The defensive aphid symbiont Hamiltonella defensa affects host quality differently for Aphelinus glycinis versus Aphelinus atriplicis

Accepted Manuscript The defensive aphid symbiont Hamiltonella defensa affects host quality differently for Aphelinus glycinis versus Aphelinus atriplicis Keith R. Hopper, Kristen L. Kuhn, Kathryn Lanier, Joshua H. Rhoades, Kerry M. Oliver, Jennifer A. White, Mark K. Asplen, George E. Heimpel PII: DOI: Reference:

S1049-9644(17)30110-X http://dx.doi.org/10.1016/j.biocontrol.2017.05.008 YBCON 3592

To appear in:

Biological Control

Received Date: Revised Date: Accepted Date:

20 January 2017 16 May 2017 19 May 2017

Please cite this article as: Hopper, K.R., Kuhn, K.L., Lanier, K., Rhoades, J.H., Oliver, K.M., White, J.A., Asplen, M.K., Heimpel, G.E., The defensive aphid symbiont Hamiltonella defensa affects host quality differently for Aphelinus glycinis versus Aphelinus atriplicis, Biological Control (2017), doi: http://dx.doi.org/10.1016/ j.biocontrol.2017.05.008

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

The defensive aphid symbiont Hamiltonella defensa affects host quality differently for Aphelinus glycinis versus Aphelinus atriplicis Keith R. Hopper1*, Kristen L. Kuhn1, Kathryn Lanier1, Joshua H. Rhoades1, Kerry M. Oliver2, Jennifer A. White3, Mark K. Asplen4, and George E. Heimpel5 1

USDA-ARS, Beneficial Insect Introductions Research Unit, Newark, Delaware, USA

2

Department of Entomology, University of Georgia, Athens, Georgia, USA

3

Department of Entomology, University of Kentucky, Lexington, Kentucky, USA

4

Natural Sciences Department, Metropolitan State University, Saint Paul, Minnesota, USA

5

Department of Entomology, University of Minnesota, St. Paul, Minnesota, USA

*Corresponding author. email: [email protected] phone: 302-731-7330 ext 238 address: USDA-ARS, Beneficial Insect Introductions Research Unit, 501 South Chapel Street, Newark, Delaware 19736,

Hopper et al. Defensive endosymbionts and aphelinid parasitoids of aphids

2

Abstract Endosymbiont interactions with hosts have important effects on fitness, including the fitness of many pest and beneficial species. Among these interactions, facultative endosymbiotic bacteria can protect aphids from parasitoids. Aphis craccivora and Acyrthosiphon pisum can harbor the symbiotic bacteria Hamiltonella defensa and its bacteriophage APSE. Infection by H. defensa defends these aphids against some but not all parasitoid species in the hymenopteran family Braconidae. Here, we report results on the effect of H. defensa on parasitism of these aphids by species in the other major lineage of aphid parasitoids, Aphelinus species in the family Aphelinidae. Parasitism of aphids infected with H. defensa/APSE by two Aphelinus species did not differ from that of uninfected aphids. While Aphelinus atriplicis showed no difference in fitness components between infected and uninfected aphids, Aphelinus glycinis actually produced more adult progeny and larger female progeny on infected than on uninfected aphids. Aphelinus glycinis may increase host quality for itself by changing the titer of nutritional versus protective bacteria in such a way that aphids infected with H. defensa can be made more suitable for parasitoid development than uninfected aphids. Our results and reasoning suggest that these Aphelinus species may be less prone to harm by H. defensa/APSE that affect eggs because they have anhydropic, heavily chorionated eggs, which may not absorb toxins during embryogenesis. Keywords: Aphelinus atriplicis, Aphelinus glycinis, Aphis craccivora, Acyrthosiphon pisum, Hamiltonella defensa, APSE, aphid endosymbionts

Hopper et al. Defensive endosymbionts and aphelinid parasitoids of aphids

3

1. Introduction Insect-associated microbes can exert major effects on host fitness, often by enhancing resource acquisition or providing protection against biotic or abiotic threats (Moran et al., 2008; for reviews, see: O'Neill et al., 1997; Oliver and Martinez, 2014; Stouthamer et al., 1999). Aphid species have emerged as important models for the study of nutritional and protective symbioses (Baumann et al., 1995; Douglas, 1998; Moran, 2001; Oliver et al., 2010; Oliver et al., 2014). Almost all aphid species harbor the obligate bacterial symbiont, Buchnera aphidicola Munson et al. (Gammaproteobacteria: Enterobacteriaceae), which supplies essential nutrients lacking in their phloem diet (Moran et al., 2008), and many are also infected with one or more maternally transmitted facultative symbionts (Oliver et al., 2010). Facultative symbionts are not generally required for aphid survival or reproduction, but often confer conditional costs and benefits (Oliver et al. 2010). From the perspective of the bacteria, the relationship with insect hosts is obligate because the bacteria are not found free-living in the environment. In some contexts, facultative symbionts impose constitutive costs on the aphids they infect (Oliver et al., 2008; Sakurai et al., 2005; Vorburger and Gouskov, 2011), or function as reproductive parasites (Simon et al., 2011), but in others they provide diverse benefits, including greater thermal tolerance (Montllor et al., 2002; Russell and Moran, 2006), wider host plant range (McLean et al., 2011; Tsuchida et al., 2004; Wagner et al., 2015), and defense against specialized natural enemies such as fungal pathogens (Lukasik et al., 2013; Parker et al., 2013; Scarborough et al., 2005) and parasitoid wasps (Oliver et al., 2003; Schmid et al., 2012; Vorburger et al., 2010). Given the importance of aphid-specialized parasitoid species in biological control programs (Heimpel et al., 2004; Hopper et al., 1998), the presence of defensive symbionts in either target or non-target species may affect the benefits and risks of biological control introductions (Ferrari et al., 2004; Vorburger, 2014). The best-developed cases for symbiont-mediated defense against parasitoids involve the pea aphid, Acyrthosiphon pisum (Harris) (Hemiptera: Aphididae) and the black-bean aphid Aphis fabae (Scopoli) (Oliver et al., 2014; Vorburger, 2014). In the pea aphid, which harbors at least eight secondary symbionts (Russell et al., 2013), the symbiont Hamiltonella defensa Moran et al. (Gammaproteobacteria: Enterobacteriaceae) provides partial to complete protection against the braconid wasp Aphidius ervi Haliday (Martinez et al., 2014; Oliver et al., 2009; Oliver et al., 2005; Oliver et al., 2003) and likely its congener Aphidius eadyi Starý et al. (Hymenoptera: Braconidae) (Ferrari et al., 2004; Oliver et al., 2003). However, for the aphid to

