Infection and transmission of the facultative endosymbiont Arsenophonus in the spider Pardosa pseudoannulata

Background The infection of insect pests and their parasitoids with the facultative endosymbiont Arsenophonus has been well studied, whereas the infection of the predators of insect pests with this symbiont is less well understood. We studied this symbiont in the spider Pardosa pseudoannulata, an important predator of rice planthoppers, collected from rice paddies in 8 geographical regions in China. Results Diagnostic PCR showed that Arsenophonus was present in 4 of the 8 sampled populations of spiders. In these 4 populations, the infection incidence was between 2.1% and 16.7% and signicantly differed between the Lishui population and the other three populations. Investigation of symbiont infection in both females and their offspring indicated that Arsenophonus can be vertically transmitted in P. pseudoannulata; the average transmission rate was approximately 25%. Remarkably, Arsenophonus was present in approximately 11% of the offspring of uninfected females, indicating that nonmaternal transmission occurs in this spider. Conclusions Populations of the spider P. pseudoannulata are infected at low rates with the symbiont Arsenophonus. This symbiont can be vertically transmitted from female spiders to offspring but may also be transmitted by male spiders or other vectors. Arsenophonus populations from different geographical regions exhibit genetic differentiation.


Background
Endosymbiotic bacteria in arthropods have received considerable attention due to their multiple roles in host ecology and evolution (Nyffeler & Benz, 1987 & Duron, 2012). Similar to Wolbachia, the genus Arsenophonus is a clade of symbiotic bacteria with a vast host distribution. The known host spectrum of Arsenophonus encompasses diverse insect groups, including parasitic wasps, triatomine bugs, psyllids, white ies, aphids, ticks, ant lions, hippoboscids, streblids, bees, lice and others, and noninsect taxa, including two plant species (Gherna et al., 1991;Chiel et al., 2007;Hansen et al., 2007;Nováková et al., 2009;Wagner et al., 2015). Arsenophonus can be vertically and horizontally transferred in parasitoids (Gherna et al., 1991;Huger et al., 1985). For example, the horizontal transfer of Arsenophonus infection in Nasonia wasps occurs after multiparasitism events in which two species of parasitoids share a pupal host (Huger et al., 1985;Duron et al., 2010). However, whether Arsenophonus infection and transmission occur in spiders is unknown. Does Arsenophonus infection occur in spiders? If so, does the infection vary among geographical populations of the host, and how does transmission occur? In this study, we attempt to answer the above questions by focusing on the spider Pardosa pseudoannulata, which is an important predator in rice paddies (Liu et al., 1999). We compared Arsenophonus infections among different geographical populations of P. pseudoannulata from rice paddies in China and investigated the transmission of Arsenophonus in this spider.

Methods
Spider collection pseudoannulata individuals were collected from 8 sites of rice paddies in China between August 2015 and October 2018 (Table 1). At each site, each individual was collected at least 30 m from other collected individuals, and the spiders were collected from at least ve rice paddies at each site. Spiders without egg bags were chosen for direct symbiont detection. Spiders carrying egg bags were chosen for the investigation of symbiont transmission. The spiders were fed on small brown planthoppers uninfected with any known bacterial symbionts (the symbiont-free planthoppers were the same as those previously described  and were reared separately in laboratory conditions at 27±1 °C with a 14:10 h (L:D) photoperiod. Numbers followed by '*' are the numbers of adult spiders with egg bags.

Detection of Arsenophonus infection in P. pseudoannulata
To compare Arsenophonus infection among different geographic populations of P. pseudoannulata, the presence of the symbiont in spiders without egg bags collected from different sites was evaluated by diagnostic PCR using speci c primers (23SF 5'CGTTTGATGAATTCATAGTCAAA3', 23SR 5'GGTCCTCCAGTTAGTGTTACCCAAC3') (Thao et al., 2000). To rule out the possibility of detecting the bacteria from spider gut, we only used spider legs to extract DNA to detect the presence of Arsenophonus. DNA from the legs of each P. pseudoannulata was extracted using the Wizard® Genomic DNA Puri cation Kit (Promega). DNA quality was examined using a spectrophotometer. Only DNA samples with an OD260/OD280 ratio ranging from 1.6 to 1.9 were used for PCR to detect symbiont species. Both positive and negative controls were used. The presence of Arsenophonus was assessed following previously described methods (Thao et al., 2000). To verify the presence of the symbionts and perform comparisons among hosts, the PCR products were isolated and sent to Sangon Biotech Co. (Shanghai, China) for sequencing. The sequences were deposited in the National Center for Biotechnology Information (NCBI) under accession numbers MN368165, MN368167 and MN368167.

