Wolbachia are widespread endosymbionts found in a large variety of arthropods. While these bacteria are generally transmitted vertically and exhibit weak virulence in their native hosts, a growing number of studies suggests that horizontal transfers of Wolbachia to new host species also occur frequently in nature. In transfer situations, virulence variations can be predicted since hosts and symbionts are not adapted to each other. Here, we describe a situation where a Wolbachia strain (wVulC) becomes a pathogen when transfected from its native terrestrial isopod host species (Armadillidium vulgare) to another species (Porcellio d. dilatatus). Such transfer of wVulC kills all recipient animals within 75 days. Before death, animals suffer symptoms such as growth slowdown and nervous system disorders. Neither those symptoms nor mortalities were observed after injection of wVulC into its native host A. vulgare. Analyses of wVulC's densities in main organs including Central Nervous System (CNS) of both naturally infected A. vulgare and transfected P. d. dilatatus and A. vulgare individuals revealed a similar pattern of host colonization suggesting an overall similar resistance of both host species towards this bacterium. However, for only P. d. dilatatus, we observed drastic accumulations of autophagic vesicles and vacuoles in the nerve cells and adipocytes of the CNS from individuals infected by wVulC. The symptoms and mortalities could therefore be explained by this huge autophagic response against wVulC in P. d. dilatatus cells that is not triggered in A. vulgare. Our results show that Wolbachia (wVulC) can lead to a pathogenic interaction when transferred horizontally into species that are phylogenetically close to their native hosts. This change in virulence likely results from the autophagic response of the host, strongly altering its tolerance to the symbiont and turning it into a deadly pathogen.
Characterizing the causes of a virulence increase when a parasite jumps from one host species to another is fundamental to the understanding of disease emergence. In this context, we studied the bacterium Wolbachia wVulC, a natural symbiont of one terrestrial isopod species that becomes a pathogen when transfected into individuals of another species. Before death, recipient animals suffer various symptoms including nervous system disorders caused by the multiplication of wVulC. Interestingly, the quantification of wVulC loads showed similar titers in the individuals from both the recipient and native species. The difference between the two host species lies in the way they respond to the invasion of wVulC and not in their resistance per se: While the recipient host species exhibits an acute autophagic response leading to central nervous system cells disorganization, this phenomenon was not observed in the native host species, which seems to better tolerate the bacterium. Together, our results show that tolerance can be a better evolutionary strategy to counteract parasite damage than to activate a putative resistance pathway which, as a double-edged sword, can arm the host itself and increase the virulence of a parasite.
Citation: Le Clec'h W, Braquart-Varnier C, Raimond M, Ferdy J-B, Bouchon D, Sicard M (2012) High Virulence of Wolbachia after Host Switching: When Autophagy Hurts. PLoS Pathog 8(8): e1002844. doi:10.1371/journal.ppat.1002844
Editor: Serap Aksoy, Yale School of Public Health, United States of America
Received: March 13, 2012; Accepted: June 22, 2012; Published: August 2, 2012
Copyright: © 2012 Le Clec'h et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This research was funded by the CNRS, the Université de Poitiers and the Agence Nationale de la Recherche (ADaWOL ANR-09-JCJC-0109-01). WL was supported by a grant from Région Poitou-Charentes. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Wolbachia pipientis are intracellular α-proteobacteria which are extensively distributed among arthropods and filarial nematodes. These bacteria are vertically transmitted from female hosts to their offspring. The presence of Wolbachia induces a number of reproductive perturbations for their hosts including cytoplasmic incompatibility (CI) , , male killing , modification of parthenogenesis  and feminization , all enhancing the vertical spread of infection in host populations. The Wolbachia, in their wide range of hosts, are involved in diverse interactions in terms of costs and benefits to their host that can be placed on a continuum from mutualism to parasitism . However, the global picture is that Wolbachia are weakly virulent to their native hosts , . This low virulence can be explained by the evolutionary consequences of vertical transmission that theoretically leads to the attenuation of symbiont virulence through a co-evolutionary process with its host , . However, the general lack of congruence between Wolbachia and host phylogenies suggests that Wolbachia frequently colonize new hosts through horizontal transfer ,  and that this mode of spreading may be fundamental to their evolutionary dynamics .
Immediately after transfer into a new host species, the virulence of Wolbachia is expected to differ from that expressed in its native host because of a lack of host/symbiont co-adaptation, which can only occur after several host generations. Two main non-exclusive situations have been proposed to explain the emergence of a higher level of virulence when a symbiont colonizes a new host: (1) maladapted symbionts can overexploit their host, by multiplying too fast or in a “wrong” compartment (e.g. in organs other than gonads, where Wolbachia are typically found), (2) maladapted hosts could respond to the presence of an unknown symbiont by employing inefficient and very costly immune defences . In the first situation, the virulence comes primarily from an overexploitation of host resources by the parasites, and the new host can be considered as less resistant because parasite burden is higher in some tissues of the new host than in the native one. In the second situation, the immune system normally dedicated to the defence of organisms becomes a double-edged sword, turning itself against the host. A disproportionately strong response can lead to an immunopathology that can in turn deeply harm the hosts. Therefore, being able to tolerate the multiplication of parasites to a certain extent by not activating some immune pathways can constitute a good strategy to limit virulence . It is thus possible that through the co-evolutionary process taking place between the Wolbachia and their hosts, especially as prevalence is often high, hosts quickly select either a way to attempt to control Wolbachia proliferation or at the opposite, a way to better tolerate the presence of the symbionts. Being able to decrease invaders load would reduce their cost for the hosts while increasing tolerance would avoid both the direct (i.e. production of immune effectors) and indirect costs (i.e. the damage caused by immune effectors) of the immune response.
