mouse [11]. That is to say, hybrid susceptibility may result from a speci®c alteration of response to helminths or it may be the expression of a global decreasing of hybrid ®tness towards parasites. Consequently, the pressure exerted by the parasites on hybrids may be a strong or a weak component explaining the whole hybrid `break-down'. The diculty in testing these hypotheses is that there are few natural parasites of the house mouse that are not helminths and that allow experimental work in the laboratory. The present study is an attempt to characterise hybrid susceptibility using the mouse protozoan Trypanosoma musculi. In this paper, we analyse the susceptibility to this protozoan of wild mice from inside and outside the hybrid zone. We then compare these results with those previously obtained with A. tetraptera [6]. Trypanosoma musculi is an extracellular protozoan of the blood regarded as slightly pathogenic for its host [12]. It is a natural parasite of the house mouse which presents an heteroxenous life-cycle. Its intermediate host is a ¯ea [13]. In the laboratory, direct infection from mouse to mouse is achieved by i.p. injection of dividing trypanosomes. The time-course of infection in all inbred strains studied is very constant and reproducible [12]. Experimental infections of inbred mice last 20±25 days [14]. Blood parasitaemia can be divided into four stages: a pre-patent
period (3±5 days), an exponential multiplicative phase (approximately 5 days), a plateau phase (7±10 days) and an exponential immune elimination phase (5±7 days) [14]. The dierences in resistance level among inbred strains were obtained by comparing the peak parasitaemia of the plateau phase [14, 15]. The clearance of parasites is followed by a long-lasting refractoriness to reinfection [12]. The parasite isolate used in this study is the Partinico II strain [16] provided by Professor Vincendeau (University of Bordeaux, France). We have cloned this parasite in the laboratory of Dr. Tibayrenc (UMR 9926, ORSTOM-CNRS) to minimise the eect of genetic variability. The parasite clone is maintained by weekly passages in male mice of the susceptible strain: C3H/ OuJIco (C3) (Ia Credo, France). Nine strains of the two sub-species M. m. domesticus and M. m. musculus and of their natural hybrids were tested in this study. The founder individuals of these strains were trapped in dierent localities within the geographic range of the two subspecies and of the hybrid zone. The mice were provided by the `Wild Mouse Repository' (UPR 9060, CNRS) and are currently maintained by random breeding in small colonies for a few generations (from ®ve to 25) in order to maintain a genetic polymorphism. The strains used in this study were: (i) M. m. domesticus mice: BZO, DMZ and DJO (from Algeria, Morocco and Italy, respectively); (ii) M. m. musculus mice: MHT and MPB (from Hungary and Poland, respectively); and (iii) hybrid mice from the Danish and Bulgarian parts of the hybrid zone. The mean introgression index (I) indicates the degree of hybridisation of individuals of a speci®c locality. It is calculated as the mean percentage of M. m. domesticus alleles at 10 diagnostic loci (isoenzymes) between the two taxa [5]. These strains are: DDO (I = 89%) and MDH (I = 2%) from Denmark, and MBS (I = 8%) and MBT (I = 7%) from Bulgaria. Mice tested in this study were all pinworm free and infected when 1 month old. They were housed in a conventional animal facility, grouped according to sex and maintained in the same conditions during the experimental period.
J.-M. Derothe et al. / International Journal for Parasitology 29 (1999) 1011±1016
Samples of blood were collected from the tail veins of C3 mice 7- to 12-days p.i. This blood was diluted at a concentration of 5104 parasites ml ÿ 1 in a sterile sodium chloride solution (0.9%) and 0.5 ml of the ®nal dilution was injected i.p. into each tested mouse [12]. Every 3 days from the time of infection until the clearance of the parasites, the parasitaemia of each infected mouse was evaluated by snipping the tail vein and collecting a blood sample. The number of trypanosomes per millilitre of blood was determined with a haemocytometer (Thoma). The response of each mouse to T. musculi was evaluated by the maximum value of parasitaemia obtained during the course of infection (corresponding to the plateau-phase) [14, 15]. The normality and homoscedasticity of the data were tested. Even after they were Log-transformed, their variances appeared highly heterogeneous (Bartlet test, w2 = 26.965; df = 8; P = 0.0007). As homoscedasticity cannot be corrected by any classical transformation, only nonparametric tests were used. As a sex-linked dierence in resistance to pinworms has been reported [17, 18], we ®rst tested for such a dierence within each strain and subsequently between the nine strains. A randomsampling test [17±19] was then performed. This test was written in Turbo Pascal language. The variable D was de®ned as the dierence between the medians of male and female samples of a par-
1013
