Distribution and recombination of Wolbachia endosymbionts in Korean coleopteran insects

Abstract

Background

Wolbachia are among the most prevalent endosymbiotic bacteria and induce reproductive anomalies in various invertebrate taxa. The bacterium has huge impacts on host reproductive biology, immunity, evolution, and molecular machinery. However, broad-scale surveys of Wolbachia infections at the order scale, including the order Coleoptera, are limited. In this study, we investigated the Wolbachia infection frequency in 201 Coleopteran insects collected in Korea.

Results

A total of 26 species (12.8%) belonging to 11 families harbored Wolbachia. The phylogenetic trees of based on partial 16S rRNA gene sequences and partial Wolbachia surface protein (wsp) gene sequences were largely incongruent to that of their hosts. This result confirms that Wolbachia evolved independently from their hosts,

Conclusion

Phylogenetic trees suggest that complex horizontal gene transfer and recombination events occurred within and between divergent Wolbachia subgroups.

Background

Wolbachia are highly prevalent endosymbiotic bacteria known to induce reproductive anomalies, such as cytoplasmic incompatibility, feminization, male-killing, and parthenogenesis, in various arthropod species, nematodes, isopods, and mites (Stouthamer et al. 1997, Werren and Windsor 2000; Zug and Hammerstein 2012). Bacteria in the genus have major impacts on host reproductive biology, immunity, evolution, and molecular machinery. Despite the controversy over the taxonomic status, the prevailing view is that Wolbachia in various hosts should be considered a single species divided into 14 supergroups (Glowska et al. 2015; Lindsey et al. 2016; Lo et al. 2002; Ramírez-Puebla et al. 2016). The genus has been a focus of research owing to its potential to control host populations, especially mosquito populations, by artificially infecting cytoplasmic incompatibility-inducing strains (Bourtzis et al. 2014; Werren 2008; Xi et al. 2006). Recent success in field trials has provided a new and safe approach for the control of Aedes aegypti, a vector of fatal viruses, such as dengue, Zika, and chikungunya (O'Neill 2016; O'Neill et al. 2018). Similar approaches for control are not necessarily limited to mosquito species but can be applied to virtually any host pest (Rostami et al. 2016).

Despite its ecological and evolutionary importance and application, in Korea, infection surveys of the bacterium have been limited to specific taxa (Choi et al. 2015; Jeong et al. 2009a and b; Jeong et al. 2012; Park et al. 2016). Since the order, Coleoptera is the largest group of insects and includes pest species affecting economic activity in Korea (see Moon and Lee 2015), we determined to investigate the Wolbachia infection frequency in the insects collected in Korea.

To our knowledge, this is the first intensive survey of Wolbachia infection at the order level in Korea.

Results and discussion

Wolbachia infection frequency

In the survey, 26 out of 201 coleopteran species were infected with Wolbachia and the genes are annotated (Tables 1 and 2, and Additional file 1: Table S1). Among 27 families, we detected Wolbachia in 10. For 18 families, we examined fewer than 4 species. Among families for which at least 5 species were examined, Attelabidae showed the highest infection frequency (4 out of 9 species), followed by Oedemeridae (1 out of 7 species), Chrysomelidae (10 out of 39 species), and Curculionidae (2 out of 15 species) (Table 1). On the other extreme, infection was not detected in any Carabidae samples. Recent two in-depth reviews show that the Wolbachia infection frequency in beetle species is about 38% and 27% respectively (Kajtoch et al. 2019; Kajtoch and Kotásková 2018). In our analysis, only 12.9% of beetle species harbored the bacterium. This discrepancy may be explained by a difference in the sampling method. Since we examined a single specimen per species, partial infections within populations were not resolved. Further tests should include multiple specimens for each species. All gene sequences named after the host insect species were deposited at GenBank and the accession numbers are in Table 2 and Additional file 1: Table S1. There have been studies on the Wolbachia infection status in various coleopteran insects (Werren et al. 1995; Kajtoch and Kotásková 2018). Incongruence in infection frequency from them may be caused by geographical variation and taxonomic composition.

