(C) PLOS One
This story was originally published by PLOS One and is unaltered.
 . . . . . . . . . .



IMD-mediated innate immune priming increases Drosophila survival and reduces pathogen transmission [1]

['Arun Prakash', 'Institute Of Ecology', 'Evolution', 'School Of Biological Sciences', 'University Of Edinburgh', 'Edinburgh', 'United Kingdom', 'Florence Fenner', 'Biswajit Shit', 'Ashoka University']

Date: 2024-06 

Invertebrates lack the immune machinery underlying vertebrate-like acquired immunity. However, in many insects past infection by the same pathogen can ‘prime’ the immune response, resulting in improved survival upon reinfection. Here, we investigated the mechanistic basis and epidemiological consequences of innate immune priming in the fruit fly Drosophila melanogaster when infected with the gram-negative bacterial pathogen Providencia rettgeri. We find that priming in response to P. rettgeri infection is long-lasting and sexually dimorphic response. We further explore the epidemiological consequences of immune priming and find it has the potential to curtail pathogen transmission by reducing pathogen shedding and spread. The enhanced survival of individuals previously exposed to a non-lethal bacterial inoculum coincided with a transient decrease in bacterial loads, and we provide strong evidence that the effect of priming requires the IMD-responsive antimicrobial-peptide Diptericin-B in the fat body. Further, we show that while Diptericin B is the main effector of bacterial clearance, it is not sufficient for immune priming, which requires regulation of IMD by peptidoglycan recognition proteins. This work underscores the plasticity and complexity of invertebrate responses to infection, providing novel experimental evidence for the effects of innate immune priming on population-level epidemiological outcomes.

When we are vaccinated, our immune response is able to respond quickly if we are ever exposed to the same pathogen in the future. Unlike humans, the immune systems of invertebrates, such as insects, are not capable of the same type of specific immune memory. However, much work has shown that insects previously exposed to an inactivated pathogen will fare better if they are re-infected–a phenomenon broadly called “immune priming”. How insects are able to do this is an exciting focus of current research. We investigated immune priming in the fruit fly, a powerful model system for infection and immunity. We found that exposure to an inactivated form of the bacterial fly pathogen Providencia rettgeri resulted in flies surviving a subsequent live infection. This effect lasted several days, was stronger in male flies, and was seen in different fly genetic backgrounds. We uncovered that priming requires a specific immune response in the fly fat-body (the equivalent to a fly ‘liver’) that produces an antimicrobial protein called Diptericin. We also found that primed flies were able to keep pathogen growth lower, and that this reduced their ability to spread the infection to other flies.

Here, we focus on immune priming in Drosophila when infected with the gram-negative bacterial pathogen Providencia rettgeri to investigate the occurrence, generality, duration, and mechanistic basis of immune priming during systemic infection. Furthermore, motivated by the theoretical predictions about the role of immune priming on epidemiological dynamics, we also designed transmission experiments to enable us to test how immune priming could affect each of these behavioural and immunological components of pathogen spread.

Beyond its underlying mechanisms, an important but understudied aspect of immune priming is its potential consequences for pathogen transmission [ 17 , 25 – 27 ]. Epidemiological modelling predicts that priming should affect the likelihood of pathogen persistence, destabilise host–pathogen population dynamics, and that these effects depend on the degree of protection conferred by priming [ 26 ]. Further theoretical work suggests that priming can either increase or decrease infection prevalence depending on the extent to which it affects the pathogen’s colonization success and the host’s ability to clear or tolerate the infection [ 25 ]. While primed individuals may live longer, thus extending the infectious period, their pathogen burden may be lower, which could lead to lower pathogen shedding and less severe epidemics. Immune priming is therefore expected to have a significant impact on the outcome of pathogen transmission by directly modifying important epidemiological parameters, but the strength and direction of these effects is not intuitive to predict.