Hopper et al. Defensive endosymbionts and aphelinid parasitoids of aphids

4

be protected, H. defensa must itself carry bacteriophages, named APSE (from the initials of Ac. pisum secondary endosymbionts), (Degnan and Moran, 2008; Moran et al., 2005; Oliver et al., 2009). Different APSE strain types (e.g. APSE-2, APSE-3, APSE-4) encode substantively different toxin-analogs that are hypothesized to harm the internally developing wasps at different stages of development (Degnan and Moran, 2008). In the black bean aphid, H. defensa confers protection against attack by a different braconid, Lysiphlebus fabarum (Marshall) (Hymenoptera: Braconidae), although the level of defense varies with aphid age and the strains of bacterium and parasitoid, revealing the possibility of a co-evolutionary arms race (Schmid et al., 2012). Infection by H. defensa has been documented in 15-40% of aphid species screened (Henry et al., 2015), including the cowpea aphid, Aphis craccivora Koch (Hemiptera: Aphididae), which harbors H. defensa at varying frequencies in natural populations (Brady et al., 2014; Brady and White, 2013). Infection by H. defensa defends Ap. craccivora against some but not all parasitoid species in the family Braconidae (Asplen et al., 2014; Desneux et al., 2009). Infection by H. defensa completely eliminates parasitism by Binodoxys communis (Gahan) and Binodoxys koreanus Starý, but has no effect on parasitism by Lysiphlebus orientalis Starý and Rakhshani and Aphidius colemani Viereck (Hymenoptera: Braconidae), which indicates at least genus-level specificity of protective effects by H. defensa (Asplen et al., 2014). Similarly, in pea and black bean aphids, H. defensa confers protection against some, but not all braconid parasitoids (Cayetano and Vorburger, 2015; Martinez et al., 2016). Here, we report results on the effect of H. defensa on parasitism of Ap. craccivora and Ac. pisum by species in the other major lineage of aphid parasitoids, species of Aphelinus (Aphelinidae), for which the reported effects of H. defensa/APSE are mixed. Recent research has shown that Aphis fabae Scopoli (Hemiptera: Aphididae) infected with H. defensa is not protected from parasitism by Aphelinus chaonia (Hymenoptera: Aphelinidae), although the APSE strain was not identified (Cayetano and Vorburger, 2015). However, other research has shown that Ac. craccivora infected with H. defensa were protected from parasitism by Aphelinus abdominalis (Dahlman) (Hymenoptera: Aphelinidae) when the aphids came from one host plant species (Lotus pedunculatus Cavanilles; Fabaceae), but not when they came from two other host plant species (Ononis spinosa L. and Medicago sativa L.; Fabaceae) (McLean and Godfray, 2015). In the present research, we compared parasitism of aphids infected with H. defensa/APSE with parasitism of uninfected aphids sharing a common aphid genotype. We measured parasitism

Hopper et al. Defensive endosymbionts and aphelinid parasitoids of aphids

5

of Ap. craccivora by Aphelinus atriplicis Kurdjumov and Aphelinus glycinis Hopper and Woolley (Hymenoptera: Aphelinidae), species that differ strongly in host range. We measured parasitism of Ac. pisum by Aphelinus atriplicis only because Aphelinus glycinis does not parasitize this aphid species. Aphelinus atriplicis is native to countries around the Black Sea, including the Ukraine and the Republic of Georgia, but was introduced into the USA to control the Russian wheat aphid, Diuraphis noxia (Kurdjumov) (Hemiptera: Aphididae) in the early 1990s (Heraty et al., 2007; Hopper et al., 1998). Aphelinus glycinis is a recently described species native to Asia (Hopper et al., 2012), which is being released in the USA to control the soybean aphid, Aphis glycines Matsumura (Hemiptera: Aphididae). Aphelinus glycinis has a very narrow host range, being limited to some species in the genus Aphis, whereas Aphelinus atriplicis has a very broad host range, parasitizing aphid species in at least six genera distributed across the phylogeny of aphids (KRH, unpublished data). All Aphelinus species are koinobiont endoparasitoids of aphids (Hagen and VanDenBosch, 1968; Viggiani, 1984), and their hosts continue to develop after oviposition for a week at 20°C, at which point Aphelinus larvae kill their hosts but leave the host exoskeleton intact, causing it to harden and turn black in a process called mummification (Christiansen-Weniger, 1994). Aphelinus females prefer 2-4th instar aphids for oviposition, but will oviposit in all stages, including alate adults (Rohne, 2002). Because both Ac. pisum and Ap. craccivora develop from the beginning of the first instar to adult in 7 days at 20C (Lu and Kuo, 2008; Soffan and Aldawood, 2014), the stage in which female parasitoids oviposit has little effect on the aphid stage and size that is mummified. Adult Aphelinus eat nectar and honeydew, but adult females also feed on aphid hemolymph to get amino acids and fats for egg production and survival (Hagen and VanDenBosch, 1968; Wu and Heimpel, 2007). Females of Aphelinus glycinis carry a mean of 8 [7-9] eggs or 12 [12-13] eggs one day after emergence, depending on whether they had four or six ovarioles (KRH, unpublished data). However, like other Aphelinus species (e.g., Perng and Liu, 2002; Wu and Heimpel, 2007), they produce far more eggs throughout their lives. In the laboratory, Aphelinus atriplicis can parasitize ~100 Diuraphis noxia during a lifetime with a median longevity of one week (KRH, unpublished data). Aphelinus eggs are heavily chorionated with substantial yolk, and these eggs absorb little or nothing from their hosts during embryogenesis, i.e. before hatching as 1st instars. Indeed, our