To address the question of the causes of drastic change in virulence of a symbiont, we studied a situation where a Wolbachia-host interaction becomes pathogenic when the bacteria (wVulC) are transferred from its native terrestrial isopod species host (Armadillidium vulgare) to another one (Porcellio d. dilatatus) , . In its native host A. vulgare, wVulC not only colonizes gonads for vertical transmission but is also found in somatic tissues , . This extensive colonization could play a role in the virulence of wVulC detected in several A. vulgare's life history traits: Reduction in hemocyte load, phenoloxydase activity, reproduction and survival , . However, though wVulC is clearly involved in a conflicting interaction with A. vulgare having negative impacts on several of its life history traits, this virulence is weak and Wolbachia-infected individuals usually live for years . The aim of the present study was to understand how Wolbachia wVulC becomes a pathogen when introduced in the new host Porcellio d. dilatatus. To do so, we focused our work on the analysis of the different symptoms that appear prior death of the recipient host Porcellio d. dilatatus. For this, several host life history traits have been measured (gain of weight, mobility, burying behaviour). In parallel, we assessed Wolbachia wVulC multiplication in gonads, hemocytes (i.e immune cells) and CNS (i.e. nerve cells and neighbouring adipocytes). Moreover, we observed gonads and CNS cells by electron microscopy. We demonstrate that infection with the strain wVulC causes acute disease to P. d. dilatatus with symptoms of paralysis that indicate the existence of nervous system disorders. We suggest that these symptoms and the death that follows are linked to the autophagic reaction observed in the CNS cells of P. d. dilatatus but not in A. vulgare which seems thus more tolerant to its native Wolbachia strain it co-evolved with.
Quantification of wVulC and wDil in inoculum
In order to quantify the dose of Wolbachia injected into recipient hosts that could vary due to the donor host species (A. vulgare for wVulC and P. d. dilatatus for wDil), we performed quantification by qPCR on the wsp gene. We found that the quantities of Wolbachia cells inoculated in recipient hosts significantly differed between treatments (i.e. wVulC versus wDil). This is because the natural amount of Wolbachia present in the ovaries of A. vulgare, the donor host species of wVulC (mean dose ± se = 1.66×105±6.30×104 wsp copies/injected µL) is ten times higher than in Porcellio d. dilatatus, the donor host species of wDil (mean dose ± se = 1.14×104±3.60×103 wsp copies/injected µL); (comparison by t-test = 2.9956 df = 4 p = 0.0401). To evaluate the influence of the injected dose in the effect observed on the recipient hosts, we tested the effect of wVulC with a very small inoculum (3.02×103±2.60×103 wsp copies/injected µL) by diluting the initial inoculum a hundred times. Moreover, to take into account the differences between injected doses in a statistical framework, the effect of the factor “dose” was tested at first to absorb as much variance as possible, and then it was therefore possible to test the effect of each strain of Wolbachia per se independently from the difference in injected dose.
wsp gene sequencing from inoculum and transfected animals
To verify that only the expected Wolbachia strain was present (i.e. wDil or wVulC) in the inoculum and in transfected A. vulgare and P. d. dilatatus, we amplified and sequenced 610 bp of the Wolbachia surface protein gene (wsp) in (i) all inoculum and (ii) recipient animals at 60 days Post Injection (PI). In all the cases, the chromatograms obtained showed well resolved peaks indicating that only one matrix has been sequenced per host or inoculum. The comparison of all the obtained sequences with the reference wsp sequences of wDil and wVulC showed no variation. In all cases we therefore found the strain of Wolbachia which was expected in the given sample.
No significant effect of Wolbachia injection on Armadillidium vulgare's life history traits
After the injections of wVulC or wDil, no significant effects were detected for A. vulgare females on all the different life history traits measured (gain of weight, activity and survival) (Table 1 for statistics; Figure 1). Moreover, no nervous disorders symptoms such as seizures and tremors were observed for A. vulgare.
There were no significant differences between animals injected with wVulC, wDil or control treatment for all these life history traits: Survival (A), Gain of weight (B) and Mobility (C).
Perturbation of Porcellio d. dilatatus's life history traits by wVulC
Gain of weight.
A global analysis showed that 60 days post-injection, most of the variance in gain of weight between samples was explained by differences among treatments (i.e. control, injection of wVulC and injection of wDil) but not by difference between sex or the injected dose of Wolbachia (Table 2). The same model ran after removing the control treatment from the dataset, demonstrates that variations in gain of weight depends on the Wolbachia strain (wVulC versus wDil): wVulC induced a stronger growth slowdown than wDil (respective means for gain of weight ± se = 1.1000±0.08 and 1.4900±0.05 after 60 days following injection; Figure 2A; Table 2). The animals injected with the 1/100 diluted wVulC inoculum suffered a similar growth slowdown compared to those injected with the pure suspension of wVulC (ANOVA, F = 0.1384 df = 1,15 p = 0.7150; Figure 2A).
Animals were injected by pure suspension of wVulC, diluted wVulC inoculum (only for females), wDil or control treatment and monitored during 75 days post-injection. (A) Gain of weight: Both Wolbachia strains negatively impact this parameter compared to the control treatment. However, wVulC (i.e pure or 1/100 diluted inoculum) induces a lower growth rate than the native strain wDil. (B) Mobility: this parameter is dramatically reduced for animals injected by wVulC compared to those from other treatments. On the opposite, injection of wDil has no significant effect on the host's mobility. (C) Burying behaviour: The level 0 corresponds to the surface of the substrate, negative values correspond to the number of buried animals while positive values correspond to the number of animals found at the surface of the substrate. For control treatment animals, not a single individual rose up at the surface and for individuals injected by wDil only few animals were spotted at the surface. When inoculated by wVulC, all animals rose up to the substrate after 60 days. (D) Survival (a) data for females and (b) data for males: The injection of wVulC kills all the individuals. However, the males die quicker than the females. No extra-mortalities were reported for injection with wDil compared to control treatment.
A global analysis showed that 60 days post-injection, the variance in mobility between animals was due mostly to differences among treatments (i.e. injection of wVulC or wDil or control) and did not relate to sex or the amount of Wolbachia cells injected (Table 2). A strong differential effect on mobility was induced according to the injected strain (wVulC or wDil). Between 30 and 75 days, the mobility of individuals injected by wVulC observed during 180 s strongly dropped from 115±3.55 s to 0 s while animals injected with the native strain wDil or the control treatment did not exhibit reduction in mobility (150±2.45 s.; Figure 2B; Video S1). Moreover, starting from 45 days post-injection, P. d. dilatatus individuals injected with wVulC turned to asthenia (i.e. they seemed “lax”), suffering from seizures and leg tremors. These symptoms were observed after injection of both pure suspension of wVulC and 1/100 diluted suspension of wVulC (Figure 2B; Video S1).
