Biology.anu.edu.au

Increased male mating rate in Drosophila is associated withWolbachia infection F. E. CHAMPION DE CRESPIGNY,*  T. D. PITT* & N. WEDELL*  *School of Biology, The University of Leeds, Leeds, UK Centre for Ecology and Conservation, University of Exeter, Cornwall Campus, Penryn, Cornwall, UK The maternally inherited bacterium Wolbachia pipientis infects 25–75% of arthropods and manipulates host reproduction to improve its transmission.
One way Wolbachia achieves this is by inducing cytoplasmic incompatibility (CI), where crosses between infected males and uninfected females are inviable. Infected males suffer reduced fertility through CI and reduced spermproduction. However, Wolbachia induce lower levels of CI in nonvirgin males.
We examined the impact of Wolbachia on mating behaviour in male Drosophilamelanogaster and D. simulans, which display varying levels of CI, and show thatinfected males mate at a higher rate than uninfected males in both species.
This may serve to increase the spread of Wolbachia, or alternatively, may be abehavioural adaptation employed by males to reduce the level of CI. Mating athigh rate restores reproductive compatibility with uninfected females resultingin higher male reproductive success thus promoting male promiscuity.
Increased male mating rates also have implications for the transmission ofWolbachia.
vioural adaptations that may enable infected individuals to avoid or reduce the detrimental effects of Wolbachia.
The maternally inherited bacterium Wolbachia pipientis is The flies Drosophila simulans and D. melanogaster are one of the most prevalent endosymbionts of arthropods.
both infected with Wolbachia that induce cytoplasmic It is estimated to infect between 20% and 75% of incompatibility (CI). When the sperm of an infected male terrestrial species (Werren et al., 1995; Jeyaprakash & fertilize the ova of an uninfected female the resulting Hoy, 2000; Stevens et al., 2001) and is renowned for embryos undergo abnormal mitosis and die (Lassy & manipulating host reproduction in order to improve its Karr, 1996; Tram & Sullivan, 2002). All other crosses own transmission. Wolbachia achieves this in a variety of remain viable. Reducing uninfected female fitness indirectly increases the spread of Wolbachia because the Stouthamer, 1997; Hurst & Jiggins, 2000) and frequently relative fitness of infected females that transmit Wolba- inflicts physiological and fitness costs on its host. Because chia is higher. Theoretical models predict CI-inducing of its potential impact on host fitness, Wolbachia is Wolbachia to spread rapidly to fixation within its host thought to influence both sexual selection and host population (Caspari & Watson, 1959; Frank, 1997); reproductive strategies (Zeh & Zeh, 1996, 1997; Hatcher, however, this is rarely seen in natural Drosophila popu- 2000; Charlat et al., 2003). However, little is known lations (Hoffmann et al., 1990; Hoffmann & Turelli, 1997; about the consequences of Wolbachia infection on either Vala et al., 2004). Intermediate infection frequencies may host reproductive behaviour or the utilization of beha- result from incomplete maternal transmission of theparasite, natural host curing events and incomplete CI Correspondence: Fleur E. Champion de Crespigny, Centre for Ecology and induction. CI usually results in high levels of embryonic Conservation, The University of Exeter, Cornwall Campus, Penryn TR10 mortality but there is significant variation between species. The source of this variation is thought to be Tel.: +44 (0)132 637 1852; fax: +44 (0)132 625 3638;e-mail: f.decrespigny@exeter.ac.uk varying bacterial levels in the testes (Clark & Karr, 2002).
ª 2 0 0 6 T H E A U T H O R SJ O U R N A L C O M P I L A T I O N ª 2 0 0 6 E U R O P E A N S O C I E T Y F O R E V O L U T I O N A R Y B I O L O G Y F . E . C . D E C R E S P I G N Y E T A L .
In D. simulans, almost 100% of offspring resulting from success (Greenspan & Ferveur, 2000; Moehring & incompatible crosses die (Hoffmann & Turelli, 1988) Mackay, 2004). Given the relationship between mating whereas in D. melanogaster, CI induction is generally success and fitness, there should be strong selection on relatively low, and is only evident when males are young mating behaviours or adaptations that improve male (Reynolds & Hoffmann, 2002; Clark et al., 2003). In mating success. Therefore, as Wolbachia affects host addition to inducing CI, Wolbachia can impose physiolo- fitness, males may evolve mating strategies, such as gical costs on hosts. Infected D. simulans males produce variable mating rates, to counter the detrimental effects approximately 40% fewer sperm than uninfected males, reducing their fertility (Snook et al., 2000). It is not yet Mating rates may be of particular importance to clear whether infected D. melanogaster males experience infected males. There is some evidence that male mating similar fertility reductions but this seems likely, partic- frequency affects the ability of Wolbachia to manipulate ularly when males are young and Wolbachia induce sperm. In both D. simulans and D. melanogaster, repeated mating leads to a decline in the level of CI induced by Alternatively, it has been suggested that hosts become males (Karr et al., 1998; Reynolds & Hoffmann, 2002).
