Many insect species are closely associated with multiple endosymbionts that can affect the feeding, reproduction, and distribution of their hosts. Some of these associations have borne dependencies between the host and symbiont. For example, some herbivorous insects rely exclusively on nitrogen-poor substrates and require their symbionts for nutritional compensation [1], such as aphids with Buchnera [2], whiteflies with Portiera [3], and psyllids with Carsonella [4]. Furthermore, some blood-feeding insects have similar associations, such as tsetse flies with Wigglesworthia [5] and bedbugs with Wolbachia [6]. Such nutritional endosymbionts contribute significantly to the diversification of insect diets. Some endosymbionts have non-trophic effects on their hosts. Wolbachia, Rickettsia, Spiroplasma, and Cardinium induce reproductive phenotypes, such as cytoplasmic incompatibility (CI), male killing (MK), parthenogenesis induction (PI), and feminization (Fem), which are considered to be “selfish strategies” for the endosymbionts [7,8,9,10,11,12]. Some other endosymbionts are known to contribute to improving host fitness by increasing reproduction and development [13], conferring tolerance to thermal stress [14], and conferring resistance to pathogens [15].

Symbiotic bacteria often co-infect an individual in the same host population and show considerable variation in their infection patterns [16, 17]. The main factors that shape such patterns and symbiont community structure include host species [18], host plant [19], and geography [17]. For example, Macrolophus (Hemiptera: Miridae) are known to harbor one strain of Wolbachia and two species of Rickettsia (relatives of Rickettsia bellii and Rickettsia limoniae). Macrolophus pygmaeus harbors all these symbionts, whereas Macrolophus melanotoma (syn. Macrolophus caliginosus) harbors only Wolbachia and R. limoniae [18, 20, 21]. Although M. pygmaeus populations are geographically separated, their microbiomes are homogeneous, whereas the microbiomes of M. melanotoma are diverse [18, 21]. To fully understand the ecology and evolution of such species, it is important to understand how this variation in the symbiotic microbiota is involved in host adaptation. This is especially important given the potential role that many predatory insects play as biological control agents.

The small green mirid, Nesidiocoris tenuis (Hemiptera: Miridae), is a cosmopolitan species commonly used in the control of agricultural pests [22, 23]. They are zoophytophagous, which allows them to survive by feeding not only on arthropods but also on plants, which can augment their biological control activities but can also cause damage to crops [23,24,25]. N. tenuis are often found on Sesamum indicum (sesame) and Cleome hassleriana (cleome) in warm regions of Japan [23, 25]. In N. tenuis, two genera of symbionts, Wolbachia and Rickettsia, have been detected in Israeli populations and commercially available strains [26, 27]. Of these, the infection frequency of Rickettsia was found to be high (93–100%) in Israeli populations [26], whereas the infection frequency of Wolbachia remains unknown. Caspi-Fluger et al. [26] suggested that Rickettsia plays a nutritional role in zoophytophagous N. tenuis due to its high prevalence and abundance in adults and localization in the gut.

The aim of the present study was to elucidate the population structure of N. tenuis in Japan in terms of microbiome composition. We revealed the diversity of the microbiome in N. tenuis by 16S rRNA amplicon sequencing and diagnostic PCR assay. We also investigated whether the infection frequencies of Wolbachia, Rickettsia, and Spiroplasma were correlated with geography, climate, host plant, and host sex. These results reveal the complex relationships between N. tenuis and its symbionts, which may potentially contribute to improve the use of this species as a biological control agent.

Materials and Methods

Insect Collection

In total, 360 wild-caught adults of N. tenuis were collected from Sesamum indicum (sesame) or Cleome hassleriana (cleome) from 15 farms in Japan between 2017 and 2021 (Table S1). All individuals were stored in 99.5% ethanol at − 80 °C until DNA extraction was performed.

DNA Extraction

The 360 DNA samples were extracted from the whole insect bodies using the Wizard® Genomic DNA Purification Kit (Promega Corporation, Madison, WI, USA) according to the manufacturer’s protocol. DNA was dissolved in 100 μL of Tris-EDTA (pH 8.0) and stored at − 30 °C until use.