Table 1 Infection frequency at the family level

Table 2 Annotation of Wolbachia from beetle species

Phylogeny of Wolbachia

Phylogenetic trees based on 16S rRNA and wsp were largely incongruent (Fig. 1). This result confirms that Wolbachia evolved independently from their hosts, as indicated by Kajtoch and Kotásková (2018).

Fig. 1

Phylogenetic analysis of the host CO1 gene and the partial 16 s rRNA gene from Wolbachia infecting the coleopteran host insects: Chrysomelidae, Carabidae, Languriidae, Attelabidae, Cerambycidae, Endomychidae, Oedemeridae, Byturidae, Brenthidae, Psephenidae

Interestingly, some species, such as Lixus maculates, were superinfected with 5 strains of Wolbachia. These strains likely diverged after a single infection, as evidenced by the monophyletic clustering of each gene (Fig. 2). However, Anadastus praeustus, Smaragdina semiaurantiaca, and Aulacophora nigripennis were superinfected based only on wsp gene diversity. For Wolbachia-infecting Basilepta pallidula, Baris dispilota, Apion collare, and Byturus aestivus, we detected incongruence between 16S rRNA and wsp phylogenies, as observed for Byturus unicolor in this study (Fig. 2). This finding indicates that the genes evolved independently.

Fig. 2

Phylogenetic analysis of the partial 16 s rRNA gene and the partial wsp gene from Wolbachia infecting the coleopteran insects (for annotation, see Table 2)

The Wolbachia strain in Anadastus praeustus showed five wsp alleles. This can potentially be explained by synonymous substitutions in 16S rRNA and wsp, rather than by recombination (data not shown). Wolbachia strains infecting Lixus maculates told a different story. The 16S rRNA gene of Wolbachia-infecting Lixus maculatus could be assigned to two main subgroups (three in subgroup A and two in B). However, the wsp sequences exhibited high similarity and were all assigned to subgroup A (Fig. 2). This implies that strains in the two subgroups infected the host and accumulated synonymous substitutions after recombination. Under the assumption that the 16S rRNA and wsp phylogenies are strictly congruent, strains that acquired wsp of subgroup B were selected against. Since the two genes are functional, this suggests that the strain with wsp classified as subgroup A has deleterious mutations in essential genes other than wsp. Another less likely explanation is that these strains were simply lost due to drift with no sex ratio drive; however, this does not explain how only particular strains were affected. Previously Wolbachia was suspected to induce parthenogenesis and oocyte formation in the host species (Chen et al. 2012; Rodriguero et al. 2010; Son et al. 2008). However, we were unable to infer the effect of infection on hosts’ life history due to the lack of general biology of the Korean coleopteran insects.

Conclusions

In this study, 12.8% of Korean beetle species are infected with Wolbachia. Phylogenetic trees suggest that complex horizontal gene transfer and recombination events occurred within and between divergent Wolbachia subgroups resulting in largely incongruent phylogenetic relationships.

Materials and methods

Study materials

Coleopteran species were mostly collected by light trapping at night between May and August 2008 in various locations in South Korea (Additional file 1: Table S1). They were brought into the laboratory and identified carefully. Voucher specimens were deposited at the NAAS repository. Genomic DNA was extracted from a leg of the specimens using the DNeasy Mini Kit (Qiagen, Hilden, Germany) or the AccuPrep® Genomic DNA Extraction Kit (Bioneer, Daejeon, Korea) and stored at − 80 °C until use.

Determination of Wolbachia infection

All PCRs were performed using a total volume of 20 μl in AccuPower® PCR PreMix (Bioneer) and a thermocycler (Biometra, Tubingen, Germany).

The quality of genomic DNA was determined by PCR using a universal arthropod 28S rRNA primer set (forward: TAC CGT GAG GGA AAG TTG AAA and reverse: AGA CTC CTT GGT CCG TGT TT) or a universal CO1 primer set (forward: GGT CAA CAA ATC ATA AAG ATA TTG G and reverse: TAA ACT TCA GGG TGA CCA AAA AAT CA) (Jeong et al. 2009a, 2009b). The temperature profile for 28S rRNA diagnostic PCR was 2 min of pre-denaturation at 94 °C, followed by 38 cycles of 30 s at 94 °C, 50 s at 58 °C, 90 s at 72 °C, and a final extension for 10 min at 72 °C.