There are many ways in which invertebrates may enhance their immune responses upon reinfection [ 6 , 17 , 18 ]. In Drosophila, the priming response during infection with the gram-positive bacterial pathogen Streptococcus pneumoniae was shown to be dependent on haemocytes and phagocytosis, while the Toll-pathway—the main pathway involved in clearance of gram-positive bacteria—was shown to be insufficient for successful priming [ 9 ]. Increased phagocytic activity in primed individuals has also been shown to play a key role in priming during Pseudomonas aeruginosa infection in Drosophila [ 19 ], while in the woodlouse, prior exposure to heat-killed bacteria led to increased phagocytosis by haemocytes upon reinfection [ 20 ]. In other Drosophila work, PGRP-LB (peptidoglycan recognition protein LB, a negative regulator of the Immune deficiency (IMD)-pathway) was identified as a key mediator of transgenerational immune priming against infection with parasitoid wasps (Leptopilina heterotoma and Leptopilina victoriae). Here, downregulation of PGRP-LB was necessary to increase haemocyte proliferation, required for wasp encapsulation by haemocytes in the offspring [ 21 ]. By contrast, in response to infection with Drosophila C virus (DCV), Drosophila progeny can produce a DCV-specific priming response by inheriting viral cDNA from the infected adult flies which is a partial copy of the virus genome [ 22 , 23 ]. The benefits of priming are not always associated with increased pathogen clearance, as shown during Enterococcus faecalis infection, where protection was explained by increased infection tolerance, not increased clearance [ 24 ].

Immunisation using attenuated or inactivated pathogens is one of the most successful public health practices to reduce the incidence of infectious diseases [ 1 ]. Immunisation works because humans and other vertebrate animals have evolved an acquired immune response capable of specific immune memory, which ensures a strong, precise, and effective response to a secondary infection [ 2 ]. Insects possess a robust innate immune response to pathogens which includes both cellular and humoral components [ 3 – 5 ], but lack vertebrate-like specialized immune cells responsible for acquired immunity. These differences in immune physiology resulted in the long-standing assumption that invertebrates should not be capable of immune ‘memory’, though this view was clearly at odds with empirical evidence from several invertebrate host-pathogen systems [ 6 – 8 ]. A substantial body of work has revealed diverse priming responses in a range of arthropod taxa, including Dipterans: fruit flies [ 9 ], mosquitoes [ 10 ]; Coleopterans: flour beetles [ 11 , 12 ]; Lepidopterans: the greater wax-moth [ 13 ]; Hymenopterans: bumblebee [ 14 ]; Crustaceans: water fleas [ 15 ] and Arachnids: spiders and scorpions [ 16 ]. There is therefore substantial evidence that arthropods possess a form of “immune priming”, where low doses of an infectious pathogen, or even an inactivated pathogen, can lead to increased survival upon reinfection.

To address this, we used fly lines with loss-of-function in PGRP-LB, LC and LE. Regardless of which PGRP was disrupted, we observed that flies were no longer able to increase their survival following an initial exposure ( Fig 10A and S14 Table ). However, unlike similar outcomes with Relish or ΔAMP, here the lack of priming was not driven by an inability to clear bacterial loads, as microbe loads in PGRP mutants were ~100-fold lower than in Relish or AMP mutants ( Fig 10B and S15 Table ). Further, Dpt expression compared to the w 1118 control was either similar (PGRP-LC, PGRP-LE) or higher (PGRP-LB), as expected, in both primed and unprimed flies ( S5 Fig and S16 Table ). This suggests that Diptericin expression is required, but not sufficient, for successful priming, which requires adequate regulation by PGRPs. Which specific molecular signal modifies PGRP-mediated regulation of Diptericin expression following an initial priming challenge is unclear, but must lie upstream of the IMD pathway.

Following gram-negative bacterial infection, DAP-type peptidoglycans from gram-negative bacteria are recognised by the peptidoglycan receptors PGRP- LC (a transmembrane receptor) and PGRP-LE (a secreted isoform and an intracellular isoform), which then activate the IMD intracellular signalling cascade [ 36 , 38 ]. Immune regulation is achieved by several negative regulators, including PGRP-LB, a secreted peptidoglycan with amidase activity, that break down peptidoglycans into smaller, less immunogenetic fragments [ 36 , 38 ]. PGRP-LB has also been shown to be important for transgenerational immune priming in Drosophila against parasitoid wasp infection [ 21 ]. Given the role of DptB in the observed priming benefit, we therefore decided to investigate the role of PGRPs in the immune priming we observed during P. rettgeri infection.