Buried or not?
Another change in the behaviour of P. d. dilatatus observed after Wolbachia injection was the location in the rearing box substrate. The normal behaviour of aposymbiotic P. d. dilatatus is to bury themselves into the substrate. But after Wolbachia's injections, some males and females rose at the surface (Figure 2C; Video S1). The analysis of data with a binomial general linear model showed that the factor that statistically explained most of the variability in location was again the type of Wolbachia strain injected (Table 2). Between 45 and 60 days post-injection, most of the individuals injected with the pure or 1/100 diluted inoculum of wVulC strain were found at the surface (respectively 100% and 57 to 72% of the animals) while after wDil injection, only 6.7% of animals were spotted at the surface over the whole 75 days observation period. For the control treatment the situation was totally opposite to the one observed with wVulC since not a single animal was found at the surface of the substrate over the 75 days period of observation (Figure 2C; Video S1).
The above described symptoms for the P. d. dilatatus injected with the pure and diluted suspensions of wVulC (i.e. decrease in gain of weight, mobility and number of animals buried) were followed by death of all the animals (Figure 2D). Comparison of survival plots between groups of individuals of P. d. dilatatus injected with wVulC, wDil and control treatments using general mixed-effects Cox model revealed significant differences. While in control and wDil treatments, only 2 mortality cases were observed during the whole experiment, all 30 individuals injected with wVulC died within 75 days (Table 2; Figure 2D). Moreover, clear differences in survival curves were reported between males and females injected with wVulC (Table 2; Figure 2D). The males infected with wVulC died quicker than the females (the TL50 was respectively 45 days and 75 days). Since to reproduce P. d. dilatatus needs at least 50 days from insemination to the releasing of the young from the marsupium no vertical transmission (i.e from transinfected mother to the progeny) could be possible after the injection of the pathogenic Wolbachia strain wVulC. As the pathologic effects rose within the first month post injection and that the bacteria would need at least a month to well colonize the ovaries vertical transmission of wVulC in P. d. dilatatus is clearly not possible.
Wolbachia multiplication in CNS, gonads and hemocytes of P. d. dilatatus and A. vulgare
In order to assess the global replication dynamics of Wolbachia in their hosts, Wolbachia quantifications were performed in CNS, gonads and hemocytes (i.e. immune cells). These quantifications were performed at 30 days and 60 days post injection for both recipient host species. The bacterial load of wVulC in CNS, which is composed of a core of nerve cells surrounded by a sheathing of adipocytes but also in gonads and hemocytes did not significantly differ between the two host species (in CNS at 30 days PI t-test = −1.1901 df = 1 p = 0.4449 and at 60 days PI t-test = 1.048 df = 1 p = 0.4851; in gonads at 30 days t-test = −1.1758 df = 1 p = 0.4489 and at 60 days t-test = 1.4303 df = 1 p = 0.3884; in hemocytes at 30 days t-test = −0.5327 df = 1 p = 0.6884 and at 60 days t-test = 1.1216 df = 1 p = 0.4635; Figure 3A). Moreover, the bacterial loads at 60 days post-injection were similar to those obtained in the CNS, gonads and hemocytes of a naturally infected A. vulgare (i.e. which received wVulC vertically from their mothers) (in CNS t-test = −2.1411 df = 3 p = 0.1217, in gonads t-test = 2.635 df = 3 p = 0.0779 and in hemocytes t-test = 1.6100 df = 3 p = 0.2487; Figure 3A). For individuals of both recipient host species injected with wDil, at both 30 days and 60 days post-injections, we showed no significant difference between bacterial load in CNS (at 30 days t-test = 1.0107 df = 1 p = 0.4966; at 60 days t-test = 1.088 df = 1 p = 0.4732; Figure 3B). In the other tested tissues (i.e gonads and hemocytes), we obtained significant differences when comparing the wDil quantification at 60 days post-injection between transfected P. d. dilatatus and A. vulgare (in gonads t-test = 2.9053 df = 5 p = 0.0335 and in hemocytes t-test = −2.9867 df = 5 p = 0.0305; Figure 3B). In these two cases, wDil densities were higher in the native host P. d. dilatatus than in A. vulgare (Figure 3B).
(A) Comparison between (i) P. d. dilatatus in dark grey and A. vulgare in white injected with wVulC and (ii) naturally infected A. vulgare in light grey revealed no significant differences for the Wolbachia loads in CNS cells, gonads and hemocytes (i.e immune cells), at both 30 and 60 days post-injection. (B) Comparison between (i) P. d. dilatatus in dark grey and A. vulgare in white injected with wDil and (ii) naturally infected P. d. dilatatus in dim grey revealed no significant differences for the Wolbachia loads in CNS cells at both 30 and 60 days PI. However, at 60 days post-injection, for gonads and hemocytes wDil loads were significantly higher in the native host P. d. dilatatus than in A. vulgare. A global comparison between wDil and wVulC loads showed that wVulC (A) always exhibited higher bacterial loads than wDil (B) in both host species.
The quantifications also showed that the density of wDil in CNS of all injected animals from both host species or in naturally infected P. d. dilatatus was always ten to a hundred times lower than the density of wVulC injected in the same animals or in naturally infected A. vulgare (t-test = −37.0596 df = 4 p<0.0001).