‘adapted’ to harbouring Wolbachia, resulting in a mutu- The level of CI expression is thought to be related to the alistic relationship between the host and the bacterium.
rate of spermatogenesis, with faster spermatogenesis Models predict infections to become increasingly benign resulting in reduced CI (Karr et al., 1998). Although CI and levels of incompatibility to decrease over-time induction declines with male age in D. melanogaster, the (Hoffmann & Turelli, 1997). This leads to the possibility decline is more rapid under repeated mating (Reynolds & that male fitness is improved by Wolbachia infection or Hoffmann, 2002). Infected males whose sperm are not alternatively, is reduced in the absence of Wolbachia.
manipulated by Wolbachia do not suffer the fitness costs There is some evidence that Wolbachia increases female associated with CI, as crosses with uninfected females are fitness in mosquitoes, although no effect on male fitness has been reported (Dobson et al., 2002, 2004). Similarly, Here we investigate the possibility that Wolbachia infected females of some strains of D. melanogaster have infection affects male mating rate and copulatory beha- enhanced fecundity and survivorship in comparison with viour in D. melanogaster and D. simulans. Decreased sperm uninfected females. However, this does not appear to be production associated with Wolbachia infection may consistent across the species and no positive effect on result in decreased mating rates in infected males male fitness has been found (Fry & Wilkinson, 2004; Fry compared with uninfected males because of more severe sperm limitation. Alternatively, infected males may mate Reduced sperm production has important implications at the same or higher frequency than uninfected males in for various aspects of host reproduction and behaviour.
order to purge Wolbachia manipulated sperm and restore For example, Wolbachia infection reduces sperm compet- reproductive compatibility with uninfected females. Male itive ability in D. simulans (Champion de Crespigny & mating rates are examined in both D. simulans and Wedell in press), presumably because infected males are D. melanogaster to allow comparison of the impact of limited by sperm production. Similarly, male stalk-eyed Wolbachia infections that induce high or low levels of CI flies carrying meiotic driving genes produce fewer sperm on male reproductive behaviour. High and low levels of and suffer reduced sperm competitive ability because of CI are predicted to exert differing selective pressures on smaller ejaculates and incapacitation of sperm by the hosts to evolve mechanisms to avoid the manipulations seminal fluid of normal males (Wilkinson & Fry, 2001; of Wolbachia. Consequently, we expect any variation in Fry & Wilkinson, 2004). Sperm production may addi- mating frequency or copulatory behaviour to be greatest tionally limit the number of copulations achieved by in D. simulans where Wolbachia induces high levels of CI males if copulation rate is regulated by the availability of and thus imposes larger fitness costs. We show that sperm (Wedell et al., 2002). Although this possibility has infected males of both species mate at higher rates than not been examined directly in Drosophila it is thought uninfected males. As predicted, we find the greatest that accessory gland secretions may limit copulation difference in mating frequency in D. simulans where frequency in D. melanogaster males (Lefevre & Jonsson, Wolbachia induces higher levels of CI.
Various aspects of male mating behaviour, such as D. melanogaster (but less frequently in D. simulans). Most studies report a strong positive correlation between malemating success and fitness in D. melanogaster (Partridge & The D. simulans stock arose from an iso-female line of flies Farquhar, 1983; Partridge et al., 1985, 1987a,b). In infected with Wolbachia that was originally collected from addition, there is significant heritable intra-specific vari- Riverside, California and maintained in laboratory pop- ation in mating behaviour, such as courtship initiation ulations for at least 2 years. Uninfected flies were and duration, which affect a male’s mating ability or ª 2 0 0 6 T H E A U T H O R S d o i : 1 0 . 1 1 1 1 / j . 1 4 2 0 - 9 1 0 1 . 2 0 0 6 . 0 1 1 4 3 . x J O U R N A L C O M P I L A T I O N ª 2 0 0 6 E U R O P E A N S O C I E T Y F O R E V O L U T I O N A R Y B I O L O G Y 9 months prior to use in the experiment: two consecutive sequences were confirmed to be homologous with generations were raised on food containing 0.03% the expected Wolbachia strain in each species: wRi in tetracycline hydrochloride. Approximately eighty infec- D. simulans (Braig et al., 1998) and wMel in D. melano- ted females (enclosed with eighty males) laid eggs on the gaster (Zhou et al., 1998). All individuals screened for tetracycline medium. Similar numbers of first generation Wolbachia showed the correct infection status after PCR.
offspring interbred and laid eggs on new tetracyclinemedium. More than 150 female (and >150 male) generation two offspring founded the uninfected popu-lation. The D. melanogaster derived from an iso-female Eggs were collected from stock populations and the line established from an OregonR (OreR) laboratory larvae reared at 25°C with a 12:12-h light–dark cycle in strain infected with Wolbachia. Similar to D. simulans, density-controlled vials (25 larvae per vial containing uninfected flies were obtained by antibiotic treatment, 6 mL of food medium). Approximately 700 larvae of each however in D. melanogaster, three generations were raised infection status were collected daily. Vials were inspected on tetracycline medium and the curing process was every 6 h (three times during the light cycle) for newly undertaken approximately 18 months prior to the eclosed flies. All vials were emptied of newly emerged experiment. Between 50 and 100 flies of each sex flies, combined within infection groups (i.e. infected and contributed offspring to each generation. Given the uninfected combined separately) and the new adults lengths of time between tetracycline treatment and were chilled on ice, sexed and the sexes placed in experiment for both species, it is unlikely that uninfected separate, density-controlled vials (£40 adults per vial) flies suffer from reduced gut flora or other side effects of containing food. This process maximizes randomization antibiotic treatment. Stock populations, consisting of of adult flies and eliminates any effect of larval rearing thousands of infected and uninfected flies, were main- vial from the experimental design. Before being used in tained in the same way with a constant supply of food/ experiments, all vials containing females were inspected medium on which to lay eggs. This, in combination with for larvae as an additional confirmation of virginity low genetic variation caused by inbred iso-female lines, reduces the likelihood that populations experienceddifferent selection regimes and/or have diverged as a result of genetic drift. Therefore, any difference betweenflies of opposing infection status can be attributable to the To determine male mating rate, 40 infected and unin- presence/absence of Wolbachia. Flies were maintained on fected D. melanogaster males and 35 infected and unin- a standard low yeast Drosophila medium. One litre of fected D. simulans males were placed in individual vials medium consists of 10 g of Agar, 85 g of sugar, 60 g of containing a standard amount of fly medium and oats, 20 g of yeast, 1 g of Nipagin and 1 L of water.
