Ecological genetics of Adalia beetles: variability and symbiotic bacteria in european populations of the ten-spot ladybird beetle Adalia decempunctata

Full Text

INTRODUCTION

Adalia decempunctata (Linnaeus, 1758), is a ladybird with ten spots on the elytra. It is a widespread Palearctic species and occurs in Europe from Scandinavia to Italy and from Portugal to the Urals. The easternmost record for this species is Yekaterinburg in Russia, with no reliable distribution records east of the Urals.

Adalia ladybirds, including A. decempuncta, are among the most morphologically variable Coccinellidae [1–3]. Previous studies on a highly variable species closely related to A. decempuncta, (A. bipunctata [Linnaeus 1758]), revealed significant polymorphism within this species not only in morphology, but also in mitochondrial DNA [4–7]. In A. bipunctata, 18 mitochondrial haplotypes were found [5, 6]. The mitochondrial diversity in A. decempuncta has not yet been studied.

More than 60% of insect species are infected with symbiotic bacteria, which often impair host reproduction [8]. In ladybirds, symbionts cause androcide, which is the death of male offspring, and accordingly, a shift in the sex-ratio of offspring toward females [9]. Coccinellidae ladybirds are especially susceptible to infection with symbiotic bacteria; 13 out of the 30 European species studied to date were found to be infected with bacterial symbionts of Wolbachia, Spiroplasma, and Rickettsia [9]. Bacteria are transmitted to offspring via transovarial transmission, along with the mother’s cytoplasm, which leads to the joint inheritance and spread of mitochondrial DNA and symbiont. Previous studies on the two-spot ladybird, A. bipunctata, found an association between certain mtDNA haplotypes and infection status with either Rickettsia or Spiroplasma [4, 6]. Rickettsia infection has also been documented in the ten-sot ladybird, A. decempunctata, in Germany [10], Sweden [11], and Great Britain [12]. However, neither Wolbachia nor Spiroplasma infection have been found to date in this species [10–12].

The objective of this work was to: 1) characterize the genetic diversity and uncover phylogeographic patterns in A. decempunctata populations sampled at nine European cities; 2) determine the frequency of bacterial infection, and assess the diversity and phylogenetic relationships between symbiotic bacteria taxa; and 3) compare both the genetic and morphological diversity between A. decempunctata and a closely related species, A. bipunctata.

MATERIALS AND METHODS

Ladybirds were collected in Moscow, Prague, Paris, Florence, Rome, and Hamburg in 2012 and 2015, and in Feodosia and Yalta in 2017. Additionally, mtDNA and Rickettsia diversity were examined in A. decempunctata from Sweden, where infection with bacterial symbionts had previously been verified [11]. Beetles were identified to species using morphological characters, according to the pattern of the elytra and pronotum [2]. Beetles were tested for infection with symbiotic bacteria (n = 199) and partial COI was sequenced for 92 individuals. Two individuals of Coccinella sp., living sympatrically with A. decempunctata in the town of Kem (Karelia, Russia), were also examined using mtDNA and bacterial DNA markers. Bacterial symbionts were sequenced from specimens of the two-spot ladybird A. bipunctata, collected in 2015 from Buryatia and Karelia in Russia. Further individuals of A. bipunctata (84 individuals, collected June 2009, St. Petersburg), were available from a previous study [6, 7] and were used in comparisons of color polymorphism and mtDNA variability with A. decempunctata from Prague (n = 116).