Hopper et al. Defensive endosymbionts and aphelinid parasitoids of aphids

6

criterion for egg maturity is lack of uptake of neutral red dye because mature eggs of Aphelinus have a chorion that is impervious to this relatively small molecule (molecular weight = 289), whereas immature eggs stain red (Hopper et al., 2013). Given the yolky and heavily chorionated eggs of Aphelinus species, we hypothesized that they might not be affected by H. defensa and its APSE phage. Furthermore, we hypothesized if H. defensa/APSE toxins affected later stages, Aphelinus atriplicis might be more susceptible to the effects of Hamiltonella/APSE than Aphelinus glycinis, given that the latter species specializes on aphids in the genus Aphis whereas Aphelinus atriplicis is a broad generalist that may not have evolved resistance to H. defensa/APSE. 2. Materials and Methods 2.1 Insect sources and rearing conditions Aphelinus glycinis was collected in the Peoples Republic of China under a Memorandum of Understanding between their Ministry of Agriculture and the United States Department of Agriculture (USDA). Aphelinus atriplicis was collected by employees of the USDA, Agricultural Research Service (ARS), in the Republic of Georgia with the permission of that government. The parasitoids were imported into the USDA, ARS, Beneficial Insect Introductions Research Unit containment facility in Newark, Delaware, under permits from the USDA, Animal and Plant Health Inspection Service (Permit Numbers P526P-08-02142 and P526P-09-01929). No specific permissions were required to collect Ap. craccivora or Ac. pisum because these are cosmopolitan aphids that occur in the field throughout North America. None of the species collected or studied are endangered or protected. Aphis craccivora is a cosmopolitan pest of legumes and is anholocyclic, i.e. reproduces asexually, throughout most of its range (Blackman and Eastop, 2006). Acyrthosiphon pisum is an almost cosmopolitan pest of legumes and is holocyclic, i.e. alternates by asexual and sexual reproduction in temperature regions (Blackman and Eastop, 2006). Aphis craccivora in these experiments were collected originally in Kentucky, USA, in 2009, and Ac. pisum were collected in Utah, USA, in 2007. The aphids were cultured on fava bean, Vicia faba L., in 12 cm pots in plant growth chambers at 20oC and a 16L:8D h photoperiod. Isofemale lines were initiated from single female aphids on individual V. fava plants. Lines from the original populations tested positive for infection with only H. defensa using Denaturing Gradient Gel Electrophoresis (DGGE) analysis of PCR-amplified fragments of 16S rRNA and confirmed with diagnostic PCR (Russell et al., 2013).

Hopper et al. Defensive endosymbionts and aphelinid parasitoids of aphids

7

APSE type 3 was found in H. defensa in Ac. pisum and APSE type 4 was found in H. defensa in Ap. craccivora. Aphid lines without H. defensa were generated by a selective curing technique in which aphids were fed an artificial diet that was treated with equal parts of the antibiotics gentamycin, defatoxiamine and ampicillin for 3 days, after which the aphids were placed on V. fava plants (Dykstra et al., 2014). Clearing of H. defensa from offspring of aphids from these lines was confirmed by PCR (Russell et al., 2013). Cured (AC1H-) and uncured lines (AC1H+) from Ap. craccivora and cured (AS3H-) and uncured lines (AS3H+) from Ac. craccivora were used for the experiments reported here. They were shipped to Newark, Delaware, USA, in spring 2012, where they were maintained for 40-50 generations prior to these experiments. Both cured and uncured lines were tested using diagnostic PCR for the presence of H. defensa and APSE prior to assays. Aphelinus atriplicis was collected as mummified Diuraphis noxia on wheat near Tbilisi, Republic of Georgia, in 2000; A. glycinis was collected as mummified Aphis glycines Matsumura (Hemiptera: Aphididae) on soybean near Xiuyan, Liaoning Province, Peoples Republic of China, in 2007. Aphelinus glycinis was described from the culture used in this study (Hopper et al., 2012); A. atriplicis was described by Kurdjumov (1913), and the culture used in this study was included in a molecular phylogeny of species in the Aphelinus varipes complex (Heraty et al., 2007). Cultures were established in the quarantine facility at the USDA-ARS, Beneficial Insect Introductions Research Unit, Newark, Delaware, USA. The cultures were divided into 4-6 subcultures, and each subculture was maintained with an adult population size >200 and sex ratio of 1:1 males:females. Aphelinus atriplicis was reared on D. noxia on barley, Hordeum vulgare L. (Poaceae); A. glycinis was reared on Aphis glycines on soybean, Glycine max (L.) (Fabaceae). The culture of D. noxia was started in 1998 with aphids from southeastern Wyoming, and the culture of A. glycines was started in 2008 with aphids from Newark, Delaware. Neither D. noxia nor A. glycines are known to harbor H. defensa, and diagnostic PCR assays confirmed the absence of H. defensa in our colonies of these species. Although A. glycines harbors another bacterium, Arsenophonus, this secondary symbiont apparently does not protect against parasitoids (Wulff et al., 2013). All cultures were reared on appropriate host plant species in plant growth chambers at 20°C, 50-70% relative humidity, 16:8 h (L:D) photoperiod. Vouchers are maintained at -20°C in molecular-grade ethanol at the Beneficial Insect Introduction Research Unit, Newark, Delaware, USA.

Hopper et al. Defensive endosymbionts and aphelinid parasitoids of aphids

8

2.2 Parasitism and parasitoid fitness on aphids infected with Hamiltonella defensa versus uninfected aphids To measure the effect of infection with H. defensa harboring APSE on parasitism of Ap. craccivora and Ac. craccivora, as well as on parasitoid fitness components, we exposed individual, mated females of Aphelinus atriplicis or Aphelinus glycinis that were <5 days old to either aphids infected with H. defensa or uninfected aphids on V. faba variety Windsor. Females were drawn at random from the replicated sub-populations used for rearing. Seeds were planted in 12 cm diameter x 12 cm tall pots with the same soil mix and fertilizer used for insect rearing and plants and insects were kept under the same conditions as described for insect rearing prior to use in experiments. We put each female parasitoid in a cage (polystyrene, 10 cm diameter by 22 cm tall, with eight 2.5 cm holes in the sides and a 12 cm hole in the top covered with fine-mesh screening) enclosing the foliage of 2-3 young fava bean plants in a pot with about 100 aphids, including 1-4 instar nymphs and adults. The aphids were transferred as colonies on excised leaves from our rearing cultures, and the excised leaves were placed on the experimental plants to allow the aphids to settle and feed two days before the parasitoids were added. Female parasitoids were removed after 7 days. Aphelinus females rarely superparasitize unless they are host limited (Bai and Mackauer, 1990; Hagen and VanDenBosch, 1968). Their lifetime fecundities are 100-300, and given that we exposed them to 100 aphids at the beginning of the experiments, and the aphids would produce >2000 progeny during the 7 day exposure, there were far more aphids than they could parasitize during this period so the likelihood of superparasitism is quite low. The density of aphids, amount of plant material, and cage size meant that parasitoids were not limited by search rate. Aphids parasitized by these Aphelinus species mummify about 7 days after being parasitized and adult parasitoids emerge 10 days after mummification. To ensure that aphids parasitized throughout the exposure period had time to mummify but would not yet have emerged as adults, we collected mummified aphids 7 days after the end of the exposure of aphids to parasitoids and held them for adult parasitoid emergence. After the adults emerged, we recorded the number of mummified aphids, the number of mummies from which adults emerged, and the number and sex of adult parasitoids. We dried the adult progeny of each female at 50oC for one hour and weighed the sexes separately on a microbalance. Because we scored parasitism after the larval parasitoids killed and mummified their hosts, which occurs during the parasitoid third instar,