Further, 72-hours following a lethal secondary live P. rettgeri infection, we found that primed w 1118 females and males showed increased DptB levels compared to unprimed individuals ( Fig 9A and 9B ). The increased DptB expression also resulted in increased bacterial clearance in males. In case of females, despite higher Dpt levels at 72-hours following secondary live infection in primed flies, we did not detect any difference in bacterial clearance between primed and unprimed flies. We also tested the expression pattern of other IMD-responsive AMPs such as Attacin-C and Drosocin in females. Overall, we found that the expression of Attacin-C and Drosocin was not different between primed and unprimed females ( Fig 9C for AttC and Dro and S13 Table ).

(A) Mean±SE (standard error) of Diptericin-AMPs expression at different time points (hours) for females and males wild-type w 1118 flies—naïve or unhandled flies (time 0h); flies exposed to initial heat-killed P. rettgeri (blue); primed flies that are initially exposed to heat-killed P. rettgeri followed by challenge with P. rettgeri (dark red); unprimed flies that receive 1xPBS during primary exposure followed by live P. rettgeri during secondary exposure (grey) [n = 5 groups of 3 pooled flies per sex/treatment/timepoint]. (B). Diptericin (A+B), Attacin-C and Drosocin AMPs expression 12-hours and 24-hours after exposure to live P. rettgeri in w 1118 female flies. Diptericin data is the same as shown in panel A (top right panel), shown here for clarity and for comparison with Attacin-C and Drosocin. Asterisks ‘*’ indicates that primed and unprimed individuals are significantly different (p<0.05). The error bars represent standard error.

As our results indicated that AMPs, especially Diptericins play a key role in Drosophila priming against P. rettgeri, we wanted to test if the priming effect we had observed was a result of constant upregulation of AMPs after initial heat-killed exposure allowing rapid bacterial clearance during the secondary exposure, or if AMP expression returned to a baseline level within the 96-hours between priming and the live infection. To do this, we measured Dpt gene expression at 18-hours and 72-hours following initial heat-killed exposure, and then 12-hours, 24-hours and 72-hours following secondary live P. rettgeri infection. DptB expression increased 18-hours after initial heat-killed exposure but returned to baseline levels by 72-hours ( Fig 9A and 9B and S13 Table ) indicating that flies did mount an immune response to the heat killed bacteria, but that this was resolved by the time they were infected with the live bacteria at 96-hours.

In response to gram-positive Streptococcus pneumoniae infection, previous work described the role of haemocytes in immune priming through increased phagocytosis [ 9 ]. Subsequent work has also shown that reactive oxygen species (ROS) burst from haemocytes is important for immune priming during Enterococcus faecalis infection [ 37 ]. Since our results pointed to a Diptericins being required for immune priming against P. rettgeri, we wanted to determine if Diptericin expression in either the fat body or haemocytes was more important for immune priming. Using-tissue specific Diptericin-B UAS-RNAi knockdown, we found that male flies with knocked-down DptB in fat bodies no longer showed immune priming compared to the background or control iso-w 1118 , while knocking down DptB in haemocytes resulted in a smaller but still significant increase in survival following an initial exposure ( Fig 8 and S12 Table ). This would therefore support that immune priming requires DptB expression in the fat body, while haemocyte-derived DptB is not important for priming. It is worth noting that it is possible that haemocytes contribute to immune priming through phagocytosis or melanisation, or via cross-talk with the IMD pathway, and this remains a question for future research.

Since Diptericins have been shown previously to play a key role in defence against P. rettgeri, we then used a Dpt mutant (lacking Diptericin-A and B) and AMPs +Dpt transgenic fly lines (flies lacking all known AMPs except Diptericin) to test whether Diptericins are required and sufficient for priming in both females and males. The survival benefit of priming disappeared in flies lacking Dpt A+B across both females and males and both primed and unprimed Dpt mutants exhibited increased bacterial load ( Fig 7Biii and S11 Table ). Notably, the priming response was recovered completely in male flies that lacked other AMPs but possessed functional Diptericins (AMPs +Dpt ). However, the same effect was not seen in females (Sex × Treatment effect p<0.01; Fig 7Aiv for survival, Fig 7Biv for bacterial load; S10 Table for survival and S11 Table for bacterial load).

