Background

The phenomenon of symbiosis can hardly be overestimated: apparently, there is no multicellular organism in nature without symbiotic relations. The crucial issues of symbiosis are the identification of evolutionary stages of establishing interactions between partners and the revealing of molecular mechanisms of these interactions at the cellular and gene level. One of the most widespread prokaryotic symbionts of invertebrates is the intracellular α-proteobacteria Wolbachia pipientis that infects no less than a 40% of terrestrial arthropods [1]. Wolbachia is maternally transmitted and is able to manipulate host sex determination or reproductive systems in order to help Wolbachia spread in host populations [2, 3]. On the other hand, Wolbachia infection could be beneficial to its host [4,5,6,7,8]. Differences in the phenotypical manifestations of the infection can be due to singularities of the host organism physiology, including processes of the endocrine regulation of growth, development and fitness. Indeed, there is a lot of research that connects biochemical changes in hosts and phenomenon of host resistance to viral infections with Wolbachia symbiont [6, 9,10,11,12]. However, molecular mechanisms underlying the basis of Wolbachia-host interactions as well as physiological mechanisms by which Wolbachia promotes adaptation of the host organism remain largely unknown. There are several model organisms to study these issues on, and Drosophila melanogaster is the most studied one.

Field populations of D. melanogaster are ubiquitously infected with Wolbachia in the frequency range of 30–60% [13,14,15,16,17,18,19]. Wolbachia symbionts of D. melanogaster have monophyletic origin with divergence time ~ 8 Kya [20]. Several lineages/genotypes/strains of Wolbachia were identified via different approaches. According to the phylogeny reconstruction of full genome sequences, the symbiont diversity in D. melanogaster includes the I-VI and VIII clades [20, 21]. In terms of polymorphism of certain genome markers the six genotypes were revealed, i.e. wMel, wMel2, wMel3, wMel4, wMelCS, wMelCS2 [18, 22]. Regarding the Wolbachia effect or its source (fly stock), several strains were investigated but two of them (wMel and wMelPop) are the most significant, especially in our discourse. Thus, wMel strain is regarded as a common monophyletic group that covers all of the diversity of bacteria isolated from D. melanogaster; whereas wMelPop is a certain pathogenic variant of wMel strain that causes early death of flies [23]. These classifications are consistent as follows: Wolbachia diversity can be reduced to unique, monophelytic wMel strain [24,25,26], which can be divided into several genotypes, among which wMelPop strain is just a variant of wMelCS genotype. The wMel genotype (not to be mixed with wMel strain) includes I-V and VIII clades and is the most widespread [18, 20, 27]. The wMel2 genotype belongs to VIII clade, wMel4 – to III clade, wMelCS and wMelCS2 belong to VI clade [18, 20, 21, 27].

One of the physiological mechanisms that promote adaptation and could be potentially influenced by Wolbachia is a non-specific neuroendocrine stress reaction. In insects, it includes several components, such as juvenile hormone, ecdysone, insulin and biogenic amines, in particular – dopamine (DA) [28,29,30]. DA plays three different roles in Drosophila: a neurotransmitter passing the nerve impulse through the synaptic cleft; a neuromodulator affecting the neighbouring neurons and modifying neurotransmitter action; and a neurohormone that is transported by the haemolymph and acts remotely [31]. Under the stress the DA level in Drosophila rises quickly and steeply, impacting survival [32,33,34]. The activity of alkaline phosphatase (ALP), an enzyme regulating the pool of DA precursor tyrosine, is shown to decrease under stress following the rise of the DA level that down-regulates it [35, 36]. As to the basal DA level under normal conditions, it is determined, at least particularly, via DA-dependent arylalkylamine N-acetyltransferase (DAT) activity [37, 38]. Here we study an effect of several Wolbachia genotypes on Drosophila heat stress resistance and the DA metabolism in order to evaluate the role of Wolbachia diversity in the symbiont influence on the host adaptability.

Methods

Drosophila melanogaster strains and rearing

To examine Wolbachia effect on physiological and biochemical traits of D. melanogaster, the nuclear background of Bi90 isofemale strain and different cytoplasmic backgrounds were used. Bi90 strain was established from wild-caught female of “Bishkek 2004” population and interbred for more than 300 generations, thereby it could be considered a nearly isogenic line. This strain was earlier characterized by Wolbachia infection and mtDNA [18, 39, 40]. One pair of flies from Bi90 strain was isolated to get Bi90 branch, which was treated with tetracycline for 3 generations to make Wolbachia-free Bi90T strain [6, 27]. Bi90T strain was used in making conplastic strains and as a control in experiments.