Autophagic vesicles accumulation in nervous cells of P. d. dilatatus infected by wVulC
For A. vulgare individuals injected with either wDil or wVulC but also for those which were naturally infected with wVulC, nerve cells from the CNS and adipocytes but also gonads observed by transmission electron microscopy at both 30 and 60 days post-injections were normal (i.e. nucleus and organelles exhibited normal shapes, no autophagic phagosomes and only few lysosomes were present in the cytoplasm). The same cell types observed in P. d. dilatatus injected with its native Wolbachia strain wDil at both 30 and 60 days post-injections were also normal. By contrast, strong alterations of the structure of the cells were observed only in the CNS and adipocytes but not in gonads of P. d. dilatatus injected with wVulC. At 30 days post injection, in both adipocytes and nerve cells, we observed some Wolbachia cells and high densities of autophagosomes and autolysosomes which are characteristic of autophagic process (Figure 4A and 5A). Moreover, in one nerve cell, we specifically observed the formation of an autophagosome around a Wolbachia cell (Figure 4A). At 60 days post injection, the autophagic process observed in the nerve cells and adipocytes clearly amplified compared to what was observed at 30 days post-injections. The CNS cells were severely damaged. Both nerve cells and adipocytes were filled with autophagic vesicles and the organelles were less visible because of a complete disorganization of the cytoplasm (Figure 4B and 5B). To have an additional confirmation of the occurrence of autophagy, we applied the antibody labelling kit LC3B (Invitrogen) on adipocytes of A. vulgare and P. d. dilatatus injected with wVulC or wDil. This approach confirmed a high autophagic activity in the adipocytes collected in P. d. dilatatus individuals that were injected with wVulC (Figure 5C). For such individuals, we observed highly labeled spherical structures that would reflect the incorporation of the LC3B protein in the phagophores. This kind of phagophore was very rarely observed for A. vulgare and for other treatments in P. d. dilatatus (Figure 5D).
These cells are localised in the CNS ganglion. (A) At 30 days post-injection, some autolysosomes and bacteria are visible: autophagic process is already quite high compared to control [N: Nucleus, V: Vacuole, Ap: Autophagosomes (Autophagic hallmark), Al: Autolysosomes, Wo: Wolbachia, *: a Wolbachia which seems to be surrounded by an autolysosome, Arrow head: magnified area]. (B) At 60 days post-injection, the nerve cells are severely damaged by the accumulation of autophagic vesicles (N: Nucleus, V: Vacuole, Ap: Autophagosomes (Autophagic hallmark), Al: Autolysosomes, Arrow head: magnified area and probably a Wolbachia in autophagic degradation).
These adipocytes are localised around the nerve cells of the CNS and constitutes the sheathing of this organ. (A) At 30 days post-injection, autolysosomes, autophagosomes but also bacteria were observed: autophagic process was already higher compared to the control (Ad: adipocyte, Ap: Autophagosomes (Autophagic hallmark), Al: Autolysosomes, Wo: Wolbachia, Arrow head: magnified area). (B) At 60 days post-injection, the adipocytes were severely damaged by the accumulation of autophagic vesicles (Ad: adipocyte, Ap: Autophagosomes, Al: Autolysosomes, Wo: Wolbachia, Arrow head: magnified area). (C) With the antibody labelling anti-LC3B, the protein LC3B implicated in autophagic processes was detected in adipocytes of P. d. dilatatus injected with wVulC at 45 days post-injection. The highly labeled spherical structures would reflect the incorporation of the LC3B protein in the phagophores. (D) With adipocytes of P. d. dilatatus injected by wDil but also in other tested situations, only really few of these structures were detected by epiflorescent microscopy.
While Wolbachia are typically transmitted vertically within host populations, it is increasingly recognized that horizontal transfer between host species plays an important role in the evolutionary dynamics of these endosymbionts in arthropods . In this study, we aimed at assessing the potential change in virulence of a Wolbachia strain (wVulC) immediately after horizontal transfer into a new host, as well as characterizing the causes of this virulence change. We show that after experimental transfer of wVulC from its native terrestrial isopod host species (A. vulgare) to a closely related one (P. d. dilatatus), the virulence increases so much that it becomes a pathogen. Such a high virulence is remarkable and is reminiscent of the observations made for another Wolbachia strain, wMelPop which causes early death of both its native and foreign hosts –. In the case of wMelPop, this high virulence has been proposed to be caused by an abnormally high bacterial load in host cells, especially in nerve ones . Along with high loads of wMelpop in nerve cells, important behavioural modifications have been recorded after horizontal transfer followed by vertical transmission in mosquitoes, suggesting an alteration of the central nervous system (CNS) , . In the present study, the characterization of symptoms such as paralysis, seizures and leg tremors observed prior to the death of P. d. dilatatus injected by wVulC, led us to hypothesize that the sickness induced by this Wolbachia strain could also involve nervous system disorders.
The quantifications of the Wolbachia in the main organs of both A. vulgare and P. d. dilatatus individuals that received the bacteria either vertically from their mothers or horizontally by experimental injection strengthened this hypothesis. We found that wVulC which is virulent in P. d. dilatatus exhibited bacterial loads in the CNS that were at least 10 times higher than those reached by the avirulent wDil. Therefore, similarly to what was observed for wMelpop, the high bacterial loads reached by wVulC in the CNS certainly constitutes one component of its high virulence. The multiplication of intracellular bacteria in nerve cells as a cause of CNS diseases has also been extensively described in vertebrates . Interestingly, most of the intracellular bacteria that cause such diseases belong to the group of Rickettsiales, to which Wolbachia also belongs , . Some of these Rickettsiales, such as Rickettsia sp. have even been shown to be responsible for diseases like encephalitis and meningitidis in humans . It is also noteworthy that symptoms similar to those caused by wVulC and wMelpop in arthropods such as reduction of movements, uncontrolled tremors and inability to feed properly have been described in important CNS diseases caused by Rickettsiales in other animals , , .
Nevertheless high bacterial loads in CNS but also in gonads and hemocytes are unlikely to fully explain the pathogenesis caused by wVulC in P. d. dilatatus because A. vulgare individuals infected vertically or horizontally with this strain survive similar infection pattern . Indeed, wVulC reaches similar bacterial loads in the main organs of both its native and foreign host suggesting that both isopod species have an overall similar resistance . We therefore suspected that the pathogenicity of this strain towards P. d. dilatatus could be mostly due to a lower ability of this foreign host to tolerate the multiplication of wVulC compared to A. vulgare , . One hypothesis would be that A. vulgare, through vertical co-evolution with wVulC, has increased its ability to tolerate the high bacterial loads in its CNS. By contrast, P. d. dilatatus could not evolve towards tolerating high bacterial loads in its CNS because it vertically co-evolved with a strain (wDil) that exhibits much lower bacterial loads in this organ than wVulC.