maintained at 25°C over night. All males were 1 day In our flies, offspring mortality in incompatible D.
old on the first day they were exposed to females. The simulans crosses is typically >95% when males are young following day, three virgin (3–6 days old) females were and virgin. However, infected males taken at random placed, using a pipette, into each vial. Both infected and from the stock population usually produce some viable uninfected females were randomly allocated to males offspring when mated to uninfected (incompatible) after being mixed prior to the experiment. Both personal females. This indicates that CI induction declines in observation and previous studies of mating preferences nonvirgin, older males. Offspring mortality is similarly have found no evidence of assortative mating by males high in incompatible D. melanogaster crosses when males based on female infection status (Hoffmann & Turelli, are 1- to 2-day-old virgins. However, CI declines rapidly 1988; Hoffmann et al., 1990). The females were chilled and does not appear to be induced by older nonvirgin on ice to prevent their escape and facilitate transfer to the vials. They typically became active within a minute of The infection status of flies of both species was their introduction. The time females were present in the confirmed by polymerase chain reaction (PCR). DNA vials was recorded and the vials were observed constantly was extracted from whole flies with a salt extraction.
for 6 h (D. melanogaster) or 4 h (D. simulans) each day for Wolbachia was detected using universal Wolbachia specific three sequential days. Drosophila simulans were observed primers (Wsp 81F and Wsp 691R) that amplify the for a shorter period of time owing to their rapidity of surface protein gene (Zhou et al., 1998). To confirm that mating and there being a limited supply of females. The the PCR correctly amplified Wolbachia DNA, the PCR experiment was started at the same time each day product of three infected D. simulans and three infected (within 2 h of ‘lights on’). When males commenced D. melanogaster samples was purified and the DNA their third, sixth or ninth copulation, three additional sequenced. The PCR product was purified using a Qiagen virgin females were added to the vial via pipette. These purification kit (Quiquick PCR purification) and sent to females usually recovered from the chilled transfer Lark Technologies (Houston, TX, USA) for sequencing.
process before the male had finished copulating. There- The sequences were compared using Blast and the fore, males had constant access to virgin females with ª 2 0 0 6 T H E A U T H O R S d o i : 1 0 . 1 1 1 1 / j . 1 4 2 0 - 9 1 0 1 . 2 0 0 6 . 0 1 1 4 3 . xJ O U R N A L C O M P I L A T I O N ª 2 0 0 6 E U R O P E A N S O C I E T Y F O R E V O L U T I O N A R Y B I O L O G Y F . E . C . D E C R E S P I G N Y E T A L .
which to mate. We recorded the times at which males commenced each mating and, in D. simulans, the time copulations were completed. Copulation duration wasnot recorded in D. melanogaster for logistical reasons. At the end of the observation period, the females were removed from each of the vials and the time of removal Mating rate 0.5
recorded. The experiments were performed in a constanttemperature room (25°C) with one artificial light source After the experiment was completed all males were frozen for size measurement. Both wings were dissected and male size was measured as wing length: the distancebetween the intersection of the anterior cross vein and Fig. 1 Mean ± SE mating rates (number of copulations per hour) of longitudinal vein 3 (L3) and the intersection of the L3 infected and uninfected Drosophila simulans and D. melanogaster. The with the distal wing margin (Partridge et al., 1987b).
mating rates presented are the average for each male across the Wings were mounted on microscope slides and measured using the biometrics program, MicroMeasure (version 3).
The average size of both wings was calculated wherepossible. However, some wings were damaged during the apparent in infected males (Friedman test: v2 ¼ 17.15, dissection process and these were discarded from the P < 0.001) as the mating rate of uninfected males remained similar throughout the experiment (Friedmantest: v2 ¼ 0.50, P ¼ 0.779). There was a significant effect of male size on mating rate (REML VCA: Wald 6.05, P ¼ 0.014) but there was no interaction between In order to take in to account variation in exposure time Wolbachia and male size (REML VCA: Wald1,61 ¼ 0.06, to females, male mating rates (the number of copulations P ¼ 0.805), and no difference in size between infected per hour of female exposure) were calculated for each and uninfected males (Independent samples t-test: t59 ¼ male as an estimate of his mating capacity. Mating rates 1.90, P ¼ 0.850). Interestingly, there is no effect of male over the 3 days of the experiment were analysed in size on mating rate when mating rate is averaged over REML variance components analyses with male size as a covariate and the residuals of the models were inspected In addition to mating at higher rate, on average, for normality. Average mating rates were analysed in infected males copulated for greater lengths of time general linear models with male size as a covariate. Data (Independent samples t-test: t61 ¼ 3.315, P ¼ 0.002) are presented as mean ± SE unless otherwise specified.
than uninfected males. The mean copulation duration Males that did not copulate at all during the 3 days of the for infected males was 26 ± 0.5 min compared with experiment were excluded from the analyses. This 22 ± 1.0 min for uninfected males. Furthermore, the resulted in the exclusion of seven uninfected but no average time interval between mating (the time between infected D. simulans males. All D. melanogaster males the completion of one copulation and the start of the copulated on at least 1 day of the experiment. The data next) throughout the experiment was significantly shor- were analysed using SPSS 11.5 and GENSTAT (7th edition).