DNA was isolated from dry adult insects or those preserved in 70%–90% ethanol using the DIAtom™ DNA Prep Kit (Isogen, Moscow). Infection with bacterial symbionts was assessed by PCR using specific primers for three bacterial symbionts: the Spiroplasma 16S rRNA small subunit gene [13], the Rickettsia gltA gene [10], and the Wolbachia wsp gene [14]. The polymorphic region of the cytochrome oxydase subunit I gene (COI) of mtDNA was amplified using the primers C1j-1951 and C1N-2618 [4]. The second internal transcribed spacer (ITS2) of the rRNA gene ribosomal cluster was amplified with the primers 5.8S and 28S [15]. PCR products were isolated from agarose gels using the Clean up kit (Eurogen, Russia). DNA fragments were sequenced on an ABI 310 sequencer using the forward and reverse primers and the ABI PRISM BigDye Terminator Cycle Sequencing kit (Applied Biosystems, Foster City, CA, USA). Sequences were deposited in GenBank under the accession numbers KJ645081-KJ645085 and MK932842-MK932845 (mitochondrial haplotypes), KJ645086-KJ645094 (ITS2), and MK932846-MK932850 (Rickettsia gltA).

Data analysis was performed using MEGA6 [16] and DnaSP v5 [17]. A maximum likelihood dendrogram for the Rickettsia gltA sequences was constructed using the Tamura-Nei model and 1000 bootstrap iterations in the MEGA6 program [16]. A mtDNA haplotype network was constructed in NETWORK ver. 4.6.1.6 [18]. Sequence divergence between mitochondrial haplotypes was calculated as the average number of nucleotide substitutions per site between two sequences [16]. Following Zhivotovsky, intra-population diversity was determined using the indicator µ (average number of morphs) [19]. When analyzing the occurrence of symbionts in the samples, confidence intervals (%) were calculated using samples of ten or more individuals using the Clopper–Pearson method, as in [20–22].

RESULTS

DNA polymorphism of A. decempunctata

92 A. decempunctata from across the nine collection sites were successfully sequenced for COI (616 bp) (Table 1). Eight mitochondrial haplotypes were found, which differed at nine variable sites. All mutations were related to A-G or C-T transitions and were synonymous.

Three mitochondrial haplotypes were rather common and were detected at two or more collection sites. The haplotype H1 occurred in all nine populations and in 74% (n = 68) of individuals (Table 1). In the haplotype network, this particular variant is ancestral (root) for all other mitochondrial types of A. decempunctata (Fig. 1). All haplotypes were interconnected by one or two sequential mutations (Fig. 1). Two hypothetical mitochondrial haplotypes, between H6 and H2 and between H1 and H8, did not occur in the studied samples. The evolutionary sequence divergence between mtDNA haplotypes did not exceed 0.8%, in agreement with the estimated level of intraspecific difference.

 

Fig. 1. Intraspecific polymorphism of mtDNA haplotypes of A. decempunctata. Eight variable haplotypes are represented on the network in proportion to their occurrence in the population

 

Table 1

Place and year of collection of investigated A. decempunctata, the number of individuals with different mitochondrial haplotypes

Local population	Collection place	Year	Number of analysed individuals	H1	H2	H3	H4	H5	H6	H7	H8
Czech Republic, Prague	Ruzyne, Crop Research Institute (50°09’ N 14°30’ E)	2012	29	19	7	2	1	0	0	0	0
France, Paris	Near the metro station Cite (48°51’ N 2°20’ E)	2012	3	3	0	0	0	0	0	0	0
Russia, Moscow	“Neskuchny sad” park (55°43’ N 37°35’ E)	2015	10	8	1	0	0	0	1	0	0
Italy, Rome	Monte Pincio (41°91’ N 12°48’ N)	2015	20	18	0	0	0	0	0	1	1
Italy, Florence	Parco dell Anconella (43°76’ N 11°30’ E)	2015	9	3	5	0	0	1	0	0	0
Germany, Hamburg	Danziger Strasse (53°56’ N 10°01’ E)	2015	6	5	1	0	0	0	0	0	0
Sweden, Stockholm	Kungens Kurva Kurva (59°16’ N 17°54’ E)	2001	5	2	0	3	0	0	0	0	0
Feodosia (Crimea)	Embankment (45°01’44.3’’ N 35°22’38.7’’ E)	2017	5	5	0	0	0	0	0	0	0
Yalta (Crimea)	Embankment (44°29’14.7’’ N 34°09’37.9’’ E)	2017	5	5	0	0	0	0	0	0	0
Total			92	68	14	5	1	1	1	1	1