Five D. melanogaster strains with different Wolbachia infections were used in the study: Bi90 strain, that harboured wMel [18], and four conplastic strains which had been produced by 20 backcrosses of Bi90T males with appropriate source of Wolbachia. Wolbachia donor strains were also characterized for infection (wMelCS, wMel2, wMel4 and wMelPop) and mtDNA [18, 22, 41] (Table 1). Two independent runs were performed to make each conplastic strain, and finally two strains of ‘certain Wolbachia’-cytoplasmic/Bi90 nuclear background were created.

Table 1 Sources of ‘certain Wolbachia’ infections used in the study
Full size table

All strains were kept at 25 °C, 12:12 h photoperiod, in a standard Drosophila medium (agar-agar, 7 g L−1; corn grits, 50 g L−1; dry yeast, 18 g L−1; sugar, 40 g L−1). Flies hatched within 3–4 h were pooled for experiments.

Viability analysis

Viability analysis under heat stress was designed as follows: 1 day before the experiment, females were separated from males, and 5 flies were placed in a vial (25–48 vials in each group under study). Before and after the experiments, the flies were kept at 25 °C. To determine the viability under heat stress, the vials with flies were transferred from 25° to 38 °C for 4 h, and then were returned to 25 °C. 24 h later surviving flies were counted and survival rates were calculated as the percentage of survivors in each vial.

To estimate the effect of L-dihydroxyphenylalanine (L-DOPA) treatment on the stability against heat stress, five 4-day-old female flies were placed in vials (17–42 vials in each group under study) in which the bottom and 1 cm of the wall were covered with filter paper soaked with 0.5 mL of the nutrition medium. The medium contained 5% sucrose, 2% yeast and 1% L-DOPA (Sigma-Aldrich, USA). After 48 h, the vials were transferred from 25 °C to 38 °C for 2 h 45 min, and then returned to 25 °C. Survivors were counted in 24 h.

Enzyme activity assays

To perform ALP and DAT activity measurements a spectrophotometric method was used. To measure ALP activity, flies (10–50 in each group under study) were homogenised on ice in 0.1 M Tris-phosphate buffer (Sigma-Aldrich, USA), pH 8.6 (1 fly in 20 μl) and centrifuged for 5 min at 13,030 g. Enzyme activity in the supernatant was determined using α-naphthylphosphate as substrate. After centrifugation, the supernatant was transferred to Eppendorf microtube (1.5 ml, Axygen Inc., USA) to which 1 ml of reaction mixture (100 ml 0.1 M Tris-phosphate buffer, pH 8.6, 100 mg α-naphthylphosphate, 100 mg fast blue RR salt (Chemapol, Czech Republic), 230 μl 10% MnCl, 230 μl 10% MgCl, 0.5 g polyvinylpyrrolidone (ICN, Russia), and 2 g NaCl) was added. Incubation was carried out at room temperature in the dark for 25 min, and the reaction was interrupted by the addition of 3 ml of ice-cold distilled water.

To measure DAT activity, flies (10–38 in each group under study) were homogenised on ice in 0.05 M Tris-HCl buffer (Sigma-Aldrich, USA), pH 7.2 (2 flies in 120 μl) and centrifuged 5 min at 13,030 g. Enzyme activity in the supernatant was determined using DA (Sigma-Aldrich, Switzerland) as substrate. The components of the reaction mixture were added to a cuvette as follows: 300 μl of 0.05 M Tris-HCl, pH 7.2, 50 μl of acetyl CoA (0.5 mM, Sigma-Aldrich, USA) in 0.05 M Tris-HCl, pH 7.2, 25 μl of 12 mM N-phenylthiourea (Fluka, China) in 0.05 M Tris-HCl, pH 7.2, 25 μl of 40 mM DA in 0.001 N HCl, 50 μl of the supernatant, and 50 μl of 2.4 mM 5,5-dithiobis(2-nitrobenzoic acid) (Fluka, USA) in 0.05 M Tris, pH 7.2. The samples were incubated for 2 min at room temperature in the dark.