ter for infected males than uninfected males (Independ-ent samples t-test: t59 ¼ )2.342, P ¼ 0.025). Typically,infected males began mating with a new female 42 ± 2.0 min after they finished mating with the previ-ous one, whereas the interval between mating for an Wolbachia had a significant effect on male mating rates In summary, Wolbachia-infected male D. simulans (i) (REML VCA: Wald1,61 ¼ 24.73, P < 0.001). Infected mate for longer periods of time and (ii) have a shorter males mated significantly more frequently than unin- interval between mating, than uninfected males. As a fected males on each day of the experiment and when result, infected males mate more frequently than unin- mating rates were averaged for the duration of the fected males in a given time period. In contrast to experiment (GLM: F1,61 ¼ 29.55, P < 0.001; Fig. 1). On uninfected males, infected males also increase their average, infected males mated 49.01% more frequently mating rate over the 3 days of the experiment.
than uninfected males. In addition, mating rates in-creased significantly over the three experimental days (REML VCA: Wald2,61 ¼ 16.14, P < 0.001). Althoughthere was no interaction between Wolbachia infection Similar to D. simulans, Wolbachia infection in D. melano- and experimental day, increasing mating rates were only gaster had an effect on mating rate with infected males ª 2 0 0 6 T H E A U T H O R S d o i : 1 0 . 1 1 1 1 / j . 1 4 2 0 - 9 1 0 1 . 2 0 0 6 . 0 1 1 4 3 . x J O U R N A L C O M P I L A T I O N ª 2 0 0 6 E U R O P E A N S O C I E T Y F O R E V O L U T I O N A R Y B I O L O G Y behaviour. Wolbachia-infected males of both D. simulansand D. melanogaster mate at a higher rate than uninfected males. Furthermore, infected male D. simulans copulatefor longer periods of time and the interval between mating is shorter than for uninfected males. Based onthese results it seems unlikely that male mating rate islimited by sperm production: despite producing fewer sperm (Snook et al., 2000), infected male D. simulans mate at higher rate than uninfected males.
There are two possible explanations for these findings: either Wolbachia or the male host benefit from increased male mating rate. Parasite mediated manipulation of host behaviour in order to improve transmission is well documented (Moore & Gotelli, 1996). Wolbachia would benefit from elevated male mating rate providing it islinked to an increased rate of CI induction, which Fig. 2 The relationship between male body size [wing length (mm)] causes Wolbachia to spread more rapidly through and average (over 3 days of experiment) male mating rate (number populations. However, this depends on the level of CI of copulations per hour) in infected and uninfected Drosophila induced at each mating. If high levels of CI are induced in crosses between infected males and uninfectedfemales irrespective of male mating history, then mating at a higher rate than uninfected males on each offspring of uninfected females will be killed, promoting day of the experiment (REML VCA: Wald1,50 ¼ 7.21, the spread of Wolbachia. If CI instead declines rapidly to P ¼ 0.007) and when mating rates were averaged for the zero through repeated male mating, then reduced CI experiment as a whole (GLM: F1,75 ¼ 6.667, P ¼ 0.012; under high male mating rate will undermine the Fig. 1). On average, infected males mated 15.9% more transmission advantage to Wolbachia. Empirical data from both D. melanogaster and D. simulans, suggest the D. simulans, there was no effect of experiment day on decline in CI associated with mating history is rapid and mating rates. Although there is no difference in body size dramatic (Karr et al., 1998; Reynolds & Hoffmann, between infected and uninfected males (Independent 2002). Therefore, repeated male mating could reduce samples t-test: t49 ¼ 1.132, P ¼ 0.263), and no effect of the transmission advantage of Wolbachia. However, male size on mating rates (REML VCA: Wald1,50 ¼ 0.60, there is likely to be an optimal rate at which the spread P ¼ 0.440), there was a significant interaction between of Wolbachia through high mating rate is balanced by Wolbachia infection and male size on mating rate (REML the reduction of CI as a consequence of male mating VCA: Wald1,50 ¼ 6.49, P ¼ 0.011). The mating rate of rate per se. This will depend on the precise relationship uninfected males increases with male size (Pearson: r ¼ between male mating rate and the level of CI induced; 0.519, n ¼ 27, P ¼ 0.006; Fig. 2), but there is no Wolbachia will be favoured by high male mating rate if it relationship between male size and mating rate in overall results in higher rate of CI in crosses with infected D. melanogaster males (Pearson: r ¼ )0.174, uninfected females, whereas Wolbachia will lose its transmission advantage if repeated male mating is There was no difference in the average time interval associated with additional reductions in CI. Further between consecutive mating (measured here as the documentation of the consequences of repeated mating difference between the starting times of the mating) by infected males for CI induction in uninfected females throughout the experiment (Mann–Whitney U: Z ¼ is required, and ideally the relationship between the )0.085, n ¼ 75, P ¼ 0.932) between infected and unin- rate of CI induction in relation to male mating rate and In summary, infected D. melanogaster males copulate at The decrease in CI induction associated with increased a higher rate than uninfected males. Because there is no male mating frequency could instead benefit the male difference between infected and uninfected males in the host. Nonvirgin D. simulans males express approximately time interval between consecutive mating, it appears that 50% less CI than virgin males (Karr et al., 1998) whereas uninfected males become reluctant to copulate with nonvirgin D. melanogaster males express approximately females earlier than infected males.