Amplicon Sequencing

For the selected 96 samples (Table S1), hypervariable V3/V4 regions of the 16S rRNA gene were amplified using the KAPA HiFi HotStart ReadyMix (Kapa Biosystems Inc., Wilmington, MA, USA) with V3V4_F primer (5′-TCGTCGGCAGCGTCAGATGTGTATAAGAGACAGCCTACGGGNGGCWGCAG-3′) and V3V4_R primer (5′-GTCTCGTGGGCTCGGAGATGTGTATAAGAGACAGGACTACHVGGGTATCTAATCC-3′). The reactions were initiated by denaturation at 95 °C for 3 min, followed by 25 cycles of 30 s at 95 °C, 30 s at 55 °C, 30 s at 72 °C, and a final extension step of 5 min at 72 °C. After purification of the PCR products using AMPure XP beads (Beckman Coulter Inc., Brea, CA, USA), eight cycles of a second PCR were performed to add barcode sequences to each product using the TG Nextera XT Index Kit v2 Set A (Illumina Inc., San Diego, CA, USA). All barcoded amplicons were pooled in equal concentrations and sequenced on the Illumina MiSeq platform using the MiSeq Reagent Nano Kit v3 (600 cycles) according to the manufacturer’s recommended protocol ( to produce 300-bp paired-end reads.

Raw Sequencing Data Analysis

The Illumina sequence data were processed using QIIME2 ver. qiime2-2020.11 [28]. The Illumina reads were demultiplexed based on the barcode sequences using “qiime demux emp-paired.” Denoizing and clustering were performed to obtain representative sequences and the feature table using “qiime dada2 denoise-paired” command. Taxonomic assignment to the representative sequences was then performed using “qiime feature-classifier classify-blast.” Sequences not identified as bacteria and all features with an abundance of < 0.01% were filtered out for further analysis. Data visualization was performed using the “qiime metadata tabulate” command, and the “qiime taxa barplot” command was used to generate a taxonomic bar plot. Alpha- and beta-diversity analyses were performed using the “qiime diversity alpha-rarefaction” and “qiime diversity core-metrics-phylogenetic” commands.

Diagnostic PCR for Rickettsia, Wolbachia, and Spiroplasma

Diagnostic PCR for the insect reproductive manipulators Rickettsia, Wolbachia, and Spiroplasma was performed on 360 N. tenuis individuals from 15 farms in Japan (Table S1). PCR was performed using the Go Taq Green Master Mix (Promega) with 528-F (5′-ACTAATCTAGAGTGTAGTAGGGGATGATGG-3′) and 1044-R (5′-GTTTTCTTATAGTTCCTGGCATTACCC-3′) for Rickettsia [29], wsp81F (5′- TGGTCCAATAAGTGATGAAGAAAC-3′) and wsp691R (5′- AAAAATTAAACGCTACTCCA-3′) for Wolbachia [30], and Spiro_Nt_124F (5′-GACGGTACCTTACCAGAAAG-3′) and Spiro_Nt_409R (5′-TTCGTGCCTAAACGTCAGTG-3′) for Spiroplasma in N. tenuis. Reactions were initiated by denaturation at 95 °C for 3 min, followed by 35 cycles of 30 s at 95 °C, 30 s at 60 °C (for 528-F/1044-R) or 55 °C (for wsp81F/wsp691R) or 56 °C (for Spiro_Nt_124F/Spiro_Nt_409R), 60 s at 72 °C, and a final extension step of 10 min at 72 °C. DNA was detected by electrophoresis on a 2% agarose gel prestained with Midori Green Xtra (Nippon Genetics Co., Ltd., Tokyo, Japan) in Tris-acetate-EDTA buffer.

Molecular Phylogenetic Analysis

Partial 16S rRNA sequences of Wolbachia, Rickettsia, and Spiroplasma isolated through amplicon sequencing were used for phylogenetic analyses. The datasets were registered in DDBJ (accession numbers: LC769520–LC769523). Phylogenetic trees based on the nucleotide sequences were constructed using the maximum likelihood method in MEGA 7.0 [31]. Kimura’s two-parameter model, evaluated with the best-fit method, was applied for the calculation [32].