Transmission of Arsenophonus in P. pseudoannulata
To study the transmission of Arsenophonus in P. pseudoannulata, spiders with egg bags were used to establish isofemale lines. All the adults and offspring were reared separately, and both adults and offsprings were used to detect symbiont infection.
After the spider eggs hatched, the adult females were used to detect the presence of all known six symbionts found in rice planthoppers and spiders, including Arsenophonus, Acinetobacter, Wolbachia, Serratia, Cardinium and Spiroplasma; the methods used were as previously described for Arsenophonus

Statistics
To explore the genetic relationships among the Arsenophonus sequences detected in different spider populations, phylogenies were constructed using maximum-likelihood and neighbor-joining methods. The sequences were imported and aligned using MEGA7, and genetic distance was estimated with the Kimura 2-parameter model.  (Fig. 1).
Coinfection of other symbionts with Arsenophonus in female P. pseudoannulata Five out of six known symbiont taxa in spiders and rice planthoppers, Wolbachia, Arsenophonus, Acinetobacter, Serratia, and Cardinium, were found in females of P. pseudoannulata carrying egg bags collected in Jurong and Nanjing, Jiangsu Province, China; Spiroplasma was not detected. In the 21 female spiders with egg bags, 13 patterns of symbiont infection were detected. Six out of 21 isofemales were found to be infected with Arsenophonus, and 4 infection patterns were observed, including single infection with Arsenophonus and coinfection with one, three or four symbionts (Cardinium, Serratia, Acinetobacter and Wolbachia). The other 15 females were not infected with Arsenophonus but were infected with one, two or three other symbionts. Among the 21 females, 17 (81.0%), 6 (28.6%), and 6 (28.6%) females were infected with Serratia, Wolbachia, and Cardinium, respectively, and besides Arsenophonus, these symbionts were the dominant taxa in the spiders.