60% less CI than virgin males (Reynolds & Hoffmann,2002). Reduction in CI induction of this magnitude dramatically improves the fitness of infected malesbecause they regain reproductive compatibility with the These results are the first demonstration of an association uninfected females in the population. This creates a between Wolbachia that induce CI and male mating strong selective advantage on males to mate at high rate ª 2 0 0 6 T H E A U T H O R S d o i : 1 0 . 1 1 1 1 / j . 1 4 2 0 - 9 1 0 1 . 2 0 0 6 . 0 1 1 4 3 . xJ O U R N A L C O M P I L A T I O N ª 2 0 0 6 E U R O P E A N S O C I E T Y F O R E V O L U T I O N A R Y B I O L O G Y F . E . C . D E C R E S P I G N Y E T A L .
when infected with Wolbachia, promoting higher levels of for host fitness may be dependent on interactions between host and Wolbachia genotypes (Olsen et al., The question remains why uninfected males mate at 2001; Fry et al., 2004). If arising by chance, it is interest- lower rate? Mating is known to be costly in Drosophila.
ing that we find increased male mating rates in two For instance, multiple mating in D. melanogaster reduces species of fly where there is no a priori reason to expect male fertility (Lefevre & Jonsson, 1962; Markow et al., mutations to manifest in the same behavioural response.
1978), suppresses immune function (McKean & Nunney, Although it seems unlikely that male mating rates are 2001) and reduces longevity (Partridge & Farquhar, dependent on host/Wolbachia genome interactions, extra- 1983). Infected males may trade off the costs associated polation of our findings to other species/strains infected with high mating frequency against the benefits of with Wolbachia must be treated with caution.
reducing CI. Because uninfected males are compatible It is possible that the association between Wolbachia and with both infected and uninfected females, there is no male mating rates arises from a benefit of infection.
additional fitness gain from mating at higher frequency Wolbachia infections are predicted to become increasingly through restoring compatibility. Therefore, mating at benign and potentially confer benefits to the host over- higher frequency may generate greater costs than time (Hoffmann & Turelli, 1997). However, this seems benefits in uninfected males. Many studies suggest that unlikely because, although fitness benefits of infection Drosophila males are incapable of fertilising eggs beyond have been demonstrated in females of some species three or four consecutive mating within a short period of (Dobson et al., 2002, 2004; Fry et al., 2004), fitness time (Lefevre & Jonsson, 1962; Markow et al., 1978).
advantages to males have so far not been found. On the Infected males in our study mated as many as 11 times contrary, there is substantial evidence of significant fitness within a 4- to 6-h period. Hence increased mating rates costs of Wolbachia infection to male Drosophila (O’Neill could be disadvantageous for uninfected males, but may et al., 1997; Snook et al., 2000). Similarly, it is unlikely be a response triggered by the presence of Wolbachia in that these results arise from differences in maturation time infected males that ultimately increases their fitness. The between infected and uninfected males in either species.
fitness consequences for infected and uninfected males of There is no apparent difference in development time or varying mating history are currently being investigated.
onset of reproduction between infected and uninfected How males might recognize their infection status and males (F.E. Champion de Crespigny, personal observa- adjust their mating behaviour remains unknown. How- tions). Infected and uninfected males are both capable of ever, behavioural plasticity in terms of mating strategies fertilising eggs at 1 day. Additionally, no consistent effect such as copulation duration, ejaculate size, nuptial gift of infection status on male longevity has been found in size and mating rates is well-documented in many taxa either D. melanogaster (Fry et al., 2004) or D. simulans and attributed to a variety of reasons including the risk of (Champion de Crespigny & Wedell, unpublished data).
sperm competition (Cook & Gage, 1995; Wedell & Cook, Hence it is unlikely that increased mating rates by infected 1999; del Barco-Trillo & Ferkin, 2004; Garcia-Gonzalez & males represent a terminal investment in reproduction. A Gomendio, 2004), variation in female quality (Gage, final possibility is that the lower mating rate of uninfected 1998; Engqvist & Sauer, 2001) and in response to males could be a consequence of infection loss, if mating parasitic infections (Polak & Starmer, 1998). Plastic rate is a co-adaptation to the presence of Wolbachia. More behavioural responses to Wolbachia by Drosophila seem simply, mating rate may be an unselected consequence of likely when one considers that host populations typically adaptation to the presence of Wolbachia, and curing flies evolve under intermediate infection frequencies and of their infection could result in a simultaneous loss of mating capacity. However, this premise relies on a strong (Hoffmann & Turelli, 1997). Because infected females linkage between mating rate and infection, i.e. consistent frequently produce both infected and uninfected off- spring, plastic behavioural responses to Wolbachia infec- Mating success in D. melanogaster is frequently posi- tion are likely to have a selective advantage to males. It is tively associated with male body size (Partridge & unknown whether this might apply to other species Farquhar, 1983; Partridge et al., 1987a,b). There was no infected with CI-inducing Wolbachia, however plastic difference in size between infected and uninfected males behavioural adaptations, e.g. mate choice have been in this study so male size cannot explain the higher demonstrated in female spider mites infected with CI-inducing Wolbachia (Vala et al., 2004).