Hopper et al. Defensive endosymbionts and aphelinid parasitoids of aphids

9

this measure of parasitism is a combination of acceptance of aphids for oviposition and suitability of aphids for parasitoid survival to third instar. 2.3 Design structure and statistical analysis The experiments on parasitism of each aphid species by each wasp species were designed as randomized complete-blocks blocks: location within rearing chamber was the blocking factor and each cage with a female parasitoid, aphids, and plants was an experimental unit. There were 48 experimental units each for Aphelinus glycinis on Ap. craccivora and Aphelinus atriplicis on Ap. craccivora and on Ac. pisum (24 each for Hamiltonella-infected versus uninfected aphids). Where the female was missing or dead after 7 days, the experimental unit was dropped from the analysis. For each analysis, block was included in the initial model but was not significant and so was removed from the final model. For each parasitoid and aphid species combination, we tested the effects of H. defensa infection on the number of aphids parasitized (mummified), the number of adult parasitoid progeny, progeny sex ratio (proportion males), and progeny size (µg dry mass) of males and females. Because the dependent variables often had non-normal distributions with variances proportional to means, we used generalized linear models with appropriate distributions (e.g. binomial, normal, gamma, Poisson, negative binomial) for the dependent variables to test for effects of model factors using glm or glm.nb in R version 3.3.3 (R_Core_Team, 2014). The negative binomial distribution gave the best fit for the number of parasitized aphids and the number of adult progeny. The normal distribution fit well for sex ratio, and the gamma distribution fit well for weights of females and males. For each of the chosen distributions, the residual deviance divided by error degrees of freedom did not differ significantly from 1, which should be the case for distributions that fit well (Littell et al., 1996). For the figures, least square means and confidence intervals were converted back to the original distributions, using the appropriate inverse link functions. Because the confidence intervals were often asymmetrical, where appropriate in the text, we report means and asymptotic 95% confidence levels using the following format: mean [lower confidence level – upper confidence level]. Data are archived on the Ag Data Commons website (data.nal.usda.gov; DOI 10.15482/USDA.ADC/1356635). 3. Results As hypothesized, H. defensa infection of Ap. craccivora did not affect the number of aphids parasitized by Aphelinus atriplicis or Aphelinus glycinis (Fig. 1a, Table 1). However, H. defensa

Hopper et al. Defensive endosymbionts and aphelinid parasitoids of aphids

10

infection of Ap. craccivora increased the fitness of Aphelinus glycinis, which produced 60 percent more adult progeny on infected aphids than on uninfected aphids (Fig. 1, Table 1). Infection of Ac. pisum by H. defensa did not affect the number of mummies or adult progeny produced by Aphelinus atriplicis (Fig. 1b, Table 1c). Although these experiments were not designed to test the effects of Hamiltonella For Aphelinus glycinis, 42 females were recovered out of 48 tested (2 were not recovered on uninfected aphids and 4 were not recovered on infected aphids). For Aphelinus atriplicis on Ap. craccivora, 38 females were recovered out of 48 tested (6 were not recovered on uninfected aphids and 4 were not recovered on infected aphids). Given that the females were 1-5 days old at the beginning of the experiments, these survival rates exceed expectations from experiments on longevity of Aphelinus atriplicis, in which adults of this parasitoid had a median longevity of one week (KRH, unpublished data). However, for A. atriplicis on Ac. pisum, only 29 females were recovered out of 48 tested, and only 9 out of 24 females were recovered on infected aphids so that Aphelinus atriplicis adults appeared to have survived significantly less well on infected Ac. pisum (χ2 = 8.7, P = 0.003). These parasitoid species differed in numbers of Ap. craccivora mummified (model deviance = 9.7, df = 1; residual deviance = 89.1, df = 78; P = 0.002), with Aphelinus glycinis parasitizing more than twice as many aphids (47 [32-68]) than Aphelinus atriplicis (19 [13-28]) and producing twice as many adult progeny on this aphid species (32 [23-45]) than Aphelinus atriplicis (19 [13-27]). Given that Aphelinus glycinis is a specialist on aphid species in the genus Aphis and Aphelinus atriplicis has a very broad host range, this result is consistent with the pattern that specialists often do better on their hosts than generalists on the same hosts. The numbers of mummified Ap. craccivora and adult progeny varied among females of both parasitoid species (Fig. 2), but the variation was the same magnitude for infected and uninfected aphids so it is unlikely to result from genetic variation in parasitoid susceptibility to H. defensa or APSE. Thus this variation cannot results from genetic variation in susceptibility to H. defensa and its APSE phage, as has been found in other systems (Rouchet and Vorburger, 2012). The coefficient of variation (standard deviation/mean) in number of mummified Ap. craccivora was over twice as large for Aphelinus atriplicis than for Aphelinus glycinis (1.2 versus 0.5, respectively), as was the coefficient for variation in number of adult progeny (1.3 versus 0.6, respectively). This was in large part because Aphelinus atriplicis variances were significantly

Hopper et al. Defensive endosymbionts and aphelinid parasitoids of aphids

11

larger than Aphelinus glycinis variances (for number of mummified aphids: t = 2.5, P <0.007; for number of adult progeny: t = 2.5, P <0.01). About half of Aphelinus atriplicis females produced <10 mummified aphids (53%) and adult progeny (58%), whereas other females produced the number of mummies and adults expected if they had used their full egg loads. This suggests segregating variation in Aphelinus atriplicis for parasitism of Ap. craccivora, independent of infection with H. defensa. Such variation has been observed in parasitoids with broad host ranges (Hopper et al., 1993). Sex ratios were not affected by H. defensa infection (Fig. 3a; Table 1). However, sex ratios of Aphelinus atriplicis on Ac. pisum and Ap. craccivora and A. glycinis on Ap. craccivora were female biased (χ2 = 33.7, P < 0.0001; χ2 = 25.2, P < 0.0001; χ2 = 49.3, P < 0.0001, respectively). Female biased sex ratios are common in Aphelinus, perhaps because they attack gregarious hosts so that local mate competition leads to fewer males per aphid colony. In any case, females of both species made the same allocation of fertilized (female) versus unfertilized (male) eggs, regardless of H. defensa infection. Neither female nor male masses of Aphelinus atriplicis were affected by H. defensa infection (Fig. 3B,C; Table 1). For Aphelinus glycinis the pattern was more complex: male mass was not affected by H. defensa infection, but female mass was 10 percent greater on Ap. craccivora infected with H. defensa than on uninfected aphids (Fig. 3B,C; Table 1A1A). 4. Discussion The heritable bacterial symbiont H. defensa is known to confer resistance in several aphid species to certain braconid and aphelinid parasitoids of aphids (Asplen et al., 2014; Cayetano and Vorburger, 2015; Martinez et al., 2016),. For Aphelinus atriplicis, we found no significant effects of H. defensa infection on the numbers of parasitized aphids or the numbers of adult progeny. However, poor recovery of Aphelinus atriplicis females at the end of experiment from the treatment with infected Ac. pisum suggests that exposure to H. defensa or APSE while host feeding may cause adult mortality. Unexpectedly, in the second species examined, Aphelinus glycinis, infection with H. defensa actually increased the number of adult progeny and the size of female progeny. Thus, defence against Aphelinus glycinis was compromised by the presence of H. defensa at least to the extent that parasitoid fitness was increased. While there is an increasing awareness of the roles of heritable symbionts in conferring protection against natural enemies