Transmission of Arsenophonus
The transmission of Arsenophonus between generations of P. pseudoannulata was examined in the 21 established isofemales. Six of the 21 isofemales were established from spiders with egg bags infected with Arsenophonus, and the remaining 15 were established from spider with egg bags lacking the symbiont. In the 6 isofemales from Arsenophonus-infected mothers, the symbiont was found in the next generation. The highest incidence of infection with Arsenophonus in the next generation was 46.7%, and the lowest was 17.5%. There were no signi cant differences in the incidence in offspring among these 6 isofemales (χ 2 =7.477, df=5, P=0.188). The average transmission rate was 25.03±4.78% in the Arsenophonus-infected mother spiders ( Fig. 2A).
In the 15 isofemales established from noninfected P. pseudoannulata females, the symbiont was found in the next generation of 12 isofemales. The highest incidence of infection with Arsenophonus in the next generation was 25%, and the lowest was 5.8%. No signi cant difference in infection incidence in offspring was found among the 12 isofemales (χ 2 =11.577, df=11, P=0.396). The infection rate of offspring from noninfected mothers was 11.11±2.13% (Fig. 2B), which was signi cantly lower than that of offspring of Arsenophonus-infected mothers (t=3.108, df=19, P=0.006).
Genetic differentiation of Arsenophonus among different geographical populations of P. pseudoannulata Based on the 23S rRNA gene sequences of Arsenophonus, a close genetic relationship was found among the NC, LS and NJ spider populations from Jiangxi and Jiangsu Provinces (with a maximum distance of nearly 700 km among the 3 sites). The analysis revealed low genetic differentiation of Arsenophonus among these three geographic populations (Fig. 3). . In this study, we found that Arsenophonus was present in the spider P. pseudoannulata, which is a main predator of rice planthoppers. We found that 62.5% (5 out of 8) of the tested spider populations were infected with Arsenophonus at low infection frequencies (with an average infection rate of 3.8% across the 8 sites, with a total of 292 spiders), which is similar to the infection level (< 30%) reported in rice planthoppers (Jiang et al., 2017). We also found that Arsenophonus could coinfect with other symbionts, including Cardinium, Serratia, Acinetobacter and Wolbachia, in P. pseudoannulata. Multi-infection has been found in the dwarf spider Oedothorax gibbosus, with coinfection with Wolbachia, Rickettsia and Cardinium detected (Vanthournout & Hendrickx, 2015). A high diversity of Arsenophonus bacteria has been found in different insect species; in some cases, a single insect species carries multiple Arsenophonus strains. Fox example, Arsenophonus was found to be genetically differentiated in Bemisia tabaci (Thao et al., 2004;Mouton et al., 2012). However, in the spider P. pseudoannulata, no genetic differentiation of Arsenophonus was found among the different geographic populations of the spider. We only investigated genetic differences among three geographical population of P. pseudoannulata. Therefore, to con rm whether there is genetic differentiation of Arsenophonus in this or other spiders, investigations of more populations or species of spider need to be conducted. Bacterial symbionts can be transmitted vertically and horizontally, and these transmission modes allow symbionts to persist for long periods and rapidly spread in host populations as well as colonize novel hosts (Bright & Bulgheresi, 2010). The widely distributed symbionts Wolbachia and Cardinium can be transferred both vertically and horizontally; they have extensive histories of movement both within and among species (Baldo et al., 2006;Raychoudhury et al., 2009;Russell et al., 2009;Oliver et al., 2010). Arsenophonus, as a clade of intracellular symbionts with a broad distribution, has also been found to be transmitted from parents to offspring or from one species to another. In the parasitoid Nasonia vitripennis, Arsenophonus can be transmitted to the next generation from parents (Huger et al., 1985;Gherna et al., 1991). In addition, in N. vitripennis, this symbiont can be transmitted to another species of parasitoid, N. giraulti (Duron et al., 2010). In the present study, we found that in the predator P. pseudoannulata, Arsenophonus could be maternally transmitted to the next generation, but the transmission rate was low (approximately 25%). Moreover, we found the symbiont in the offspring of spiders that were not infected with Arsenophonus, at an infection rate of approximately 11%, indicating that Arsenophonus in spiders has one or more transmission modes in addition to maternal transmission; possibilities include paternal transmission and spider parasitoid transmission. Although Arsenophonus has been detected in strawberry and beet crops (Zreik et al., 1998;Bressan et al., 2008), it has not been found in rice plants (Jiang et al., 2017). Therefore, we infer that Arsenophonus can be transmitted by parents of the spider, i.e., both maternally and paternally. Due to the di culty of identifying the fathers of spiders in the eld, we could not con rm a role of male spiders in Arsenophonus transmission in the present study. The possibility of male transmission needs to be veri ed in future work.
The symbiotic relationships between Arsenophonus and their hosts have attracted increasing interest. In N. vitripennis, a natural enemy of insect pests, Arsenophonus can induce death of male embryos (Gherna et al., 1991), and in rice planthoppers, this symbiont can improve host resistance to the entomopathogenic fungi Metarhizium anisopliae (Zhu et al., 2017). Arsenophonus has also been found or speculated to be involved in insect resistance (Pang et al., 2018). For example, it was found to decrease the resistance of rice brown planthoppers to the chemical insecticide imidachloprid (Liu & Guo, 2019). A recent study found that Arsenophonus affected the amino acid requirements of aphids (Tian et al., 2019). However, to the best of our knowledge, the effects of Arsenophonus on spiders remain unknown. Thus, studies of Arsenophonus infection and transmission characteristics in spiders can help reveal the biological roles and evolution of this symbiont in the predators of insect pests. Studies of the interactions between Arsenophonus and their hosts are urgently needed.

Conclusions
Spider P. pseudoannulata populations are infected with the symbiont Arsenophonus at low rates. This symbiont can be vertically transmitted from female spiders to offspring but also may be transmitted by male spiders or other vectors. Arsenophonus populations are genetically differentiated among different geographical regions of spiders.

Declarations
Ethics approval and consent to participate The spiders used in our studies were collected from rice paddies in different locations in China, and no speci c permissions were required for these locations. We con rm that these locations are not privatelyowned or protected in any way, and our eld studies do not involve endangered or protected species.

Consent to publish
Not applicable.
Availability of data and materials The datasets used and analysed during the current study are available from the corresponding author on reasonable request.