Symbiont‐conferred reproduction and fitness benefits can favour their host occurrence

Abstract Double infections of Wolbachia and Spiroplasma are frequent in natural populations of Tetranychus truncatus, a polyphagous mite species that has been a dominant species in China since 2009. However, little is known about the causes and ecological importance of such coexistences. In this study, we established T. truncatus strains with different infection types and then inferred the impact of the two endosymbionts on host reproduction and fitness. Double infection induced cytoplasmic incompatibility, which was demonstrated by reduction in egg hatchability of incompatible crosses. However, doubly infected females produced more eggs relative to other strains. Wolbachia and Spiroplasma did not affect host survival, whereas doubly infected females and males developed faster than other strains. Such reproduction and fitness benefits provided by double infections may be associated with the lower densities of each symbiont, and the quantitative results also confirmed competition between Wolbachia and Spiroplasma in doubly infected females. These symbiont‐conferred beneficial effects maintain stable prevalence of the symbionts and also help drive T. truncatus outbreaks in combination with other environmental factors.

Endosymbionts have traditionally been grouped into primary and secondary symbionts. Primary symbionts are mutualists that tend to have a nutritional value, which is the case in the typical example of aphids and Buchnera aphidicola (Baumann, 2005). In contrast, secondary symbionts are often facultative symbionts from the host's perspective and have a much broader array of effects. Most symbionts are exclusively or predominantly vertically transmitted; however, the often-found incongruence between host and symbiont phylogenies indicates that these infections are generally short-lived on macroevolutionary timescales. In theory, once a symbiont invades a novel host, it must immediately be capable of replication and colonization of the female germ line. Hurst and Darby (2009) also speculated that high rates of horizontal transmission, reproductive manipulation, and direct fitness benefits to the host will favour maintenance of secondary symbionts within the host population. The high prevalence of several secondary symbionts without obvious parasitic phenotypes can sometimes be linked to such direct fitness effects; for example, increased performance and sex ratio bias of infected whiteflies sufficiently explain the spread of Rickettsia symbionts (Himler et al., 2011).
Among the secondary symbionts, Wolbachia is by far the most widely documented and infects a large proportion of arthropod species (Werren, Baldo, & Clark, 2008). Wolbachia is notorious for its reproductive parasitism, which ensures its spread despite lowering host fitness. However, even for reproductive parasites, it can be beneficial to enhance host fitness. Indeed, recent years have seen a growing body of evidence on Wolbachia-associated fitness benefits (reviewed in Zug & Hammerstein, 2015). Spiroplasma is a wall-less bacterium associated with diverse arthropods and plants in which it has commensal, pathogenic, or mutualistic effects (Cisak et al., 2015).
In Drosophila neotestacea, recent spread and mutualism account for the positive association between Wolbachia and Spiroplasma (Jaenike et al., 2010). However, the ecological and evolutionary importance of such coexistences remains underexplored.
It is mainly distributed in East and Southeast Asia, and there is report of its presence in the USA (Migeon & Dorkeld, 2006-2017  , and double infections of Wolbachia and Cardinium are frequent in some natural populations (Zhao, Zhang, & Hong, 2013).
Recently, high prevalence of Spiroplasma was identified in T. truncatus, and there was a substantial positive association between Wolbachia and Spiroplasma in some populations (Zhang, Chen, Yang, Qiao, & Hong, 2016).
Given the high rate of co-transmission of these two symbionts and recent T. truncatus outbreaks, we investigated the mutualistic phenotypes of these two symbionts within T. truncatus. We established different T. truncatus strains with the same host genetic background and investigated the effect of Wolbachia alone or in combination with Spiroplasma on T. truncatus reproduction and fitness. Furthermore, the densities of the two symbionts were monitored and compared during host development to infer their competition relationship.

| Wolbachia and Spiroplasma prevalence in T. truncatus
Adult T. truncatus were collected from six populations in northern China ( Figure 1). Individual mites were either immediately analyzed in the laboratory to assess infection frequency (see below) or individual females were reared as isofemale lines on leaves of common bean (Phaseolus vulgaris L.) placed on a water-saturated sponge mat in Petri dishes (dia. 9) at 25 ± 1°C, 60% humidity, and under L16-D8 conditions. Subsequently, 16-24 field-collected individual mites were screened for infection with Wolbachia and/or Spiroplasma. DNA was extracted from single mites using a genomic DNA extraction kit (TaKaRa, Dalian, China). All DNA samples were first PCR-screened for the mitochondrial gene COI as a control for quality (Navajas, Gutierrez, Lagnel, & Boursot, 1996). Wolbachia and Spiroplasma presence was detected using PCR amplifications of wsp and 16S rRNA, respectively.