Hopper et al. Defensive endosymbionts and aphelinid parasitoids of aphids

12

(Oliver et al. 2014), underappreciated is the notion that symbiont infection can also decrease host performance in particular antagonistic interactions. In another recent example of this phenomenon, Wolbachia, which is now known to confer protection against a range of viral pathogens (Hamilton and Perlman, 2013), also increases the susceptibility of the African armyworm, Spodoptera exempta (Walker) to infection by baculoviruses (Graham et al., 2012). At least two hypotheses could explain why H. defensa has not evolved to provide resistance against parasitism by these Aphelinus species. First, these Aphelinus species may rarely parasitize these aphids so that there has been little selection for H. defensa/APSE to protect against these parasitoids. There are no field data on parasitism rates of Ap. craccivora or Ac. pisum by Aphelinus glycinis or Aphelinus atriplicis. However, field surveys tend to show lower levels of parasitism of aphids by aphelinids than by braconids (Pons et al., 2011). Second, the barrier of heavily chorionated eggs in Aphelinus may be difficult to overcome so that even if selection were intense, there may be no segregating variation in H. defensa to provide a response. APSE-4, including the specific haplotype harbored by H. defensa in Ap. craccivora in the experiments reported here, has shiga-like toxin homologs (including a possible functional correlate of stxA, the cytotoxic alpha unit) which can destroy cells (Degnan and Moran, 2008). APSE-3 found in the Ac. craccivora line used here has a gene coding for a homology of the YD-repeat toxin gene (Degnan and Moran, 2008). However, toxins must contact the cells, which may not happen in Aphelinus before egg hatch. These toxins are >300 amino acids long (MW > 33,000) and thus may not pass through the Aphelinus chorion given that neutral red dye (MW = 289) does not pass. Given that all Aphelinus have anhydropic, heavily chorionated eggs that take a long time to hatch and may not absorb toxins during embryogenesis, these Aphelinus species may not be affected by H. defensa/APSE that we tested, if the mechanism of aphid defense involves toxins that affect parasitoid eggs. This has been suggested for Praon pequodorum (Hymenoptera: Braconidae) which is not affected by H. defense/APSE(strains 2 and 3) in Ac. pisum and has heavily chorionated eggs that take three times longer to rupture than Aphidius ervi eggs that are killed in Ac. pisum infected with H. defense/APSE (Martinez et al., 2016). One caveat is that we only examined two H. defensa/APSE strains in two aphid species, and it is possible that other strains, using different protective mechanisms, provide protection against Aphelinus parasitoids. H. defensa strain diversity appears to be very limited in North American A. craccivora (Dykstra et al., 2014), suggesting widespread susceptibility to Aphelinus in this aphid. Ap. craccivora shows low

Hopper et al. Defensive endosymbionts and aphelinid parasitoids of aphids

13

levels of genetic variation (Angelella, 2015), and although aphids from locust versus alfalfa differ in performance on these plants, the differences are associated with infection by Arsenophonus (Wagner et al., 2015). However, in Ac. pisum, there are numerous H. defensa/APSE strains that confer varying levels of protection against Ap. ervi (Oliver et al. 2014) and some do provide protection against Aphelinus abdominalis (McLean and Godfray, 2015). The differences in susceptibility to H. defensa among braconids (Asplen et al., 2014; Cayetano and Vorburger, 2015; Martinez et al., 2016; Vorburger et al., 2009) must have a different explanation than that given above. Braconids have hydropic eggs that absorb materials from the host hemolymph so braconid embryos would be exposed to H. defensa/APSE toxins. In research on Ac. pisum infected with H. defensa, the aphids from L. pedunculatus that were protected from parasitism by Aphelinus abdominalis had H. defensa with an APSE whose sequences for two genes were not clearly associated with those from previously reported sequences from other APSE (McLean and Godfray, 2015). Perhaps this APSE strain produces a toxin that affects the larval stage of parasitoids. Neither of the above hypotheses can explain why Aphelinus glycinis produced more adult progeny and larger female progeny on aphids infected with H. defensa than on uninfected aphids. Dykstra et al. (2014) showed no difference in weight between infected versus uninfected aphids, using the same Ap. craccivora, the same strain of H. defensa, and the same APSE type used in the experiments reported here. Thus the difference in Aphelinus glycinis size on infected and uninfected aphids cannot result simply from a difference in the size of infected versus uninfected aphids. However, infection with endosymbionts, including H. defensa, can influence the pool of available metabolites circulating in the aphid hemocoel (Burke et al., 2010), thus one possible explanation is that the aphid nutritional profile is altered in a way that specifically benefits A. glycinis. Alternatively, many aphid secondary symbionts (although not specifically H. defensa) confer protection against microbial pathogens (Lukasik et al., 2013; Oliver et al., 2010), thus H. defensa may eliminate an Aphelinus glycinis competitor. A third explanation may be found in the complex interactions between parasitoids, aphid nutritional and protective bacterial symbionts, and bacteriophages. Parasitism by Aphidius ervi increased the titer of Buchnera aphidicola and APSE but decreased the titer of H. defensa in Acyrthosiphon pisum lines susceptible to parasitism (Martinez et al., 2014). Aphelinus glycinis may increase host quality by changing the titer of

Hopper et al. Defensive endosymbionts and aphelinid parasitoids of aphids

14

nutritional versus protective bacteria in such a way that aphids infected with H. defensa can be made more suitable for parasitoid development than uninfected aphids. Heritable symbionts, which are widespread among insects, are now associated with a wide range of protective roles, including protection against parasitoids in several aphid species and Drosophila (Oliver et al., 2014). Building on these initial reports are findings that protective symbioses are often dynamic, with symbiont or strain specificity to particular natural enemy species or genotypes (Asplen et al., 2014; Rouchet and Vorburger, 2014). Adding to this complexity, we found that not only does H. defensa not confer protection against two aphelinid parasitoids, but it actually increases the fitness of one aphelinid species. Although negative effects on host fitness in the presence of natural enemies can constrain the spread of heritable symbionts with host populations (Dykstra et al., 2014; Oliver et al., 2014), this increase in parasitoid fitness may not affect the frequency of defensive symbionts because the parasitoids appear to attack infected and uninfected aphids with equal frequency. Finding symbiont specificity to particular natural enemies informs the sustainable management of pests in agricultural systems. While braconid susceptibility to aphid secondary symbionts and their phages has led to suggestions for artificial selection to avoid these consequences (Vorburger, 2014), our results are good news for biological control using Aphelinus species because such measures may not always be needed. Acknowledgements This research was funded by the United States Department of Agriculture, Agriculture Research Service and by Award 2009-02237 from the United States Department of Agriculture, Agriculture and Food Research Initiative to KRH KMO JAW MKA GEH.

