Diversity Loss and Restructuring of the Microbiota in a Globally Invasive Lady Beetle

The invasion success of Harmonia axyridis in North America and Europe has aroused a growing interest among students of invasion biology. This study was focused on the microbial community of H. axyridis with the aim of gaining insights into their potential roles in the invasion. We hypothesized that microbial communities of H. axyridis in the non-native range may decrease in diversity but restructured in composition as compared to those in the native regions. We collected lady beetle samples from eleven geographical areas across the native range mainland China and from seven areas in the non-native range USA. We applied next-generation amplicon sequencing to the samples to estimate microbial communities and examined differences in their composition and diversity between the host’s native and non-native ranges. We found that H. axyridis in the non-native range had lower microbial community richness and evenness than it had in the native range, but its microbiota composition of the amplicon sequence variants and genera differed markedly between the non-native and native ranges. The ndings support our hypotheses concerning microbial community in H. axyridis across the native and non-native range. This study provides new information for our understanding of potential roles of the microbiota in the invasion success of H. axyridis.


Introduction
Alien invasive species pose increasing ecological challenges with escalating globalization (Mack et al., 2000;Wardle et al., 2011;Hill et al., 2016). They threaten biodiversity, jeopardize endangered species, and deform extent biogeochemical cycles in non-native regions (Kenis et al., 2009). Understanding species invasiveness is one of major challenges long facing researchers in invasion biology. Given the ubiquitous associations between insects and microbial symbionts, the co-introduction of microbial associates by invasive insects may be quite common . The microbiome can in uence exotic animal species in broad aspects, ranging from life history traits, defenses against natural enemies, and tolerance to abiotic stresses (Oliver et al. 2003;Duron et al. 2008;Dunbar et al. 2007;Douglas, 2018). The enhancement of any aspect may contribute to the successful invasion of alien species (Floate et (Himler et al., 2011); symbiotic bacterial community inside galleries of the red turpentine beetle Dendroctonus valens increases the host's overall tness and facilitates its invasion (Cheng et al., 2018). To explain how mutualism facilitates the invasion of introduced species, several alternative hypotheses have been framed (Klock et al. 2015). The enhanced mutualism hypothesis suggests that introduced species become invasive with the aid of forming novel effective mutualism in non-native ranges (Richardson et al., 2000); the accompanying mutualist hypothesis posits that introduced species are facilitated during the invasion by mutualistic symbionts concurrent with introductions from their native ranges (Rodriguez-Echeverria, 2010); the Generalist Host Hypothesis predicts that introduced species are generalists in terms of association with mutualists and thus less constrained by absence of speci c partners (Parker, 2001). These hypotheses gain more support from invasive plants than from invasive invertebrate animals.
Harmonia axyridis (Coleoptera: Coccinellidae) is widely appreciated as an effective predator suppressing pest aphid populations in its native range Asia (Seo andYoun 2002, Ali et al., 2016). Due to its potential in biological pest control, it was introduced into North America and then from there to Europe; it has spread far and wide in exotic continents, causing mounting concerns for its potential roles in the decline of endemic lady beetle populations (Koch 2003, Majerus et al., 2006, Brown et al., 2008van Lenteren 2012, Sloggett et al., 2012;Roy et al., 2016). Understanding the invasiveness of H. axyridis has become a daunting challenge for the students in invasion biology (Roy et al., 2016;Li et al., 2021b). Lady beetles harbor rich and diverse microbial symbionts (Weinert et al., 2007). Laboratory studies have found that some symbionts are in uential on H. axyridis tness such as chemical defenses (Schmidtberg et al. 2018) and body size (Elnagdy et al. 2013). To examine the role of microorganisms in the invasion of H. axyridis, a laboratory study suggests that pathogenic microspores carried by invasive H. axyridis can be exploited as leverage in competition with endemic lady beetles (Vilcinskas et al., 2013). Preliminary investigations of the prevalence of common maternally-inherited bacteria inhabiting H. axyridis in the native and non-native range found sporadic infections across the ranges (Nakamura et al. . Taking these processes into consideration, we hypothesized that the microbiota of H. axyridis in the non-native range may have a lower diversity than those in the native range. It is generally held that the composition of microbiota is not a stand-still but can vary with novel environments their hosts invaded (Richardson et al., 2000;Kloch et al., 2015). For invasive species, their expanding ranges create new opportunities for horizontal acquisition of new symbionts from native species (Amsellem et asl., 2017). A study found that local habitat affects the gut microbiota of H. axyridis (Tiede et al., 2017). So, we hypothesized that the microbiota of H. axyridis in non-native regions may be restructured in composition and thus not similar with that in native regions. To test these two hypotheses, we compared microbial community in H. axyridis collected from 11 geographical areas covering wide climatic zones across mainland China and 7 across the United States of America. We compare diversity and composition of the microbiota inhabiting H. axyridis between its native and non-native ranges.

