Institut Cavanilles de Biodiversitat i Biologia Evolutiva, Universitat de València, València, Spain,1 Laboratoire de Biologie Fonctionnelle Insectes et Interactions, UMR INRA/INSA de Lyon, Villeurbanne Cedex, France2
Received 19 May 2004/ Accepted 7 July 2004
| ABSTRACT |
|---|
|
|
|---|
-proteobacterium that belongs to the Wolbachia genus. Using in situ hybridization with specific bacterial probes designed for symbiont 16S rDNA sequences, we have shown that Wolbachia was represented by only a few minute bacteria surrounding the S symbionts. Moreover, the observed B. aphidicola and the S symbionts had similar sizes and were housed in separate specific bacterial cells, the bacteriocytes. Interestingly, in contrast to the case for all aphids examined thus far, the S symbionts were shown to occupy a similarly sized or even larger bacteriocyte space than B. aphidicola. These findings, along with the facts that C. cedri harbors the B. aphidicola strain with the smallest bacterial genome and that the S symbionts infect all Cinara spp. analyzed so far, suggest the possibility of bacterial replacement in these species. | INTRODUCTION |
|---|
|
|
|---|
In aphids (Hemiptera: Aphididae), the primary endosymbiont Buchnera aphidicola, one of the
-Proteobacteria, is maternally transmitted by entering the embryos of each generation. The insect life cycle is complex, involving alternations of sexual and asexual reproduction. During their reproductive phase, aphids contain within their body cavity a bilobed structure called the bacteriome that consists of 60 to 90 uninucleate, polyploid bacteriocytes. Inside the bacteriocytes, B. aphidicola is enclosed within host-derived vesicles (4).
In nature, all but a few aphid populations harbor B. aphidicola (4, 32). Exceptions include certain aphid species of the tribe Ceratiphidini that possess extracellular yeast-like symbionts belonging to the subphylum Ascomycotina in the abdominal hemocoel instead of B. aphidicola. The latter association was interpreted as being a symbiont replacement, with an intracellular bacterium being replaced with an extracellular fungus (14). The possibility of natural endosymbiont replacement in aphids has also been postulated by other authors (9, 28).
In addition to harboring B. aphidicola, some aphid populations harbor other intracellular bacteria, which are commonly referred to as secondary symbionts (S symbionts). Typically, S symbionts are present in various lineages of aphids. However, in most aphids examined thus far, S symbionts are only represented in small numbers and occupy a limited bacteriocyte area compared to B. aphidicola (4, 17). Moreover, the S symbionts are not always confined in specific bacteriocytes and can be found in gut tissues, glands, body fluids, and cells that are surrounding, or even invading, the primary bacteriocytes. At present, five different S symbionts have been found. They include three different lineages within the
-proteobacteria, provisionally called PASS or R type (7, 39, 50), PABS or T type (12, 39), and PAUS or U type (39), one lineage within the
-proteobacteria, referred to as the PAR (pea aphid rickettsia) symbiont (8, 27, 48), and a Spiroplasma symbiont (16). S symbionts seem to be the result of multiple independent infections (37), and although they are usually maternally inherited (4, 6), their transmission may also occur horizontally from one host to another.
Wolbachia is an obligatory intracellular
-proteobacterium that infects diverse groups of insects and most species of filarial nematodes (3, 46, 52, 54). In arthropods, it causes reproductive alterations to the host, such as cytoplasmic incompatibility (CI), parthenogenesis, genetic male feminization, male killing, and virulence in Drosophila melanogaster (5, 26, 34, 36, 52, 54). Although there have been established cases of horizontal transfer, Wolbachia organisms are inherited mainly through the maternal lineage of their hosts by vertical transmission. A phylogenetic analysis of worldwide Wolbachia strains by the use of fast-evolving genes such as ftsZ and wsp split invertebrate Wolbachia strains into two major clades, designated A and B, and further divided them into subgroups (55, 57). Genomic size determination has shown that Wolbachia genomes are much smaller than the genomes of free-living bacteria (ranging from 0.95 to 1.5 Mb) (43) and closer to the genome sizes of other intracellular bacteria (1, 19, 40, 45, 51). Contrary to the genomes of these endosymbionts, the recently sequenced genome of Wolbachia pipientis wMel revealed the existence of very large amounts of repetitive DNA and mobile genetic elements, suggesting that they may have played a key role in shaping the evolution of Wolbachia (56).