The main factor explaining the different virulence of wVulC observed between the two isopod species would thus be the way infected tissues respond to the multiplication of wVulC. In agreement with this hypothesis, electron microscopy and LC3B labelling revealed marked differences in the autophagic response of the host cells between A. vulgare and P. d. dilatatus. While in the former species both adipocytes and nerve cells did not show any sign of deregulated autophagy, in P. d. dilatatus, we observed that these two different cell types were completely filled with autophagic vesicles, resulting in a profound disorganization of their cytoplasm. These observations suggest that in addition to high bacterial loads in the CNS cells, the disproportionate autophagic response of these cells is likely to be a major component of the CNS disease caused by wVulC in P. d. dilatatus.
Autophagy even if really conserved among eukaryotes has only quite recently been considered as a way to regulate intracellular parasites . A role for autophagy in the response to Wolbachia infection has not previously been clearly described, but a recent paper described the presence of autophagosomes putatively involved in the degradation of dying wMelpop in ovaries of D. melanogaster . Moreover, Anaplasma phagocytophilum, a pathogen closely related to Wolbachia, has been shown to manipulate its host's autophagic machinery . Given the importance of autophagy as a multi-pronged defense against intracellular microbes, we propose that the pathogenic effect of wVulC onto P. d. dilatatus could be an immunopathology characterized by Wolbachia-induced disturbance of host autophagic processes, as already described for other intracellular pathogens , . The mechanism that causes the activation of an apparently unregulated autophagic response in this novel host is not known. However, for Anaplasma phagocytophilum, a previous study has unveiled the importance of the type IV secretion system (T4SS) in the induction and subversion of autophagy , . In the genome of wVulC strain, genes corresponding to the minimum components for a typical and functional type IV secretion system (T4SS) were identified , . This suggests that wVulC factors that trigger autophagy may use the T4SS system as an effector translocator to reach their targets. We can thus hypothesize that the strong autophagic response in P. d. dilatatus that causes nerve cells destruction could be due to reaction of the host towards the abnormally intense multiplication of wVulC in its CNS but also to the production of particular toxins possibly shuttled via T4SS. Moreover, it cannot be excluded that the expression of some wVulC effectors could change when the bacteria are transfected from native to new host and that this differential expression could play a role in the difference observed in host reactions.
In this study, we show that the Wolbachia strain wVulC becomes highly virulent when it jumps from its native host (A. vulgare) to a new host (P. d. dilatatus). This increase in virulence is linked to an apparently unregulated autophagic reaction in the new host which is not observed in the same conditions for the native one. These results suggest that A. vulgare, which vertically co-evolved with wVulC better tolerates the presence of this bacterium in its cells. In conclusion, our experiments uphold the hypothesis according to which tolerance could be a better evolutionary strategy to counteract parasite damage than activating a putative resistance pathway such as autophagy, which as a double-edged sword can hurt the host and increases the virulence of the parasites.
Materials and Methods
All experimental procedures and animal manipulations did not require an ethics statement.
All the animals used in this experiment were grown at 20°C in plastic breeding boxes, in natural photoperiod, on moistened potting mix derived from peat from sphagnum moss (pH = 6.4 and conductivity = 50.0 mS/m) with dead lime-tree leaves as a food source. In such laboratory conditions, animals can normally live up to 3 years, whether they are infected or not by Wolbachia. The recipient hosts of all experiments were aposymbiotic (i.e. without any Wolbachia) Armadillidium vulgare (originating from the city of Helsingor in Denmark and reared in the laboratory since 1991) and Porcellio dilatatus dilatatus which were collecting at the village of Rom in Deux-Sèvres (France) in 1988 and reared since then in the laboratory. As donors of the native strain wDil, we used another lineage of P. d. dilatatus naturally infected by Wolbachia. These animals were originated from the Sainte-Marguerite Island in Alpes-Maritimes (France) and reared in the laboratory since 2007. As donors of the Wolbachia strain wVulC, symbiotic Armadillidium vulgare originating from the city of Helsingor (Denmark) and reared in the laboratory since 1991 were used.
Wolbachia injection experiments
Aposymbiotic A. vulgare (females of 6 months-old) and P. d. dilatatus (females and males of 6 months-old) were infected by wVulC, wDil or by a control treatment. For each batch of injection, ovary suspensions were prepared with the ovaries of (i) 5 A. vulgare symbiotically associated with wVulC; (ii) 5 P. d. dilatatus symbiotically associated with wDil or (iii) 5 A. vulgare that did not host Wolbachia for control treatment. The ovaries were collected and crushed into 1 ml of Ringer solution. The resulting suspension was filtered through a 1.2 µm pore membrane, and 1 µL of each filtrate was injected in a small hole pierced in each individual cuticle, using a thin glass needle, into the general cavity, at the posterior part of animals . This protocol was applied to inject 3 independent batches of individuals per treatments (i.e. injection of wVulC, wDil or control). Animals were then used for life history traits measurements, Wolbachia quantifications, electron microscopy and LC3B antibody labeling.
Animals injected for life history trait measurements
Thirty individuals per treatment for P. d. dilatatus (i.e. 15 males and 15 females) and 15 females per treatment for A. vulgare were monitored for life history measurements. To be able to demonstrate that the effect of wVulC was not due to higher doses injected compared to wDil, 15 P. d. dilatatus females were injected with a wVulC suspension diluted 100 times with filtered sterile Ringer. For all these animals, life-history traits were recorded during 75 days after the injection.
Animals injected for Wolbachia quantifications
Twelve individuals per treatment for P. d. dilatatus (i.e. 6 males and 6 females) and 6 females per treatment for A. vulgare were injected for Wolbachia quantifications. Half of the animals were used for Wolbachia quantification at 30 days post-injection and the other half for Wolbachia quantification at 60 days post injection.
Animals used for electron microscopy and LC3B labeling
For each host species 18 individuals per treatment were injected to observe the adipocytes and the nerve cells but also oocytes by electron microscopy at 30 (6 individuals for each treatment) and 60 days (6 individuals for each treatment) post-injection and to perform LC3B labeling antibody kit to characterize autophagic vesicles in adipocytes at 45 days post injections (6 individuals for each treatment).