| Crossing experiments
To test whether Wolbachia and Spiroplasma cooperate to cause cytoplasmic incompatibility (CI) in T. truncatus, crossing experiments were performed (listed as male × female, Figure 2). Single females in the teleiochrysalis stage (the last developmental stage before adult emergence) were placed with a 1-day-old adult virgin male from the same culture on the same leaf disk. Each cross used 16 leaf disks. Males were discarded 2 days after the females reached adulthood. The mated females were allowed to oviposit for 5 days. The eggs on the leaf disks were checked daily to determine hatchability and the sex ratio (% females). Prior to analyses, data were first tested for normality (Kolmogorov-Smirnov test, SPSS 17.0) and homogeneity of group variances (Levene's test, SPSS 17.0). Which possible, square root, logarithmic or arcsine transformations were performed to attain normality and homogeneity of variance. In this study, log transformation was used for the number of eggs laid per female, and arcsine square root transformation was used for egg hatchability and sex ratio. Data were then analyzed with one-way analysis of variance (ANOVA), and the means were compared using the Tukey HSD test (SPSS 17.0).
The male-killing phenotype of Spiroplasma in arthropods results in the production of female-biased offspring sex ratios. Here, the sex ratio data of the four T. truncatus strains were analyzed to determine if male-killing occurred.

| Survival assay
To test for potential effects of symbionts on host survival, we measured the survival of different T. truncatus strains by placing eight virgin females and eight virgin males from the same strains on the same leaf.
Three leaves were used for each line, and females were transferred to F I G U R E 2 Results of crosses between different Tetranychus truncatus strains. Number of eggs (a), egg hatch percent (b), and female proportion of offspring (c) are shown. Results are mean ± SEM, and a and b represent statistically different groups (Tukey-HSD test, p < .05) fresh leaves every 3 days. The number of dead females was recorded daily. Survivor curves for individual hosts were compared using the Kaplan-Meier method and log-rank test (Dobson, Rattanadechakul, & Marsland, 2004).

| Development assay
The effect of symbionts on mite development was assessed. Thirty virgin and 30 mated females (which produce male and female offspring, respectively) were placed on a leaf disk and allowed to lay eggs for 8 hr. The eggs were individually moved to new small leaf disks.
The small disks were monitored every 8 hr, and the stage of mite was recorded until adulthood. The development time of every stage was calculated. Log transformation was used for the total development time of each mite line, and differences were analyzed using the Mann-Whitney test (SPSS 17.0).

| Wolbachia and Spiroplasma densities
The Wolbachia and Spiroplasma densities in individual mites were estimated by real-time quantitative PCR (QPCR).
QPCR was carried out with the ABI PRISM 7300 Sequence Detection System (Applied Biosystems). We used primers designed to amplify a 141-bp fragment of wsp from Wolbachia The number of molecules in all samples is determined from the threshold cycles in the PCR based on a standard curve. Statistical analysis was performed using the Mann-Whitney U test.

| Double infections of Wolbachia and Spiroplasma are frequent in T. truncatus
Among 132 wild-caught adult mites that were collected from six sites, 16.7% and 3.8% were infected with only Wolbachia or Spiroplasma, respectively, and 79.5% were infected with both symbionts. Double infections with high prevalence were present in all screened populations ( Figure 1). The infections of Cardinium and Rickettsia were not detected in these examined samples.
However, identical Spiroplasma 16S rRNA sequences were observed among infected individuals.

| Wolbachia and Spiroplasma have diverse effects on host reproduction
To The effects of Wolbachia and Spiroplasma on host fecundity were also measured. The doubly infected females mostly showed the highest egg production, except for the crosses that involved the singly Wolbachiainfected males, in which singly Spiroplasma-infected females produced more eggs. In general, egg production of singly infected females and uninfected females did not significantly differ across all crosses.

| Wolbachia and Spiroplasma do not affect host survival
The longevities of four female lines were compared to discover the effects of infection. As shown in Figure 3, there was no significant effect of Wolbachia and Spiroplasma on female host survival (log-rank test, p = .87).
Additionally, time to adulthood was significantly shorter in uninfected than singly infected females (p < .05), and the latter two had no signifi-

| Wolbachia and Spiroplasma densities are higher in singly than doubly infected females
The Wolbachia density steadily increased (except for 2-day-old to 4-day-old adults of the singly infected strain) as adult females developed in both strains. At three developmental stages, 2, 4, and 6 days after emergence, the Wolbachia densities in singly infected females were significantly higher in doubly infected females ( Figure 5A, p < .01). A similar pattern was observed for Spiroplasma, which revealed that Spiroplasma density dynamics were significantly affected by coinfecting Wolbachia (Figure 5B, p < .001).

| DISCUSSION
In this study, we found that double infections of Wolbachia and  (Hoffmann, Hercus, & Dagher, 1998;Werren, 1997). In this respect, empirical studies have revealed that facultative symbionts can spread quickly within the host population when they provide a large fitness benefit at no or relatively little cost (Himler et al., 2011;Jaenike et al., 2010;Oliver, Campos, Moran, & Hunter, 2008). In such cases, reproductive manipulations and/or providing fitness benefits of double infections would be predicted in T. truncatus.