Statistical Analysis

The PCR-based presence or absence of each bacterium within a mirid individual was analyzed based on a generalized linear model (GLM) with a binomial distribution (with a logit link function). Latitude, longitude, annual mean temperature, host plant, and sex were analyzed as explanatory variables. Data with unidentified sex were excluded from the analysis. Based on the GLM, an analysis of variance (ANOVA) was performed to evaluate the effects of each explanatory variable. Geographical and climatic factors that showed significant correlations with the GLM analysis were plotted to explicitly evaluate differential infection frequency. In addition, graphical visualization and Fisher’s exact test for the presence/absence of each bacterium were performed for each host plant. Geographical data were obtained from the Geospatial Information Authority of Japan (, and climatic data were obtained from the Automated Meteorological Data Acquisition System administered by the Japan Meteorological Agency ( Co-infection of Rickettsia, Wolbachia, and Spiroplasma was analyzed using an association screening approach as previously described [33]. The envelope function from the boot package in R software was used to estimate the 95% confidence envelope for the distribution profile of the combination counts, simultaneously including all infection patterns. A global test based on the 95% confidence envelope was then performed. All of the above analyses were performed using R version 4.2.2 [34].

Vertical Transmission Analysis

An isofemale line, K11, co-infected with Wolbachia and Rickettsia, was established from a female collected from population no. 1 in 2022 (Table S1). In the laboratory, K11 was reared using Ephestia kuehniella eggs (purchased in a frozen state from Agrisect Inc., Ibaraki, Japan) as the food source and Crassula ovata leaves as the oviposition substrate. The founder female and one of her G2 generation offspring were subjected to DNA extraction and amplicon sequencing as described above. All breeding was performed at 25 ± 1 °C with a light:dark regime of 14:10 h.


Microbiomes of N. tenuis Inferred from 16S rRNA Gene Amplicon Sequencing

For the 96 N. tenuis individuals, the microbiomes were analyzed by amplicon sequencing of the hypervariable V3/V4 region of 16S rRNA, and a total of 4,625,099 reads were clustered into 77 operational taxonomic units (OTUs). The nine major OTUs (> 25,000 total reads and > 2000 reads per observed sample) were Rickettsia sp., two strains of Wolbachia sp., Providencia sp., Serratia marcescens, Pseudochrobactrum sp., Lactococcus lactis, Stenotrophomonas sp., and Spiroplasma sp., in order of frequency (Fig. 1, Table S2). Assuming that a mirid individual has the bacterium when it represented more than 1% of the tags analyzed, 69 out of 96 individuals had Rickettsia (71.9%), 30 individuals had Wolbachia sp. A (wNtenA, 31.3%), 4 individuals had Wolbachia sp. B (wNtenB, 4.2%), 25 individuals had Providencia (26.0%), 26 individuals had Serratia marcescens (27.1%), 18 individuals had Pseudochrobactrum (18.8%), 10 individuals had Lactococcus lactis (10.4%), 9 individuals had Stenotrophomonas (9.4%), and 5 individuals had Spiroplasma (5.2%).

Fig. 1
figure 1

Proportion of bacterial sequences in 96 N. tenuis individuals collected from 14 regions in Japan. Sequences were obtained by amplicon sequencing of the hypervariable V3/V4 region of 16S rRNA. Assigned bacterial taxa are color coded as shown in the box on the right. Sequences with less than 25,000 total reads or 2000 reads per observed sample are categorized as “others.” The numbers at the bottom represent the geographic populations shown in Table S1

Infection Status of Rickettsia, Wolbachia, and Spiroplasma Inferred from Diagnostic PCR

We further investigated the prevalence of Rickettsia, Wolbachia, and Spiroplasma in 360 individuals from 15 populations of N. tenuis. We found that 293 of the 360 individuals (81.4%) were infected with at least one bacterium. Rickettsia was the most prevalent being detected in 251/360 individuals (69.7%), followed by Wolbachia in 142/360 individuals (39.4%) and Spiroplasma in 22/360 individuals (6.1%), and the frequency of infection varied between populations (Fig. 2a, 2b; Table S3). Some N. tenuis were co-infected with multiple bacteria; 104 individuals were doubly infected with Rickettsia and Wolbachia, 9 individuals were doubly infected with Rickettsia and Spiroplasma, 1 individual was doubly infected with Wolbachia and Spiroplasma, and 4 individuals were triply infected (Fig. 2b).