Measurements of the different life history traits
For each condition and host species the life history traits were recorded for 75 days post-injection (until the last individuals infected by wVulC died for P. d. dilatatus). The first measurement was performed at t = 0 then 30 days after the injection. The following measurements were then performed every 15 days until the final one after 75 days. The gain of weight was calculated for each animal at each point by weighting them on a precision balance (d = 0.001 g) and dividing the weight at the time t = x by the weight at time t = 0. The mobility test was performed by measuring the time during which an individual moves in a glass Petri dish onto a period of 180 s. For P. d. dilatatus, only we determined the number of individuals located at the surface versus the number of buried animals after opening the breeding box by counting the number of animals found at the substrate and those which were buried. The survival rate was measured by counting the number of live animals for each treatment.
Tissue samples and DNA extraction for Wolbachia quantification
Total DNA was extracted from the Central Nervous system (CNS) (i.e. nerve cells and neighbouring adipocytes), the gonads and the hemocytes of each individual as described by Kocher et al.,  after dissections. For each sample, the concentration and quality (ratios OD 260/280 nm and 260/230 nm) of the extracted DNA were measured using the Nanodrop 1000 spectrophotometer (Thermo).
Quantification of Wolbachia in host's tissues
The quantification of Wolbachia by quantitative PCR (qPCR) was performed at 30 and 60 days post-injection to follow the time course of the infections before the death of all P. d. dilatatus infected by wVulC but after apparition of the first symptoms. All the qPCR amplifications were performed with DNA sampled from the CNS (i.e. nerve chord = nerve cells and neighbouring adipocytes), the gonads and the hemocytes. To allow a comparison between animals injected by Wolbachia and those which received the bacteria vertically from their mother; we also performed qPCR quantification on respectively 5 females A. vulgare naturally infected with wVulC and 5 females P. d. dilatatus naturally infected with wDil. All the qPCR reactions were performed using Roche LIGHTCYCLER 480 under the following conditions in 10 µL: 5 µL of SYBRGreen MasterMix (Roche), 0.5 µL of 10 µM specific primers wsp208f (5′- TGG-TGC-AGC-ATT-TAC-TCC-AG-3′) and wsp413r (5′-TCG-CTT-GAT-AAG-CAA-AAC-CA-3′), which amplified 205 bp of a single-copy of the gene wsp, 3 µL of sterile water and 1 µL of DNA (between 10 ng and 80 ng of DNA). The thermal cycling used an initial denaturation period of 10 min at 95°C, followed by 45 cycles of denaturing temperature at 95°C for 10 s, the annealing temperature for the reaction was 60°C for 10 s and 72°C for 20 s. A melting curve (65°C to 97°C) was recorded at the end of each reaction in order to check that the PCR product was unique. Efficiency of the PCR reaction was calculated. Standard curve was plotted using 7 dilutions of wsp purified PCR product (wsp copies.µL-1: 2.63×100, 2.63×101, 2.63×102, 2.63×103, 2.63×104, 2.63×105, 2.63×106, 2.63×107). wsp copy number was estimated by calculation in reference to the standard curve. The total DNA quantity (i.e. host+Wolbachia) of each sample was used to normalize wsp gene copy number. The results are thus given in number of wsp copies by ng of total DNA. For each condition (individual*organ*Wolbachia strain), two independent technical replicates were performed.
wsp gene amplification and sequencing for wVulC strain
In the inoculum and in animals injected with Wolbachia 60 days post-injection, we verified that only the expected Wolbachia strain was present (i.e. wDil or wVulC). To do so, a 610 bp fragment of the wsp region of the Wolbachia DNA was amplified using the specific primers wsp81F and wsp691R (81F 5′-TGG TCC AAT AAG TGA TGA AGA AAC and 691R 5′- AAA AAT TAA ACG CTA CTC CA) . The PCR cycling conditions were 95°C for 2 min followed by 35 cycles (95°C for 1 min, 55°C for 1 min, 72°C for 1 min) and 72°C for 5 min. Double strand PCR products were purified using 2 µl of a mix containing two enzymes: 0.05 µl of Antarctic Phosphatase (AnP, 1 u/µl; New England Biolabs, NEB, USA) and 0.1 µl of Exonuclease I (0.5 u/µl; NEB, USA) and incubation at 37°C for 1 hour and 80°C for 20 min. Purified DNA fragments were sequenced using the Big Dye v3.1 Terminator Sequencing kit (PE Applied Biosystems, Foster City, CA, USA), visualizing results on an ABI 310 automated sequencer. For each sample, wsp sequences were obtained from both strands and compared between them and with the reference sequence of wDil and wVulC.
To date, the most reliable way to characterize autophagy is the visualization by TEM of the formation of double membrane cytosolic vesicles called autophagosomes which sequesters cytoplasm and delivers it to the lysosome for degradation and are considered as the hallmark of autophagy , . To observe CNS cells and gonads by TEM, 6 animals from each host species (i.e A.vulgare and P. d. dilatatus) and from each treatment (injected by wVulC, injected by wDil, or injected by an ovary suspension containing no Wolbachia) were individually sampled 30 and 60 days post-injection. Tissues were fixed (9% glutaraldehyde, 0.3 M sodium cacodylate, 3% NaCl, v/v/v) for 2 hours at 4°C. Tissues were washed (0.3 M sodium cacodylate, 3% NaCl, 0.8 M sucrose, v/v/v) for 2 hours at 4°C. Post-fixation was performed into 4% OsO4, 0.3 M sodium cacodylate, 5.5% NaCl for 45 min. Tissues were subsequently dehydrated through a graded series of acetone solutions, infiltrated and embedded in resin (Spurr, Polyscience Inc.). Thick sections (0.5 µm) were stained with 1% toluidin blue for light microscopy observation. Thin sections (90 nm) were contrasted by incubation in 1% uranyl acetate in 50% ethanol for 1 min, then stained with lead citrate . Sections were observed using a transmission electron microscope (JEOL 100C).