ticular strain. For within-strain analysis, the probability of obtaining a greater D-value by chance was computed (one-way test). For the global level test (all the strains), the probability of obtaining a greater mean D-value (calculated as the mean of the D-values of the nine strains) by chance was computed. Twenty-thousand iterations were performed for each test. Heterogeneity of response to the parasite between the strains was determined using the Kruskal±Wallis (KW) test. The Noether's test is a posthoc test for multiple comparisons between samples based on the Kruskal±Wallis mean ranks; it takes into account the type-one error [19, 20]. In all cases except MPB, males exhibited higher parasitaemia than females (Table 1). A signi®cant sex-linked dierence was detected in three of nine strains (Table 1). Some of these signi®cant results might be due to type-one error because of the number of tests performed. Because this could not be determined, males (m) and females (f) of the concerned strains were separated in order to carry out the analysis with homogeneous samples. Moreover, when the random sampling test was applied to the whole set of data (the nine strains), there was a general tendency for males to exhibit a greater parasite load than females (P = 0.019). The KW test detected a high signi®cant heterogeneity of responses among the strains
Table 1 Comparison of parasitaemia in male and female mice experimentally infected with Trypanosoma musculi
M. m. domesticus M. m. musculus Hybrids
Strain
Number of males
Median of male parasite load(a)
Number of females
Median of female parasite loadsa
Pb
BZO DMZ DJO MHT MPB DDO MBS MBT MDH
15 11 11 22 13 17 12 12 18
330.0 645.0 675.0 1020.0 510.0 150.0 630.0 135.0 765.0
15 17 16 11 13 16 17 18 7
225.0 600.0 330.0 570.0 690.0 75.0 600.0 97.5 420.0
0.0928 0.4214 0.0045 0.1183 0.9371 0.0119 0.4695 0.1673 0.0302
a Number 105 parasites ml ÿ 1 of blood. bResult of the random sampling test applied between males and females of each strain. See text for more information. Signi®cant results are indicated in bold type (a = 0.05).
1014
J.-M. Derothe et al. / International Journal for Parasitology 29 (1999) 1011±1016
Table 2 Comparison of parasitaemia within wild derived strains of mice (see text for strain designation). Only signi®cant results of the Noether's test are given
DDOf MBT DDOm BZO
MPB
DMZ
MBS
MHT
MDHm
DJOm
5.67a 5.23 4.20 3.78
6.05 5.69 4.57 4.21
6.36 6.07 4.88 4.58
7.11 7.00 5.61 5.46
6.24 5.85 4.89 4.55
5.78 5.29 4.58 4.19
a
z: Result of the Noether's test; theoretical z-value = 3.6; a = 0.05.
(Hcorr. = 146.668; df = 11; P < 0.0001) and the Noether's tests (Table 2) discriminated among three categories of strains as far as susceptibility to the trypanosome was concerned: the resistant strains (median below 320105 parasites ml ÿ 1 of blood): DDOf, DDOm, MBT, BZO; the susceptible strains (more than 600105 parasites ml ÿ 1): MPB, DMZ, MBS, MHT, MDHm, DJOm; and the intermediate strains (from 320105 to 600105 parasites ml ÿ 1: DJOf, MDHf). These intermediate mice presented insigni®cant dierences in levels of parasitaemia with both resistant and susceptible strains whereas resistant and susceptible strains were signi®cantly dierent from each other (Table 2). Hybrid strains were found in all three categories. This work shows a sex-linked dierence in parasite burden previously demonstrated in many ®eld and experimental studies on mammals [17, 18, 21, 22]; see [23] for reviews). In this study, as in most models, the males have greater parasite loads than the females. There are two main hypotheses to explain this male susceptibility: (i) eco-ethological dierences between the two genders (feeding research, mate competition, territory defence, etc) could lead the males to be more exposed to infection than the females [24]. However, this parameter cannot be involved in this study because all the mice were kept under the same experimental conditions; (ii) the high level of testosterone of the males could impair their immunological status [25±27]. The involvement of testosterone is not clearly demonstrated yet [28], more speci®c studies on sex-linked dierences in parasitism correlating the constitutive level of testosterone and the level of infection
have to be conducted with dierent parasite models. However, concerning this study, the male susceptibility is an indirect proof that genetic mechanisms, modulated by the physiological status of the individuals, are involved in the response to T. musculi infection. The main result of this study is that there was no dierence in susceptibility to T. musculi between the hybrids and the M. m. domesticus/ M. m. musculus mice. This is in contrast with previous studies which demonstrated a higher susceptibility of hybrids to the pinworm A. tetraptera when compared with the M. m. musculus/ M. m. domesticus mice [6, 9]. Three hypotheses can be proposed to explain this dierence. The ®rst is based on one analysis of the genetic determinism of the resistance to trypanosome infection in two inbred strains of mice, the derived recombinant inbred strains and the ®rst generations of hybrids [29]. From this study, the infection appears to be controlled by a major dominant gene [29]. If the determinism of the resistance is the same in laboratory and wild-derived strains of mice, then hybrid susceptibility due to recombination of two divergent genomes is not expected. However, the data on the genetic determinism in inbred strains appear incomplete (two strains). Moreover, inbred strains are not samples from natural populations of mice. They are issued for mixing between dierent subspecies of the complex M. musculus [30], and they have been selected in the modi®ed environment of the laboratory. So, they present speci®c characteristics they do not share with wild-derived mice, especially concerning resistance to pathogens [17].