Hopper et al. Defensive endosymbionts and aphelinid parasitoids of aphids

15

References Angelella, G.M., 2015. Tracking plant virus infections through multiple dimensions: A search for sources of nonpersistent virus vectors and reservoirs at local and regional scales. Ph.D. dissertation. Purdue University, West Lafayette, Indiana. Asplen, M.K., Bano, N., Brady, C.M., Desneux, N., Hopper, K.R., Malouines, C., Oliver, K.M., White, J.A., Heimpel, G.E., 2014. Specialisation of bacterial endosymbionts that protect aphids from parasitoids. Ecological Entomology 39, 736-739. Bai, B., Mackauer, M., 1990. Host discrimination by the aphid parasitoid Aphelinus asychis (Hymenoptera: Aphelinidae): When superparasitism is not adaptive. Canadian Entomologist 122, 363-372. Baumann, P., Baumann, L., Lai, C.Y., Roubakhsh, D., Moran, N.A., Clark, M.A., 1995. Genetics, physiology, and evolutionary relationships of the genus Buchnera - intracellular symbionts of aphids. Annual Review of Microbiology 49, 55-94. Blackman, R.L., Eastop, V.F., 2006. Aphids of the world's herbaceous plants and shrubs. Volume 2: the aphids. Wiley & Sons, Ltd., West Sussex, U.K. Brady, C.M., Asplen, M.K., Desneux, N., Heimpel, G.E., Hopper, K.R., Linnen, C.R., Oliver, K.M., Wulff, J.A., White, J.A., 2014. Worldwide populations of the aphid Aphis craccivora are infected with diverse facultative bacterial symbionts. Microbial Ecology 67, 195-204. Brady, C.M., White, J.A., 2013. Cowpea aphid (Aphis craccivora) associated with different host plants has different facultative endosymbionts. Ecological Entomology 38, 433-437. Burke, G., Fiehn, O., Moran, N., 2010. Effects of facultative symbionts and heat stress on the metabolome of pea aphids. Isme Journal 4, 242-252. Cayetano, L., Vorburger, C., 2015. Symbiont-conferred protection against Hymenopteran parasitoids in aphids: how general is it? Ecological Entomology 40, 85-93. Christiansen-Weniger, P., 1994. Morphological observations on the preimaginal stages of Aphelinus varipes (Hym., Aphelinidae) and the effects of this parasitoid on the aphid Rhopalosiphum padi (Hom., Aphididae). Entomophaga 39, 267-274. Degnan, P.H., Moran, N.A., 2008. Diverse phage-encoded toxins in a protective insect endosymbiont. Applied and Environmental Microbiology 74, 6782-6791.

Hopper et al. Defensive endosymbionts and aphelinid parasitoids of aphids

16

Desneux, N., Barta, R.J., Hoelmer, K.A., Hopper, K.R., Heimpel, G.E., 2009. Multifaceted determinants of host specificity in an aphid parasitoid. Oecologia 160, 387-398. Douglas, A.E., 1998. Nutritional interactions in insect-microbial symbioses: Aphids and their symbiotic bacteria Buchnera. Annual Review of Entomology 43, 17-37. Dykstra, H.R., Weldon, S.R., Martinez, A.J., White, J.A., Hopper, K.R., Heimpel, G.E., Asplen, M.K., Oliver, K.M., 2014. Factors limiting the spread of the protective symbiont Hamiltonella defensa in the aphid Aphis craccivora. Applied and Environmental Microbiology 80, 5818-5827. Ferrari, J., Darby, A.C., Daniell, T.J., Godfray, H.C.J., Douglas, A.E., 2004. Linking the bacterial community in pea aphids with host-plant use and natural enemy resistance. Ecological Entomology 29, 60-65. Graham, R.I., Grzywacz, D., Mushobozi, W.L., Wilson, K., 2012. Wolbachia in a major African crop pest increases susceptibility to viral disease rather than protects. Ecology Letters 15, 993-1000. Hagen, K.S., VanDenBosch, R., 1968. Impact of Pathogens, Parasites, and Predators on Aphids. Annual Review of Entomology 13, 325-384. Hamilton, P.T., Perlman, S.J., 2013. Host defense via symbiosis in Drosophila. Plos Pathogens 9. Heimpel, G.E., Ragsdale, D.W., Venette, R., Hopper, K.R., O'Neil, R.J., Rutledge, C.E., Wu, Z.S., 2004. Prospects for importation biological control of the soybean aphid: Anticipating potential costs and benefits. Annals of the Entomological Society of America 97, 249-258. Henry, L.M., Maiden, M.C.J., Ferrari, J., Godfray, H.C.J., 2015. Insect life history and the evolution of bacterial mutualism. Ecology Letters 18, 516-525. Heraty, J.H., Woolley, J.B., Hopper, K.R., Hawks, D.L., Kim, J.-W., Buffington, M., 2007. Molecular phylogenetics and reproductive incompatibility in a complex of cryptic species of aphid parasitoids. Molecular Phylogenetics and Evolution 45, 480-493. Hopper, K.R., Coutinot, D., Chen, K., Mercadier, G., Halbert, S.E., Kazmer, D.J., Miller, R.H., Pike, K.S., Tanigoshi, L.K., 1998. Exploration for natural enemies to control Diuraphis noxia in the United States. In: Quissenbery, S.S., Peairs, F.B., Eds.), A response model for an introduced pest-- The Russian wheat aphid. Entomological Society of America, Lanham, MD, pp. 166-182. Hopper, K.R., Prager, S.M., Heimpel, G.E., 2013. Is parasitoid acceptance of different host species dynamic? Functional Ecology 27, 1201-1211.