Fig. 2
figure 2

Infection frequencies of Rickettsia, Wolbachia, and Spiroplasma in each population of N. tenuis based on the diagnostic PCR assay. a Infection frequencies of Rickettsia (left panel), Wolbachia (center panel), and Spiroplasma (right panel). The frequencies of positive (black) and negative (white) individuals are shown with bar graphs. b Venn diagrams illustrating the co-infection status of Rickettsia, Wolbachia, and Spiroplasma. Each inner circle indicates the number of N. tenuis individuals infected with Rickettsia (red), Wolbachia (blue), and Spiroplasma (green). Overlapping circles indicate multiple infections. The outer circles represent the total number of N. tenuis individuals that were examined. Population numbers correspond to those in Table S1

Correlation of Rickettsia, Wolbachia, and Spiroplasma with Latitude, Temperature, and Host Plants

GLMs showed that the infection frequency of Rickettsia was significantly correlated with latitude and annual mean temperature (Table 1; Fig. 2a). Regression analyses showed a higher frequency of Rickettsia at lower latitude and higher temperature (Fig. 3a). Furthermore, GLMs indicated that the host plant significantly affected the infection frequency of Wolbachia and Spiroplasma (Table 1). Wolbachia infection frequency was significantly higher on C. hassleriana (cleome) than on S. indicum (sesame), while Spiroplasma was not found on C. hassleriana (Fig. 3b). The association screening approach showed that no significant association was detected between the co-infection status of Rickettsia, Wolbachia, or Spiroplasma from 360 individuals of N. tenuis, and this was also the case when the analysis was run by area (Table 2).

Table 1 Correlation between geographic, climatic, and host factors and endosymbiont infections in natural populations of N. tenuis in Japan. The generalized linear model (GLM) incorporated the effects of geographic (latitude and longitude), climatic (average of annual temperature), host plant species, and sex variables of insects on the presence/absence of each endosymbiont with binomial error and logit-link function. Based on the GLM, an ANOVA was performed for each endosymbiont to estimate the P-value for each explanatory variable using chi-squared tests
Fig. 3
figure 3

Relationship between infection frequencies of each symbiont (Rickettsia, Wolbachia, or Spiroplasma) in N. tenuis and each variable (latitude, temperature, or host plant). a A generalized linear model (GLM) with binomial error and logit-link function was plotted to estimate the effects of the correlation between Rickettsia and latitude or annual mean temperature for those significant differences detected (Table 1). The difference in deviance between the null hypothesis and the estimated model explained by each GLM is shown as ΔD, and the 95% confidence intervals are shaded in gray. b Infection frequencies of each endosymbiont in the host plants Sesamum indicum (sesame) and Cleome hassleriana (cleome). Error bars indicate 95% bootstrap percentiles (10,000 replicates). Asterisks indicate significant differences based on Fisher’s exact test (*P < 0.05; ***P < 0.0005)

Table 2 Co-infection status for Rickettsia, Wolbachia, and Spiroplasma of N. tenuis by area as seen through association screening analysis. The number of data points is shown as N, and the P-value from the global test based on the 95% confidence envelope is shown as P

Molecular Phylogenetic Analysis of Rickettsia, Wolbachia, and Spiroplasma

To infer the phylogenetic position of Rickettsia, Wolbachia, and Spiroplasma, nucleotide sequences (360, 360, and 384 bp, respectively) obtained through amplicon sequencing were subjected to maximum-likelihood tree reconstruction. In the Rickettsia phylogeny, the Rickettsia in the N. tenuis from Japanese populations was identical to that from the Israeli population [26], which was closely related to Rickettsia bellii (Fig. 4a). Of the two Wolbachia isolates in N. tenuis, one is the major isolate (wNtenA), which was detected in 30 out of 96 individuals, and the other is the minor isolate (wNtenB), which was detected in 4 out of 96 individuals. In the Wolbachia phylogeny, wNtenA and wNtenB both belonged to the Wolbachia supergroup B (Fig. 4b). wNtenA was closely related to the Wolbachia from the whitefly Bemisia tabaci, and wNtenB was closely related to those from Macrolophus pygmaeus, Cadra cautella, and Culex pipiens. In the Spiroplasma phylogeny, the Spiroplasma in N. tenuis fell into the Citri-Poulsonii clade, a large group consisting of S. citri, S. melliferum, S. kunkelli, S. penaei, S. insolitum, S. leucomae, S. phoeniceum, and S. poulsonii (Fig. 4c).