Given the important role of the IMD pathway for the priming response ( Fig 6 ), next, we tested whether mutants with defective IMD signalling or unable to produce specific antimicrobial peptides were capable of immune priming. We first used a ΔAMP transgenic line, which lacks most of the known Drosophila AMPs (10 AMPs in total). ΔAMPs flies are extremely susceptible to the majority of microbial pathogens, including gram-negative bacteria [ 34 ], and we confirmed that both primed and unprimed ΔAMP flies succumb to death at a similar rate following systemic infection, and both primed and unprimed ΔAMP flies also exhibited elevated bacterial loads (measured 24-hours after the secondary pathogenic exposure) ( Fig 7Ai for survival, Fig 7Bi for bacterial load, S10 and S11 Tables , relative to control w 1118 flies–compare with previous figure Fig 6Bi ), showing that AMPs are needed for the priming response against P. rettgeri. To investigate which AMPs are required for priming against P. rettgeri infection, we used a Group-B transgenic line, lacking major IMD regulated AMPs (including Diptericins and Attacins and Drosocin) but have all upstream IMD signalling intact. Primed Group-B flies showed mortality similar to unprimed Group-B flies ( Fig 7Aii and S11 Table ) and exhibited increased bacterial loads across both the sexes irrespective of being primed or not ( Fig 7Bii and S11 Table ). Thus, despite being able to produce other AMPs, removal of IMD-regulated AMPs completely eliminated the priming effect, indicating that AMPs regulated by the IMD-pathway are required for immune priming against P. rettgeri infection.

The inducible AMPs regulated by the IMD signalling pathway play a crucial role in resisting gram-negative bacterial infection such as P. rettgeri [ 36 ]. Therefore, we wanted to investigate whether the IMD signalling pathway and IMD-responsive AMPs contribute to immune priming. To address this, we used several transgenic fly lines (CRISPR knockouts and UAS-RNAi knockdowns) with functional absence of or knockdown of different regulatory and effector components of the IMD-signalling pathway. As all CRISPR/cas9 AMP mutants were isogenized onto the iso-w 1118 background, we first confirmed that the priming response in iso-w 1118 was the same as the w 1118 used in the previous experiments (Figs 6A and S4 ). First, we tested a Relish loss-of-function mutant Rel E20 , a key regulator of the IMD immune response. As expected, we found that Relish mutants did not show any bacterial clearance, died at faster rate, and therefore did not show any benefit of a priming treatment ( Fig 6Aii for survival and Fig 6Bii for bacterial load, S10 Table for survival and S11 Table for bacterial load). Loss-of—function of spätzle (spz), a key regulator in the Toll pathway, showed enhanced survival following initial heat-killed exposure, and their mortality rates were the same as in controls (w 1118 ) ( Fig 6Aiii and S10 Table ) indicating that Toll-pathway does not contribute to priming during P. rettgeri infection ( Fig 6Aiii and S10 Table ). In both lines where priming induced a survival benefit, the magnitude of effect was similar for males and females (Sex × Treatment effect >0.6; S10 Table ).

In fruit flies, the production of AMPs during antibacterial immunity is mediated by the Immune deficiency (IMD) and Toll pathways. In both pathways, pathogens are recognised by peptidoglycan receptors (PGRPs), initiating a signalling cascade that culminates in the activation of the NF-κB-like transcription factors (Dorsal in Toll or Relish in IMD), resulting in the upregulation of AMP genes. The Toll-pathway generally recognises LYS-type peptidoglycan found in gram-positive bacteria and fungi. The IMD-pathway recognises DAP-type peptidoglycan found in gram-negative bacteria and produces AMPs such as Diptericins, Attacins and Drosocin among others. AMPs work with a high degree of specificity, so that only a small subset of the total AMP repertoire provides the most effective protection against specific pathogens [ 34 ], although this specificity has been shown to be greatly reduced during aging [ 35 ].