Cinara cedri belongs to the subfamily Lachninae of aphids and lives in colonies disposed on branches on the gymnosperm Cedrus atlantica and Cedrus deodora. Gil et al. (18) estimated the genome size of the B. aphidicola strain associated with C. cedri to be 450 kb by pulsed-field gel electrophoresis, making it the smallest bacterial genome reported so far.
In the present paper, we describe and characterize two other bacteria found in the aphid C. cedri in addition to B. aphidicola: an S symbiont (PASS, or R type) that was previously described for aphids and Wolbachia. Furthermore, based on the relative abundance of the primary and secondary endosymbionts, we provide additional evidence for endosymbiont competition and replacement during evolution.
| MATERIALS AND METHODS |
|---|
|
|
|---|
Electron microscopy. Adults, embryos, and eggs from aphids were dissected under a microscope in 0.9% NaCl, prefixed in 0.1 M phosphate buffer (PB) (pH 7.4) at 4°C for 24 h, and washed several times in 0.1 M PB. They were then postfixed in 2% osmium tetroxide in 0.1 M PB for 90 min in darkness, dehydrated in ethanol, and embedded in araldite (Durcupan; Fluka). Semithin sections (1.5 µm) were cut with a diamond knife and stained with toluidine blue. Ultrathin (0.05 µm) sections were cut with a diamond knife, stained with lead citrate, and examined under a transmission electron microscope (JEOL-JEM1010).
16S rDNA amplification and sequencing of B. aphidicola, S symbiont, and Wolbachia from C. cedri.
To amplify the 16S ribosomal DNAs (rDNAs) of both B. aphidicola and the S symbiont from C. cedri, we used the universal
-proteobacterial primers 16Sup1 (5'-AGAGTTTGATCATGGCTCAGATTG-3') and 16Slo1 (5'-TACCTTGTTACGACTTCACCCCAG-3'). The PCR conditions were 94°C for 2 min, followed by 30 cycles of 94°C for 15 s, 60°C for 30 s, and 72°C for 1 min. To assess the presence of P (primary) and S symbionts, we digested the PCR products with the SalI restriction enzyme, which specifically recognizes B. aphidicola, and three diagnostic enzymes, SacI, XbaI, and ClaI, that discriminate among R-, T-, and U-type S symbionts, respectively (37, 39). To further confirm the restriction analysis results, we cloned the PCR products into a pGEM T-vector (Promega) and sequenced them by using the universal primers T7 and SP6.
To detect the presence of
-proteobacteria, we designed a pair of specific primers from database alignments of 16S rDNAs by using an alignment of sequences from the following organisms: the PAR symbiont, Rickettsia japonica, and Wolbachia from Sitophilus oryzae, Drosophila mauritiana, and Drosophila sechellia (Table 1). The designed primers were as follows: 16SWup (5'-GCCTAACACATGCAAGTCGAA-3') and 16SWlo (5'-AGCTTCGAGTGAAACCAATTCCC-3'), corresponding to positions 24 to 45 and 1379 to 1357, respectively, of the 16S rDNA partial sequence of the Wolbachia strain associated with D. sechellia (accession number U17059). PCR mixtures consisted of 1.5 U of Taq DNA polymerase (Promega), a 200 µM concentration of each deoxynucleoside triphosphate, a 300 nM concentration of each primer, and 10 ng of DNA template in a final volume of 50 µl. An additional positive control of Wolbachia from S. oryzae (20) was used. The PCR amplification conditions were 94°C for 2 min, followed by 35 cycles of 94°C for 20 s, 63°C for 30 s, and 72°C for 90 s. The 1.3-kb PCR product obtained was purified (Qiaquick PCR purification kit; Qiagen) and directly sequenced by using the PCR primers. Simultaneous sequencing of the positive control discarded the possibility of a contamination by Wolbachia from S. oryzae. Searches in databases and a subsequent BlastN search confirmed the homology with Wolbachia within the
-proteobacterial 16S rDNAs. To resolve the phylogenetic position of the Wolbachia strain associated with C. cedri in the different subgroups into which Wolbachia has been divided, we amplified and sequenced a fragment of the wsp gene by using the primers wsp 81F (5'-TGGTCCAATAAGTGATGAAGAAAC-3') and wsp 691R (5'-AAAAATTAAACGCTACTCCA-3') according to the method described by Zhou et al. (57).