LC3B labeling antibody
The LC3B protein plays a critical role in autophagy. Normally, this protein resides in the cytosol, but following cleavage and lipidation with phosphatidylethanolamine, LC3B associates with the phagophore. This localization can be used as a general marker for autophagic membranes. To visualize the LC3B, we used the LC3B Antibody Kit for Autophagy (Invitrogen). As this kit was designed to work with cell coming from cell culture it was not possible to perform it on the full CNS. We thus only performed LC3B labeling on adipocytes from the nerve chord which form a loose tissue that can be more easily penetrated by antibodies. To perform this labeling, we first deposited individually part of fat tissue containing adipocytes collected in: 6 A. vulgare naturally infected by wVulC, 6 A. vulgare injected by wVulC, 6 P. d. dilatatus injected by wDil and 6 P. d. dilatatus injected by wVulC (at 45 days post-injection) onto microscope slides harboring a 200 µL drop of 3.7% formaldehyde in PBS. The fat tissue samples were then incubated in fixative for 15 minutes at room temperature. The fixative was then removed and the adipocytes washed three times with PBS. After the last washing, 200 µL of 0.2% Triton X-100 in PBS were deposited onto the cells. The permeabilization buffer was then removed and 200 µL of the primary antibody (0.5 µg/mL in blocking BSA buffer) were deposited onto the cells and incubated for 1 hour at room temperature. The primary antibody was then removed and the samples washed again three times with PBS. The samples were then placed with 200 µL of an anti-rabbit secondary antibody (0.5 µg/mL in blocking BSA buffer) and incubated for 1 hour at room temperature. Samples were mounted in Citifluor (AFl antifading, Cititfluor, England). Detection was performed with a Carl Zeiss epifluorescent microscope (Axio Observer.Z1) with Apotome and AxioVision 4.8.1 software (Zeiss) equipped with a 63X objective (oil immersion).
All statistical analyses were performed using R software (version 2.10.1, 2009-12-14). For all life-history traits, we first adjusted a model to our data which includes sex, dose (i.e. the number of Wolbachia injected), and treatments (control, wVulC or wDil). We then re-adjusted the same model after having removed the control treatment from the data set, which allowed us to compare between the effects of wVulC and wDil. As the experiments were performed on three independent groups of individuals, a randomized block effect has been taken into account in the models when possible. This last analysis showed that the inter-block variance was very low indicating that measures were repeatable between blocks. The gain of weight and mobility were compared between treatments at 60 days post-injection with a Gaussian linear mixed-effects model with randomized block effect fit by maximum likelihood (ML) (lme function in nlme package ). For comparison of the animals' location in the boxes between treatments through the whole experiment, a generalized linear model with binomial error and logit link function was performed. In these analyses, the data were collected from simultaneous observations of all the individuals from a same rearing box. The comparison was thus made between groups of 6 boxes and not between individuals, it was thus not necessary to add a random effect in this model. Survival curves obtained with the different treatments were compared using a general mixed-effects Cox model with randomized block effect (coxme fonction in kinship package). Wolbachia density in tissues of A. vulgare, P. d. dilatatus and naturally infected animals were compared with t-test because data distribution followed a normal distribution and variances of the samples were homogeneous.
Behavioral film showing the symptoms encountered in P. d. dilatatus after the injection of the wVulC strain. Animals exhibited surfacing behavior, lack of mobility and several symptoms as seizures, leg tremors and paralysis.
We thank all the technical staff of the UMR CNRS 7267 and S. Guillot for technical assistance. We also thank Dr. J. Bertaux for helping to design the project, Dr. F.J. Richard for her assistance in the establishment of the mobility test and Dr. C. Gilbert for comments on previous version of the manuscript.
Conceived and designed the experiments: MS DB WL CBV. Performed the experiments: WL MR MS CBV. Analyzed the data: WL JBF MR MS. Wrote the paper: WL MS JBF CBV DB.
- 1. Serbus LR, Casper-Lindley C, Landmann F, Sullivan W (2008) The genetics and cell biology of Wolbachia-host interactions. Annu Rev Genet 42: 683–707.
- 2. Werren JH (1997) Biology of Wolbachia. Annu Rev Entomol 42: 587–609.
- 3. Jaenike J (2007) Fighting back against male-killers. Trends Ecol Evol 22: 167–169.
- 4. Huigens ME, Luck RF, Klaassen RH, Maas MF, Timmermans MJ, et al. (2000) Infectious parthenogenesis. Nature 405: 178–179.
- 5. Bouchon D, Cordaux R, Grève P (2008) Feminizing Wolbachia and the evolution of sex determination in isopods. In: Bourtzis K, Miller T, editors. Insect symbiosis. pp. 273–294.
- 6. Werren JH, Baldo L, Clark ME (2008) Wolbachia: master manipulators of invertebrate biology. Nat Rev Microbiol 6: 741–751.
- 7. Poinsot D, Mercot H (1997) Wolbachia infection in Drosophila simulans: does the females host bear a physiological cost? Evolution 51: 180–186.
- 8. Engelstädter J, Hurst GDD (2009) The Ecology and Evolution of Microbes that Manipulate Host Reproduction. Annu Rev Ecol Evol Syst 40: 127–149.
- 9. Lipsitch M, Siller S, Nowak MA (1996) The evolution of virulence in pathogens with vertical and horizontal transmission. Evolution 50: 1729–1741.
- 10. Bonds MH (2006) Host life-history strategy explains pathogen-induced sterility. Am Nat 168: 281–293.
- 11. Kraaijeveld K, Franco P, De Knijff P, Stouthamer R, Van Alphen JJM (2011) Clonal genetic variation in a Wolbachia-infected asexual wasp: horizontal transmission or historical sex? Mol Ecol 20: 3644–3652.
- 12. Werren JH, Zhang W, Guo LR (1995) Evolution and phylogeny of Wolbachia: reproductive parasites of arthropods. Proc Biol Sci 261: 55–63.
- 13. Lipsitch M, Moxon ER (1997) Virulence and transmissibility of pathogens: what is the relationship? Trends Microbiol 5: 31–37.
- 14. Schneider DS, Ayres JS (2008) Two ways to survive infection: what resistance and tolerance can teach us about treating infectious diseases. Nat Rev Immunol 8: 889–895.
- 15. Juchault P, Legrand JJ, Martin G (1974) Action interspécifique du facteur épigénétique féminisant responsable de la thélygénie et de l'intersexualité du Crustacé Armadillidium vulgare (Isopode Oniscoïde). Annales d'Embryologie et de Morphogenèse 7: 265–276.