Given this priming effect on bacterial loads ( Fig 4B ) we hypothesised that priming could directly impact the amount of bacterial shedding, and therefore have a direct impact on pathogen transmission. We measured the shedding of single flies 4-hours after infection by live P. rettgeri exposure. We found that male flies previously primed with a heat-killed bacterial inoculum shed less bacteria than unprimed flies. However, both primed and unprimed females showed increased bacterial shedding following oral P. rettgeri infection ( Fig 5A and S9 Table ). These effects on bacterial shedding are likely to have a direct effect on the spread of infection in groups of individuals. In transmission assays, primed donor males also spread very little pathogen upon re-infection with oral P. rettgeri, compared to the unprimed male treatment where successful transmission was detected recipient flies ( Fig 5B and S9 Table ). However, both primed and unprimed females showed equivalent levels of bacterial spread following oral P. rettgeri infection ( Fig 5B and S9 Table ). While multiple traits, including host activity levels and contact rates may influence pathogen transmission [ 32 , 33 ], we did not find any difference in the locomotor activity of primed and unprimed flies ( S3 Fig ). Our results therefore provide evidence that, in male flies, immune priming reduces pathogen transmission by directly decreasing bacterial shedding from infectious flies.

Apart from increasing survival during infections, immune priming may also have consequences to pathogen spread and transmission [ 27 ]. Despite having enhanced survival, primed individuals may also extend the infectious period or their pathogen burden may be lower, which would lead to less severe epidemics [ 27 ]. To investigate how immune priming affects epidemiological parameters, we tested whether primed individuals have reduced bacterial shedding and spread. Given the oral-faecal nature of bacterial transmission and that our previous results all related to systemic infections, we first established that a survival benefit of priming also occurs under oral infection. Following an initial oral exposure to a heat-killed P. rettgeri culture, after a 72-hour period we exposed female and male w 1118 flies orally to a lethal dose of live P. rettgeri. Primed w 1118 males, but not females, showed increased survival decreased internal bacterial loads after priming via the oral route of infection. ( Fig 4A and 4B and S6 and S7 Tables ).

(A) Survival curves of male and female wildtype (w 1118 ) flies with (i) systemic (ii) oral priming where flies initially exposed to heat-killed P. rettgeri or unprimed flies exposed to a sterile solution in the first exposure, this is followed by live P. rettgeri pathogen after 72-hour time gap [systemic infection dose = 0.75 OD (~45 cells/fly) and oral P. rettgeri infection dose = 25 OD 600 ]. (n = 7–9 vials/sex/treatment/infection route) (B). Mean bacterial load measured as colony forming units at 24 hours following (i) systemic and (ii) oral priming, followed by live P. rettgeri infection for male and female w 1118 . Different letters in panel-B denotes primed and unprimed individuals are significantly different, tested using Tukey’s HSD pairwise comparisons.

Previous studies have shown that the priming response to fungi and gram-positive bacteria can be explained by increased clearance of bacterial pathogens in primed individuals [ 9 , 31 ]. We therefore examined whether the increased survival following a prior challenge we observed was a result of greater bacterial clearance in the primed individuals, or if the primed flies were simply better able to tolerate the bacterial infection [ 24 ]. To investigate this, we repeated the priming experiment with w 1118 as it showed the strongest priming phenotype in both sexes ( Figs 1C and 3A ), and measured the bacterial load at 24-hours and 72-hours post-infection for both primed and unprimed female and male flies. Again, we observed a clear priming effect in both sexes ( Fig 3A and S4 Table ). In primed males the systemic infection resulted in higher survival (Sex effect, p = 0<001), but the effect of priming on the survival benefit was the same in both sexes (Sex × Treatment effect p = 0.34; S4 Table ) and also in decreased bacterial loads at 24-hours post-infection when compared to unprimed individuals ( Fig 3 and S4 Table for survival and S5 Table for load). However, by 72-hours post-infection, bacterial loads had dropped in both primed and unprimed male flies and there was no detectable effect of priming on the bacterial loads ( Fig 4B and S5 Table ).