|
Maximum likelihood (ML), maximum parsimony (MP), and distance (neighbor-joining) algorithms were used to perform phylogenetic analyses. The ML and MP algorithms were performed with PAUP 4.0, using the evolution model given by Modeltest (35), and the distance algorithm was performed with Mega software (22), using the Kimura-2 parameter model and pairwise deletion. Bootstrap analyses were done with 1,000 replications, except for the analysis of wsp with the ML algorithm, which was done with 300 replications.
In situ hybridization. All experimental procedures for in situ hybridization were conducted under RNase-free conditions. For digoxigenin-labeled probes, we followed a protocol similar to a previously described one (17). For rhodamine-labeled probes, we followed another previously described procedure (20).
Four 5'-end-labeled digoxigenin probes were used. EUB338 was used previously (17), and the remainder were designed for the present work to specifically detect B. aphidicola (BuCc), the S symbiont (SCc), and Wolbachia (WCc) in C. cedri. Two 5'-end-labeled rhodamine probes (W1 and W2) were also used (20). They are available upon request.
| RESULTS |
|---|
|
|
|---|
|
PCR and sequencing of endosymbiont 16S rDNA.
Amplification of endosymbiont 16S rDNAs with universal
-proteobacterial primers yielded a band of approximately 1,500 bp. Digestion of the PCR product with diagnostic enzymes that discriminate among B. aphidicola and R-, T-, and U-type S symbionts (39) revealed the presence of two different 16S rDNA fragments that corresponded to two symbionts, specifically B. aphidicola and an R-type (or PASS) S symbiont (data not shown). Cloning, sequencing, and phylogenetic analyses of the two 16S rDNA bands further confirmed these results (see below).
Amplification with the specific primers designed to amplify
-proteobacteria yielded a band of 1,259 bp that was sequenced and compared with selected
-proteobacterial 16S rDNA sequences from databases by the use of BlastN.
Phylogenetic positions of C. cedri endosymbionts.
Phylogenetic trees obtained after applying three different phylogenetic procedures yielded similar topologies (Fig. 2). Figure 2a shows the topology of a tree obtained with 16S rDNAs from B. aphidicola associated with Acyrthosiphon pisum (BAp) and from three S symbionts (R, T, and U types) associated with different aphids (37) and with the two 16S rDNA
-proteobacterial sequences obtained from C. cedri. The sequence corresponding to B. aphidicola (BCc) clustered with the BAp sequence, while the second one clustered with the sequence corresponding to the S symbiont associated with Cinara tujafilina, which was previously assigned to the R type (or PASS) (37). This sequence was called SCce (R) (Table 1).