Although no relationship was found between body size Despite finding increased mating rates in Wolbachia- and mating rate in infected D. melanogaster, mating rates infected males in two species of Drosophila, without were positively correlated with male size in uninfected further survey in different taxa it is not possible to flies. This supports previous findings (Partridge & evaluate the generality of our results. Our findings may Farquhar, 1983; Partridge et al., 1987a,b), although it is be specific to our particular Drosophila strains. This could evident that Wolbachia infection overrides any effect of arise by chance or because, at least in the case of male size as infected males copulate at consistently high D. melanogaster, the consequences of Wolbachia infection ª 2 0 0 6 T H E A U T H O R S d o i : 1 0 . 1 1 1 1 / j . 1 4 2 0 - 9 1 0 1 . 2 0 0 6 . 0 1 1 4 3 . x J O U R N A L C O M P I L A T I O N ª 2 0 0 6 E U R O P E A N S O C I E T Y F O R E V O L U T I O N A R Y B I O L O G Y Male mating rates in natural populations are unlikely males of mating at high rates. The frequency of unin- to be as extreme as we, and others find in the laboratory fected females in a population manipulated by Wolbachia (Bateman, 1948; Lefevre & Jonsson, 1962). This is will affect the selective advantage of males mating at high primarily because of access to receptive females. Because frequency. When the infection invades a population and we were specifically interested in male mating behaviour the majority of females are uninfected, the selective we ensured male mating rate was not limited by female advantage to infected males is greater than when availability in this study. Although little is known about Wolbachia is near fixation. However, maternal transmis- male mating rates in the field it seems unlikely that males sion failure of Wolbachia, and natural curing are thought have a constant supply of receptive females. Males to provide a constant source of uninfected individuals probably wait by food sources/oviposition sites and court and this may be sufficient to maintain the selective all females that arrive at the site (Markow, 1988). Female advantage of high mating rates in infected males.
availability will depend on population size and will- Although surveys of levels of CI-inducing Wolbachia ingness to mate. It is likely that males will court infections within wild populations of other species are previously mated females more frequently than virgin generally lacking, a significant number reveal interme- females (Partridge et al., 1987b; Gromko & Markow, diate infection frequencies (see Vala et al., 2004 for a 1993). Our D. simulans females remate readily after 24 h review). This is in contrast to male-killing Wolbachia in (although this could also be a laboratory phenomenon) butterflies for example, where it can reach almost 100% so it is possible that many of the nonvirgin females prevalence in some populations (Dyson & Hurst, 2004).
arriving at a site will be willing to mate (but see Gromko Consequently, it would be interesting to investigate male & Markow, 1993). However, even if males do not mating rates in other species manipulated by CI-inducing copulate frequently, this may not alter the selective advantage of mating at high rate. Previous work by Karr In conclusion, this is the first study to identify a potential et al. (1998) and Reynolds & Hoffmann (2002) indicates male behavioural adaptation that may have evolved in that only relatively few mating (compared with the response to the selfish manipulations of CI-inducing number our males performed) are required to substan- Wolbachia. We suggest that Wolbachia infected males mate at high frequency in order to reduce the induction of CI.
This study provides an opportunity to compare the This restores reproductive compatibility with uninfected impact of fitness variation resulting from high and low females, increasing male fitness. The selective advantage of CI-inducing Wolbachia infections on the strength of this strategy depends critically on the availability of selection on host behavioural adaptations. Because CI receptive, uninfected females and further work is required induction in D. simulans is higher than in D. melanogaster, to elucidate the exact mechanisms involved and to confirm a stronger response is expected. This prediction is CI as the driving force behind increased male mating rates.
corroborated by the finding that a greater difference in This study highlights the potential importance of Wolbachia mating rate between uninfected and infected males is present in D. simulans (50%) compared with D. melano-gaster (16%). However, despite the typically low levels of CI in D. melanogaster, males still display a behaviouralresponse associated with Wolbachia infection.
We thank A. Bretman, Z. Lewis and J. Craven for tech- The findings of this study have important implications nical assistance and T. Tregenza, D. Hosken, D. Hodgson for Wolbachia population dynamics. Reducing CI induc- and two anonymous reviewers for helpful comments.
tion may undermine the transmission advantage of This work was funded by a Leeds University ORS Wolbachia and consequently decrease its rate of spread, award to F. de Crespigny and a Royal Society Fellowship although this will depend on the precise relationship between male mating rate and mating history and theoverall level of CI induced. Furthermore, this process could extend the length of time Wolbachia remains at lowfrequency in populations when it may be vulnerable to Bateman, A.J. 1948. Intra-sexual selection in Drosophila. Heredity stochastic events, such as natural curing, which may cause its loss from populations (Champion de Crespigny Braig, H.R., Zhou, W., Dobson, S.L. & O’Neill, S.L. 1998. Cloning et al., 2005). In contrast to theoretical predictions, host and characterization of a gene encoding the major surfaceprotein of the bacterial endosymbiont Wolbachia pipientis. J.
populations of Drosophila sometimes display intermediate levels of infection in the wild (Hoffmann et al., 1990; Caspari, E. & Watson, G.S. 1959. On the evolutionary Turelli & Hoffmann, 1995; Vala et al., 2004). Loss of CI importance of cytoplasmic sterility in mosquitoes. Evolution induction through increased male mating rates may contribute to the maintenance of such infection poly- Champion de Crespigny, F.E., Butlin, R.K. & Wedell, N. 2005.
morphisms in the wild as the presence of uninfected Can cytoplasmic incompatibility inducing Wolbachia promote females maintains the selective advantage to infected the evolution of mate preferences? J. Evol. Biol. 18: 967–977.
