The Mediterranean flour moth, Ephestia kuehniella, is infected with A-group Wolbachia (wKue), and the almond moth, Cadra cautella, is doubly infected with A- and B-group Wolbachia, which are designated as wCauA and wCauB, respectively. In both insects, the Wolbachia populations increased greatly during embryonic and larval stages. The Wolbachia population doubled every 3.6 days on average in E. kuehniella larvae, whereas those of wCauA and wCauB doubled every 2.1 days in C. cautella larvae. The populations of wCauA and wCauB that had been transferred into the E. kuehniella background increased at similar rates to that of wKue in the natural host E. kuehniella, suggesting that the host genetic background influences Wolbachia proliferation. To examine whether the populations of the two Wolbachia variants in double infection is regulated collectively or independently, we measured the infection load in the ovaries of three transfected E. kuehniella lines in different infection states: single infection with wCauA, single infection with wCauB, and double infection. The density of each Wolbachia variant did not differ significantly between the singly and doubly transfected hosts, suggesting independent regulation.
Wolbachia is a group of Rickettsia-like bacteria that are present in a wide range of arthropods and filarial nematodes. Wolbachia resides intracellularly and is inherited maternally by transovarial transmission. In arthropod hosts, the infection can cause various reproductive alterations, such as cytoplasmic incompatibility (CI), thelytokous parthenogenesis, male killing and feminization (reviewed in Werren, 1997; Bourtzis and Braig, 1999; Stouthamer et al., 1999). These reproductive alterations enhance the spread of Wolbachia in the host population by increasing the number of infected females either directly (feminization and parthenogenesis) or indirectly by increasing the relative fitness of infected females against uninfected counterparts (male killing and CI).
Since Wolbachia is inherited vertically from generation to generation and establishes a long term association with its host, its proliferation in a host is expected to be controlled so that the density is maintained high enough to ensure transovarial transmission while being low enough not to cause pathology. The discovery of a virulent Wolbachia variant in Drosophila melanogaster may imply the importance of such regulation of Wolbachia proliferation. This variant, named popcorn, proliferates to abnormally high density especially in nervous tissue, and reduces the host lifespan (Min and Benzer, 1997).
The density of Wolbachia has often been measured in conjunction with the intensity of CI. Many studies have suggested a positive correlation between the infection load and CI strength (e.g., Breeuwer and Werren 1993; Bressac and Rousset, 1993; Turelli and Hoffmann, 1995; Noda et al., 2001). It has also been shown that Wolbachia density varies depending on both bacterial genotype and host species. For example, two Wolbachia variants, wRi and wHa, differ widely in density although they are carried by the same host species, Drosophila simulans (Bourtzis et al., 1996). The influence of host genetic background on the infection load was revealed by artificial transfers of Wolbachia in Drosophila, in which the same Wolbachia variant was maintained at higher densities in D. simulans than in D. melanogaster (Boyle et al., 1993; Poinsot et al., 1998).
The density of Wolbachia is determined by the original infection in host egg and subsequent bacterial proliferation. To examine whether Wolbachia proliferation is determined by the bacterial genotype or is regulated by the host, it is of use to analyze the infection dynamics in various host-Wolbachia combinations in which either the host or bacterial effect can be extracted. The involvement of Wolbachia genotype in the determination of proliferation patterns has recently been demonstrated in the adzuki bean beetle, Callosobruchus chinensis, and in D. simulans. In C. chinensis three Wolbachia variants harbored by the same multiply infected host exhibited different population growth patterns (Ijichi et al., 2002). In D. simulans, the popcorn variant transferred from D. melanogaster and wRi showed different infection dynamics (McGraw et al., 2002). On the other hand, a comparison of popcorn infection between D. melanogaster and D. simulans revealed that host species affects the bacterial proliferation rate (McGraw et al., 2002). Thus it appears that Wolbachia proliferation is controlled by a complex interaction of bacterium and host.