 

The genetic diversity within A. decempunctata based on COI is summarized in Table 2. The haplotypic diversity was highest in individuals from Prague and Florence. At three sampling sites: Paris, Feodosia, and Yalta, the number of individuals studied was low, and only H1 haplotype was found. The results of the Tajima’s D and Fu’s Fs tests were not statistically significant (Table 2) and indicated that the detected mutations are neutral in nature. Our results demonstrate that, in general, A. decempunctata in Europe retain a rather high level of haplotype diversity with a low level of nucleotide variability (Table 2). The ITS2 region sequenced was 900 bps in length and was identical in all individuals studied.

 

Table 2

Genetic diversity in A. decempunctata populations based on analysis of the COI gene

Local population	Н	Hd	К	Pi	Tajima’s D	Fu’s Fs statistic
Czech Republic, Prague	4	0.52463	1.3399	0.00218	0.40983	1.970
France, Paris	1	0	0	0	–	–
Russia, Moscow	3	0.37778	0.75556	0.00123	–1.03446	–0.046
Italy, Rome	3	0.19474	0.3	0.00049	–1.72331	–1.143
Italy, Florence	3	0.63889	1.88889	0.00308	1.15206	1.658
Germany, Hamburg	2	0.33333	1	0.00163	–1.23311	1.609
Sweden, Stockholm	2	0.6	0.6	0.00098	1.22474	0.626
Feodosia (Crimea)	1	0	0	0	–	–
Yalta (Crimea)	1	0	0	0	–	–
Total	8	0.43168	1.01027	0.00164	–1.08219	–2.116
Note. Н – number of haplotypes, Hd – haplotype diversity, К – average number of nucleotide differences, Pi – nucleotide diversity (PiJC), Tajima’s D and Fu’s Fs statistics – neutrality tests (not significant, p > 0.05).

 

For analysis of mtDNA of A. decempunctata, a sequence similar to the most variable part of the COI gene in ladybirds of close species A. bipunctata was used [4].When comparing mtDNA variability between A. decempunctata and A. bipunctata, the latter was found to have higher diversity (Table 3).

 

Table 3

A. decempunctata and A. bipunctata intrapopulation diversity

Species, local population	N	The average number of morphs (drawing on the elytra), μ	The average number of mtDNA haplotypes, μ	Intrapopulation diversity of mtDNA, Hd
A. decempunctata, Prague	116	2.850 ± 0.055	3.362 ± 0.408	0.556 ± 0.086
A.bipunctata, Saint Petersburg	84	3.010 ± 0.127	9.290 ± 1.007	0.749 ± 0.079

 

Variability of color pattern (morphological polymorphism)

Three morphs were found in the local population of A. decempunctata from Prague (n = 116 individuals), which was the most densely sampled. These morphs were: bimaculata, decempustulata, and typica; they differed in elytra patterns, and had abundances of 23, 74, and 44 individuals, respectively. Among the populations of A. bipunctata included in our study, that of St. Petersburg was distinguished by having a very high number of melanic forms [6]. Due to the study, we could compare the levels of intraspecific variability of mtDNA and morphological characters (color variability) in two closely related species – A. decempunctata and A. bipunctata. A. bipunctata was found to be more variable in mtDNA diversity and have the same diversity in color patterns (Table 3).