ª 2 0 0 6 T H E A U T H O R S d o i : 1 0 . 1 1 1 1 / j . 1 4 2 0 - 9 1 0 1 . 2 0 0 6 . 0 1 1 4 3 . xJ O U R N A L C O M P I L A T I O N ª 2 0 0 6 E U R O P E A N S O C I E T Y F O R E V O L U T I O N A R Y B I O L O G Y F . E . C . D E C R E S P I G N Y E T A L .
Champion de Crespigny, F.E. & Wedell, N. 2006. Wolbachia Hurst, G.D.D. & Jiggins, F.M. 2000. Male-killing bacteria in infection reduces sperm competitive ability in an insect. Proc.
insects: mechanisms, incidence, and implications. Emerg. Infect.
Charlat, S., Hurst, G.D.D. & Merc¸ot, H. 2003. Evolutionary Jeyaprakash, A. & Hoy, M.A. 2000. Long PCR improves consequences of Wolbachia infections. Trends Genet. 19: 217– Wolbachia DNA amplification: wsp sequences found in 76% of sixty-three arthropod species. Insect Mol. Biol. 9: 393–405.
Clark, M.E. & Karr, T.L. 2002. Distribution of Wolbachia within Karr, T.L., Yang, W. & Feder, M.E. 1998. Overcoming Drosophila reproductive tissue: implications for the expression cytoplasmic incompatibility in Drosophila. Proc. R. Soc. B 265: of cytoplasmic incompatibility. Integr. Comp. Biol. 42: 332–339.
Clark, M.E., Veneti, Z., Bourtzis, K. & Karr, T.L. 2003. Wolbachia Lassy, C.W. & Karr, T.L. 1996. Cytological analysis of fertilisation distribution and cytoplasmic incompatibility during sperm and early embryonic development in incompatible crosses of development: the cyst as the basic cellular unit of CI Drosophila simulans. Mech. Dev. 57: 47–58.
expression. Mech. Dev. 120: 185–198.
Lefevre, G. & Jonsson, U.B. 1962. Sperm transfer, storage, Cook, P.A. & Gage, M.J.G. 1995. Effects of risks of sperm displacement and utilization in Drosophila melanogaster. Genetics competition on the numbers of eupyrene and apyrene sperm ejaculated by the moth Plodia interpunctella (Lepidoptera: Markow, T.A. 1988. Reproductive behavior of Drosophila mel- Pyralidae). Behav. Ecol. Sociobiol. 36: 261–268.
anogaster and D. nigrospiracula in the field and in the laborat- del Barco-Trillo, J. & Ferkin, M. 2004. Male mammals respond to ory. J. Comp. Psychol. 102: 169–173.
a risk of sperm competition conveyed by odours of conspecific Markow, T.A., Quaid, M. & Kerr, S. 1978. Male mating experience and competitive courtship success in Drosophila Dobson, S.L., Marsland, E.J. & Rattanadechakul, W. 2002.
melanogaster. Nature 276: 821–822.
Mutualistic Wolbachia infection in Aedes albopictus: accelerating McKean, K.A. & Nunney, L. 2001. Increased sexual activity cytoplasmic drive. Genetics 160: 1087–1094.
reduces male immune function in Drosophila melanogaster. Proc.
Dobson, S.L., Rattanadechakul, W. & Marsland, E.J. 2004.
Natl. Acad. Sci. USA 98: 7904–7909.
Fitness advantage and cytoplasmic incompatibility in Wolba- Moehring, A.J. & Mackay, T.F.C. 2004. The quantitative genetic chia single- and superinfected Aedes albopictus. Heredity 93: basis of male mating behavior in Drosophila melanogaster.
Dyson, E.A. & Hurst, G.D.D. 2004. Persistence of an extreme Moore, J. & Gotelli, N.J. 1996. Evolutionary patterns of altered sex-ratio bias in a natural population. Proc. Natl. Acad. Sci. USA behavior and susceptibility in parasitized hosts. Evolution 50: Engqvist, L. & Sauer, K.P. 2001. Strategic male mating effort and Olsen, K., Reynolds, K.T. & Hoffmann, A.A. 2001. A field cage cryptic male choice in a scorpionfly. Proc. R. Soc. B 268: 729–735.
test of the effects of the endosymbiont Wolbachia on Drosophila Frank, S.A. 1997. Cytoplasmic incompatibility and population melanogaster. Heredity 86: 731–737.
structure. J. Theor. Biol. 184: 327–330.
O’Neill, S.L., Hoffmann, A.A. & Werren, J.H. 1997. Influential Fry, C.L. & Wilkinson, G.S. 2004. Sperm survival in female stalk- Passengers: Inherited Microorganisms and Arthropod Reproduction.
eyed flies depends on seminal fluid and meiotic drive.
Partridge, L., Ewing, A. & Chandler, A. 1987a. Male size and Fry, A.J., Palmer, M.R. & Rand, D.M. 2004. Variable fitness mating success in Drosophila melanogaster: the roles of male and effects of Wolbachia infection in Drosophila melanogaster.
female behaviour. Anim. Behav. 35: 555–562.
Partridge, L. & Farquhar, M. 1983. Lifetime mating success of Gage, M.J.G. 1998. Influences of sex, size, and symmetry on male fruitflies (Drosophila melanogaster) is related to their size.
ejaculate expenditure in a moth. Behav. Ecol. 9: 582–597.
Garcia-Gonzalez, F. & Gomendio, M. 2004. Adjustment of Partridge, L., Hoffmann, A.A. & Jones, J.S. 1987b. Male size and copula duration and ejaculate size according to the risk of mating success in Drosophila melanogaster and D. psuedoobscura sperm competition in the golden egg bug (Phyllomorpha under field conditions. Anim. Behav. 35: 468–476.
laciniata). Behav. Ecol. 15: 22–30.