- 16. Bouchon D, Rigaud T, Juchault P (1998) Evidence of widespread Wolbachia infection in isopod crustaceans: molecular identification and host feminization. Proc Biol Sci 265: 1081–1090.
- 17. Chevalier F, Herbinière-Gaboreau J, Bertaux J, Raimond M, Morel F, et al. (2011) The immune cellular effectors of terrestrial isopod Armadillidium vulgare: Meeting with their invaders, Wolbachia. PLoS One 6: e18531.
- 18. Rigaud T, Souty Grosset C, Raimond R, Mocquard JP, Juchault P (1991) Feminizing endocytobiosis in the terrestrial crustacean Armadillidium vulgare Latr. (Isopoda): Recent acquisitions. Endocytobiosis and Cell Res 7: 259–273.
- 19. Braquart-Varnier C, Lachat M, Herbinière J, Johnson M, Caubet Y, et al. (2008) Wolbachia mediate variation of host immunocompetence. PLoS One 3: e3286.
- 20. Sicard M, Chevalier F, De Vlechouver M, Bouchon D, Grève P, et al. (2010) Variations of immune parameters in terrestrial isopods: a matter of gender, aging and Wolbachia. Naturwissenschaften 97: 819–826.
- 21. Min KT, Benzer S (1997) Wolbachia, normally a symbiont of Drosophila, can be virulent, causing degeneration and early death. Proc Natl Acad Sci U S A 94: 10792–10796.
- 22. Reynolds KT, Thomson LJ, Hoffmann AA (2003) The effects of host age, host nuclear background and temperature on phenotypic effects of the virulent Wolbachia strain popcorn in Drosophila melanogaster. Genetics 164: 1027–1034.
- 23. McMeniman CJ, Lane RV, Cass BN, Fong AWC, Sidhu M, et al. (2009) Stable introduction of a life-shortening Wolbachia infection into the mosquito Aedes aegypti. Science 323: 141–144.
- 24. McGraw EA, Merritt DJ, Droller JN, O'Neill SL (2002) Wolbachia density and virulence attenuation after transfer into a novel host. Proc Natl Acad Sci U S A 99: 2918–2923.
- 25. Moreira LA, Saig E, Turley AP, Ribeiro JMC, O'Neill SL, et al. (2009) Human probing behavior of Aedes aegypti when infected with a life-shortening strain of Wolbachia. PLoS Negl Trop Dis 3: e568.
- 26. Drevets DA, Leenen PJM, Greenfield RA (2004) Invasion of the Central Nervous System by Intracellular Bacteria. Clin Microbiol Rev 17: 323–347.
- 27. Treadwell TA, Holman RC, Clarke MJ, Krebs JW, Paddock CD, et al. (2000) Rocky Mountain spotted fever in the United States, 1993–1996. Am J Trop Med Hyg 63: 21–26.
- 28. Ratnasamy N, Everett ED, Roland WE, McDonald G, Caldwell CW (1996) Central nervous system manifestations of human ehrlichiosis. Clin Infect Dis 23: 314–319.
- 29. Corby-Harris V, Habel KE, Ali FG, Promislow DEL (2007) Alternative measures of response to Pseudomonas aeruginosa infection in Drosophila melanogaster. J Evol Biol 20: 526–533.
- 30. Levine B, Klionsky DJ (2004) Development by self-digestion: molecular mechanisms and biological functions of autophagy. Dev Cell 6: 463–477.
- 31. Zhukova MV, Kiseleva E (2012) The virulent Wolbachia strain wMelPop increases the frequency of apoptosis in the female germline cells of Drosophila melanogaster. BMC Microbiol 12: Suppl 1S15.
- 32. Maeda K, Markowitz N, Hawley RC, Ristic M, Cox D, et al. (1987) Human infection with Ehrlichia canis, a leukocytic rickettsia. N Engl J Med 316: 853–856.
- 33. Levine B, Deretic V (2007) Unveiling the roles of autophagy in innate and adaptive immunity. Nat Rev Immunol 7: 767–777.
- 34. Levine B, Kroemer G (2008) Autophagy in the pathogenesis of disease. Cell 132: 27–42.
- 35. Niu H, Yamaguchi M, Rikihisa Y (2008) Subversion of cellular autophagy by Anaplasma phagocytophilum. Cell Microbiol 10: 593–605.
- 36. Rikihisa Y, Lin M (2010) Anaplasma phagocytophilum and Ehrlichia chaffeensis type IV secretion and Ank proteins. Curr Opin Microbiol 13: 59–66.
- 37. Félix C, Pichon S, Braquart-Varnier C, Braig H, Chen L, et al. (2008) Characterization and transcriptional analysis of two gene clusters for type IV secretion machinery in Wolbachia of Armadillidium vulgare. Res Microbiol 159: 481–485.
- 38. Pichon S, Bouchon D, Cordaux R, Chen L, Garrett RA, et al. (2009) Conservation of the Type IV Secretion System throughout Wolbachia evolution. Biochem Biophys Res Comm 385: 557–562.
- 39. Rigaud T, Juchault P (1995) Success and failure of horizontal transfers of feminizing Wolbachia endosymbionts in woodlice. J Evol Biol 8: 249–255.
- 40. Kocher TD, Thomas WK, Meyer A, Edwards SV, Pääbo S, et al. (1989) Dynamics of mitochondrial DNA evolution in animals: amplification and sequencing with conserved primers. Proc Natl Acad Sci U S A 86: 6196–6200.
- 41. Braig HR, Zhou W, Dobson SL, O'Neill SL (1998) Cloning and characterization of a gene encoding the major surface protein of the bacterial endosymbiont Wolbachia pipientis. J Bacteriol 180: 2373–2378.
- 42. Klionsky DJ, Abeliovich H, Agostinis P, Agrawal DK, Aliev G, et al. (2008) Guidelines for the use and interpretation of assays for monitoring autophagy in higher eukaryotes. Autophagy 4: 151–175.
- 43. Reynolds ES (1963) The use of lead citrate at high pH as an electron opaque strain in electron microscopy. J Cell Biol 17: 208–212.
- 44. Lindstrom MJ, Bates DM (1988) Newton-Raphson and EM Algorithms for Linear Mixed-Effects Models for Repeated-Measures Data. J Am Stat Assoc 83: 1014–1022.