J.-M. Derothe et al. / International Journal for Parasitology 29 (1999) 1011±1016
The second hypothesis is based on the fact that the four hybrid strains used in this study were slightly introgressed (11% at the most for DDO (I = 89%)) while two out of four previously tested with A. tetraptera showed 30 to 40% introgression [6]. If slight introgression occurred in the hybrid genomes tested in this study, one may propose that there was no breakdown of the co-adapted gene systems controlling parasitaemia in these strain genomes. The absence of dierences of susceptibility to the murine trypanosome between hybrids and unintrogressed mice would suggest that mice regarded as `hybrids' in this study have unrecombined gene systems of resistance. However, two of these hybrid strains (MDH and DDO) were previously shown to be highly susceptible to pinworm infection, leading the authors to suggest that the functional gene associations had been broken-up in their genomes by recombination [6]. The third hypothesis to explain this dierence is that pinworms exert enough constraints on the mouse to induce the selection of dierent coadapted gene systems in the two sub-species genomes while T. musculi do not. Dierences in the life traits of the two kinds of parasites may support this hypothesis. Although the geographical range of T. musculi is not well-known, it seems to be restricted to the Mediterranean area, West African shore [31] and Panama [32]; and moreover, the transmission of T. musculi depends on the occurrence of its intermediate host while pinworms have a direct cycle. The trypanosome is then not supposed to be as broadly distributed in mouse populations as the cosmopolitan pinworms. Only some Mediterranean populations of M. m. domesticus, when the intermediate host would be present, could be infected by the protozoan and then would show adaptations to limit the parasite infection. Experimental data suggest that after the ®rst infection, the mouse shows a de®nitive refractoriness to reinfection [12, 31]. The trypanosome impact on its host is then limited in time during the mouse life. On the contrary, pinworms can reinfest one host throughout its life [33] and thus may represent a real selective pressure on it [17].
1015
To understand the level of divergence of the two mouse subspecies genomes, further studies have to be conducted with some of these parasite models. They will enable us to characterise the hybrid susceptibility and its involvement in limiting the hybridisation process between the two subspecies of mice. This is an essential step in understanding how the `durable interactions' [34] between a host and its parasites are involved in the evolution of host populations. Acknowledgements We are especially grateful to Professor Vincendeau (University of Bordeaux, France) for providing us a sample of the Partinico II strain; Dr Tibayrenc (UMR 9926 CNRS-ORSTOM, France) for allowing us to clone the parasite strain in his laboratory; and Dr Bonhomme (UPR 9060 CNRS, France) for providing mice. Thanks to Christian Barnabe for his help in cloning the parasites. References [1] Blows MW. The genetics of central and marginal populations of Drosophila serrata. II. Hybrid breakdown in ®tness components as a correlated response to selection for dessication resistance. Evolution 1993;47:1271±85. [2] Arnold ML, Hodges SA. Are natural hybrids ®t or un®t relative to their parents?. Trends Ecol Evol 1995;10:67± 71. [3] Alibert P, Fel-Clair F, Manolakou K, Britton-Davidian J, Auray JC. Developmental stability, ®tness, and trait size in laboratory hybrids between European subspecies of the house mouse. Evolution 1997;51:1284±95. [4] Moulia C. Parasitism of plant and animal hybridsÐare facts and fates the same?. Ecology 1999;80:392±406. [5] Vanlerberghe F, Boursot P, Nielsen JT, Bonhomme F. A steep cline for mitochondrial DNA in Danish mice. Genetic Res 1988;52:185±93. [6] Moulia C, Le Brun N, Dallas J, Orth A, Renaud F. Experimental evidence of genetic determinism in high susceptibility to intestinal pinworm infection in mice: a hybrid zone model. Parasitology 1993;106:387±93. [7] Dod B, Jermiin LS, Boursot P et al. Counterselection on sex chromosomes in the Mus musculus European hybrid zone. J Evol Biol 1993;6:529±46. [8] Sage RD, Heyneman D, Lim KC, Wilson AC. Wormy mice in a hybrid zone. Nature 1986;324:60±3.