|
-proteobacterial sequences (Fig. 2b) clearly showed that the third bacterium found in C. cedri corresponds to the Wolbachia strain (wCce in Table 1). Due to the paucity of 16S rDNA nucleotide substitution in Wolbachia, the tree was useful to differentiate between Rickettsia and Wolbachia but not between Wolbachia strains. To accomplish this task, we performed a phylogenetic analysis as described by Zhou et al. (57), including the new sequence. As seen in Fig. 2c, the Wolbachia strain from C. cedri (wCce) clustered with strains wCon and wStri, which are associated with Tribolium confusum and Laodelphax striallus, respectively (Table 1). In situ endosymbiont localization. To physically localize C. cedri-associated bacteria, we performed in situ hybridization with specific oligonucleotide probes for 16S rDNAs. Since it was stated previously that insect tissues generally emit strong autofluorescence (17), we developed nonfluorescent in situ hybridization methods using digoxigenin-labeled probes. Using the eubacterial universal EUB338 16S rDNA probe, we succeeded in identifying and localizing the bacteriocytes of C. cedri in both adults and embryos (Fig. 3a). Two other digoxigenin-specific probes, BuCc and SCc, identified two types of bacteriocytes, those filled with B. aphidicola (Fig. 3b) and those filled with S symbionts (Fig. 3c). This observation confirmed the results obtained by electron microscopy and demonstrated the bacterial specificity of the probes. Furthermore, an intriguing bacterial space distribution was detected in the three figures, which exhibit three serial sections. Whereas B. aphidicola bacteriocytes were found outside of the bacteriome (Fig. 3b), S-symbiont bacteriocytes were concentrated in the interior (Fig. 3c). Interestingly, Fig. 3b and c and other examined slides (data not shown) show that the S symbiont seems to cover at least 60% of the total bacteriome surface.
|
| DISCUSSION |
|---|
|
|
|---|
The association between aphids and B. aphidicola is very ancient, and the congruence of phylogenetic trees of aphids and B. aphidicola indicates a unique infection event about 84 to 164 million years ago, followed by the coevolution of both partners (30, 53). During the accommodation to symbiotic life, B. aphidicola has suffered a drastic genome reduction as well as many important molecular and biochemical changes (4, 10, 29, 31).
In contrast to B. aphidicola, S symbionts seem to be the result of multiple independent infections, and in addition to their maternal transmission, they can also be horizontally transmitted from one host to another (37, 39), therefore not sharing a long evolutionary history with their hosts. Most studies on the presence of S-symbiont occurrences in aphids have been conducted on members of the tribe Macrosiphini from the subfamily Aphidinae. For the subfamily Lachninae, S symbionts have been visualized by histological techniques in Stomaphis yanonis, Nippolachnus piri, and Cinara pini (15, 17) and amplified by PCR from three new species belonging to the Cinara genus (C. cupressi, C. maritimae, and C. tujafilina). In these three species, the S symbiont was characterized as PASS (or R type) (37), which is the same bacterium found associated with C. cedri in the present work. This new finding provides support for the previously stated hypothesis of a stable association between R-type S symbionts and the Cinara host clade (37).
The different techniques used to visualize endosymbionts in aphids have shown that P and S symbionts exhibit different morphologies, and in all cases the sizes of B. aphidicola bacteriocytes overwhelmed those of S-symbiont bacteriocytes. In the present study, however, we found that the P and S symbionts associated with C. cedri have similar sizes (Fig. 1 and 3b and c), and the S symbiont seems to occupy a similar amount of or more space (up to 60%) than B. aphidicola. Therefore, since B. aphidicola associated with C. cedri possesses the smallest bacterial genome reported so far (450 kb), the possibility of a symbiont replacement in an advanced stage can be suggested.
A comparison of the three B. aphidicola sequenced genomes (40, 45, 51) revealed a high degree of gene order conservation since the last common symbiotic ancestor (for a different perspective, see reference 38). However, wide variations in genome size have been found, since B. aphidicola strains from several aphid subfamilies showed differences of up to 200 kb (18), indicating that genome degradation is still an ongoing process that is probably related to variations in the host lifestyle. Therefore, the evolution of the B. aphidicola genome appears to be degenerative rather than adaptive (29). This degenerative process raises the question of whether these endosymbionts are approaching a minimal genome stage for symbiotic life or are being driven toward extinction. In the latter case, S-symbiont competition with such a reduced and degraded B. aphidicola genome may compensate for the loss of the ability of B. aphidicola to support host fitness and eventually may replace the P symbiont.