Wolbachia-induced CI phenotypes have been observed in some
Tetranychus species (Gotoh, Noda, & Hong, 2003;Zhao, Chen, Ge, Gotoh, & Hong, 2013;Zhu et al., 2012). However, there was no evidence of Wolbachia-mediated CI in T. truncatus, which is consistent with a previous finding (Zhao, Chen, et al., 2013). Singly Wolbachiainfected females may suffer fitness costs, such as lower hatch percent, but they also have a higher female offspring proportion. Wolbachia strains that infect T. truncatus are phylogenetically diverse , and non-CI Wolbachia were present in this study. Some likely reasons have been explained for the lack of CI (Ros & Breeuwer, 2009 (Zhao, Chen, et al., 2013), T. piercei (Zhu et al., 2012), and Bryobia sarothamni (Ros & Breeuwer, 2009), in which double infections induced strong CI and single infections (Wolbachia or Cardinium) also induced CI. Together, we infer that interactions between Wolbachia and Spiroplasma resulted in CI induction, and the underlying mechanism merits further study.
Symbionts' complex effects on their hosts result from longterm interactions and coevolution. Symbionts may manipulate their host physiological process and affect their fitness traits. wMelPop Wolbachia is well known for shortening the life of its Drosophila host (Min & Benzer, 1997). In Drosophila hydei, Spiroplasma showed enhanced survival when attacked by Leptopilina heterotoma wasps (Xie, Vilchez, & Mateos, 2010). Negative effects of symbionts on host survival were also observed (Costopoulos, Kovacs, Kamins, & Gerardo, 2014).
In this study, symbionts did not influence T. truncatus survival, but significantly regulated female fecundity. Relative to singly infected and uninfected females, doubly infected females produced more eggs. Fecundity advantage provided by symbionts is not surprising, and other studies report a similar phenomenon (Dobson et al., 2004;Himler et al., 2011;Zug & Hammerstein, 2015). These findings are concordant with a hypothesis that vertically transmitted symbionts cause their hosts to produce more infected daughters than are produced by uninfected females to facilitate their own spread (Werren & O'Neill, 1997). The infection density of symbionts is among the most important factors that influence their biological effects, such as intensity of reproductive phenotypes, level of fitness effects, and fidelity of vertical transmission. Oliver et al. (2006) proposed that over-proliferation of symbionts may be harmful to the aphid host. In T. truncatus, the high density of symbionts may result in fecundity cost to singly infected females. Additionally, QPCR estimates also indicated that numbers of each of the two symbionts in singly infected females are significantly higher than those in doubly infected females.
In addition, double infections appeared to enhance female and male mites' developmental speed. Whatever the mechanism, it is evident that more generations will be produced within the doubly infected mite strains under certain circumstances. As a result, T. truncatus occurrence will directly benefit from these manipulations, and such ma- When several symbionts are simultaneously present within the same host, symbiont-symbiont interactions can take place and substantially affect infection densities. The symbionts may compete for available resources and space in the host body or they may share the resources and habitats by regulating their own exploitation so as not to damage the whole symbiotic system. Alternatively, symbiont infection density is governed by symbiont genotype, host genotype, and environment. For example, Goto, Anbutsu, and Fukatsu (2006) revealed asymmetrical interactions between Wolbachia and Spiroplasma endosymbionts that coexist in the same insect host, D. melanogaster.
In this study, the QPCR result provided evidence of a competitive relationship between Wolbachia and Spiroplasma in doubly infected females of T. truncatus. The two symbionts are maternally transmitted and mostly located in the ovary; thus, their expansions are limited by resources and space in cases of co-infection.
In summary, double infections of Wolbachia and Spiroplasma are beneficial to their mite host by inducing CI, and increasing fecundity and developmental rate. These benefits can help maintain stable double infections and also drive T. truncatus outbreaks in combination with other environmental factors.
F I G U R E 5 Density dynamics of Wolbachia and Spiroplasma during host development in singly and doubly infected female Tetranychus truncatus. Copy numbers per ml were determined by quantitative PCR. Asterisks indicate statistically significant differences (Mann-Whitney U test, *p < .05; **p < .01; ***p < .001; NS, not significant)