The symbionts of A. decempunctata and other ladybirds

Potential infection with the bacterial symbionts Rickettsia, Wolbachia, or Spiroplasma, was assessed in A. decempunctata from Prague (n = 116), Moscow (n = 24), Rome (n = 20), Florence (n = 10), Hamburg (n = 6), Paris (n = 3), Feodosia (n = 12), and Yalta (n = 8). Out of the 199 individuals tested, only one individual (from Feodosia) was found to be infected with Rickettsia. Our analysis was supplemented by infection status data for 18 individuals of A. decempunctata from Stockholm (Table 4). Neither infection with Wolbachia nor Spiroplasma was detected in any individual. Amplification of the mtDNA COI locus was used as a DNA quality control; A. decempunctata and A. bipunctata individuals infected with bacterial symbionts were used as positive controls. The confidence intervals established in this study enabled the assessment of the level of infection. Apart from cases where Rickettsia was detected, the confidence interval for each infection was the same (Table 4). The low number of individuals sampled in each collection precludes a definite statement on the level of bacterial infection in the A. decempunctata population as a whole. The CI(95%) for the sample from Prague, where 116 individuals were tested, was 0–3. The infection with symbiotic bacteria could be lower than 3%, and we did not detect this within our sampling.

 

Table 4

Local population, number of tested A. decempunctata, the number of individuals, infected with Wolbachia, Spiroplasma and Rickettsia

Local population	N	Number of cases (% infection)			95% confidence interval *,**
		Wolbachia	Spiroplasma	Rickettsia
Czech Republic, Prague	116	0	0	0	0–3
France, Paris	3	0	0	0	–
Russia, Moscow	24	0	0	0	0–14
Italy, Rome	20	0	0	0	0–17
Italy, Florence	10	0	0	0	0–31
Germany, Hamburg	6	0	0	0	–
Sweden, Stockholm [10]	18	0	0	4 (22)	0–19*; 6–48**
Feodosia (Crimea)	12	0	0	1 (8)	0–26*; 0,2–38**
Yalta (Crimea)	8	0	0	0	–
Total	204	0	0	5	0,8–5
Note. *CI is calculated for samples ≥ 10; ** CI is specified for Rickettsia.

 

To study the genetic diversity of bacterial symbionts, the citrate synthase gene (gltA) of Rickettsia from A. decempunctata from Feodosia and Stockholm was sequenced. In addition, gltA was sequenced from individuals of A. bipunctata fasciatopunctata from UlanUde (Buryatia), and A. bipunctata and Coccinella sp. from Kem (Karelia). The DNA sequences obtained were compared with those available for ladybirds in GenBank. Rickettsia from A. decempunctata from Feodosia and Stockholm differ by two nucleotide substitutions. Those from Feodosia cluster with other Rickettsia sequences from A. bipunctata, A. b. fasciatopunctata, and Coccinella sp. (Fig. 2). Those from Stockholm were identical to sequences from Germany and Great Britain (Fig. 2). The gltA sequences for A. bipunctata form two distinct clusters on the dendrogram (Fig. 2). Namely, a clade of A. bipunctata Rickettsia (FJ666763, AJ269519) and a second clade of A. bipunctata Rickettsia which is one substitution different from A. decempunctata Rickettsia (FJ666768, AJ269522).

 

Fig. 2. Phylogenetic tree of Rickettsia based on gltA gene sequences. Hosts of the intracellular symbiotic bacteria Rickettsia and the places of their collection are indicated. The sequences obtained in this work are marked with black diamonds. Other sequences are selected from GenBank for comparison, registration numbers are given. Rickettsia from Ixodes colasbelcouri was used as an outgroup

 

DISCUSSION

We conducted a large-scale study on the variability of nuclear and mtDNA in 92 individuals of A. decempunctata across nine collection sites in the Palearctic and found evidence for Rickettsia infection in individuals from Stockholm and Feodosia.