Next, we asked if priming occurred in two other commonly used Drosophila lines, Canton-S and Oregon-R (Ore-R) as observed with w 1118 ( Fig 2A–2D ). Given the widespread effects of the endosymbiont Wolbachia on Drosophila immunity [ 28 – 30 ] we also tested whether the presence of Wolbachia had any effect on immune priming by comparing the priming response of Oregon-R (OreR), a line originally infected with Wolbachia strain wMel, henceforth OreR Wol+ and a Wolbachia-free line OreR Wol- that was derived from OreR Wol+ by antibiotic treatment [ 30 ]. Females and males of the four Drosophila lines were treated first with heat-killed P. rettgeri, followed by infection with a systemic infection with live P. rettgeri 96-hours after the first treatment. Since the 96-hour time gap between priming and live P. rettgeri treatments showed maximum priming response (difference in survival) for both males and females of the w 1118 line (Figs 1C and 2A ), we kept the 96-hours timepoint as the consistent time-gap between priming and live infection throughout the study. w 1118 did not show differences between sexes in the priming responses at this timepoint ( S2 Table ; Sex × Treatment effect p = 0.93). Canton-S flies showed increased survival following priming and we observed a stronger survival benefit in females (Sex effect, p<0.001)( Fig 2C and S2 Table ). Oregon-R males also exhibited increased survival following priming, but priming had no significant effect on the females, as indicated by a significant interaction between sex and priming treatment (Sex × Treatment effect p<0.001; Fig 2D and S2 Table ). We found that the presence of Wolbachia significantly improved overall survival of both males and females ( Fig 2C and 2D and S3 Table ). However, the immune priming observed in males in absence of Wolbachia (OreR Wol- ) was no longer present in flies carrying Wolbachia (OreR Wol+ ), and both sexes showed similar patterns of survival (Sex effect, p = 0.43; Fig 2B and S2 Table ).

We first examined if the length of time between the initial non-lethal exposure with heat-killed P. rettgeri and the secondary systemic pathogenic challenge with live P. rettgeri affects the extent of priming. To address this, we exposed w 1118 male and female control flies to a systemic infection of live P. rettgeri 18-hours, 48-hours, 96-hours, 1-week or 2-weeks following the initial exposure to heat-killed bacteria. We found significant sex differences in the magnitude of the priming effect (sex treatment effect p<0.05) at 18-hours, 1-week and 2-weeks treatments ( S1 Table ), reflecting time-dependent and sex-specific priming dynamics. Male w 1118 flies showed increased survival after initial priming for time points 18-hours, 48-hours and 96-hours, and still showed a significant, albeit reduced, priming response 1-week and 2-weeks after the initial exposure ( Fig 1A–1E ). Female flies did not show a priming response when infected 18-hours following the initial challenge, but the priming response increased with 48-hours and 96-hours priming intervals before completely disappearing after a week time-interval (including 2-weeks) ( Fig 1A–1E and S1 Table ).

Discussion

The observation of immune priming in invertebrates underlines the importance of whole organism research in immunity in lieu of a purely mechanistic approach to immunology, highlighting that the same phenomenology can originate in very different mechanisms [18,39]. Thus, substantial experimental evidence in both lab adapted and wild-caught arthropods suggests that immune priming is a widespread phenomenon, and is predicted to have a profound impact on the outcome of host–pathogen interactions, including infection severity and pathogen transmission [40,26,27,17].

In the present work, we investigated the occurrence, generality, and mechanistic basis of immune priming in fruit flies when infected with the gram-negative pathogen Providencia rettgeri. We present evidence that priming with an initial non-lethal bacterial inoculum results increased survival after a secondary lethal challenge with the same live bacterial pathogen. This protective response may last at least two weeks after the initial exposure, is particularly strong in male flies, and occurs in several genetic backgrounds. We show that the increased survival of primed individuals coincides with a transient decrease in bacterial loads, and that this is likely driven by the expression of the IMD-responsive AMP Diptericin-B in the fat body. Further, we show that while Diptericin is required as the effector of bacterial clearance, it is not sufficient for immune priming, which requires the regulation by at least three PGRPs (PGRP-LB, PGRP-LC, and PGRP-LE). Therefore, despite having an intact IMD signalling cascade, and being able to express Diptericin, flies lacking any one of PGRP-LB, LC or LE were not capable of increasing survival following an initial sublethal challenge. Together, our data indicates that PGRPs are necessary for regulating immune priming against P. rettgeri.