A Wolbachia-diagnostic marker (forward: CAT ACC TAT TCG AAG GGA TAG and reverse: AGC TTC GAG TGA AAC CAA TTC) was used to determine the infection status of specimens. Genomic DNA of Wolbachia-infected Cadra cautella (Pyralidae: Lepidoptera) was used as a positive control. The temperature profile for Wolbachia-specific diagnostic PCR was 2 min of pre-denaturation at 94 °C, followed by 38 cycles of 30 s at 94 °C, 45 s at 55 °C, 90 s at 72 °C, and a final extension for 10 min at 72 °C.

Positive PCR samples for 28S rRNA and/or CO1 and the Wolbachia diagnostic marker were scored as infected. Additionally, when a sample was negative for 28S rRNA and/or CO1 but positive for the Wolbachia marker, the species was scored as infected.

Genotyping of Wolbachia

Multilocus sequence typing (MLST) is widely used for genotyping and assignment to supergroups in Wolbachia (Baldo et al. 2006; but see Bleidorn and Gerth 2017). The MLST was not adopted since this study was a basic survey on Wolbachia from Korean Coleopteran insects. Instead, we chose two conventionally used genes. For samples identified as infected, PCR was performed with genomic DNA to obtain the partial 16S rRNA gene sequence and partial Wolbachia surface protein (wsp) gene sequence. For Wolbachia, nested PCR was performed following the methods described by Weeks et al. (2003). For the initial PCR, the 16S rDNA primer set was used (27f and 1513r) (Weisburg et al., 1991). The temperature profile was 4 min of pre-denaturation at 94 °C, followed by 35 cycles of 1 min at 94 °C, 1 min at 52 °C, and 1 min at 72 °C, and a final extension for 5 min at 72 °C. Then, Wolbachia-specific PCR was performed using the Wolbachia-specific 16S rDNA primer set (176f and 1012r) taken from O’Neill et al. (1992). One microliter of initial PCR product was diluted 10-fold and used as template DNA. The temperature profile for PCR was 4 min of pre-denaturation at 94 °C, followed by 35 cycles of 1 min at 94 °C, 1 min at 50 °C, and 1 min at 72 °C, and a final extension for 5 min at 72 °C. The partial wsp gene of the bacterium was obtained using the following primer set: 81f, TGG TCC AAT AAG TGA TGA AGA AAC; 691r, AAA AAT TAA ACG CTA CTC CA (Zhou et al. 1998). The temperature profile was 5 min of pre-denaturation at 94 °C, followed by 35 cycles of 1 min at 94 °C, 1 min at 55 °C, and 1 min at 72 °C, and a final extension for 5 min at 72 °C.

Phylogenetic analyses

A multiple sequence alignment of all partial 16S rRNA genes and wsp genes from Wolbachia was generated using ClustalW embedded in MEGA (ver. 6, Tamura et al. 2013). The protein-coding wsp gene sequences were translated into amino acid sequences before alignment. These sequences were then reverse-translated into DNA sequences.

The alignments were manually adjusted. All gaps were manually removed, resulting in 765 bp of the partial 16S rRNA gene, of which 31 sites were parsimony-informative, and 509 bp of wsp, of which 172 sites were parsimony-informative. The HKY + I and TVM + I + G models of evolution were used to estimate phylogenies for 16S rRNA and the wsp, respectively, the best-fit models were determined by log-likelihood ratio tests using ModelTest (ver. 3.7, Posada and Crandall 1998).

Likelihood analyses were performed using PAUP* (ver. 4.0b, Swofford 2002) by a heuristic search with 10,000 random sequence replicates and the tree-bisection-reconnection branch swapping algorithm. Nodal support was estimated using 500 bootstrap replicates with 10 random addition sequence replicates per bootstrap replicate.

The phylogenetic trees were visualized and manually edited using TreeGraph 2 (Müller and Müller 2004).