The optical density of the obtained reaction products was measured with a SmartSpec™ Plus spectrophotometer (Bio-Rad, USA) at 405 nm (DAT) and 470 nm (ALP) against the reaction zero point. For ALP activity measurements under heat stress flies were exposed to 38 °C for 1 h 40 min; the optimum exposure time was determined previously [36].

Statistics

All data are represented as means ± S.E.M. The false-discovery rate corrections for multiple comparisons were made when appropriate. The data on ALP activity, DAT activity and fly viability were analyzed by 1-way ANOVA (Strain – the simple factor) or by 2-way ANOVA (Strain – the 1st simple factor; Heat stress or L-DOPA treatment – the 2nd simple factor). Before performing the ANOVA, a Shapiro-Wilk’s W test was used to assess normality of the datasets analyzed. All datasets that failed to meet the assumptions of the ANOVA were transformed prior to analysis. The comparison of the group means was performed with the Benjamini-Hochberg stepwise post-hoc test. The results were considered significant at probability level < 0.05.

Results

The heat stress impact on viability of D. melanogaster infected with different Wolbachia genotypes

The results of an evaluation of the viability after heat stress exposure (4 h 38 °C) of 6-day-old Drosophila females of wild type strain Bi90T (uninfected control) and strains that harboured wMel, wMel2, wMel4, wMelCS and wMelPop Wolbachia variants are presented in Fig. 1. No significant difference was found in the survival rates under heat stress between females with Wolbachia genotypes wMel, wMel2 and wMel4 and uninfected control. On the contrary, the survival of females with wMelPop infection was significantly decreased compared with control females and females with wMel, wMel2 and wMel4 genotypes, whereas females with wMelCS genotype of Wolbachia demonstrated a significant increase of viability under heat stress (Fig. 1; Strain – F(5211) = 14.05, p ≪ 0.00001).

Fig. 1
figure 1

The effect of various Wolbachia infections on Drosophila heat stress resistance in comparison with uninfected (tetracycline-treated) control. The data represents survival rate of 6-day-old Drosophila females under 4 h of heat exposure (38оC). Each histogram bar represents an average value of 25–48 tests (means ± SEM). a – p < 0.01 vs uninfected and wMel, wMel2, wMel4 infected groups. b – p < 0.0001 vs wMelPop infected group

Full size image

The effects of various Wolbachia genotypes on D. melanogaster alkaline phosphatase (ALP) activity

ALP converts the inert tyrosine conjugate, tyrosine-O-phosphate, into tyrosine and thus changes in ALP activity usually correlate with changes in the DA level in flies [35, 38]. The ALP activity of 1- and 6-day-old Drosophila females infected with wMel, wMel2, wMel4, wMelCS and wMelPop Wolbachia variants and uninfected Bi90T strain were measured under normal and heat stress (1 h 40 min 38 °C) conditions. No statistical significance under normal conditions between the ALP activities of the control uninfected flies and flies with Wolbachia genotypes wMel, wMel2 and wMel4 was found (Fig. 2a, b). However, the ALP activities in 1- and 6-day-old flies with wMelPop infection were lower and in 1- and 6-day-old flies with wMelCS – higher, than in wMel, wMel2, wMel4 and Bi90T at the same age (for Day 1 – Fig. 2a; Strain – F(5213) = 269.41, p ≪ 0.00001; for Day 6 – Fig. 2b; Strain – F(5233) = 56.87, p ≪ 0.00001). The significant decrease in ALP activity following heat stress in the females of both ages of every strain under study was demonstrated (for Day 1 – Fig. 2a; Stress – F(1213) = 1270.47, p ≪ 0.00001; Strain*Stress – F(5213) = 74.22, p ≪ 0.00001; for Day 6 – Fig. 2b; Stress – F(1233) = 867.15, p ≪ 0.00001; Strain*Stress – F(5233) = 22.22, p ≪ 0.00001).