These results are largely consistent with other work showing that the pathways that signal pathogen clearance may not be the same that underlie the signalling of the priming response. For example, Cabrera et al [24] investigated priming with the Gram-positive E. faecalis and found that while the Toll pathway was required for responding to a single infection, Toll signalling was dispensable for immune priming [24]. By contrast, IMD-deficient flies were still able to clear single E. faecalis infections (due a functional Toll response) but were no longer able to increase survival following an initial exposure. That is, in support of our results here, the IMD played a distinct role in signalling priming that was independent of its role in clearance. Further, previous work found that the Toll pathway is required (though not sufficient) for priming in response to gram-positive bacterial Streptococcus pneumoniae infection [9] and here we find that the Toll pathway is not required for a functional immune priming response.

In response to infection, the expression of several AMP genes in Drosophila increases and then drops once the infection threat is resolved or controlled. Previous studies have shown that this occurs within a period of few hours during bacterial infections [41], and that in many cases, this increased AMP expression underlies the immune priming response [18,42]. In other host-pathogen systems, genome wide insect transcriptome studies have identified upregulation of several AMPs in primed individuals, for example, Attacins, Defensins and Coleoptericins in flour beetles [43,44], Cecropin, Attacin, Gloverin, Moricin and Lysozyme in silkworms [45], Gallerimycin and Galiomicin in wax-moths [46] and finally, Cecropin in tobacco moths and fruit flies [47,37]. However, there are also examples where innate immune priming involves the downregulation of AMP expression, such as priming with the gram-positive E. faecalis [24].

One key aspect of our results is that immune priming was not the result of continued upregulation of Diptericin following the initial exposure to heat-killed bacteria. Instead, we observed that the initial immune response to heat-killed bacteria had already been resolved at 72-hours, before the secondary exposure to a lethal infection at 96-hours (see Fig 9A). This second up-regulation of AMP expression was at least 10-times higher than the response to heat-killed bacteria, and was initially similar between both primed and unprimed flies (exposed to a sterile solution in the first exposure). It was only at 72-hours following the second lethal exposure that we observed significantly higher expression of Diptericin in primed flies. This difference appears to arise because unprimed flies show a faster resolution of the immune response compared to primed flies; that is, the expression of Diptericin shows a faster decline between 24-hours and 72-hours post-exposure in unprimed compared to primed flies (see Fig 9A). These patterns of gene expression provide a partial explanation for the increased survival following priming, but they do not explain why primed individuals had lower bacterial loads at 24-hours after exposure to the second lethal infection, as at this timepoint we did not detect any differences in Diptericin expression between primed and unprimed flies. It is also unclear why primed females showed increased expression of Diptericin after 24-hours, but do not show any reduction in bacterial loads at this timepoint following priming, in the way male flies did.

The sex differences we observed in priming reflect a larger pattern of sexual dimorphism in immunity present in most organisms, including Drosophila [48–50]. Here, we found that males showed better bacterial clearance after initial exposure to heat-killed P. rettgeri which enabled them to experience enhanced survival compared to females, who exhibited higher bacterial loads and greater mortality. Previous work has established several ways in which the immune responses of male and female Drosophila may differ, including Toll [51] and IMD [52] signalling or signalling of damage during enteric infections [53]. Our results indicate that the Toll pathway is not required for successful priming, so sex-differences in Toll signalling are unlikely to be important in explaining sexually dimorphic priming. However, other work as shown that disrupting the negative regulator of IMD, PGRP-LB, affected survival to a greater extent in females following E. coli infection [52]. Given that we identified a role for PGRP-mediated regulation of Dpt as key for immune priming, sex differences in PGRP regulation could potentially explain differences in priming between males and females.