E. kuehniella is infected with A-group Wolbachia (wKue) designated by Werren et al. (1995), and expresses a partial CI (Sasaki and Ishikawa, 1999). C. cautella is doubly infected with A- and B-group Wolbachia variants, which are referred to as wCauA and wCauB, respectively, and expresses complete CI (Sasaki and Ishikawa, 1999). We recently performed a transfer of Wolbachia from C. cautella to E. kuehniella, and generated three transfected lines, namely singly infected with wCauA, singly infected with wCauB, and doubly infected (Sasaki et al., 2002). These doubly infected hosts and transfected lines provide an ideal system for studying the extent to which the control of Wolbachia proliferation is influenced by bacterial genotype and host species.
In the present study, to further understand the basis for the regulation of Wolbachia proliferation, we investigated the growth pattern of the Wolbachia population in two naturally infected hosts and the transfected E. kuehniella carrying wCauA and wCauB. We also measured the infection load in the doubly transfected and singly transfected E. kuehniella lines, in order to determine whether the densities of the two Wolbachia variants in double infection are regulated collectively or independently.
MATERIALS AND METHODS
Wolbachia-infected E. kuehniella and C. cautella were originally collected in Tsuchiura, Japan. Transfer of Wolbachia from C. cautella to E. kuehniella was performed by embryonic microinjection as described by Sasaki and Ishikawa (2000). Although naturally infected C. cautella and E. kuehniella expressed CI, wCauA induced male killing at host embryonic stage when transferred into E. kuehniella. The E. kuehniella line transfected with only wCauB expressed partial CI (Sasaki et al., 2002). The transfected lines had been reared for more than 10 generations before the present study was performed.
The insects were maintained on a diet consisting of wheat bran, water, glycerol and dried yeast (100 : 5 : 10 : 5 w/w) under a 16 hr : 8 hr light : dark cycle at 25°C. To minimize the possible effects of rearing density on the infection load of Wolbachia (Sinkins et al., 1995; Hoffmann et al., 1998), the larvae were reared under uncrowded conditions (approximately 1 insect per 1 g of diet).
DNA was extracted from the whole tissues of insects or ovaries collected by dissecting adult females. Each sample was homogenized in 500 μl of a buffer (10 mM Tris-HCl, 150 mM NaCl, 10 mM EDTA, 0.1% SDS, pH=7.5) containing proteinase K at 0.1 mg/ml, and incubated at 55°C for 3 hr. The solution was extracted with phenol saturated with TE (10 mM Tris-HCl, 1 mM EDTA, pH=7.5), and then extracted with chloroform. DNA was precipitated by adding 500 μl of ethanol to 250 μl of the aqueous phase and dissolved in 500 μl of TE. When the samples were prepared from the eggs or one-day-old larvae, 20 μg of glycogen was added to the aqueous phase prior to ethanol precipitation, and DNA was finally dissolved in 200 μl TE.
Real time quantitative (RTQ)-PCR
To estimate Wolbachia densities, the copy number of the groEL gene was quantified by RTQ-PCR in a Light Cycler (Roche Diagnostics). The amplification reaction was monitored using a set of fluorescent probes specific for the PCR product.
In the measurement of wKue and wCauA, a 264 bp fragment of the groEL gene was amplified by using the primers groEL415AF [5′-CAA TGT CTC GTA CAA TTT CTC-3′ (positions 395–415)] and groEL641AR [5′-GAT TAT CAA GCT CCA CGA-3′ (positions 658–641)]. The PCR product was detected with the probes groELA1 [5′-AAA GAG TTA GAA GTT GAG CTG ACT ACT GG-3′-FITC (positions 553–581)] and groELA2 [LC-Red 640-5′-ATG CAA TTT GAT CGC GGT TAT CTC TCT CCG-3′ (positions 583-612)]. These primers and probes were designed based on the region where the sequences are identical between wKue and wCauA. The groEL gene of wCauB was amplified with the primers groEL415BF [5′-CAA TGT CTC GCA CAA TTT CTT-3′ (positions 395–415)] and groEL641BR [5′-GAT CAT CAA GCT CCA CGC-3′ (positions 658–641)], and the product was detected with the probes groELB1 [5′-GAA GTT GAA CTT ACA ACT GGT ATG CAG TT-3′-FITC (positions 562–590)] and groELB2 [LC-Red 640-5′-GAC CGT GGT TAT CTC TCT CCA TAC-3′ (positions 592–615)].