Hopper et al. Defensive endosymbionts and aphelinid parasitoids of aphids

17

Hopper, K.R., Roush, R.T., Powell, W., 1993. Management of genetics of biological-control introductions. Annual Review of Entomology 38, 27-51. Hopper, K.R., Woolley, J.B., Hoelmer, K., Wu, K., Qiao, G., Lee, S., 2012. An identification key to species in the mali complex of Aphelinus (Hymenoptera, Chalcidoidea) with descriptions of three new species. Journal of Hymenoptera Research 26, 73-96. Kurdjumov, N.V., 1913. Notes on European species of the genus Aphelinus Dalm. (Hymenoptera, Chalcidodea), parasitic upon the plant-lice. Russian Review of Entomology 13, 266-270. Littell, R.C., Milliken, G.A., Stroup, W.W., Wolfinger, R.D., 1996. Generalized linear mixed models, Chapter 10. SAS System for Mixed Models, pp. 423-460. Lu, W.-N., Kuo, M.-H., 2008. Life table and heat tolerance of Acyrthosiphon pisum (Hemiptera: Aphididae) in subtropical Taiwan. Entomological Science 11, 273-279. Lukasik, P., van Asch, M., Guo, H.F., Ferrari, J., Godfray, H.C.J., 2013. Unrelated facultative endosymbionts protect aphids against a fungal pathogen. Ecology Letters 16, 214-218. Martinez, A.J., Kim, K.L., Harmon, J.P., Oliver, K.M., 2016. Specificity of multi-modal aphid defenses against two rival parasitoids. Plos One 11, e0154670. Martinez, A.J., Weldon, S.R., Oliver, K.M., 2014. Effects of parasitism on aphid nutritional and protective symbioses. Molecular Ecology 23, 1594-1607. McLean, A.H.C., Godfray, H.C.J., 2015. Evidence for specificity in symbiont-conferred protection against parasitoids. Proceedings of the Royal Society B-Biological Sciences 282, 20150977. McLean, A.H.C., van Asch, M., Ferrari, J., Godfray, H.C.J., 2011. Effects of bacterial secondary symbionts on host plant use in pea aphids. Proceedings of the Royal Society B-Biological Sciences 278, 760-766. Montllor, C.B., Maxmen, A., Purcell, A.H., 2002. Facultative bacterial endosymbionts benefit pea aphids Acyrthosiphon pisum under heat stress. Ecological Entomology 27, 189-195. Moran, N.A., 2001. The coevolution of bacterial endosymbionts and phloem-feeding insects. Annals of the Missouri Botanical Garden 88, 35-44. Moran, N.A., Degnan, P.H., Santos, S.R., Dunbar, H.E., Ochman, H., 2005. The players in a mutualistic symbiosis: Insects, bacteria, viruses, and virulence genes. Proceedings of the National Academy of Sciences of the United States of America 102, 16919-16926. Moran, N.A., McCutcheon, J.P., Nakabachi, A., 2008. Genomics and evolution of heritable bacterial symbionts. Annual Review of Genetics 42, 165-190.

Hopper et al. Defensive endosymbionts and aphelinid parasitoids of aphids

18

O'Neill, S.L., Hoffman, A., Werren, J.H., 1997. Influential passengers: inherited microorganisms and arthropod reproduction. Influential passengers: inherited microorganisms and arthropod reproduction. Oliver, K.M., Campos, J., Moran, N.A., Hunter, M.S., 2008. Population dynamics of defensive symbionts in aphids. Proceedings of the Royal Society B-Biological Sciences 275, 293-299. Oliver, K.M., Degnan, P.H., Burke, G.R., Moran, N.A., 2010. Facultative symbionts in aphids and the horizontal transfer of ecologically important traits. Annual Review of Entomology, pp. 247-266. Oliver, K.M., Degnan, P.H., Hunter, M.S., Moran, N.A., 2009. Bacteriophages encode factors required for protection in a symbiotic mutualism. Science 325, 992-994. Oliver, K.M., Martinez, A.J., 2014. How resident microbes modulate ecologically-important traits of insects. Current Opinion in Insect Science, DOI:10.1016/j.cois.2014.1008.1001. Oliver, K.M., Moran, N.A., Hunter, M.S., 2005. Variation in resistance to parasitism in aphids is due to symbionts not host genotype. Proceedings of the National Academy of Sciences of the United States of America 102, 12795-12800. Oliver, K.M., Russell, J.A., Moran, N.A., Hunter, M.S., 2003. Facultative bacterial symbionts in aphids confer resistance to parasitic wasps. Proceedings of the National Academy of Sciences of the United States of America 100, 1803-1807. Oliver, K.M., Smith, A.H., Russell, J.A., 2014. Defensive symbiosis in the real world -advancing ecological studies of heritable, protective bacteria in aphids and beyond. Functional Ecology 28, 341-355. Papaj, D.R., 2000. Ovarian dynamics and host use. Annual Review of Entomology 45, 423-448. Parker, B.J., Spragg, C.J., Altincicek, B., Gerardo, N.M., 2013. Symbiont-mediated protection against fungal pathogens in pea aphids: a role for pathogen specificity? Applied and Environmental Microbiology 79, 2455-2458. Perng, J.J., Liu, Y.C., 2002. Age-specific survival and fecundity and their effects on the intrinsic rate of increase of Aphelinus gossypii (Hym., Aphelinidae), a parasitoid of Aphis gossypii (Hom., Aphididae). Journal of Applied Entomology 126, 484-489. Pons, X., Lumbierres, B., Antoni, R., Stary, P., 2011. Parasitoid complex of alfalfa aphids in an IPM intensive crop system in northern Catalonia. Journal of Pest Science 84, 437-445.

Hopper et al. Defensive endosymbionts and aphelinid parasitoids of aphids

19

R_Core_Team, 2014. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. Rohne, O., 2002. Effect of temperature and host stage on performance of Aphelinus varipes Förster (Hym., Aphelinidae) parasitizing the cotton aphid, Aphis gossypii Glover (Hom., Aphididae). Journal of Applied Entomology 126, 572-576. Rouchet, R., Vorburger, C., 2012. Strong specificity in the interaction between parasitoids and symbiont-protected hosts. Journal of Evolutionary Biology 25, 2369-2375. Rouchet, R., Vorburger, C., 2014. Experimental evolution of parasitoid infectivity on symbiont-protected hosts leads to the emergence of genotype specificity. Evolution 68, 1607-1616. Russell, J.A., Moran, N.A., 2006. Costs and benefits of symbiont infection in aphids: variation among symbionts and across temperatures. Proceedings of the Royal Society B-Biological Sciences 273, 603-610. Russell, J.A., Weldon, S., Smith, A.H., Kim, K.L., Hu, Y., Lukasik, P., Doll, S., Anastopoulos, I., Novin, M., Oliver, K.M., 2013. Uncovering symbiont-driven genetic diversity across North American pea aphids. Molecular Ecology 22, 2045-2059. Sakurai, M., Koga, R., Tsuchida, T., Meng, X.Y., Fukatsu, T., 2005. Rickettsia symbiont in the pea aphid Acyrthosiphon pisum: Novel cellular tropism, effect on host fitness, and interaction with the essential symbiont Buchnera. Applied and Environmental Microbioology 71, 4069-4075. Scarborough, C.L., Ferrari, J., Godfray, H.C.J., 2005. Aphid protected from pathogen by endosymbiont. Science 310, 1781-1781. Schmid, M., Sieber, R., Zimmermann, Y.S., Vorburger, C., 2012. Development, specificity and sublethal effects of symbiont-conferred resistance to parasitoids in aphids. Functional Ecology 26, 207-215. Simon, J.-C., Boutin, S., Tsuchida, T., Koga, R., Le Gallic, J.-F., Frantz, A., Outreman, Y., Fukatsu, T., 2011. Facultative symbiont infections affect aphid reproduction. Plos One 6, e21831. Soffan, A., Aldawood, A.S., 2014. Biology and demographic growth parameters of cowpea aphid ( Aphis craccivora ) on faba bean ( Vicia faba ) cultivars. Journal Of Insect Science 14, 120-120.