Another aspect related to sex differences in priming, was that males experienced a survival benefit of prior exposure is very experiment we report; female flies showed more variable response, showing a survival benefit priming repeatedly in several independent experiments (Figs 1, 2, 3, and 6) but not in others (Fig 4). Further while the priming phenotype was successfully recovered in males by restoring expression of the AMP Dtp, this was not observed in females. These variable outcomes in females remain puzzling, and are likely not the outcome of heritable sex-differences–it is hard to imagine how a priming response might evolve in males, but not females, and we did repeatedly observe it. Instead, one possibility for the observed sex-differences could relate to unmeasured environmental variation in our experimental setups–although these were always minimised as much as possible–which may interact with different nutritional demands and metabolic activities in female Drosophila. For instance, female fruit flies are often able to reallocate resources in accordance with their reproductive demands, as observed in terminal investment during infection [54–56]. Studies from other insects suggest that inducing priming responses can directly reduce the reproductive fitness in mosquitoes [57], wax moth [58], and mealworm beetles [59]. There are therefore potential trade-offs between investment in reproductive effort and investment in stronger immune responses following priming. Given that all females used in this study were mated, it is possible that such trade-offs forced a reallocation of resources towards reproduction, thereby reducing the observed magnitude of the priming response in females. Future studies may consider comparing priming responses in females with different reproductive states in order to test whether immune priming is costlier for female Drosophila.

A subsidiary finding of this work was the effect of the endosymbiont Wolbachia on immune priming. How endosymbionts that are widespread among insects are likely to influence innate immune priming is a topic of considerable interest [60]. There is abundant evidence that Drosophila carrying the endosymbiont Wolbachia are better able to survive infections, especially viral infections [61,29,62,30]. In this case, we observed that the priming response that was present in male flies cleared of Wolbachia disappeared in males carrying the endosymbiont. This effect is unlikely due to a direct effect of Wolbachia on the ability to clear P. rettgeri in primed flies, as previous work has found that Wolbachia had no effect on the ability to suppress P. rettgeri during systemic infection [63]. Indeed, if priming is the result of upregulation of AMP expression, this may suggest that Wolbachia may be actively suppressing the expression of AMPs, thereby reducing the beneficial effects of priming. However, this hypothesis would contradict work showing that some Wolbachia strains upregulate the host’s immune response and result in a reduction of pathogen growth [64,65]. Another possibility is that while both Wolbachia and P. rettgeri are Gram-negative bacteria, they could elicit the expression of competing AMP responses, leading to a less efficient priming response. Further rigorous experimental work is therefore required to fully understand this effect of Wolbachia on immune priming.

Finally, it is important to consider the implications of immune priming for disease ecology and epidemiology, and particularly how it may affect pathogen transmission. If priming acts by improving bacterial clearance via increased AMP expression, as observed in the present work, we predict that priming is likely to reduce pathogen shedding at the individual level, resulting in reduced disease transmission at the population level. Epidemiological models that have incorporated priming predict that primed individuals with enhanced survival following an initial sub-lethal pathogenic exposure are less likely to become infectious upon re-infections [27]. It remains unclear whether immune priming reduces pathogen transmissibility, varies the infectious period, or alters infection-induced behavioural changes in the host [27]. An additional level of complexity is that there is likely to be substantial within-population genetic variation in how priming affects each of these components of pathogen transmission. Insects offer a powerful system to investigate these effects, because they rely on an innate immune system that can induce an easily measurable priming response, but further, insect are also important vectors of many infectious diseases. Immune priming has thus emerged as a provocative idea to reduce the vectorial capacity of insect vectors, thereby reducing the transmission of vector-borne pathogens [17]. A better understanding of how immune priming contributes to host heterogeneity in disease outcomes would aid our understanding of the causes and consequences of variation in infectious disease dynamics.

[END]
---
[1] Url: https://journals.plos.org/plospathogens/article?id=10.1371/journal.ppat.1012308

Published and (C) by PLOS One
Content appears here under this condition or license: Creative Commons - Attribution BY 4.0.

via Magical.Fish Gopher News Feeds:
gopher://magical.fish/1/feeds/news/plosone/