Fig. 4
figure 4

Phylogenetic trees based on the 16S rRNA gene sequences of Rickettsia, Wolbachia, and Spiroplasma. These trees were generated using the maximum likelihood method based on the Kimura 2-parameter model [32] with 1000 bootstrap replicates. Bootstrap values < 50% are not shown. The symbionts from N. tenuis are shown in red. The host organisms are given in parentheses, whereas the accession number is provided after each OTU. The scale bar indicates 0.02 substitutions per site. a Phylogenetic tree of Rickettsia based on 360 nucleotide sites. MK and PI represent Rickettsia isolates that cause male killing and parthenogenesis induction, respectively. The OTUs from insect symbionts are shaded gray. The outgroups are Wolbachia pipientis and Orientia tsutsugamushi. b Phylogenetic tree of Wolbachia based on 360 nucleotide sites. Wolbachia supergroups are depicted on the right side. The outgroup is Ehrlichia ruminantium. c Phylogenetic tree of Spiroplasma based on 384 nucleotide sites. MK and CI represent Spiroplasma isolates that cause male killing and cytoplasmic incompatibility, respectively. The Citri-Poulsonii clade is shaded green. The outgroup is Erysipelothrix larvae

Vertical Transmission of Rickettsia and Wolbachia

A total of 35,102 reads were obtained from the amplicon sequence analysis of the founder female of strain K11. Two major OTUs were classified as Rickettsia (17,396 reads) and Wolbachia (17,642 reads), respectively (Fig. S1). In G2, a total of 35,397 reads were clustered into Rickettsia (11,158 reads) and Wolbachia (24,097 reads) (Fig. S1). These nucleotide sequences of Rickettsia and Wolbachia were identical to those of Rickettsia and wNtenA obtained from N. tenuis in Fig. 4, respectively.


This study demonstrated the high prevalence of Rickettsia and Wolbachia in Japanese N. tenuis populations (Fig. 1; Table S2), which is consistent with the results of a previous study on Israeli N. tenuis [26]. These symbionts induce reproductive phenotypes in insect hosts, and some of them can improve host fitness [9, 12, 13, 35]. Similarly, Spiroplasma manipulates host reproduction in some insects and can confer resistance to various parasites [10, 36, 37]. To the best of our knowledge, our study is the first to detect Spiroplasma in N. tenuis. In addition, Providencia, Serratia marcescens, Pseudochrobactrum, Stenotrophomonas, and Lactococcus lactis were found to be relatively abundant bacterial taxa in the N. tenuis population in Japan (Fig. 1; Table S2). S. marcescens and Providencia are commonly present in the environment [38, 39], and S. marcescens was also isolated from N. tenuis in a previous study [27]. Our study showed widespread infection of S. marcescens and Providencia among individuals but with low sequence reads per individual (Table S2), which may suggest opportunistic pathogenic properties of these bacteria [39, 40]. In the mosquito species Aedes aegypti, S. marcescens is present as a gut commensal bacterium that influences viral vector competence [41]. Stenotrophomonas, Pseudochrobactrum, and Lactococcus lactis have also been reported to be latent in the environment [42,43,44] and insect gut [45]. These results reveal a diversity of endosymbiotic microbes in natural populations of N. tenuis. The fact that none of the bacterial species found in this study were fixed in zoophytophagous N. tenuis suggests the absence of obligate symbionts in N. tenuis, a trait more typical for predatory arthropods rather than sap-feeding insects.

Rickettsia, Wolbachia, and Spiroplasma manipulate host reproduction in various insects [7,8,9,10, 12]. We found all possible combinations of these genera in N. tenuis individuals. Given that there was no correlation between the frequency of Rickettsia, Wolbachia, or Spiroplasma and the host sex (Table 1), it is unlikely that these symbionts induce MK, PI, or Fem in N. tenuis, which would otherwise result in a female-biased sex ratio and preferential presence of the symbiont in females. Coexisting symbionts may engage in interactions that are either negative or positive [46, 47]. Although no significant association with infection frequency was found (Table 2), further analysis of reproductive phenotypes or life history traits in various symbiont combinations is needed to understand the complex symbiotic system of N. tenuis populations and to propose the optimal biological control agent. Rickettsia, Wolbachia, or Spiroplasma have been detected in other carnivorous arthropods, such as mirids [18, 21, 26], coccinellids [48], and lacewings [7]. Feeding on other arthropods may have increased the chance of acquiring the symbionts common to prey species for N. tenuis [49].