1016
J.-M. Derothe et al. / International Journal for Parasitology 29 (1999) 1011±1016
[9] Moulia C, Aussel JP, Bonhomme F et al. Wormy mice in a hybrid zone, a genetic control of susceptibility to parasite infection. J Evol Biol 1991;4:679±87. [10] Alibert P, Renaud S, Dod B, Bonhomme F, Auray JC. Fluctuating asymmetry in the Mus musculus hybrid zone: a heterotic eect in disrupted co-adapted genomes. Proc R Soc Lond B 1994;258:53±9. [11] Moulia C, Le Brun N, Renaud F. Mouse±parasite interactions: from gene to population. Adv Parasitol 1996;38:120±67. [12] Albright JW, Pierantoni M, Albright JF. Immune and non-immune regulation of the population of Trypanosoma musculi in infected host mice. Infect Immun 1990;58:1757±62. [13] Cheng TC. General parasitology. 2nd Edn. Orlando, Florida: Academic Press, 1986. [14] Chiejina SN, Street J, Wakelin D, Behnke JM. Response of inbred mice to infection with a new isolate of Trypanosoma musculi. Parasitology 1993;107:233±6. [15] Albright JW, Albright JF. Rodent trypanosomes: their con¯ict with the immune system of the host. Parasitol Today 1991;7:137±40. [16] Viens P, Targett GAT, Leuchars E, Davies AJS. The immunological response of CBA mice to Trypanosoma musculi. I. Initial control of the infection and the eect of T-cell deprivation. Clin Exp Immunol 1974;16:279±94. [17] Derothe J-M, Loubes C, Orth A, Renaud F, Moulia C. Comparison between patterns of pinworms infection (Aspiculuris tetraptera) in wild and laboratory strains of mice Mus musculus. Int J Parasitol 1997;27:645± 651. [18] Ressouche L, Ganem G, Derothe J-M et al. Host chromosomal evolution and parasites of the house mouse Mus musculus domesticus in Scotland. Z Saugertierkd 1998;63:52±7. [19] Sokal RR, Rohlf FJ. Biometry, 3nd Edn. New York: Freeman and Company, 1995. [20] Sherrer B. Biostatistique. Paris: Gaetan Morin, 1984. [21] Yamamoto Y, Saito H, Setogawa T, Tomioka H. Sex dierences in host resistance to Mycobacterium
[22] [23] [24] [25] [26] [27]
[28] [29]
[30]
[31] [32] [33] [34]
marinum infection in mice. Infect Immun 1991;59:4089± 96. Daniels CW, Belosevic M. Comparison of the course of infection with Giardia muris in male and female mice. Int J Parasitol 1995;25:131±5. Zuk M, McKean KA. Sex dierences in parasitic infections: patterns and processes. Int J Parasitol 1996;26:1009±24. Tinsley RC. The eects of host sex on transmission success. Parasitol Today 1989;5:190±5. Alexander J, Stimson WH. Sex hormones and the course of parasitic infection. Parasitol Today 1988;4:189± 93. Bundy DAP. Gender-dependant patterns of infection and disease. Parasitol Today 1988;4:186±9. Wunderlich F, Marinovski P, Peter W et al. Testosterone and other gonadal factor(s) restrict the ecacy of genes controlling resistance to Plasmodium chabaudi malaria. Parasite Immunol 1991;13:357±67. Ansar Ahmed S, Talal N. Sex hormones and the immune systemÐpart 2, Animal data. Baillieres Clin Rheumatol 1990;4:13±31. Kongshawn P, Vargas F, Skamene E, Ghadirian E. Genetic control of resistance to infection with Trypanosoma musculi. In: Liss AR, editor. Genetic control of host resistance to infection and malignancy, 1985; pp. 517±522. Bonhomme F, GueÂnet JL, Dod B, Moriwaki K, Bu®eld G. The polyphyletic origin of laboratory inbred mice and their rate of evolution. Biol J Linn Soc 1987;30:51± 8. Hoare CA. The trypanosomes of mammals: a zoological monograph. Edinburgh: Blackwell Scienti®c Publication, 1972. Kendall AI. A new species of Trypanosome occurring in the mouse Mus musculus. J Infect Dis 1906;3:228±30. Adamson ML. Evolutionary biology of the Oxyurida (Nematoda): biofaces of a haplodiploid taxon. Adv Parasitol 1989;28:175±228. Combes C. Interactions durables: eÂcologie et eÂvolution du parasitisme. Paris: Masson, 1995.