Standard solutions were prepared from the PCR products amplified from wCauA and wCauB using the primers groEL49F [5′-GTT GCA AGA AGC CTT TCG TG-3′ (positions 29-49)] and groEL827R [5′-CCA AAA CCT GGA GCT TTT ACT G-3′ (positions 848-827)]. The PCR products were electrophoresed on 1.0% agar-ose gel, extracted from the gel using QIAquick Gel Extraction Kit (Qiagen), and quantified based on the optic absorbance at 260 nm. Ten-fold dilutions from 10 pg/μl to 1 fg/μl were used to draw standard curves.
Infection dynamics of Wolbachia in E. kuehniella and C. cautella
The changes of Wolbachia titer during host development were monitored in E. kuehniella and C. cautella by a quantitative PCR. Females were subjected to the measurements, except for one-day-old larvae in which the females and males could not be distinguished.
Fig. 1 shows the temporal infection dynamics of the Wolbachia population in terms of groEL gene copy number. In naturally infected E. kuehniella, the population of wKue increased from 5.7×105 to 2.2×108 during larval development, and thereafter remained at an almost constant level. At the larval stage, the wKue population doubled every 3.6 days (R2=0.97) on average (Fig. 1A). The proliferation rates of the two Wolbachia variants in C. cautella larvae were similar to each other: the populations of wCauA and wCauB doubled every 2.1 days (R2=0.97) and 2.1 days (R2=0.98), respectively (Fig. 1B). After pupal ecdysis, the population of wCauA increased, especially at adult stage, while that of wCauB did not.
The changes of Wolbachia density in eggs are shown in Table 1. The duration of embryonic development of E. kuehniella and C. cautella was about 6 days and 4 days at 25°C, respectively. The density of wKue doubled in 3.6 days (R2=0.96), whereas that of wCauA and wCauB doubled in 2.5 days (R2=0.54) and 1.6 days (R2=0.85), respectively. Thus, the growth rate of the Wolbachia population during the host embryonic development was higher in C. cautella than in E. kuehniella as during the larval development.
Changes of Wolbachia density during embryonic development of E. kuehniella and C. cautella
The differential proliferation rates of Wolbachia observed in the two naturally infected hosts can be accounted for by the differences in both the bacterium and host species. In order to examine the extent to which the proliferation rate is influenced by the bacterial genotype and host genetic background, the infection dynamics in a transfected E. kuehniella line that carried wCauA and wCauB was investigated (Fig. 1C). In the transfected larvae, the Wolbachia populations increased more slowly than those in C. cautella, and the growth rates were similar to that of wKue in E. kuehniella: the populations of wCauA and wCauB doubled every 4.2 days (R2=0.95) and 3.9 days (R2=0.90), respectively. The proliferation rates in eggs were not examined because wCauA causes male killing in E. kuehniella embryos.
Wolbachia density in host ovary
In Table 2, are given the Wolbachia densities in the ovaries of the two naturally infected hosts and three transfected E. kuehniella lines in different infection states: single infection with wCauA, single infection with wCauB, and double infection. The density of wCauA was significantly higher than that of wCauB in C. cautella ovary. A difference in infection load of wCauA and wCauB was also observed in comparisons of the two variants in the doubly transfected E. kuehniella line and also between the two singly transfected lines. For each of wCauA and wCauB, the infection load did not differ when compared between the doubly and singly transfected E. kuehniella lines.