Five mitochondrial haplotypes were recovered from individuals collected in Prague where sampling intensity was highest. At other collection sites, the number of mtDNA haplotypes found depended on the number of individuals studied (Table 1). For example, in Rome, 20 ladybirds were collected which had three mtDNA haplotypes; in Paris, only three ladybirds were collected and shared the same haplotype (H1). This mitochondrial variant (H1) was found at all sample sites (Table 1). Previously published COI sequences of A. decempunctata from Germany (AJ312061) and Great Britain (DQ155924, DQ155760) also belong to the mtDNA haplotype designated by us as H1. We found haplotype H2 in Prague and Florence, and also in Moscow and Hamburg. Type H3 was found in Prague and Stockholm. Overall, eight mitochondrial haplotypes were detected in this study, seven of which were previously unknown. Six of the eight mtDNA haplotypes differ in one substitution (out of 616 bp). Three of the eight haplotypes have a geographically widespread occurrence in this species. A quarter of the studied individuals in the Prague population and 40% of ladybirds from Florence have a unique H2 haplotype, which differs from the other haplotypes by three nucleotide substitutions (0.49%). Using the mutation rate for mtDNA in Drosophila (6.2 ∙ 10^–8) [23], and a generation time of 1 or 1.5 per year, implies a divergence time of 55–83 thousand years between H1 and H2. The mtDNA diversity in A. decempunctata predates the last glaciation which covered a significant part of the species range in Europe. The finding of a single nuclear (ITS2) sequence indicates the absence of barriers to cross-breeding and exchange of genetic information between individuals. The comparisons of intraspecific DNA variability and color variation in the two closely related species, A. decempunctata and A. bipunctata, showed that the latter had higher mtDNA variation, but a similar level of morphological variation.

Infection of ladybirds of the Coccinellidae family with bacterial symbionts is intensively studied, often due to the significance of these insects as predators of pests in agriculture [9, 24, 25]. In the invasive species Harmonia axyridis in Russia, Spiroplasma infection was detected in the native populations [26, 27]. In the two-spot ladybird A. bipunctata, infection with symbiotic bacteria of three genera was found in the European and Asian parts of the distribution [7, 10]. Rickettsia was found in A. decempunctata in England and Germany [10]; while Wolbachia and Spiroplasma were never reported.

Although 199 individuals of A. decempunctata from eight European cities were screened for the presence of symbiotic bacteria of three genera, Rickettsia, Wolbachia, and Spiroplasma, only individuals from Feodosia were found to have the bacterium Rickettsia. In 2001, we detected A. decempunctata specimens infected with Rickettsia in Stockholm with a 23% infection rate [11]. In the current study, some of the sample sizes per site are low. Therefore it is possible that infections with symbiotic bacteria went undetected. The infection rate in Stockholm in 2001 was high, and the calculation of the confidence interval showed that it could reach 48%. Based on our sampling in Prague, the infection rate does not exceed 3% if it is present in a given local population. It is known that in ladybirds, only part of the population is typically infected with symbiotic bacteria. Additionally, bacteria are often lost in a population [9] if there are no selective factors contributing to their spread. Spiroplasma infection was found to decrease the viability of A. bipunctata larvae [24]; in the same work, no similar effect was detected with Rickettsia and Wolbachia infections. In ladybirds, symbiotic bacteria disrupt reproduction, resulting in no male offspring, which leads to a shift toward females in the gender ratio. Under certain conditions, this gives the population some advantages, as the remaining larvae eat eggs that have stopped developing. However, in favorable living conditions, uninfected females leave more offspring than infected ones [24]. Imperfect maternal inheritance combined with the meager benefits of androcide and the absence of other benefits for infected females can induce an intense selection against individuals infected with symbiotic bacteria.