Sample collection
Adults of H. axyridis were collected at 11 geographical localities across mainland China and at 7 localities from the United States of America (Fig. 1). The sample insects from the native regions were brought to laboratory and starved for ve or six days to empty their intestines and those from the nonnative regions were starved for the similar, prior to DNA extraction. In preparation for DNA extraction, these insects had their elytra removed and body surface cleansed rst with 75% ethanol, then with 10% bleach, and nally with sterile distilled water three times, with each cleaning step for one minute (Gauthier et  The sequenced paired-end Illumina reads were rst analyzed using AfterQC (version 0.9.6) for quality control (Chen et al., 2017), and then organized in a table of amplicon sequence variants (ASV) using DADA2 (Callahan et al., 2016) implemented in QIIME2 (version 2018.8.0) (Bolyen et al., 2019). The taxonomy of all 16S rRNA gene sequences was assigned by feature-classi er plugin with the classifysklearn method (Bokulich et al., 2018) based on the Silva database (Release132, http://www.arb-silva.de) (Quast et al., 2013). To avoid the in uence of differences in sequencing depth between samples, sequences from different samples were rare ed to the same depth for further diversity estimation.
The α-diversity was estimated based on ASVs by averaging over replicated insects per location and with the location as an independent observation unit for the native and non-native range. The Chao1 estimator and the number of observed ASVs (S) were used to estimate the richness and the Shannon and Simpson indices to estimate the evenness. To determine differences in these metrics between native and nonnative ranges, the non-parametric Wilcoxon rank sum test with continuity correction was applied. The βdiversity between native and non-native ranges was estimated with the permutational multivariate
The analysis of microbiota α-diversity showed signi cant differences in the richness (Chao1: p < 0.001; S: p = 0.002) and the evenness (Simpson index: p = 0.003; but Shannon index: p = 0.08) between the native and non-native ranges (Fog 3A-D). The analysis of microbiota β-diversity showed a signi cant difference in the richness between the native and non-native ranges (permutational MANOVA: F = 11.71, R 2 = 0.117, p < 0.001), as visualized in Fig. 3E.

Discussion
In this study, we found that the microbiota in H. axyridis in the non-native range had lower richness and evenness compared to that in the native range. The ndings support our hypothesis that the microbiota of H. axyridis in non-native regions decreases in diversity than in native regions. There are two explanations for the diversity-diminished microbiota of H. axyridis in the non-native ranges. ). An examination of gut microbiome in six species (including H. axyridis) of aphidophagous ladybird beetles collected from the eld found a lower bacterial richness in exotic species (three) than in native ones (three) (Tiede et al, 2017). Here, we assume that the reduction in microbiota diversity may contribute to the invasion success of H. axyridis. This assumption remains to be tested with empirical evidence on ecological and evolutionary consequences of the microbiota in native and non-native populations of H. axyridis, for example, with comparing life history traits in both native and non-native ranges within a life-table framework (Roy et al., 2011).
We found that the microbiota composition of H. axyridis in the non-native range overlapped slightly with that in the native range. The nding supports our second hypothesis that the microbiota composition in non-native regions may be restructured and thus is not similar with that in native regions. The nding suggests that H. axyridis in the non-native range has restructured its microbiota from novel environments they invaded. Hosts can switch to associate themselves with different symbionts in the invaded range (Amsellem et al., 2017). Some of these acquired symbionts can confer their hosts the potency of invasiveness (Hulcr and Dunn 2011; Garnas, 2016; Wing eld et al. 2016). Among multiple potential pathways to acquire symbionts, host diet is a major one. Laboratory studies have found that host diet is in uential on the gut microbial community in many herbivorous arthropods (Broderick et al., 2004;Dillon and Dillon, 2004;Lundgren and Lehman, 2010;Wang et al., 2011;Taerum et al., 2013;Mason and Raffa, 2014), either through effects of food substrates on the persistence of speci c microbes, or directly from the acquisition of associated microbes (Chandler et al., 2011;Bili et al., 2016). Diet-related bacteria can play a potential role in enhancing their host in digestion processes or adaptation to novel food sources (Bouchon et al, 2016). In mobile predator insects, including H. axyridis, prey diversity and habitat type are in uential on gut microbiota (Tiede et al., 2017). A laboratory experiment showed that aphid-symbionts can be detected up to 96 hr after aphid consumption (Paula et al., 2015). There are multiple potential pathways for H. axyridis to acquire novel microbials, including cannibalism, intra-guild predation, predation of herbivorous prey and even occasionally plants. Intra-guild predation of resident lady beetles by invasive H. axyridis occurs frequently and is suggested as one of main factors for its invasion success (Alypkin et al., Pell et al., 2008;Koch and Costamagna, 2017). In addition, novel diets in invaded regions can contribute unique microbials to the microbiota in H. axyridis. So, it is likely that the restructured symbionts in H. axyridis in non-native regions may play potential roles in its invasion success.
In conclusion, this is the rst study that examines symbiotic community inhabiting H. axyridis across multiple geographical locations in both native and non-native ranges. The data support our hypotheses that H. axyridis in the non-native range harbors the symbiotic microbiota with a decreased diversity but restructured composition as opposed to that in the native range. We assume that these symbiotic characteristics may facilitate the invasiveness of H. axyridis in the non-native range. Future research should expand sample collections while controlling covariates, such as habitats, over larger areas in both native and non-native ranges to construct a global pro le of microbiota associated with H. axyridis. Declarations FUNDING This work was supported by the National Key R&D Program of China (2017YFE0104900). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

AUTHOR CONTRIBUTIONS
HL and XS performed the experiments. HL analyzed the data, prepared gures and tables and draft the paper. JZ participated in the data analysis. XZ and JO revised the manuscript. LM and BL conceived and designed the experiments, and revised the manuscript.

ACKNOWLEDGMENT
We thank PersonalBio Company (Shanghai) for helping with 16S rDNA sequencing and Jenifer White for revising earlier draft of the manuscript.

DATE AVEILABILITY
The raw data have been deposited in the Sequence Read Archive (SRA) database with an access number SRP194258.