Partridge, L., Mackay, T.F.C. & Aitken, S. 1985. Male mating Greenspan, R.J. & Ferveur, J.-F. 2000. Courtship in Drosophila.
success and fertility in Drosophila melanongaster. Genet. Res. 46: Gromko, M.H. & Markow, T.A. 1993. Courtship and remating in Polak, M. & Starmer, W.T. 1998. Parasite-induced risk of field populations of Drosophila. Anim. Behav. 45: 253–262.
mortality elevates reproductive effort in male Drosophila. Proc.
Hatcher, M.J. 2000. Persistence of selfish genetic elements: population structure and conflict. Trends Ecol. Evol. 15: 271–277.
Reynolds, K.T. & Hoffmann, A.A. 2002. Male age, host effects Hoffmann, A.A. & Turelli, M. 1988. Unidirectional incompat- and the weak expression or non-expression of cytoplasmic ibility in Drosophila simulans: inheritance, geographic variation incompatibility in Drosophila strains infected by maternally and fitness effects. Genetics 119: 435–444.
transmitted Wolbachia. Genet. Res. 80: 79–87.
Hoffmann, A.A. & Turelli, M. 1997. Cytoplasmic incompatibility Rigaud, T. 1997. Inherited microorganisms and sex determin- in insects. In: Influential Passengers Inherited Microorganisms and ation of arthropod hosts. In: Influential Passengers: Inherited Arthropod Reproduction (S. L. O’Neill, A. A. Hoffmann & J. H.
Microorganisms and Arthropod Reproduction (S. L. O’Neill, A. A.
Werren, eds), pp. 42–80. Oxford University Press, Oxford.
Hoffmann & J. H. Werren eds), pp. 81–101. Oxford University Hoffmann, A.A., Turelli, M. & Harshman, L.G. 1990. Factors affecting the distribution of cytoplasmic incompatibility in Snook, R.R., Cleland, S.Y., Wolfner, M.F. & Karr, T.L. 2000.
Drosophila simulans. Genetics 126: 933–948.
Offsetting effects of Wolbachia infection and heat shock on ª 2 0 0 6 T H E A U T H O R S d o i : 1 0 . 1 1 1 1 / j . 1 4 2 0 - 9 1 0 1 . 2 0 0 6 . 0 1 1 4 3 . x J O U R N A L C O M P I L A T I O N ª 2 0 0 6 E U R O P E A N S O C I E T Y F O R E V O L U T I O N A R Y B I O L O G Y sperm production in Drosophila simulans: Analyses of fecund- Wedell, N., Gage, M.J.G. & Parker, G.A. 2002. Sperm competi- ity, fertility and accessory gland proteins. Genetics 155: 167– tion, male prudence and sperm-limited females. Trends Ecol.
Stevens, L., Giordano, R. & Fialho, R.F. 2001. Male-killing, Werren, J.H., Windsor, D. & Guo, L. 1995. Distribution of nematode infections, bacteriophage infection, and virulence of Wolbachia among neotropical arthropods. Proc. R. Soc. B 262: cytoplasmic bacteria in the genus Wolbachia. Annu. Rev. Ecol.
Wilkinson, G.S. & Fry, C.L. 2001. Meiotic drive alters sperm Stouthamer, R. 1997. Wolbachia-induced parthenogenesis. In: competitive ability in stalk-eyed flies. Proc. R. Soc. B 268: 2559– Influential Passengers: Inherited Microorganisms and Arthropod Reproduction (S. L. O’Neill, A. A. Hoffmann & J. H. Werren, Zeh, J.A. & Zeh, D.W. 1996. The evolution of polyandry I: eds), pp. 102–124. Oxford University Press, Oxford.
intragenomic conflict and genetic incompatibility. Proc. R. Soc.
Tram, U. & Sullivan, W. 2002. Role of delayed nuclear envelope breakdown and mitosis in Wolbachia-induced cytoplasmic Zeh, J.A. & Zeh, D.W. 1997. The evolution of polyandry II: post- incompatibility. Science 296: 1124–1126.
copulatory defences against genetic incompatibility. Proc. R.
Turelli, M. & Hoffmann, A.A. 1995. Cytoplasmic incompatibility in Drosophila simulans: dynamics and parameter estimates from Zhou, W., Rousset, F. & O’Neill, S.L. 1998. Phylogeny and PCR- natural populations. Genetics 140: 1319–1338.
based classification of Wolbachia strains using wsp gene Vala, F., Egas, M., Breeuwer, J.A.J. & Sabelis, M.W. 2004.
sequences. Proc. R. Soc. B 265: 509–515.
Wolbachia affects oviposition and mating behaviour of itsspider mite host. J. Evol. Biol. 17: 692–700.
Received 16 November 2005; revised 4 April 2006; accepted 10 April Wedell, N. & Cook, P.A. 1999. Butterflies tailor their ejaculate in response to sperm competition risk and intensity. Proc. R. Soc. B266: 1033–1039.
ª 2 0 0 6 T H E A U T H O R S d o i : 1 0 . 1 1 1 1 / j . 1 4 2 0 - 9 1 0 1 . 2 0 0 6 . 0 1 1 4 3 . xJ O U R N A L C O M P I L A T I O N ª 2 0 0 6 E U R O P E A N S O C I E T Y F O R E V O L U T I O N A R Y B I O L O G Y

Source: http://biology.anu.edu.au/hosted_sites/jennions/publications/fleur3.pdf