The Rickettsia found in the present study is identical based on the partial sequence of the 16S rRNA gene to the Rickettsia sequence previously reported in N. tenuis [26], which is closely related to the R. bellii group. Previously, Rickettsia was detected in the gut lumen along the digestive tract of N. tenuis, while Wolbachia was detected in the surrounding epithelial cells [26]. A similar distribution of Rickettsia was elucidated in Macrolophus; both R. bellii and R. limoniae were found in the gut of M. pygmaeus and M. melanotoma [20, 21]. Although no correlation was found between the infection frequency of Rickettsia and the host plant, future studies should investigate the possible involvement of Rickettsia in the nutrient metabolism, including the zoophytophagous trait, of this species. It should be noted that no significant effects of Rickettsia and Wolbachia on the fitness traits of nymphal development and fecundity were detected in M. pygmaeus [18]. Although the Israeli populations harbored Rickettsia at a consistently high frequency (93–100%) [26], Japanese populations harbored Rickettsia at variable and relatively low frequencies (20.8–95.8%) (Fig. 2; Table S3). The fact that the high infection frequency of Rickettsia was associated with lower latitude and higher annual mean temperature (Fig. 3; Table 1) suggests the possibility that Rickettsia may provide positive effects to the host, such as heat tolerance, under high temperature [50] or negative effects under low temperature. Rickettsia infection is known to upregulate the expression of stress response genes in B. tabaci, which may underlie the mechanism of heat tolerance [50]. Furthermore, the supercooling point of M. pygmaeus exhibited a decrease upon the removal of its symbionts (two Rickettsia species and Wolbachia); however, it remains uncertain which bacterium influenced to the freezing susceptibility [51]. These possible effects of Rickettsia on hosts may explain the variable frequency of Rickettsia in Japanese populations of N. tenuis. Alternatively, Rickettsia may have no effect on host temperature sensitivity and our observation simply reflects the temperature sensitivity of Rickettsia itself [52].

Of the two supergroup B Wolbachia strains identified in this study, the major strain wNtenA was identical in terms of the partial 16S rRNA gene sequence to the Wolbachia strain found in B. tabaci. Despite the existence of a predator–prey relationship between N. tenuis and B. tabaci [23, 24], it is unlikely that the detected Wolbachia bacteria are exclusively derived from undigested B. tabaci remaining in the gut. This is because a large number of Wolbachia sequence reads were obtained using amplicon sequencing. Furthermore, vertical transmission was confirmed by breeding individuals under controlled laboratory conditions where they were not exposed to B. tabaci (Fig. S1). In B. tabaci, Wolbachia can be transmitted horizontally through plants and subsequently transmitted vertically to offspring [53]. The possibility that N. tenuis acquired Wolbachia from plants might be supported by the observed correlation between the frequency of Wolbachia and host plants.

The other strain, wNtenB, was identical with respect to the partial sequence of the 16S rRNA gene to the Wolbachia strain found in M. pygmaeus, which is known to induce strong CI [54]. Although strong CI is generally considered to cause widespread infection of the symbiont within the host population [55], wNtenB was rare (4 out of 96) in the N. tenuis populations in Japan. Furthermore, we did not observe co-infection of wNtenA and wNtenB, so whether they are in conflict or not remains unclear.

In the present study, we detected Spiroplsma from N. tenuis for the first time. Spiroplasma has also been detected in other hemipteran species, such as planthoppers, leafhoppers, and Orius predatory bugs [56,57,58]. In leafhoppers, Spiroplasma is transmitted horizontally between plants and insects [56]. Interestingly, we found that the infection frequency of Spiroplasma differed depending on the host plant (Fig. 3; Table 1), and the partial sequence of the 16S rRNA gene of Spiroplasma from N. tenuis was related to that of another mirid bug from Taiwan, Trigonotylus ruficornis (Fig. 4c). Future studies should aim to directly test whether Spiroplasma can be horizontally transmitted via plants.

The presence of symbionts may have important implications for the practical use of the predators as biological control agents. In particular, the high infection frequency of Rickettsia and Wolbachia may indicate their ability to manipulate host reproduction or their positive effects on host fitness. Significant differences in infection rates among host plants and geographic regions may affect the effectiveness of its use as a biological control agent, including its choice of insectary plants and its ability to propagate in the regions where it is used, both of which remain unexplored. Our findings highlight the potential importance of these symbionts, which may strongly affect the intrinsic rate of increase and confound the population dynamics of N. tenuis. We encourage future studies to determine the impact of each symbiont on this important biological control agent.