Infection density of Wolbachia in the ovary of E. kuehniella and C. cautella
In this study, we estimated the population growth rate of different Wolbachia variants in the same host and that of the same Wolbachia in different hosts. Using this approach we can determine the relative importance of bacterial geno-type and host genetic background in the determination of the proliferation rates. Comparisons of different Wolbachia variants in the same host did not show a remarkable difference in the rate of proliferation during the host larval development. The populations of wCauA and wCauB infecting the same host increased at similar rates in either C. cautella or E. kuehniella (Fig. 1, B and C). In addition, the proliferation rate of both wCauA and wCauB in the E. kuehniella back-ground was similar to that of wKue (Fig. 1, A and C). In contrast, host species had an obvious influence on Wolbachia proliferation: the populations of wCauA and wCauB increased more rapidly in C. cautella than in E. kuehniella (Fig. 1, B and C). Thus, our data suggested that host genetic background plays a major role in the determination of the Wolbachia proliferation rate, though it should be noted that Wolbachia genotype has also been shown to affect the proliferation rate in C. chinensis (Ijichi et al., 2002) and D. simulans (McGraw et al., 2002).
Interestingly, the densities of wCauA, wCauB and wKue were comparable at the late larval stage regardless of the host species they infected, which is because the duration of larval development of E. kuehniella is longer than that of C. cautella while the Wolbachia proliferation rate is lower in E. kuehniella than in C. cautella. This may imply that the host growth rate influences Wolbachia proliferation, though the mechanism for this is unclear. The duration of larval development of C. cautella and E. kuehniella can be controlled by rearing conditions such as temperature and humidity. The relationship between Wolbachia proliferation and the growth rate of the host could be addressed by examining Wolbachia density in hosts maintained under different conditions.
After pupal ecdysis, the population of Wolbachia did not change much. It stayed at almost constant levels except for wCauA in C. cautella adult. The proliferation and death of Wolbachia may be in equilibrium in the pupae and adults. Many larval cells undergo apoptosis following metamorphosis in pupae, and the fatbody tissue in adults is degraded as the ovary develops. A portion of the Wolbachia population may be lost in tissues that are degraded as host development proceeds. The increase of wCauA population in C. cautella adult could be explained by its high infection density in the ovarian tissue (Table 2), considering that the ovaries develop greatly after adult emergence and amount to approximately half of the whole tissues in weight.
The density of wCauA was higher than that of wCauB in the ovary of C. cautella and E. kuehniella (Table 2) and in C. cautella eggs (Table 1). wCauA might enter the germ line cells more efficiently or proliferate more quickly in ovarian tissue than wCauB. It is possible that the two Wolbachia variants differ in tissue tropism. Indeed, a remarkable difference in tissue tropism has been described among the three Wolbachia variants in C. chinensis (Ijichi et al., 2002) and between wRi and popcorn in D. simulans (McGraw et al., 2002).
In doubly infected hosts, two forms of density regulation can be assumed. One is that the total number of the two Wolbachia variants is regulated, which may occur if the two variants compete with each other for some limited resources in the same host cell. The other is that the presence of one Wolbachia variant does not affect the proliferation of the other variant, and the density of each variant is regulated independently. Our data on infection load in the three transfected E. kuehniella lines supported the latter possibility because the density of each of wCauA and wCauB did not differ between the singly and doubly infected lines. Since the host genetic background affects the growth rate of the Wolbachia population, it is conceivable that there may be a host factor that limits the bacterial proliferation. The proliferation of wCauA and wCauB may be limited by different host factors. It is also possible that wCauA and wCauB are spatially segregated and do not affect each other. Histological observations are needed to understand the dynamics of double infection further.
In the present study, quantitative PCR was performed with groEL as the target gene. The groEL gene is known to be present in a single copy per genome in the Rickettsiaceae group (Segal and Ron, 1996) to which Wolbachia belongs (O'Neill et al., 1992). Masui et al. (1997) analyzed the intergenic, less conserved, sequences between the groES and groEL genes in various Wolbachia variants, and reported that only one sequence was found from each Wolbachia variant, suggesting that Wolbachia also has one groE operon per genome. However, we should emphasize that gene copy number does not always reflect the number of microbial cells. Buchnera, the intracellular symbiont of aphids, has more than 100 copies of a genome in a single bacterium (Komaki and Ishikawa, 1999). The copy number of the genome of Wolbachia has not been examined. In addition, quantitative PCR counts gene copies derived from dead cells as well as those from living cells. Therefore, the groEL copy number shown in this paper should be considered as an index of relative density rather than the number of viable Wolbachia cells.