Hopper et al. Defensive endosymbionts and aphelinid parasitoids of aphids

20

Stouthamer, R., Breeuwer, J.A.J., Hurst, G.D.D., 1999. Wolbachia pipientis: Microbial manipulator of arthropod reproduction. Annual Review of Microbiology 53, 71-102. Tsuchida, T., Koga, R., Fukatsu, T., 2004. Host plant specialization governed by facultative symbiont. Science 303, 1989-1989. Viggiani, G., 1984. Bionomics of the Aphelinidae. Annual Review of Entomology 29 257-276. Vorburger, C., 2014. The evolutionary ecology of symbiont-conferred resistance to parasitoids in aphids. Insect Science 21, 251-264. Vorburger, C., Gehrer, L., Rodriguez, P., 2010. A strain of the bacterial symbiont Regiella insecticola protects aphids against parasitoids. Biology Letters 6, 109-111. Vorburger, C., Gouskov, A., 2011. Only helpful when required: a longevity cost of harbouring defensive symbionts. Journal of Evolutionary Biology 24, 1611-1617. Vorburger, C., Sandrock, C., Gouskov, A., Castaneda, L.E., Ferrari, J., 2009. Genotypic variation and the role of defensive endosymbionts in an all-parthenogenetic host-parasitoid interaction. Evolution 63, 1439-1450. Wagner, S.M., Martinez, A.J., Ruan, Y.M., Kim, K.L., Lenhart, P.A., Dehnel, A.C., Oliver, K.M., White, J.A., 2015. Facultative endosymbionts mediate dietary breadth in a polyphagous herbivore. Functional Ecology 29, 1402-1410. Wu, Z.S., Heimpel, G.E., 2007. Dynamic egg maturation strategies in an aphid parasitoid. Physiological Entomology 32, 143-149. Wulff, J.A., Buckman, K.A., Wu, K.M., Heimpel, G.E., White, J.A., 2013. The endosymbiont arsenophonus is widespread in soybean aphid, Aphis glycines, but does not provide protection from parasitoids or a fungal pathogen. Plos One 8, e62145.

Hopper et al. Defensive endosymbionts and aphelinid parasitoids of aphids

21

Figure captions Fig 1. Effects of infection of aphids by Hamiltonella defensa and its phage APSE on parasitism. (a) numbers of mummified aphids and (b) numbers of adult parasitoid progeny of the parasitoids Aphelinus atriplicis and Aphelinus glycinis. Asterisks between means indicate signficant difference (** P = 0.01). Error bars are asymptotic 95% confidence intervals. Fig 2. Distributions of numbers of mummified aphids and numbers of adult parasitoid progeny. (a) Aphelinus atriplicis and (b) Aphelinus glycinis with and without infection of Aphis craccivora by Hamiltonella defensa and its phage APSE-4. Fig 3. Effects of infection of aphids by Hamiltonella defensa and its phage APSE on fitness components of Aphelinus atriplicis and Aphelinus glycinis. (a) progeny sex ratio, (b) mass of female progeny, (c) mass of male progeny. Asterisk between means indicates signficant difference (* P = 0.05). Error bars are asymptotic 95% confidence intervals.

Table 1. Analysis of deviance for the effect of infection of aphids by Hamiltonella defensa on number of parasitized (mummified) and parasitoid fitness (see text for details of statistical analyses). (a) Aphelinus glycinis on A. craccivora Model Factor df Deviance number of parasitized 1 2.6 aphids number of adult 1 6.2 parasitoids sex ratio 1 0.5 female mass 1 2.9 male mass 1 0.2 (b) Aphelinus atriplicis on Aphis craccivora Model Factor df Deviance number of parasitized 1 0.6 aphids number of adult 1 0.6 parasitoids sex ratio 1 0.3 female mass 1 0.01 male mass 1 0.2

Residual df Deviance

P

40

44.6

0.11

40

44.9

0.01

43 43 41

20.8 32.3 27.7

0.30 0.04 0.58

Residual df Deviance

P

36

43.3

0.45

36

44.3

0.43

29 26 28

10.6 25.7 21.7

0.36 0.93 0.59

(c) Aphelinus atriplicis on Acyrthosiphon pisum Model Residual Factor df Deviance df Deviance number of parasitized 1 2.3 27 33.3 aphids number of adult 1 1.3 27 34.0 parasitoids sex ratio 1 0.12 34 6.0 female mass 1 0.002 33 9.3 male mass 1 0.0001 29 20.4

P 0.13 0.25 0.41 0.94 0.99

Fig1

(a) Aphis craccivora

60

mummified aphids

50 40

Acyrthosiphon pisum

30 20 10 0

(b) 60

adult parasitoids

50 40

H− H+

**

30 20 10 0

Aphelinus atriplicis

Aphelinus atriplicis

parasitoid species

Aphelinus glycinis

Fig2

frequency

(a) Aphelinus atriplicis

(b) Aphelinus glycinis

12

5

10

4 H− H+

8

3

6 2

4

1

2 0

0

10

20

30

40

50

60

70

80

90 100

0

0

10

number of mummified aphids

30

40

50

60

70

80

90 100

number of mummified aphids

12

8 7

10

frequency

20

6 8

5

6

4 3

4

2 2 0

1 0

10

20

30

40

50

60

70

80

number of adult parasitoids

90 100

0

0

10

20

30

40

50

60

70

80

number of adult parasitoids

90 100

(a)

sex ratio (proportion males)

Fig3

0.6 Acyrthosiphon pisum

Aphis craccivora

0.5 0.4 0.3 0.2 0.1 0.0

female mass (ug)

(b) 18 16 14 12 10 8 6 4 2 0

H− H+

*

(c) 16

male mass (ug)

14 12 10 8 6 4 2 0

Aphelinus atriplicis

Aphelinus atriplicis

Aphelinus glycinis

parasitoid species