Many aphid species lack S symbionts, which suggests that they may not be essential for host survival (4, 39). Nevertheless, different positive effects, such as rescue from heat damage (9, 27), host plant specialization and reproduction (41, 48), and resistance to parasitoid attack and other natural enemies (13, 33), have been described, indicating some putative roles for the S symbionts. The most direct evidence of S symbionts overtaking the role of B. aphidicola has been obtained by Koga et al. (21) through experiments investigating the biological effects of PASS on A. pisum strains. They have proven that infections with PASS enabled the survival and reproduction of B. aphidicola-free aphids. Interestingly, they showed that PASS invaded the bacteriocyte space in parallel with B. aphidicola elimination, establishing a novel endosymbiotic system. If this is so, then the distribution of P- and S-symbiont bacteriocytes in C. cedri could be interpreted as the P-symbiont being replaced with the S symbiont. In insects, the best evidence of an endosymbiont replacement was recently found in Sitophilus spp. of the family Dryophthoridae (24). The currently ongoing sequencing of B. aphidicola from C. cedri in our laboratory will help us to understand whether this reduced genome is still able to perform the necessary functions for symbiotic life or if some of these functions have been overtaken by the abundant S symbionts.
In spite of the high prevalence of Wolbachia throughout arthropod species, no aphid species has been described to be associated with these bacteria before. Rickettsia (PAR) is the only
-proteobacterium that was previously found in aphids, as it was found in the hemolymph of A. pisum with different levels of infection (7, 8, 27, 49). Here we report the presence of Wolbachia in the aphid C. cedri. If it is confirmed that C. cedri populations are naturally infected with Wolbachia, the phenotypic effects on the host should be further investigated. Nevertheless, according to its phylogenetic position (Fig. 2c), we hypothesize that CI is the most probable phenotypic effect (57). The other two host effects (i.e., parthenogenesis and male killing) cannot be excluded. Indeed, since noninfected C. cedri possesses sexual and asexual lineages, the presence of Wolbachia could increase the prevalence of asexual lineages, as previously described for the Hymenopteran group (42).
Genes encoding homologs of the type IV secretion system, used by many pathogenic bacteria to secrete macromolecules, are also present and expressed in Wolbachia (25). It has been postulated that Wolbachia secretes certain molecules that may participate in the expression of CI through the type IV secretion system. It would be interesting to investigate whether an active vir operon containing these genes is also present in strain wCce and whether the secretion of macromolecules affects the two other symbionts that are already established in C. cedri.
Variations in genome size in Wolbachia strains reach about 550 kb (43), which is even higher than the up to 200-kb variations found among B. aphidicola strains (18). This suggests that, similar to the case for B. aphidicola, adaptations to different host lifestyles are taking place in Wolbachia. The most reduced genomes correspond to Wolbachia strains infecting nematodes (0.95 to 1.1 Mb), which are considered to be more like classical mutualists, and the largest genome was found in the parasitic A group of Wolbachia (43). However, contrary to the gene order conservation observed in B. aphidicola strains, frequent rearrangement events during Wolbachia evolution have been observed (56), which could have important consequences on the virulence of the strains. A comparison of the physical and genetic maps of the virulent Wolbachia strain wMelPop with those of the closely related benign strain wMel have shown that the two genomes are largely conserved, with the exception of a single inversion in the chromosome (44). The rearrangements in Wolbachia likely correspond to the introduction and massive expansion of repeat element families that are absent from other obligate intracellular species (56).
Since Wolbachia has not been previously found in aphids, its presence in C. cedri may be the product of an infection by horizontal transfer from some other insect. Parasitoids have been proposed as a possible vector for transmitting Wolbachia (52). Thus, we hypothesized that the parasitoids of the subfamily Lachninae, the Aphidinae of the genus Pauesia, may be good candidates for horizontal transfer events. So far, no Wolbachia strain has been described as being present in any species of this genus.
| ACKNOWLEDGMENTS |
|---|
We thank J. M. Michelena and P. González for aphid species identification and Rosario Gil and Juan José Canales for critical reading of the manuscript. We also acknowledge the "Servicio de Secuenciación de ácidos nucléicos y proteínas" at SCSIE (Universitat de València) for sequencing support.
| FOOTNOTES |
|---|
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| Appl. Environ. Microbiol. | Infect. Immun. | Eukaryot. Cell |
|---|---|---|
| Mol. Cell. Biol. | J. Virol. | Microbiol. Mol. Biol. Rev. |
| ALL ASM JOURNALS |