Fig. 2
figure 2

The effect of various Wolbachia infections on ALP activity in comparison with uninfected (tetracycline-treated) control. a 1-day-old and b 6-day-old Drosophila females under normal conditions and upon heat stress (38оC). Each histogram bar represents an average of 10 to 50 measurements (means ± SEM). a – p < 0.001 vs uninfected and wMel, wMel2, wMel4 infected groups that were not stressed. b – p < 0.0001 vs wMelPop infected non-stressed group. c – p < 0.001 vs the control non-stressed group of the same strain

Full size image

The effects of different Wolbachia genotypes on D. melanogaster dopamine-dependent arylalkylamine N-acetyltransferase (DAT) activity

DAT also takes a part in the regulation of DA content in flies [37], so we measured DAT activities in 1- and 6-day-old Drosophila females infected with wMel, wMel2, wMel4, wMelCS and wMelPop Wolbachia variants, as well as in the uninfected Bi90T females (Fig. 3). Since DAT does not respond to stress in Drosophila [42], we measured its activity under normal conditions only. The wMelPop infection results in a significant decrease and wMelCS – in an increase of DAT activity compared with the control in both ages (for Day 1 – Fig. 3a; Strain – F(5147) = 16.06, p ≪ 0.00001; for Day 6 – Fig. 3b; Strain – F(5107) = 16.61, p ≪ 0.00001). Other Wolbachia variants under study do not affect DAT activity in either 1- or 6-day-old females. DAT determines, at least particularly, the basal level of DA [37, 38], so we have assumed that Wolbachia infection affects it.

Fig. 3
figure 3

The effect of various Wolbachia infections on DAT activity in comparison with uninfected (tetracycline-treated) control. a 1-day-old and b 6-day-old Drosophila females. Each histogram bar represents an average of 10 to 38 measurements (means ± SEM). a – p < 0.05 vs uninfected and wMel, wMel2, wMel4 infected groups. b – p < 0.01 vs uninfected and wMel, wMel2, wMel4 infected groups. c – p < 0.0001 vs wMelPop infected group

Full size image

The influence of dopamine level on the survival under heat stress of females infected with different Wolbachia variants

To find out whether the changes in the heat stress resistance of Drosophila females infected with wMelPop and wMelCS have a connection with the altered DA level, we examined their stress resistance following a pharmacological increase of DA content. Feeding with L-DOPA was shown to double the DA level in Drosophila [36], so we fed these females as well as infected with wMel and uninfected control with L-DOPA for 48 h before stress exposure. The rise of the DA level decreases the survival of all four groups under heat stress and eliminates the differences between the survival rate of control and wMel females and that of females infected with wMelPop and wMelCS (Fig. 4; Strain – F(3222) = 14.79, p ≪ 0.00001; L-DOPA – F(1222) = 174.18, p ≪ 0.00001; Strain*L-DOPA – F(3222) = 14.86, p ≪ 0.00001). It is worth noting that the survival rates in Bi90T, wMel and wMelCS females after L-DOPA feeding do not differ in this parameter from the L-DOPA-treated females with wMelPop infection (Fig. 4). The survival rates in wMelPop L-DOPA untreated females were low but still higher than in Bi90T, wMel and wMelCS females after L-DOPA feeding (Fig. 4).

Fig. 4
figure 4

The effect of L-DOPA on heat stress resistance of Drosophila females with various Wolbachia infections in comparison with uninfected (tetracycline-treated) control. The data represents survival rate of 6-day-old Drosophila females under 4 h of heat exposure (38оC) following 2 days of L-DOPA treatment. L-DOPA designates the uninfected flies and flies with various Wolbachia infections that were treated with L-dihydroxyphenylalanine. Each histogram bar represents an average value of 17–42 tests (means ± SEM). a – p < 0.001 vs uninfected and wMel infected groups that not received L-DOPA. b – p < 0.0001 vs wMelPop infected group that not received L-DOPA. c – p < 0.0001 vs the control group of the same strain that not received L-DOPA. d – p < 0.05 vs the control group of the same strain that not received L-DOPA. e – p < 0.05 vs uninfected and wMel, wMelCS infected groups that received L-DOPA

Full size image

Discussion

Here we try to reveal the influence of the Wolbachia symbiont on heat stress resistance and DA metabolism in D. melanogaster. This investigation was motivated by the reports on Wolbachia effect on insulin signaling [11, 43] and data on Wolbachia diversity in D. melanogaster [18, 20,21,22, 27]. Ikeya et al. [11] had demonstrated the increase of insulin signaling in Wolbachia-infected strains. Insulin signaling pathway interacts with the components of the neuroendocrine stress reaction and the stress-responsive c-Jun-N-terminal kinase (JNK) signaling pathway (which controls of a large number of cellular processes in response a wide range of stressors), and contribute to the fitness and increased stress tolerance [29, 30, 44,45,46]. The removal of Wolbachia from chico 2 homozygotes (chico gene codes the Drosophila orthologue of mammalian insulin receptor substrate) resulted in complete lethality [43]. Wolbachia infection was also shown to down-regulate 41% (11 of 27) of known heat shock proteins in the Drosophila S2 cell line [47].