We detected Rickettsia infection in individuals of A. decempunctata collected in Stockholm and Feodosia, which are at the northern and southern limits of the distributional range for this species. The results of this study are the first report of the presence of Rickettsia in the southern part of the species range. Previously, infection was found only in the north, in the UK, Germany [10], and Sweden [11]. We investigated the variability of the Rickettsia citrate synthase (gltA) gene, since this part of the bacterial genome has been used in other studies and it was shown to be more informative for bacterial phylogenetic analysis than 16S rRNA [10]. The differences revealed in the gltA gene of Rickettsia in A. decempunctata from Stockholm and Feodosia may have a different origin. First, the same mutations can arise by independent or unrelated mutations in a given gene, although the probability of this is low considering the low number of mutations detected. Second, in Feodosia, horizontal transfer of bacteria from the sympatric species of A. bipunctata is possible. This has already been documented between A. bipunctata and A. decempunctata in Denmark [10]. This hypothesis of horizontal transfer is also supported by the fact that three different strains of Rickettsia were found in laboratory lines of A. bipunctata, one of which clusters with Rickettsia from A. decempunctata based on a comparison of the atpA, coxA, gltA, and 16S rRNA genes [28]. Our data, together with those obtained from other ladybird species, do not support the idea of strong coevolution of the parasite and the host. Studying more DNA sequences of the Adalia symbionts and other Coccinellidae species collected in geographically remote locations will help clarify this issue in the future.

Ladybirds with seven spots on the elytra, similar in morphology to Coccinella septempunctata (Coccinellidae), were found at our collection sites. Species identification using barcoding showed that the ladybird individual with seven spots was not previously recorded. The closest species, Coccinella magnifica (KU916547) from Germany, is distinguished by four nucleotide substitutions. Therefore, until further information is received, we designate this individual Coccinella sp. (MK932845). Interestingly, in the course of this study, we first discovered the Rickettsia bacterium infection in Coccinella sp. Previously, only Wolbachia was reported in beetles of this genus [9, 24, 25]. The sequences of the gltA gene of Rickettsia in Coccinella sp. were identical to those of A. bipunctata (Fig. 2). In addition, individuals of A. bipunctata and Coccinella sp. were collected at the same site in the city of Kem (Karelia), which suggests both the possible horizontal transfer of bacterial symbionts between different species of coccinellids, and/or the contamination of Coccinella sp after eating eggs of A. bipunctata infected with Rickettsia.

Since the A. decempunctata specimens collected in Prague, Moscow, Yalta, Rome, Florence, Hamburg, and Paris were not infected with the symbiotic bacteria Rickettsia, Spiroplasma, and Wolbachia, we could not establish any relation between the mtDNA haplotype and infection with the symbiotic bacterium, as was the case with A. bipunctata. In Stockholm, the H1 haplotype was detected in one individual infected with Rickettsia and in uninfected individuals, while the H3 haplotype was found in three Rickettsia infected individuals. However, ladybirds from Prague with the H3 haplotype of mtDNA were not infected. In Feodosia, both individuals infected with Rickettsia and uninfected ones shared the H1 haplotype. Due to low sample sizes, however, a relationship between mtDNA haplotype with Rickettsia infection cannot be ruled out. The low number of nucleotide substitutions in Rickettsia sequences between A. decempunctata and A. bipunctata suggests a common origin of the symbiont in ladybirds of the genus Adalia but does not exclude subsequent horizontal transfer events between individuals of both species.

FUNDING

The current study was funded by RFBR under the research project No. 19-04-00739; the collection of material was partially performed by I.A. Zakharov according to the state order No. 011220190002. A. Honek was supported from grant no. 17-06763S of the Czech Science Foundation

About the authors

Elena V. Shaikevich

Vavilov Institute of General genetics

Author for correspondence.
Email: elenashaikevich@mail.ru
ORCID iD: 0000-0002-6504-5547
SPIN-code: 4746-3067

Doctor of Science, Main Researcher, Laboratory of Insect Genetics

Russian Federation, 3, Gubkin street, Moscow, 119991

Ilya A. Zakharov

Vavilov Institute of General genetics

Email: iaz34@mail.ru

Doctor of Science, Main Researcher, Laboratory of Insect Genetics

Russian Federation, 3, Gubkin street, Moscow, 119991

Alois Honek

Crop Research Institute

Email: honek@vurv.cz

Doctor of Science, Main Researcher

Czech Republic, 161 06 Czech Republic, Prague 6 – Ruzyně, Drnovská, 507. 

References

Supplementary files