Using data on D. melanogaster strains with uniform nuclear background but infected with different Wolbachia variants we have shown that Wolbachia genotypes wMel, wMel2 and wMel4 (of V, VIII, and III clade, respectively) do not induce alteration in the heat stress resistance and DA metabolism of the host.

However, two Wolbachia isolates under study do cause the changes in survival rate and DA metabolism of D. melanogaster host: wMelPop and wMelCS. The wMelPop infection reduces both the survival and activities of ALP and DAT, whereas the wMelCS infection increases these parameters (see Figs. 2 and 3). Previously, we found the negative correlation of the heat stress resistance with the DA level in Drosophila [48]. The DA level in D. melanogaster is negatively correlated with the level of ALP and DAT activities [36, 38]. Based on this observation, we assumed the DA level to be decreased in the flies with wMelCS infection (and to be increased in the flies with wMelPop). We verified this assumption using the treatment of the flies with the DA precursor, L-DOPA (see Fig. 4). The increase of DA level drastically reduces the survival rates of all studied strains. It is important that increased DA has been revealed to level the viability under heat stress of wMelCS-infected flies with other strains (see Fig. 4).

Low survival under heat stress of the wMelPop-infected flies could be explained by the well-known pathogenicity of this Wolbachia strain [23, 49]. But it is noteworthy that the changes in the DA metabolism are manifested in these flies prior to the mass death of flies (see Figs. 2a and 3a) [23]. The negative effect of wMelPop (in comparison with wMel- and wMelCS-infected and uninfected strains) on the level and the biosynthesis of one more biogenic amine involved in the stress reaction, octopamine, in D. melanogaster was shown by Rohrscheib et al. [50]. However, no difference in octopamine biosynthesis pathway was found between flies with wMel and wMelCS Wolbachia genotypes [50]. Perhaps this is due to various roles of DA and octopamine in flies, or with some delicate genetic differences in Wolbachia strains used in our study and in the study of Rohrscheib et al. [50].

We believe that the most interesting result, which we observed here, is the effect of wMelCS Wolbachia on D. melanogaster viability under stress and DA metabolism. Based on the study of Riegler et al. [22] who proposed the hypothesis of global replacement of Wolbachia wMelCS infection by wMel in D. melanogaster we expected to find a decreased fitness in wMelCS-infected flies compared with wMel-infected. However, we have found quite the opposite phenomenon.

The design of our study included an attempt to find phylogenetical signal of symbiont effects on the host. Previous works using genome data for both mtDNA and Wolbachia have revealed strict associations of those maternal factors and have distinguished coevolved clades of Wolbachia and mtDNA in D. melanogaster [18, 20, 21, 27]. Here we showed the influence of wMelCS, but not wMel-like, Wolbachia variants on the components of host fitness. The wMelCS-like isolates are related to Wolbachia clade VI that is the most diverged from all other clades, the time divergence of wMel and wMelCS-like variants is approximately in range 3.2–14 Kya [20]. Thus, we assume a specific influence of Wolbachia clade VI on D. melanogaster that should be verified in the following experiments.

Conclusions

Here we revealed that the effect of Wolbachia symbiont on the stress resistance and DA metabolism of the host insect depends on the symbiont’s genotype variant. We found out that wMelCS genotype demonstrates a strong positive influence on the D. melanogaster heat stress resistance, while survival rates of the flies with Wolbachia genotypes of wMel group do not differ from those of uninfected flies. This result is particularly surprising because genotypes of wMel group predominate in the nature populations all over the world and wMelCS variants are very rare. It is necessary to check whether such a fitness effect is inherent in all wMelCS variants or we are faced with a particular case similar to that of pathogenic wMelPop strain only “with an opposite sign”. Besides, we discovered that strong influence of wMelPop strain on D. melanogaster metabolism starts much earlier than mass death of flies.